**Meet the editor**

Dr Rajunor Ettarh is Professor and Associate Director of Anatomical Teaching at Tulane University School of Medicine and a Fellow of the Royal Society of Medicine in London. Previously he was a visiting professor at University of Perugia, Italy and investigator at the Conway Institute, University College Dublin. His research interests include enteric epithelial cell biology, non-ste-

roidal anti-inflammatory agents, peroxisome proliferator-activated receptors and medical education. He is a member of the American Association of Anatomists, sits on the Council of the Anatomical Society (UK) and has reviewed for several journals including Carcinogenesis and Anatomical Sciences Education.

Contents

**Preface IX** 

Rajunor Ettarh

Chapter 3 **Early Detection of Colorectal** 

Chapter 4 **Turning Intention Into Behaviour:** 

Chapter 5 **Psychological Impact and Associated** 

Hitoshi Okamura

Martina Perše

Chapter 7 **Dietary Anthocyanins: Impact on** 

**Part 3 Nutrition 101** 

Chapter 6 **Physical Activity,** 

Chapter 1 **Tumor Engineering: Finding the Brakes 3** 

**Part 2 Epidemiology and Psychology 9** 

Chapter 2 **Colorectal Carcinoma in the Young 11** 

Christos Lionis and Elena Petelos

Shahana Gupta and Anadi Nath Acharya

**The Effect of Providing Cues to Action** 

**Dietary Fat and Colorectal Cancer 103** 

Sabina Passamonti and Jovana Cvorovic

**Cancer and Population Screening Tests 45** 

Ingrid Flight, Carlene Wilson and Jane McGillivray

**Factors After Disclosure of Genetic Test Results** 

**Colorectal Cancer and Mechanisms of Action 123**  Federica Tramer, Spela Moze, Ayokunle O. Ademosun,

**on Participation Rates for Colorectal Cancer Screening 67** 

**Concerning Hereditary Nonpolyposis Colorectal Cancer 87** 

**Part 1 Introduction 1** 

### Contents

#### **Preface XIII**

#### **Part 1 Introduction 1**

	- **Part 2 Epidemiology and Psychology 9**

#### **Part 3 Nutrition 101**


X Contents


Contents VII

Chapter 19 **Follow Up and Recurrence of Colorectal Cancer 363** 

**of Metastatic Colorectal Cancer 381** Béla Pikó, Ali Bassam, Enikő Török, Henriette Ócsai and Farkas Sükösd

Chapter 21 **Resection for Colorectal Liver Metastases 409** Daniel Kostov and Georgi Kobakov

> **After Surgery for Colorectal Cancer: A Matched Case – Control Study 465**  Takatoshi Nakamura and Masahiko Watanabe

Rania B. Georges, Hassan Adwan and Martin R. Berger

**Exemplification with the Faecal Occult Blood Test 473**

**Metastasis in Colon Cancer – Based on Japanese Techniques of Resection and Handling of Resected Specimens 509**

**Cancer Liver Metastasis 441**

Miroslav Levy

Chapter 22 **Experimental Colorectal** 

**Part 6 Study Reports 463** 

Chapter 25 **Dietary Risks: Folate,** 

Chapter 27 **Minimally Invasive Robot –**

Chapter 23 **Risk Factors for Wound Infection**

Chapter 24 **Modelling and Inference in Screening:** 

Dongfeng Wu and Adriana Pérez

**Alcohol and Gene Polymorphisms 491** Zi-Yuan Zhou, Keitaro Matsuo, Wen-Chang Wang,

Chapter 26 **The Prognostic Significance of Number of Lymph Node** 

Yoshito Akagi, Romeo Kansakar and Kazuo Shirouz

Huan Yang, Kazuo Tajima and Jia Cao

**Assisted Colorectal Resections 521**  Annibale D'Annibale, Graziano Pernazza,

Vito Pende and Igor Monsellato

Chapter 20 **Panitumumab for the Treatment** 

**Part 5 Metastasis 379**

Chapter 9 **The Molecular Genetic Events in Colorectal Cancer and Diet 173**  Adam Naguib, Laura J Gay, Panagiota N. Mitrou and Mark J. Arends

Chapter 10 **Colorectal Cancer and Alcohol 199**  Seitz K. Helmut and Homann Nils

Chapter 11 **Effects of Dietary Counseling on Patients with Colorectal Cancer 211**  Renata Dobrila-Dintinjana, Dragan Trivanović, Marijan Dintinjana, Jelena Vukelic and Nenad Vanis

#### **Part 4 Management and Treatment 227**


Chapter 19 **Follow Up and Recurrence of Colorectal Cancer 363**  Miroslav Levy

#### **Part 5 Metastasis 379**

VI Contents

Chapter 8 **Polyunsaturated Fatty Acids,**

Chapter 9 **The Molecular Genetic** 

Vera Lucia Flor Silveira

Adam Naguib, Laura J Gay,

Chapter 10 **Colorectal Cancer and Alcohol 199**  Seitz K. Helmut and Homann Nils

**Part 4 Management and Treatment 227**

Joseph Ciccolini, Fréderic Fina, L'Houcine Ouafik and Bruno Lacarelle

**Cancer in Chemoprevention** 

**and Therapeutics Development 277**

**Kinetics, Signaling and Markers 301**  George D. Wilson and Bryan J. Thibodeau

**Treatment for Colorectal Cancer 327**

**Surgery in the Twenty-First Century 349**

Rieko Yamada and Kiyo Watanabe

Hiroyuki Kato, Teruhiko Sakamoto, Hiroko Otsuka,

Chapter 12 **Therapeutic Targets in Colorectal Cancer 229**

**Patients with Colorectal Cancer 245** Camilla Qvortrup and Per Pfeiffer

Chapter 14 **Pharmacogenetics and Pharmacogenomics of Colorectal**

**Cancer: Moving Towards Personalized Medicine 259**

Shubhankar Suman, Albert J. Fornace Jr. and Kamal Datta

Chapter 11 **Effects of Dietary Counseling** 

Chapter 13 **Anti-EGFR Treatment in** 

Chapter 15 **Animal Models of Colorectal**

Chapter 16 **The Stem Cell Environment:** 

Chapter 17 **Endoscopic Diagnosis and**

Ned Abraham

Chapter 18 **Peri-Operative Care in Colorectal** 

**Ulcerative Colitis and Cancer Prevention 157** Karina Vieira de Barros, Ana Paula Cassulino and

**Events in Colorectal Cancer and Diet 173**

**on Patients with Colorectal Cancer 211**  Renata Dobrila-Dintinjana, Dragan Trivanović, Marijan Dintinjana, Jelena Vukelic and Nenad Vanis

Rajunor Ettarh, Alvise Calamai and Anthony Cullen

Panagiota N. Mitrou and Mark J. Arends


#### **Part 6 Study Reports 463**


Preface

emerging targets.

of metastatic lymph nodes.

When a patient are presented with symptoms that eventually lead to a diagnosis of colorectal cancer, it is the clinician who ultimately has to deliver the management and treatment of the condition based on what is known about the disease. The clinician synthesizes and brings together an understanding of the basic scientific facts available, and the information about effective clinical management and treatment – all for the patient's maximum benefit. Is the search of answers complete? No. There is still a very long way to go, but in terms of information, we are further ahead than we were a less than decade ago. Is there more to do? Yes. Our understanding is improving, but translation of basic scientific evidence to its application in terms of clinical treatment and management of patients remains a challenge. One thing is clear, a complete understanding of colorectal cancer and how it affects patients involves the continuing cooperation between research science and clinical practice. There are a number of positive examples resulting from searching and querying: monoclonal antibody therapy, pharmacologic agents and treatments, low dose aspirin, better risk management for colorectal cancer, and continually

This second volume of the book presents two sections that address aspects of epidemiology, psychology and nutrition as they relate to colorectal cancer from a patient illness and care perspective. Section 3 deals with the clinical management and treatment of the disease, while Section 4 explores different management approaches to colorectal cancer metastasis. Section 5 presents a collection of short reports that outline findings from studies on probability modeling, dietary risks and the prognostic value

Basic scientific researchers need to know where success has been registered, where failures lie and where the challenges in patient management and care remain. This volume represents an attempt to bring together much of what is known about colorectal cancer and provide a synoptic source of information that serves as a

reference point for scientists, clinicians, researchers, students and patients.

### Preface

When a patient are presented with symptoms that eventually lead to a diagnosis of colorectal cancer, it is the clinician who ultimately has to deliver the management and treatment of the condition based on what is known about the disease. The clinician synthesizes and brings together an understanding of the basic scientific facts available, and the information about effective clinical management and treatment – all for the patient's maximum benefit. Is the search of answers complete? No. There is still a very long way to go, but in terms of information, we are further ahead than we were a less than decade ago. Is there more to do? Yes. Our understanding is improving, but translation of basic scientific evidence to its application in terms of clinical treatment and management of patients remains a challenge. One thing is clear, a complete understanding of colorectal cancer and how it affects patients involves the continuing cooperation between research science and clinical practice. There are a number of positive examples resulting from searching and querying: monoclonal antibody therapy, pharmacologic agents and treatments, low dose aspirin, better risk management for colorectal cancer, and continually emerging targets.

This second volume of the book presents two sections that address aspects of epidemiology, psychology and nutrition as they relate to colorectal cancer from a patient illness and care perspective. Section 3 deals with the clinical management and treatment of the disease, while Section 4 explores different management approaches to colorectal cancer metastasis. Section 5 presents a collection of short reports that outline findings from studies on probability modeling, dietary risks and the prognostic value of metastatic lymph nodes.

Basic scientific researchers need to know where success has been registered, where failures lie and where the challenges in patient management and care remain. This volume represents an attempt to bring together much of what is known about colorectal cancer and provide a synoptic source of information that serves as a reference point for scientists, clinicians, researchers, students and patients.

#### X Preface

The cure for colorectal cancer can only be discovered when research science and clinical evidence collectively arrive at the right cocktail of information.

#### **Dr Rajunor Ettarh**

Professor & Associate Director of Anatomical Teaching, Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans, USA

#### **Acknowledgements**

The publication of this book would not have been possible without the support of my family. I am also especially indebted to Publishing Process Manager Tajana Jevtic whose infinite patience, timely reminders, and never-ending assistance and support made the task of editing this book easier.

X Preface

**Acknowledgements** 

made the task of editing this book easier.

The cure for colorectal cancer can only be discovered when research science and

The publication of this book would not have been possible without the support of my family. I am also especially indebted to Publishing Process Manager Tajana Jevtic whose infinite patience, timely reminders, and never-ending assistance and support

Professor & Associate Director of Anatomical Teaching,

Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans,

**Dr Rajunor Ettarh** 

USA

clinical evidence collectively arrive at the right cocktail of information.

**Part 1** 

**Introduction**

## **Part 1**

### **Introduction**

**1** 

*USA* 

Rajunor Ettarh

**Tumor Engineering: Finding the Brakes** 

There is sufficiently detailed understanding about how an automobile works such that starting or running the engine can be selectively disabled. Applying this analogy to colorectal cancer, the question researchers and clinicians ask is: can colorectal cancer be prevented from starting? Once started, can the disease be prevented from running? These two aims fall broadly into the categories of prevention and treatment respectively. Many of the aspects of colorectal cancer that provide focal points for clinical management of patients of the disease are included in Figure 1 (below). This volume provides insights into aspects of disease incidence and presentation, some of the advances and developments in diagnosis and patient management, and examines prevention and therapeutic targets and regimes. This chapter provides a general overview of some of the aspects of colorectal cancer that affect clinical management of the disease and explores incidence of the disease, diagnosis

As a disease, the statistical data for colorectal cancer are disturbing. Every year, there are over 1 million new cases worldwide, half of them in men; over 200,000 new cases in Europe; and 1.5 million new cases in the United States (Jemal et al, 2010). Over 700,000 patients die

Expanded surveys and studies show that the incidence of colorectal cancer is increasing worldwide, along with cancer detection rates. Other studies suggest that these rates may also be dependent on anatomic site along the intestine at which the cancer occurs. However, although absolute numbers of patients affected by the disease is increasing in the US, the trend for colorectal cancer is downward: age adjusted incidence has declined steadily since 1976 (Ji et al, 1998; Chen et al, 2011; Eser et al, 2010; Merrill & Anderson, 2011). Genomic instability is present in 15% of colorectal cancer, and forms the basis for those who advocate the need for screening programs for colorectal cancer patients (Geiersbach et al, 2011). Incidence of colorectal cancer around the world per 100,000 of population varies between 3- 43 and is influenced by age, gender, socioeconomic status, and ethnicity (Center et al, 2009; Hao et al, 2009). Younger patients have greater susceptibility if there is an associated family history and tend to present at a more advanced stage of the disease. Long and short-term incidence of colorectal cancer is also affected by aspirin intake and this effect may be

and treatment as well as preventive screening programs.

**1. Introduction** 

**2. Epidemiology** 

each year.

*Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans,* 

### **Tumor Engineering: Finding the Brakes**

#### Rajunor Ettarh

*Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans, USA* 

#### **1. Introduction**

There is sufficiently detailed understanding about how an automobile works such that starting or running the engine can be selectively disabled. Applying this analogy to colorectal cancer, the question researchers and clinicians ask is: can colorectal cancer be prevented from starting? Once started, can the disease be prevented from running? These two aims fall broadly into the categories of prevention and treatment respectively. Many of the aspects of colorectal cancer that provide focal points for clinical management of patients of the disease are included in Figure 1 (below). This volume provides insights into aspects of disease incidence and presentation, some of the advances and developments in diagnosis and patient management, and examines prevention and therapeutic targets and regimes. This chapter provides a general overview of some of the aspects of colorectal cancer that affect clinical management of the disease and explores incidence of the disease, diagnosis and treatment as well as preventive screening programs.

#### **2. Epidemiology**

As a disease, the statistical data for colorectal cancer are disturbing. Every year, there are over 1 million new cases worldwide, half of them in men; over 200,000 new cases in Europe; and 1.5 million new cases in the United States (Jemal et al, 2010). Over 700,000 patients die each year.

Expanded surveys and studies show that the incidence of colorectal cancer is increasing worldwide, along with cancer detection rates. Other studies suggest that these rates may also be dependent on anatomic site along the intestine at which the cancer occurs. However, although absolute numbers of patients affected by the disease is increasing in the US, the trend for colorectal cancer is downward: age adjusted incidence has declined steadily since 1976 (Ji et al, 1998; Chen et al, 2011; Eser et al, 2010; Merrill & Anderson, 2011). Genomic instability is present in 15% of colorectal cancer, and forms the basis for those who advocate the need for screening programs for colorectal cancer patients (Geiersbach et al, 2011).

Incidence of colorectal cancer around the world per 100,000 of population varies between 3- 43 and is influenced by age, gender, socioeconomic status, and ethnicity (Center et al, 2009; Hao et al, 2009). Younger patients have greater susceptibility if there is an associated family history and tend to present at a more advanced stage of the disease. Long and short-term incidence of colorectal cancer is also affected by aspirin intake and this effect may be

Tumor Engineering: Finding the Brakes 5

components. Biological treatment with bevacizumab, a recombinant antibody to vascular endothelial growth factor (VEGF) receptor, cetuximab and panitumumab has improved clinical outcomes for patients, prolonged survival times and is recommended in metastatic disease (Koukourakis et al, 2011). Despite these improvements in treatment, the number of patients who develop metastatic disease is significant and the prognosis for such patients is poor. Metastatic disease is thought to be related to epigenetic mechanisms and the development of cancer stem cell-mediated chemoresistance (Anderson et al, 2011). Treatment for metastatic disease is complex and requires careful patient evaluation and selection from single and combination treatment options that include surgery for resectable metastases, chemotherapy and biological therapy. Fluoropyrimidine 5-fluorouracil (5-FU) has been joined by cetuximab, an IgG antibody whose efficacy has been documented in several clinical trials (Lee & Chu, 2007). Improving regimes have led to better 2-year

New therapeutic approaches and targets are emerging from research studies. One promising approach currently being explored is the prospect of therapeutic vaccines to

A reduction in the morbidity and mortality from colorectal cancer can only be achieved through effective screening for the disease. Screening allows for early detection of cancer and early treatment of detected cancers. It is estimated that up to 60% of deaths from colorectal cancer could be prevented by routine screening after the age 50 years (Byers, 2011; He & Efron, 2011). Approaches to screening for colorectal cancer include stool-based tests (fecal immunochemical testing FIT, fecal occult blood testing FOBT), endoscopy (sigmoidoscopy and colonoscopy) and radiologic examinations (barium radiography, and colonography) (de Wijkerslooth et al, 2011). Studies suggest that stool-based testing is more

Colonoscopy remains the gold standard for screening and while it offers advantages for treatment such as removal of premalignant lesions, this approach may not be as protective for right-sided disease as it is for left sided disease (Baxter et al, 2009; Brenner et al, 2010; Singh et al, 2010). Other advanced colonic imaging techniques include capsule colonoscopy, computed tomographic colonography, virtual colonoscopy and magnetic resonance colonography (Liu et al, 2011). All screening programs are complicated by social and community factors (such as culture, level of knowledge about the disease) that affect

Colorectal cancer remains a major health challenge. Trends for geographically distributed fluctuations in incidence point towards the need for developing strategies to tackle increasing colorectal disease in the population under age 50 years, the relationship of the disease with socioeconomic status, and the increasing incidence of the disease in Asia. Treatment options are still dictated by the stage of the disease in the patient at presentation but evidence from basic science research studies are providing a better understanding of the disease process, drivers for improvements in therapeutic options for patients, and new

combat colorectal cancer (Kabaker et al, 2011; Kameshima et al, 2011).

cost effective than colonoscopy (Hassan et al, 2011; Wilschut et al, 2011).

participation rates (O'Donnell et al, 2010; Ramos et al, 2011; Reeder, 2011).

therapeutic targets for impeding the progression of the disease.

survival rates in patients.

**5. Conclusion** 

**4. Screening for prevention** 

dependent on dosing regime and patient history (Dube et al, 2007; Flossmann et al, 2007; Rothwell et al, 2010).

Fig. 1. Basic science studies and clinical management continue to improve our understanding of colorectal cancer. This volume considers incidence, diagnosis and clinical management of the disease as well as metastatic disease. Aspects of initiation and mechanisms are dealt with in the first volume of the book.

#### **3. Diagnosis and treatment**

There has been steady improvement in survival rates in colorectal cancers that are diagnosed early. Prognosis for patients who present with late stages of the disease remains poor. Treatment options include surgery for localized tumors, chemotherapy and immunotherapy. When resectable, surgical removal of the tumor remains the treatment of choice for localized colorectal disease. Surgery may be curative or palliative and is sometimes combined with chemotherapeutic regimes to achieve pre-operative tumor shrinkage (Zhao et al, 2010). Minimally invasive approaches such as laparoscopic surgery for colonic tumors are reported to offer improved short-term clinical outcomes (Hiranyakas & Ho, 2011). Chemotherapeutic regimes include infusional combination therapies such as FOLFIRI that combine irinotecan, 5-fluorouracil and leucovorin, and FOLFOX that combines oxaliplatin, 5-fluorouracil and leucovorin (Lee & Chu, 2007; Garcia-Foncillas & Diaz-Rubio, 2010). Studies suggest that overall survival time and progression-free survival are significantly improved with the addition of cetuximab to FOLFIRI.

Better understanding of some of the molecular mechanisms in colorectal cancer has led to the development of targeted therapy that modulate specific pathways and pathway

dependent on dosing regime and patient history (Dube et al, 2007; Flossmann et al, 2007;

Fig. 1. Basic science studies and clinical management continue to improve our

mechanisms are dealt with in the first volume of the book.

significantly improved with the addition of cetuximab to FOLFIRI.

**3. Diagnosis and treatment** 

management of the disease as well as metastatic disease. Aspects of initiation and

understanding of colorectal cancer. This volume considers incidence, diagnosis and clinical

There has been steady improvement in survival rates in colorectal cancers that are diagnosed early. Prognosis for patients who present with late stages of the disease remains poor. Treatment options include surgery for localized tumors, chemotherapy and immunotherapy. When resectable, surgical removal of the tumor remains the treatment of choice for localized colorectal disease. Surgery may be curative or palliative and is sometimes combined with chemotherapeutic regimes to achieve pre-operative tumor shrinkage (Zhao et al, 2010). Minimally invasive approaches such as laparoscopic surgery for colonic tumors are reported to offer improved short-term clinical outcomes (Hiranyakas & Ho, 2011). Chemotherapeutic regimes include infusional combination therapies such as FOLFIRI that combine irinotecan, 5-fluorouracil and leucovorin, and FOLFOX that combines oxaliplatin, 5-fluorouracil and leucovorin (Lee & Chu, 2007; Garcia-Foncillas & Diaz-Rubio, 2010). Studies suggest that overall survival time and progression-free survival are

Better understanding of some of the molecular mechanisms in colorectal cancer has led to the development of targeted therapy that modulate specific pathways and pathway

Rothwell et al, 2010).

components. Biological treatment with bevacizumab, a recombinant antibody to vascular endothelial growth factor (VEGF) receptor, cetuximab and panitumumab has improved clinical outcomes for patients, prolonged survival times and is recommended in metastatic disease (Koukourakis et al, 2011). Despite these improvements in treatment, the number of patients who develop metastatic disease is significant and the prognosis for such patients is poor. Metastatic disease is thought to be related to epigenetic mechanisms and the development of cancer stem cell-mediated chemoresistance (Anderson et al, 2011). Treatment for metastatic disease is complex and requires careful patient evaluation and selection from single and combination treatment options that include surgery for resectable metastases, chemotherapy and biological therapy. Fluoropyrimidine 5-fluorouracil (5-FU) has been joined by cetuximab, an IgG antibody whose efficacy has been documented in several clinical trials (Lee & Chu, 2007). Improving regimes have led to better 2-year survival rates in patients.

New therapeutic approaches and targets are emerging from research studies. One promising approach currently being explored is the prospect of therapeutic vaccines to combat colorectal cancer (Kabaker et al, 2011; Kameshima et al, 2011).

#### **4. Screening for prevention**

A reduction in the morbidity and mortality from colorectal cancer can only be achieved through effective screening for the disease. Screening allows for early detection of cancer and early treatment of detected cancers. It is estimated that up to 60% of deaths from colorectal cancer could be prevented by routine screening after the age 50 years (Byers, 2011; He & Efron, 2011). Approaches to screening for colorectal cancer include stool-based tests (fecal immunochemical testing FIT, fecal occult blood testing FOBT), endoscopy (sigmoidoscopy and colonoscopy) and radiologic examinations (barium radiography, and colonography) (de Wijkerslooth et al, 2011). Studies suggest that stool-based testing is more cost effective than colonoscopy (Hassan et al, 2011; Wilschut et al, 2011).

Colonoscopy remains the gold standard for screening and while it offers advantages for treatment such as removal of premalignant lesions, this approach may not be as protective for right-sided disease as it is for left sided disease (Baxter et al, 2009; Brenner et al, 2010; Singh et al, 2010). Other advanced colonic imaging techniques include capsule colonoscopy, computed tomographic colonography, virtual colonoscopy and magnetic resonance colonography (Liu et al, 2011). All screening programs are complicated by social and community factors (such as culture, level of knowledge about the disease) that affect participation rates (O'Donnell et al, 2010; Ramos et al, 2011; Reeder, 2011).

#### **5. Conclusion**

Colorectal cancer remains a major health challenge. Trends for geographically distributed fluctuations in incidence point towards the need for developing strategies to tackle increasing colorectal disease in the population under age 50 years, the relationship of the disease with socioeconomic status, and the increasing incidence of the disease in Asia.

Treatment options are still dictated by the stage of the disease in the patient at presentation but evidence from basic science research studies are providing a better understanding of the disease process, drivers for improvements in therapeutic options for patients, and new therapeutic targets for impeding the progression of the disease.

Tumor Engineering: Finding the Brakes 7

Hao Y, Jemal A, Zhang X, Ward EM. (2009). Trends in colorectal cancer incidence rates by

Hassan C, Benamouzig R, Spada C, Ponchon T, Zullo A, Saurin JC, Costamagna G. (2011).

Hiranyakas A, Ho YH. (2011). Surgical treatment for colorectal cancer. *Int Surg*, Vol.96, No.2,

Jemal A, Siegel R, Xu J, Ward E. (2010). Cancer statistics, 2010. *CA Cancer J Clin*, Vol.60, No.5,

Ji BT, Devesa SS, Chow WH, Jin F, Gao YT. (1998). Colorectal cancer incidence trends by

Kabaker K, Shell K, Kaufman HL. (2011). Vaccines for colorectal cancer and renal cell

Kameshima H, Tsuruma T, Torigoe T, Takahashi A, Hirohashi Y, Tamura Y, Tsukahara T,

Koukourakis GV, Sotiropoulou-Lontou A. (2011). Targeted therapy with bevacizumab

Lee JJ, Chu E. (2007). An update on treatment advances for the first-line therapy of

Liu J, Kabadi S, Van Uitert R, Petrick N, Deriche R, Summers RM. (2011). Improved

Merrill RM, Anderson AE. (2011). Risk-adjusted colon and rectal cancer incidence rates in

O'Donnell S, Goldstein B, Dimatteo MR, Fox SA, John CR, Obrzut JE. (2010). Adherence to

based screening for colorectal cancer. *BMC Cancer*, Vol.11, No.1, pp. 408. Reeder AI. (2011). "It's a small price to pay for life": faecal occult blood test (FOBT) screening

Rothwell PM, Wilson M, Elwin CE, Norrving B, Algra A, Warlow CP, Meade TW. (2010).

follow-up of five randomised trials. *Lancet*, Vol.376, No.9754, pp. 1741-1750.

metastatic colorectal cancer. *Cancer J*, Vol.13, No.5, pp. 276-281.

for curvature estimation. *Med Phys*, Vol.38, No.7, pp. 4276-4284

the United States. *Dis Colon Rectum*, Vol.54, No.10, pp. 1301-1306.

He J, Efron JE. (2011). Screening for colorectal cancer. *Adv Surg*, Vol.45, pp. 31-44.

*Cancer Causes Control*, Vol.20, No.10, pp. 1855-1863

France. *Endoscopy*, Vol.43, No.9, pp. 780-793.

carcinoma. *Cancer J*, Vol.17, No.5, pp. 283-293.

*Sci*, Vol.102, No.6, pp. 1181-1187.

pp. 120-126.

pp. 277-300.

714.

No.8, pp. 661-666.

No.1331, pp. 11-17.

age, race/ethnicity, and indices of access to medical care, 1995–2004 (United States).

Cost effectiveness and projected national impact of colorectal cancer screening in

subsite in urban Shanghai, 1972-1994. Cancer Epidemiol Biomarkers Prev, Vol.7,

Ichimiya S, Kanaseki T, Iwayama Y, Sato N, Hirata K. (2011). Immunogenic enhancement and clinical effect by type-I interferon of anti-apoptotic protein, survivin-derived peptide vaccine, in advanced colorectal cancer patients. *Cancer* 

(Avastin) for metastatic colorectal cancer. *Clin Transl Oncol*, Vol.13, No.10, pp. 710-

computer-aided detection of small polyps in CT colonography using interpolation

mammography and colorectal cancer screening in women 50-80 years of age the role of psychological distress. *Womens Health Issues*, Vol.20, No.5, pp. 343-349. Ramos M, Llagostera M, Esteva M, Cabeza E, Cantero X, Segarra M, Martín-Rabadán M,

Artigues G, Torrent M, Taltavull JM, Vanrell JM, Marzo M, Llobera J. (2011). Knowledge and attitudes of primary healthcare patients regarding population-

for colorectal cancer, perceived barriers and facilitators. *N Z Med J*, Vol.124,

Long-term effect of aspirin on colorectal cancer incidence and mortality: 20-year

Despite the remarkable improvement in our understanding of certain aspects of colorectal cancer, the best approach to combating the disease remains a preventive one. Prevention and screening programs need to be more efficient and more effective. Cost benefit analyses preclude early adoption of newer screening methods but advances in colonoscopic and colonographic approaches are helping to reduce morbidity and mortality for colorectal cancer.

#### **6. References**


Despite the remarkable improvement in our understanding of certain aspects of colorectal cancer, the best approach to combating the disease remains a preventive one. Prevention and screening programs need to be more efficient and more effective. Cost benefit analyses preclude early adoption of newer screening methods but advances in colonoscopic and colonographic approaches are helping to reduce morbidity and mortality for colorectal

Anderson EC, Hessman C, Levin TG, Monroe MM, Wong MH. (2011). The Role of

Baxter NN, Goldwasser MA, Paszat LF, Saskin R, Urbach DR, Rabeneck L. (2009).

Brenner H, Hoffmeister M, Arndt V, Stegmaier C, Altenhofen L, Haug U. (2010). Protection

Burke CA, Choure AG, Sanaka MR, Lopez R. (2010). A comparison of high-definition versus conventional colonoscopes for polyp detection. *Dig Dis Sci*, Vol.55, pp. 1716–1720. Byers T. (2011). Examining stools for colon cancer prevention: what are we really looking

Center MM, Jemal A, Ward E. (2009). International trends in colorectal cancer incidence

Chen HM, Weng YR, Jiang B, Sheng JQ, Zheng P, Yu CG, Fang JY. (2011). Epidemiological

de Wijkerslooth TR, Bossuyt PM, Dekker E. (2011). Strategies in screening for colon

Dubé C, Rostom A, Lewin G, Tsertsvadze A, Barrowman N, Code C, Sampson M, Moher D;

Preventive Services Task Force. *Ann Intern Med*, Vol.146, No.5, pp. 365-375. Eser S, Yakut C, Özdemir R, Karakilinç H, Özalan S, Marshall SF, Karaoğlanoğlu O,

Flossmann E, Rothwell PM; British Doctors Aspirin Trial and the UK-TIA Aspirin Trial.

García-Foncillas J, Díaz-Rubio E. (2010). Progress in metastatic colorectal cancer: growing

Geiersbach KB, Samowitz WS. (2011). Microsatellite instability and colorectal cancer. *Arch* 

estimation. *Asian Pac J Cancer Prev*, Vol.11, No.6, pp. 1731-1739.

study of colorectal adenoma and cancer in symptomatic patients in China between

U.S. Preventive Services Task Force. (2007). The use of aspirin for primary prevention of colorectal cancer: a systematic review prepared for the U.S.

Anbarcioğlu Z, Üçüncü N, Akin Ü, Özen E, Özgül N, Anton-Culver H, Tuncer M. (2010). Cancer incidence rates in Turkey in 2006: a detailed registry based

(2007). Effect of aspirin on long-term risk of colorectal cancer: consistent evidence from randomised and observational studies. *Lancet*, Vol.369, No.9573, pp. 1603-

role of cetuximab to optimize clinical outcome. *Clin Transl Oncol*, Vol.12, No.8, pp.

*Cancers (Basel)*, Vol.3, No.1, pp. 319-339.

based study. *J Natl Cancer Inst*, Vol.102, pp. 89–95.

for? *Cancer Prev Res (Phila)*, Vol.4, No.10, pp. 1531-1533

1990 and 2009. *J Dig Dis*, Vol.12, No.5, pp. 371-378.

carcinoma. *Neth J Med*, Vol.69, No.3, pp. 112-119.

*Pathol Lab Med*, Vol.135, No.10, pp. 1269-1277.

rates. *Cancer Epidemiol Biomarkers Prev*, Vol.18, pp. 1688–1694.

Colorectal Cancer Stem Cells in Metastatic Disease and Therapeutic Response.

Association of colonoscopy and death from colorectal cancer. *Ann Intern Med*,

from right- and left-sided colorectal neoplasms after colonoscopy: population-

cancer.

**6. References** 

Vol.150, pp. 1–8

1613.

533-542.


**Part 2** 

**Epidemiology and Psychology**


## **Part 2**

**Epidemiology and Psychology**

8 Colorectal Cancer – From Prevention to Patient Care

Siegel RL, Jemal A, Ward EM. (2009). Increase in incidence of colorectal cancer among

Singh H, Nugent Z, Demers AA, Kliewer EV, Mahmud SM, Bernstein CN. (2010). The

Tribonias G, Theodoropoulou A, Konstantinidis K, Vardas E, Karmiris K, Chroniaris N,

van den Broek FJ, Reitsma JB, Curvers WL, Fockens P, Dekker E. (2009). Systematic review

Wilschut JA, Hol L, Dekker E, Jansen JB, van Leerdam ME, Lansdorp-Vogelaar I, Kuipers EJ,

Zhao R, Zhu J, Ji X, Cai J, Wan F, Li Q, Zhong B, Tucker S, Wang D. (2010). A phase II study

from colorectal cancer. *Jpn J Clin Oncol*, Vol.40, No.1, pp. 10-16.

Vol.18, No.6, pp. 1695-1698.

*Colorectal Dis*, 12, e260–e266.

124–135

cancer. *Gastroenterology*, 139, pp. 1128–1137.

*Gastroenterology*, Vol.141, No.5, pp. 1648-1655.

young men and women in the United States. *Cancer Epidemiol Biomarkers Prev*,

reduction in colorectal cancer mortality after colonoscopy varies by site of the

Chlouverakis G, Paspatis GA. (2010). Comparison of standard vs high-definition, wide-angle colonoscopy for polyp detection: a randomized controlled trial.

of narrow-band imaging for the detection and differentiation of neoplastic and non-neoplastic lesions in the colon (with videos). *Gastrointest Endosc*, Vol.69, pp.

Habbema JD, van Ballegooijen M. (2011). Cost-effectiveness Analysis of a Quantitative Immunochemical Test for Colorectal Cancer Screening.

of irinotecan and capecitabine for patients with unresectable liver-only metastases

**2** 

*India* 

**Colorectal Carcinoma in the Young** 

*1Department of Surgery, Medical College & Hospitals Kolkata,* 

*Institute of Post Graduate Medical Education and Research, Kolkata,* 

Colorectal cancer (CRC) is the most common malignancy of the gastrointestinal tract. In the United States, it is the third most commonly diagnosed cancer, next only to breast and lung. It is the second most common cause of cancer-related death both in the USA and in the UK. (www. cancer. org, O'Connell et. al. 2004a, Leff et. al. 2007). Its incidence has risen rapidly in Asia to pose a problem (Yuen et. al. 1997, Huang et. al. 1999, Mohandas et. al. 1999, Yiu et. al. 2004, Goh et. al. 2005,Gupta et. al. 2010). Sung et. al. (2005) in a review on CRC in Asia stated that many Asian countries, e. g., China, Japan, South Korea, Singapore have experienced an increase of two to four times in CRC incidence during the past few decades. In Hong-Kong CRC is the second most common cancer and the third most common cause of cancer death (Yuen et. al. 1997). Tamura et. al. (1996) in a Japanese study reported that age adjusted incidence for CRC per 100,000 population were 12. 6 and 8. 7 for males and females respectively in 1974, 20 and 13. 6 in 1980, 42. 5 and 25. 6 in 1991. Bae et. al. (2002) estimated on the basis of Korean data, that the expected number of cancer deaths in Korea showed an increasing trend for CRC, although the same did not hold for all cancers. In Iran, age adjusted CRC incidence per 100,000 population per year increased from 1. 61 in 1970-80 to 4. 2 in 1990- 2000 in men and 2. 35 to 2. 72 for women (Hosseini et. al. 2004). The rising trend is more striking in affluent than in poorer societies and differs substantially amongst ethnic groups. Changes in dietary habits and lifestyle are recognized causes. Genetic characteristics of a population mediate the effect of life style change into disease propensity (Lin et. al. 2010). Although the common perception is that it is a disease of an older person, there have been many reports from different parts of the world on CRC in the young adults (Bulow 1980, Denmark; Ohman 1982, Sweden;Jarvinen and Turunen 1984, Finland; Ibrahim and Karim 1986, Lebanon; Adloff et. al. 1986, France; Isbister and Fraser 1990, New Zealand; Yuen et. al. 1997, Hong-Kong; Fante et. al. 1997, Italy; Ashenafi 2000, Ethiopia; deSilva et. al. 2000, Srilanka; Paraf and Jothy 2000, Canada; Turkiewicz et. al. , 2001, Australia; Singh et. al. 2002a, Nepal; Kam et. al. 2004, Malayasia; Frizis et. al. 2004, Greece; Guraya and Eltinay2006, Saudi Arabia; Fazeli et. al. 2007, Iran; Karsten et. al. 2008, USA; Gupta et. al. 2010, India). O'Connell et. al. (2004a) have reviewed the literature. The proportion of patients in the young group in a population of CRC patients was significantly larger in reports from Asia

**1. Introduction** 

and Africa, as compared to the Western reports.

 \*

Corresponding Author

Shahana Gupta1\* and Anadi Nath Acharya2

*2Department of Surgery,* 

### **Colorectal Carcinoma in the Young**

Shahana Gupta1\* and Anadi Nath Acharya2

*1Department of Surgery, Medical College & Hospitals Kolkata, 2Department of Surgery, Institute of Post Graduate Medical Education and Research, Kolkata, India* 

#### **1. Introduction**

Colorectal cancer (CRC) is the most common malignancy of the gastrointestinal tract. In the United States, it is the third most commonly diagnosed cancer, next only to breast and lung. It is the second most common cause of cancer-related death both in the USA and in the UK. (www. cancer. org, O'Connell et. al. 2004a, Leff et. al. 2007). Its incidence has risen rapidly in Asia to pose a problem (Yuen et. al. 1997, Huang et. al. 1999, Mohandas et. al. 1999, Yiu et. al. 2004, Goh et. al. 2005,Gupta et. al. 2010). Sung et. al. (2005) in a review on CRC in Asia stated that many Asian countries, e. g., China, Japan, South Korea, Singapore have experienced an increase of two to four times in CRC incidence during the past few decades. In Hong-Kong CRC is the second most common cancer and the third most common cause of cancer death (Yuen et. al. 1997). Tamura et. al. (1996) in a Japanese study reported that age adjusted incidence for CRC per 100,000 population were 12. 6 and 8. 7 for males and females respectively in 1974, 20 and 13. 6 in 1980, 42. 5 and 25. 6 in 1991. Bae et. al. (2002) estimated on the basis of Korean data, that the expected number of cancer deaths in Korea showed an increasing trend for CRC, although the same did not hold for all cancers. In Iran, age adjusted CRC incidence per 100,000 population per year increased from 1. 61 in 1970-80 to 4. 2 in 1990- 2000 in men and 2. 35 to 2. 72 for women (Hosseini et. al. 2004). The rising trend is more striking in affluent than in poorer societies and differs substantially amongst ethnic groups. Changes in dietary habits and lifestyle are recognized causes. Genetic characteristics of a population mediate the effect of life style change into disease propensity (Lin et. al. 2010).

Although the common perception is that it is a disease of an older person, there have been many reports from different parts of the world on CRC in the young adults (Bulow 1980, Denmark; Ohman 1982, Sweden;Jarvinen and Turunen 1984, Finland; Ibrahim and Karim 1986, Lebanon; Adloff et. al. 1986, France; Isbister and Fraser 1990, New Zealand; Yuen et. al. 1997, Hong-Kong; Fante et. al. 1997, Italy; Ashenafi 2000, Ethiopia; deSilva et. al. 2000, Srilanka; Paraf and Jothy 2000, Canada; Turkiewicz et. al. , 2001, Australia; Singh et. al. 2002a, Nepal; Kam et. al. 2004, Malayasia; Frizis et. al. 2004, Greece; Guraya and Eltinay2006, Saudi Arabia; Fazeli et. al. 2007, Iran; Karsten et. al. 2008, USA; Gupta et. al. 2010, India). O'Connell et. al. (2004a) have reviewed the literature. The proportion of patients in the young group in a population of CRC patients was significantly larger in reports from Asia and Africa, as compared to the Western reports.

<sup>\*</sup> Corresponding Author

Colorectal Carcinoma in the Young 13

being made. Intelligent choice of treatment protocol, surgical as well as chemotherapeutic is also influenced by research on molecular genetics of CRC (Liang and Church 2010). Hereditary CRC usually occurs at a relatively young age, between 25 and 55 years in individuals with family history of CRC. Individuals who inherit the predisposing cancer gene have a greater chance of developing the disease (Murday and Slock 1989, Lynch et. al. 1991, Lynch and de la Chapelle 2003, Ewart Toland, 2012). The importance of family history in determining susceptibility to CRC in the young has been stressed in the literature (St. John et. al. 1993, Fuchs et. al. 1994, Turkiewicz et. al. 2001). There exist literature reports that identify genetic factors in younger CRC patients which differ from those in older patients and may be responsible for greater cancer susceptibility of the younger patients (Farrington

In this essay, we focus on the issue of CRC in young age, with particular reference to developing countries. The relative incidence figures of CRC in the young patients as compared to older patients in different parts of the world are given. These figures, in greater detail are given in the Indian context (section 2). Disease stage at presentation and tumour characteristics of younger patients, often in comparison with the older ones in different countries are then summarized (section 3). A brief reference to de novo cancer in Asians (section 4) is followed by a discussion of some recent genetic studies in the young (section 5). Section 6 contains a discussion on prediosposing factors and section 7 has focus on prognosis in the young. The paper concludes (section 8) with a brief reference to the effect of

The relative incidence of CRC in the younger group varies significantly from one country to another. As cited above, it is typically 2-3% in the West. Other European figures are: Fante et. al. (Italy): 1%; Endreseth et. al. (Norway):6%**;** Ohman (Sweden): 4%; Adloff et. al. (France): 3%; Yilmazlar et. al. (Turkey): 20%. The corresponding figures are much higher from several Asian and African countries: Nath et. al. (India): 35. 6%, <40 yrs; Gupta et. al. (India): 39%,<40 years; Singh et. al. (South Asia): 23%,<40 years (with a maximum incidence in 40-60 years, a decade earlier than Western figures): study period 1975-1981; Soliman et. al. (Egypt): 35. 6%,<40 yrs; Ashenafi (Ethiopia): mean age 47 years (61. 4% <50years, 36% <40 yrs,16% <30 yrs) in two 5 year periods with a 10 year gap; Guraya and Eltinay (Saudi Arabia): study period 1999-2004,63% <40 yrs, mean age 44years, peak incidence 30-39 years; Hosseini et. al. (Iran): 70% (<60 years):study period 1990-2000; Chew et. al. (Singapore):25% <40 years; Singh et. al. (Nepal): 28. 6% <40 years; de Silva et. al. (Sri Lanka): 19. 7% <40 years. Some of these references are detailed in Table 1. In Egypt, more than half of all CRC patients are below-50, patients under-30 constitute 22% of the population of all CRC patients (Soliman et. al. 1997). Qing et.al. (2003) in a comparative study of American and Chinese patients (1990- 2000) reported that the mean age at diagnosis of 690 American patients was 69 years (20-91 years) and that of 870 Chinese patients was 48. 3 years (13-84 years); peak incidence was 70- 79 years in Whites and 50-59 years in Orientals. The conclusion is that the Orientals are affected by the disease at a younger age. The same theme emerges from recent data from several Indian hospitals which includes our own recent work (Gupta et. al. 2010). In a period spanning 8 years (2000-2008), we found the ratio of under-40 to above- 40 years age group to be 0. 64. The study group comprised of 305 patients in SSKM Hospital, Kolkata, India, a premier referral Hospital. The values reported by three premier Oncology centers located in two cities in India and in another report by Pal (2006), based on work done

et. al. 1998, Chan et. al. 1999, Morris et. al. 2007, Berg et. al. 2010, Lin et. al. 2010).

recent molecular genetic research on treatment protocol.

**2. Incidence amongst young adults** 

The definition of 'Young adults' varies, to a small extent, in the literature. Majority of articles defined 'young' as <40 years, although upper limits of 50 years, 35 years and 30 years have also been used. O'Connell et. al. (2004a) estimated the average value of incidence of CRC in the young adults (<40 years) in the population of all CRC patients as 7% and adjusted it to 6%, when outliers were removed. It has been suggested (Hamilton 2005) that the adjustment was 'too small' and a more realistic estimate was an average of 2. 2%. Leff et. al. (2007) gave an estimate of 2-3%. About 0. 1% of all CRC patients were diagnosed <20 years of age, ~1% between 20-34 years, ~4% between 35-44 years and a further ~12% between 45-54 years. These average figures reflect the extent of the problem in the West. The figures from Asian and African countries are considerably higher, a quarter or a half of a study group of CRC patients may belong to the under-40 group (Ashenafi 2000, Ethiopia; deSilva et. al. 2000, SriLanka; Singh et. al. 2002a, Nepal; Guraya and Eltinay 2006, Saudi Arabia; Gupta et. al. 2010, India). Numerical values given later will establish that the problem of CRC in the young adult in the developing world is alarming.

We now cite reports, from the West (USA, France, Scotland) and from Asia (Iran, Hong Kong), in which the incidence of the disease amongst the young adults has been studied in the same population over a period of time. O'Connell et. al. (2003) noted that in the USA, colon cancer incidence in older patients (60 + years) remained stable in the period 1973-1999 while rectal cancer incidence decreased by 11%. In the group of younger patients (20-40 years) colon cancer incidence increased by 17%, while rectal cancer incidence rose by 75% in the period 1973-1999. The improvement in the older age group is a reflection of more efficient cancer screening in the USA, a result of improved awareness of the disease. It is possible that relative ignorance about the problem of CRC in the young adult is responsible for the fact that the problem has worsened over the years. Other issues namely difference in molecular genetics, may also be present. In Iran, Hosseini et. al. (2004) defined the younger group as <60 years, compared figures in two 10 year periods 1970-1980 and 1990-2000 and found an increased proportion of < 60 years CRC patients (in a population of all CRC patients) in the latter decade, 37. 5% as against 70%. An increase in proportion of the young CRC patients was noted over a prolonged time span. Mitry et. al. (2001) from France reported that below-45 age standardized incidence rates doubled in the period 1976-1982 and then again in the period 1983-1989, in both genders and stabilized thereafter. In Hong-Kong, the overall incidence in > 50 years group increased at a rate of 4% a year during 1978-87, whereas in Scotland a higher overall incidence remained stable during this period (Yuen et. al. 1997).

O'Connell et. al. (2004b) in a study of American patients found that young (20-40 years) colon cancer patients tend to have later-stage and higher-grade tumours. However they have equivalent or better 5 year cancer-specific survival compared to 60+ older group, an apparently paradoxical result. Although most reports agree on a more severe advanced disease at presentation in the young (Adloff et. al. 1986, Cusack et. al. 1996, Nath et. al. , 2009) and many also agree with the opinion that prognosis is not poorer in the young (Jarvinen and Turunen 1984, Turkiewicz et. al. 2001, Karsten et. al. 2008) some reports (Moore et. al. 1984, Adkins et. al. 1987, Okuno et. al. 1987, Singh et. al. 2002a) do not share the view that prognosis is 'equivalent or better'. Inspite of this difference in assessment, a favourable prognosis in many studies should inspire more aggressive detection and treatment for the young.

The genetic basis of CRC has been investigated in recent years. A satisfactory understanding of the disease, tumour characteristics, relationship of disease susceptibility with age and issues related to survival rely on an understanding of the link between molecular genetics and disease. A complete resolution of this relation is a tall order, but a modest beginning is

The definition of 'Young adults' varies, to a small extent, in the literature. Majority of articles defined 'young' as <40 years, although upper limits of 50 years, 35 years and 30 years have also been used. O'Connell et. al. (2004a) estimated the average value of incidence of CRC in the young adults (<40 years) in the population of all CRC patients as 7% and adjusted it to 6%, when outliers were removed. It has been suggested (Hamilton 2005) that the adjustment was 'too small' and a more realistic estimate was an average of 2. 2%. Leff et. al. (2007) gave an estimate of 2-3%. About 0. 1% of all CRC patients were diagnosed <20 years of age, ~1% between 20-34 years, ~4% between 35-44 years and a further ~12% between 45-54 years. These average figures reflect the extent of the problem in the West. The figures from Asian and African countries are considerably higher, a quarter or a half of a study group of CRC patients may belong to the under-40 group (Ashenafi 2000, Ethiopia; deSilva et. al. 2000, SriLanka; Singh et. al. 2002a, Nepal; Guraya and Eltinay 2006, Saudi Arabia; Gupta et. al. 2010, India). Numerical values given later will establish that the problem of CRC in the

We now cite reports, from the West (USA, France, Scotland) and from Asia (Iran, Hong Kong), in which the incidence of the disease amongst the young adults has been studied in the same population over a period of time. O'Connell et. al. (2003) noted that in the USA, colon cancer incidence in older patients (60 + years) remained stable in the period 1973-1999 while rectal cancer incidence decreased by 11%. In the group of younger patients (20-40 years) colon cancer incidence increased by 17%, while rectal cancer incidence rose by 75% in the period 1973-1999. The improvement in the older age group is a reflection of more efficient cancer screening in the USA, a result of improved awareness of the disease. It is possible that relative ignorance about the problem of CRC in the young adult is responsible for the fact that the problem has worsened over the years. Other issues namely difference in molecular genetics, may also be present. In Iran, Hosseini et. al. (2004) defined the younger group as <60 years, compared figures in two 10 year periods 1970-1980 and 1990-2000 and found an increased proportion of < 60 years CRC patients (in a population of all CRC patients) in the latter decade, 37. 5% as against 70%. An increase in proportion of the young CRC patients was noted over a prolonged time span. Mitry et. al. (2001) from France reported that below-45 age standardized incidence rates doubled in the period 1976-1982 and then again in the period 1983-1989, in both genders and stabilized thereafter. In Hong-Kong, the overall incidence in > 50 years group increased at a rate of 4% a year during 1978-87, whereas in Scotland a higher overall incidence remained

O'Connell et. al. (2004b) in a study of American patients found that young (20-40 years) colon cancer patients tend to have later-stage and higher-grade tumours. However they have equivalent or better 5 year cancer-specific survival compared to 60+ older group, an apparently paradoxical result. Although most reports agree on a more severe advanced disease at presentation in the young (Adloff et. al. 1986, Cusack et. al. 1996, Nath et. al. , 2009) and many also agree with the opinion that prognosis is not poorer in the young (Jarvinen and Turunen 1984, Turkiewicz et. al. 2001, Karsten et. al. 2008) some reports (Moore et. al. 1984, Adkins et. al. 1987, Okuno et. al. 1987, Singh et. al. 2002a) do not share the view that prognosis is 'equivalent or better'. Inspite of this difference in assessment, a favourable prognosis in

many studies should inspire more aggressive detection and treatment for the young.

The genetic basis of CRC has been investigated in recent years. A satisfactory understanding of the disease, tumour characteristics, relationship of disease susceptibility with age and issues related to survival rely on an understanding of the link between molecular genetics and disease. A complete resolution of this relation is a tall order, but a modest beginning is

young adult in the developing world is alarming.

stable during this period (Yuen et. al. 1997).

being made. Intelligent choice of treatment protocol, surgical as well as chemotherapeutic is also influenced by research on molecular genetics of CRC (Liang and Church 2010). Hereditary CRC usually occurs at a relatively young age, between 25 and 55 years in individuals with family history of CRC. Individuals who inherit the predisposing cancer gene have a greater chance of developing the disease (Murday and Slock 1989, Lynch et. al. 1991, Lynch and de la Chapelle 2003, Ewart Toland, 2012). The importance of family history in determining susceptibility to CRC in the young has been stressed in the literature (St. John et. al. 1993, Fuchs et. al. 1994, Turkiewicz et. al. 2001). There exist literature reports that identify genetic factors in younger CRC patients which differ from those in older patients and may be responsible for greater cancer susceptibility of the younger patients (Farrington et. al. 1998, Chan et. al. 1999, Morris et. al. 2007, Berg et. al. 2010, Lin et. al. 2010).

In this essay, we focus on the issue of CRC in young age, with particular reference to developing countries. The relative incidence figures of CRC in the young patients as compared to older patients in different parts of the world are given. These figures, in greater detail are given in the Indian context (section 2). Disease stage at presentation and tumour characteristics of younger patients, often in comparison with the older ones in different countries are then summarized (section 3). A brief reference to de novo cancer in Asians (section 4) is followed by a discussion of some recent genetic studies in the young (section 5). Section 6 contains a discussion on prediosposing factors and section 7 has focus on prognosis in the young. The paper concludes (section 8) with a brief reference to the effect of recent molecular genetic research on treatment protocol.

#### **2. Incidence amongst young adults**

The relative incidence of CRC in the younger group varies significantly from one country to another. As cited above, it is typically 2-3% in the West. Other European figures are: Fante et. al. (Italy): 1%; Endreseth et. al. (Norway):6%**;** Ohman (Sweden): 4%; Adloff et. al. (France): 3%; Yilmazlar et. al. (Turkey): 20%. The corresponding figures are much higher from several Asian and African countries: Nath et. al. (India): 35. 6%, <40 yrs; Gupta et. al. (India): 39%,<40 years; Singh et. al. (South Asia): 23%,<40 years (with a maximum incidence in 40-60 years, a decade earlier than Western figures): study period 1975-1981; Soliman et. al. (Egypt): 35. 6%,<40 yrs; Ashenafi (Ethiopia): mean age 47 years (61. 4% <50years, 36% <40 yrs,16% <30 yrs) in two 5 year periods with a 10 year gap; Guraya and Eltinay (Saudi Arabia): study period 1999-2004,63% <40 yrs, mean age 44years, peak incidence 30-39 years; Hosseini et. al. (Iran): 70% (<60 years):study period 1990-2000; Chew et. al. (Singapore):25% <40 years; Singh et. al. (Nepal): 28. 6% <40 years; de Silva et. al. (Sri Lanka): 19. 7% <40 years. Some of these references are detailed in Table 1. In Egypt, more than half of all CRC patients are below-50, patients under-30 constitute 22% of the population of all CRC patients (Soliman et. al. 1997). Qing et.al. (2003) in a comparative study of American and Chinese patients (1990- 2000) reported that the mean age at diagnosis of 690 American patients was 69 years (20-91 years) and that of 870 Chinese patients was 48. 3 years (13-84 years); peak incidence was 70- 79 years in Whites and 50-59 years in Orientals. The conclusion is that the Orientals are affected by the disease at a younger age. The same theme emerges from recent data from several Indian hospitals which includes our own recent work (Gupta et. al. 2010). In a period spanning 8 years (2000-2008), we found the ratio of under-40 to above- 40 years age group to be 0. 64. The study group comprised of 305 patients in SSKM Hospital, Kolkata, India, a premier referral Hospital. The values reported by three premier Oncology centers located in two cities in India and in another report by Pal (2006), based on work done

Colorectal Carcinoma in the Young 15

In the younger group: Dukes' A:2,B:8, C: 28, D: 7 patients.

Higher incidence of advanced disease, especially in second or third decades.

Advanced stage: T-3/4 lesion in 87. 8% of young/63% in older group (p=0. 002).

Less localized, more aggressive disease in terms of stage in the young (20-49yrs)

Dukes' C and D more frequent at a lower age (p=0. 03). Mean Age A/B-1 67. 7 yrs, B-2 70. 1yrs, C/D 63. 9yrs

19: poorly differentiated,19: well or moderately differentiated tumour

Greater incidence of mucinous variety (32. 3% in young vs. 8. 6% in the whole study group). Poorly differentiated tumour: 98%; distant metastases in one-third patients. Vascular (24%) and perineural (11%) invasion in the young.

Poorly differentiated, (p=0. 003), mucin secreting/ signet ring (p=0. 005), more common in the young

Incidence of poorly differentiated tumour in young (<50yrs) (i) twice as high as well differentiated ones in the young (ii)60% higher than that for well differentiated cancers in the old

> Grade not related to age

705 patients; 45 patients, <35yrs

3. 2% of 1909 patients <40yrs

Younger group: 41 patients <40yrs Older group : >60 yrs

All age groups Young:20-49yrs

1987-1991 57. 2% <70yrs

<sup>9</sup>Adkins,

<sup>10</sup>Moore,

<sup>11</sup>Karsten. ,

12

1998-2005

Fairley, 1998-2001 Cancer Registry (NPCR,SEER)\*

<sup>13</sup>Lichtman

1973-1984

1967-1981


62 patients, <40yrs

37 patients, <40yrs

<40yrs

2609 patients,<50yrs age; 183 aged<40 yrs. Comparison between<30yr and 30-39yr age group and with yet older age group

69 patients: 20- 39yrs

all age groups

55 patients, <30yrs

801 patients including <40yrs group

**Period of study Age profile Disease stage Tumour characteristics** 

Dukes' A:8%, B:20%, C:23%, D:48%.

Dukes' C:37%, D: 22%.

Dukes' C & D: 65. 6%.

96% of < 40 years group had carcinoma extending beyond colonic wall.

> 67% Dukes' C and D

Advanced stage more frequent in the young.

> 62% Dukes' C and D

Advanced signs and stages more frequent in the young.

Half of stage D patients and 20% of lower stage patients (p=0. 037) had high grade lesions.

Mucinous tumour : 42%; moderate and poor differentiation : 84%

Poorly differentiated tumour in 41%, signetring cell tumours in 11. 1%, infiltrating tumour leading edges in 69% of young patients. Aggressive tumour biology with higher frequency in <40yr patients (p<0. 001), potentially metastatic.

Moderate and poorly differentiated neoplasms (80%) and mucinous variety (33%) in young.

Mucinous variety (28%).

Greater frequency of mucinous tumour in the young.

Poorly differentiated /mucinous variety: 33%

Greater frequency of mucinous variety in the young

**Sr. No**. **Reference,** 

1968-91

1976-97

3 Cusack 186 patients,

<sup>1</sup>Lee

<sup>2</sup>Minardi,

<sup>4</sup>Bedikian,

<sup>5</sup>Beckman

8 Howard

7

1944-1977

1943-1977

Cozart, Unusual Case Registry 1992-93

6 Varma A review:


Colorectal Carcinoma in the Young 17

The most powerful predictor of outcome for young adults, as it is for older patients is disease stage. Two staging systems are in use and are cited in Table 1-3. One is the tumournode-metastases (TNM) staging system of the American Joint Committee on Cancer (AJCC). Microscopic extent of tumour invasion (T stage) and nodal involvement (N stage) from histological assessment are combined with assessment for metastatic disease (M stage) to specify a tumour stage. Brief description of TNM stages are: Tumour stages (T): Tumour in T1, invades submucosa, T2: invades muscularis propria, T3 and T4 are more extensive, T3 indicates invasion through muscularis propria into subserosa or into nonperitonealised pericolic or perirectal tissues while T-4 invades adjacent organs. Regional Lymph node stages: N1: 1-3 positive nodes, N2:4 or more positive nodes. Distant metastases stages (M): M1: Distant metastases present. The other classification system known as Dukes' system is: A: limited to bowel wall, B: penetration of bowel wall, C: lymph node involvement, D:

Mucinous adenocarcinoma is one of the histological subtypes of colorectal cancers. It accounts for 5-15% of all primary CRC and is defined as a tumour with >50% of its body showing a mucinous pattern on histological examination and with a large amount of extra cellular mucin produced by secreting acini. This is distinct from signet ring adenocarcinoma, a rare variant in which mucin remains inside the cell, which is well known for its aggressiveness. It has been suggested that mucinous adenocarcinoma behaves differently from more common histological subtypes of CRC. However, its clinical implications remain unclear. According to published series, mucinous adenocarcinoma affects younger patients, is more frequent in proximal part of the colon and tends to present at a more advanced stage

In Table 1, 2 and 3 we tabulate data on disease stage and tumour characteristics, in particular its mucinous nature, of CRC patients in the USA (Table 1), in Europe, inclusive of

Several reports cited in Table 1 (Sr No 1,2,3,5,7,9,14) were entirely on features of CRC in the younger patients. In several other reports (Sr No. 4,6,8,10-13,15), both the younger and the older patient groups were studied and comparative features were assessed. The size of the younger group was mostly ~50, was ~200 in two reports (Sr No 3 and 4) and was 1025 in the work of Dozois et. al. (2005) (Sr no 14), the largest cohort of young CRC patients. In reports that included older and younger patients, older patients were much larger in number (Sr No 4,8-10). In all studies that were on younger patients alone, a high incidence of advanced stage (C+D: >70%)was reported. In studies that included both groups, the frequency of advanced disease in the young was as high or higher (Sr No. 4,11,15). In all of them, advanced disease stages were found to be more frequent in the young than in the old. In studies on younger patients alone, a significant proportion of patients had aggressive lesions, namely mucinous, poorly differentiated tumours with infiltrating leading edge. The frequency of aggressive tumour biology varied from one study to another but remained significant in all of them. In the comparative studies (Sr No. 6,8,10-12,15), the younger patients showed a higher frequency of aggressive tumour biology. Only one report (Sr No. 13) concluded that grade was not

Turkey and the UK and Australia (Table 2) and in Asia and Africa (Table 3).

**3. Disease stage and tumour characteristics in the young adults** 

distant metastatic disease present (Fry et. al. 2008).

(Negri et. al. 2005).

related to age.


\*NPCR: National Program of Cancer Registries; SEER: Surveillance, Epidemiology and End Results

Table 1. Summary of references in the literature on stage and tumour characteristics in the USA

in the same referral hospital where Gupta et. al. (2010) worked are 0. 58, 0. 63, 0. 45, 0. 62. Average of these five ratios is 0. 52, which is equivalent to ~34% of < 40 years CRC patients amongst all CRC patients. This figure is of the same order as the values from several Asian and African countries cited above. They are also substantially larger than values recorded in National Cancer Registry (PBCR) in four Indian metropolises. The PBCR ratio is 0. 20 and has remained stable over 16 years (1988-2004).

The difference between PBCR values and those reported by five premier hospitals in India, irrespective of their location and specialty, cited by Gupta et. al. (2010) has a clear message. The concern and facilities for cancer detection in the premier hospitals is greater than those in district hospitals. The data of the district hospitals are reflected in the PBCR values. This is the reason for the larger proportion of under-40 patients reported by the premier hospitals. The reason for delay in diagnosis of a young patient in either the premier hospitals or the district hospitals, particularly in the developing world, is that unless there is a family history these patients are not screened. So cancers are usually symptomatic at presentation. Even when symptoms occur, they may initially be misdiagnosed. Rectal bleeding for example is often put down to an anorectal cause. O'Connell et. al. (2004a) report an average delay in diagnosis of 6. 2 months, the reasons for which include a delay in presentation on the part of the patients, limited access to care and misdiagnosis on the part of the physician. This delay is larger in the developing world. Minimizing delay in diagnosis means not taking such symptoms lightly. Rectal bleeding usually has an anorectal cause, but when no such cause is obvious and the bleeding persists, colonoscopy is mandatory, regardless of patient's age. The same concern must apply to other less obvious symptoms.

In a review on CRC in Asia, Sung et. al. (2005) placed India at the bottom of the list amongst Asian countries, in order of decreasing CRC incidence. The data we provide does not contradict this assessment, but if relative incidence in the young is an indication, India has joined the rest of Asia.

70% colon, 60% rectal: stage C&D

Advanced stage C&D: ~90% in young ~50% in general population

Colon Cancer: Mucinous(11%) & Signet cell (2%) Grade 2+3 for both rectum & colon cancer: ~87%

Poor differentiation: 21% in young, 8% in general population

1025 patients,<50yrs; Mean age 42. 4±6. 4 years 51% colon, 49% rectal (largest cohort of young-onset patients without genetic predisposition)

> <40 years group: 56 patients

\*NPCR: National Program of Cancer Registries; SEER: Surveillance, Epidemiology and End Results Table 1. Summary of references in the literature on stage and tumour characteristics in the

in the same referral hospital where Gupta et. al. (2010) worked are 0. 58, 0. 63, 0. 45, 0. 62. Average of these five ratios is 0. 52, which is equivalent to ~34% of < 40 years CRC patients amongst all CRC patients. This figure is of the same order as the values from several Asian and African countries cited above. They are also substantially larger than values recorded in National Cancer Registry (PBCR) in four Indian metropolises. The PBCR ratio is 0. 20 and has

The difference between PBCR values and those reported by five premier hospitals in India, irrespective of their location and specialty, cited by Gupta et. al. (2010) has a clear message. The concern and facilities for cancer detection in the premier hospitals is greater than those in district hospitals. The data of the district hospitals are reflected in the PBCR values. This is the reason for the larger proportion of under-40 patients reported by the premier hospitals. The reason for delay in diagnosis of a young patient in either the premier hospitals or the district hospitals, particularly in the developing world, is that unless there is a family history these patients are not screened. So cancers are usually symptomatic at presentation. Even when symptoms occur, they may initially be misdiagnosed. Rectal bleeding for example is often put down to an anorectal cause. O'Connell et. al. (2004a) report an average delay in diagnosis of 6. 2 months, the reasons for which include a delay in presentation on the part of the patients, limited access to care and misdiagnosis on the part of the physician. This delay is larger in the developing world. Minimizing delay in diagnosis means not taking such symptoms lightly. Rectal bleeding usually has an anorectal cause, but when no such cause is obvious and the bleeding persists, colonoscopy is mandatory, regardless of patient's age. The same

In a review on CRC in Asia, Sung et. al. (2005) placed India at the bottom of the list amongst Asian countries, in order of decreasing CRC incidence. The data we provide does not contradict this assessment, but if relative incidence in the young is an indication, India has

<sup>14</sup>Dozois,

15

USA

1976-2002

Behbehani 11 yrs period pre-1980

remained stable over 16 years (1988-2004).

concern must apply to other less obvious symptoms.

joined the rest of Asia.

#### **3. Disease stage and tumour characteristics in the young adults**

The most powerful predictor of outcome for young adults, as it is for older patients is disease stage. Two staging systems are in use and are cited in Table 1-3. One is the tumournode-metastases (TNM) staging system of the American Joint Committee on Cancer (AJCC). Microscopic extent of tumour invasion (T stage) and nodal involvement (N stage) from histological assessment are combined with assessment for metastatic disease (M stage) to specify a tumour stage. Brief description of TNM stages are: Tumour stages (T): Tumour in T1, invades submucosa, T2: invades muscularis propria, T3 and T4 are more extensive, T3 indicates invasion through muscularis propria into subserosa or into nonperitonealised pericolic or perirectal tissues while T-4 invades adjacent organs. Regional Lymph node stages: N1: 1-3 positive nodes, N2:4 or more positive nodes. Distant metastases stages (M): M1: Distant metastases present. The other classification system known as Dukes' system is: A: limited to bowel wall, B: penetration of bowel wall, C: lymph node involvement, D: distant metastatic disease present (Fry et. al. 2008).

Mucinous adenocarcinoma is one of the histological subtypes of colorectal cancers. It accounts for 5-15% of all primary CRC and is defined as a tumour with >50% of its body showing a mucinous pattern on histological examination and with a large amount of extra cellular mucin produced by secreting acini. This is distinct from signet ring adenocarcinoma, a rare variant in which mucin remains inside the cell, which is well known for its aggressiveness. It has been suggested that mucinous adenocarcinoma behaves differently from more common histological subtypes of CRC. However, its clinical implications remain unclear. According to published series, mucinous adenocarcinoma affects younger patients, is more frequent in proximal part of the colon and tends to present at a more advanced stage (Negri et. al. 2005).

In Table 1, 2 and 3 we tabulate data on disease stage and tumour characteristics, in particular its mucinous nature, of CRC patients in the USA (Table 1), in Europe, inclusive of Turkey and the UK and Australia (Table 2) and in Asia and Africa (Table 3).

Several reports cited in Table 1 (Sr No 1,2,3,5,7,9,14) were entirely on features of CRC in the younger patients. In several other reports (Sr No. 4,6,8,10-13,15), both the younger and the older patient groups were studied and comparative features were assessed. The size of the younger group was mostly ~50, was ~200 in two reports (Sr No 3 and 4) and was 1025 in the work of Dozois et. al. (2005) (Sr no 14), the largest cohort of young CRC patients. In reports that included older and younger patients, older patients were much larger in number (Sr No 4,8-10). In all studies that were on younger patients alone, a high incidence of advanced stage (C+D: >70%)was reported. In studies that included both groups, the frequency of advanced disease in the young was as high or higher (Sr No. 4,11,15). In all of them, advanced disease stages were found to be more frequent in the young than in the old. In studies on younger patients alone, a significant proportion of patients had aggressive lesions, namely mucinous, poorly differentiated tumours with infiltrating leading edge. The frequency of aggressive tumour biology varied from one study to another but remained significant in all of them. In the comparative studies (Sr No. 6,8,10-12,15), the younger patients showed a higher frequency of aggressive tumour biology. Only one report (Sr No. 13) concluded that grade was not related to age.

Colorectal Carcinoma in the Young 19

49 patients all < 40 yrs: 67% in 31-40 yrs, 2 in their teens

Three groups <40, 41-50, 51-55 years: ~1%, 6%, 6% of 1298 patients

237 patients; 46 below 40yrs

> 61/2384 below 40 years

Table 2. Summary of references in the literature on stage and tumour characteristics from

Two of the reports from Europe listed in Table 2 (Sr.No.2,7) are entirely on young patients. One of these ( Sr No. 2) has the largest study group of young onset patients (~250), while the other reports have ~100-150 (Sr Nos. 5,6,8) or less ~50 (Sr No. 1,4,7,10)young patients. The report, Sr No. 3 is on a much smaller population of 11 patients. A significant frequency of more advanced (C+D) tumour in the young (50-60%) was reported in several studies (Sr No. 2,3,5-7). This frequency was larger (76%) in a study from Turkey (Sr No. 9). Comparative assessment showed a higher frequency of advanced stages in the young as compared to that in the older patients (Sr No. 3-6). Significant frequency (~ ≤50%) of high grade tumours were reported in the young in several publications (Sr Nos. 3,5,7,9,10). Higher frequency of high grade tumours in the young as compared to the older group were cited in several other papers (Sr No. 1,3,5,7). Three studies (Sr No. 1,8,10) however, reported no difference in disease stage and one report (Sr No. 8) found no difference in tumour grade, between the younger and older patient groups. A significant occurrence of premalignant conditions in the young was reported in

**Age profile Disease stage Tumour** 

Among all patients: 60% Dukes' C&D. Among patients at risk (family history /predisposing factor): 56% Dukes' C

Stage did not differ

76%of the young: Dukes C&D

Distribution of stage not significantly different in younger and older group

**characteristics** 

Among all patients: 59% moderately & 22% poorly differentiated . Among patients at risk: 53% moderately & 20% poorly differentiated

Histological features did not differ

48% tumours are poorly differentiated or mucinous in young.

35%tumours in the young are poorly differentiated

**Country Sr** 

**No.** 

UK 7 Leff

Italy 8 Fante

Turkey 9 Yilmazlar

Australia 10 Turkiewitz

only one paper (Sr No. 2).

**Reference, Period of study** 

1982-1992

1984-1992

1986-1993

1971-1999

Europe (inclusive of UK & Turkey) and Australia


1037 patients; 3% <40yrs

249 patients,

Two groups: 11 young <40yrs; 45 old > 80 yrs.

48/1061 patients are <40 years (21-39 years)

2283 patients with rectal cancer <70 years, <45 yrs: 132, 45-49 yrs: 153 50-69 yrs:1998

181 patients, 45 of them < 50 yrs

**Age profile Disease stage Tumour** 

No significant difference in stage between <40 and >40 Yrs group.

Dukes' C 54. 5% in the young and 44. 4% in the elderly group.

Dukes' A same proportion in young and old, Dukes' B fewer, Dukes' C more in young

Dukes' C&D : under 45: 73/132(~55%) 45-69yrs: 998/2152(~46%)

Dukes' C &D 54% in < 50 yrs group 46% in 51-70 yrs 36% in > 70 yrs

<40yrs 53% Dukes' C and D.

**characteristics** 

Greater frequency of mucinous and poorly differentiated carcinoma in the young

Premalignant condition more common in young

Undifferentiated tumour: 36. 3% of the young and 8. 8% of elderly.

Higher frequency of poorly differentiated tumours (27 vs. 15%) & N-2 stage (37 vs. 15%) with distant metastases (38 vs. 20%); 56% of under-40 years: developed metastases (20-26% of older group) after tumour resection

**Country Sr** 

**No.** 

France 1 Adloff

Finland 2 Jarvinen

Greece 3 Frizis

Sweden 4 Ohman

Norway

**Reference, Period of study** 

1973-1980

1970-1979

1994-2003

1950-1979

<sup>5</sup>Endreseth 1993-1999

<sup>6</sup>Berg

2010


Table 2. Summary of references in the literature on stage and tumour characteristics from Europe (inclusive of UK & Turkey) and Australia

Two of the reports from Europe listed in Table 2 (Sr.No.2,7) are entirely on young patients. One of these ( Sr No. 2) has the largest study group of young onset patients (~250), while the other reports have ~100-150 (Sr Nos. 5,6,8) or less ~50 (Sr No. 1,4,7,10)young patients. The report, Sr No. 3 is on a much smaller population of 11 patients. A significant frequency of more advanced (C+D) tumour in the young (50-60%) was reported in several studies (Sr No. 2,3,5-7). This frequency was larger (76%) in a study from Turkey (Sr No. 9). Comparative assessment showed a higher frequency of advanced stages in the young as compared to that in the older patients (Sr No. 3-6). Significant frequency (~ ≤50%) of high grade tumours were reported in the young in several publications (Sr Nos. 3,5,7,9,10). Higher frequency of high grade tumours in the young as compared to the older group were cited in several other papers (Sr No. 1,3,5,7). Three studies (Sr No. 1,8,10) however, reported no difference in disease stage and one report (Sr No. 8) found no difference in tumour grade, between the younger and older patient groups. A significant occurrence of premalignant conditions in the young was reported in only one paper (Sr No. 2).

Colorectal Carcinoma in the Young 21

5436 patients 7% <40 years

287 patients 35. 6% < 40 yrs

305 patients 40% < 40 yrs

91 patients 28. 6% < 40 yrs

305 patients 19. 7% < 40 yrs

**Age profile Disease Stage Tumour** 

Dukes' stage improves with age (A & B 31% < 30 years, 49% > 80 years)

Advanced T stage (T 0-2: 18. 9% T -3: 62. 3% T-4: 19. 7% vs. 34. 5%, 56. 0%, 9. 5%) and N-stage (N 0: 31. 1%, N1: 41%, N2: 27. 8% vs. 53. 9%, 26. 7%, 17. 2% )

60% presented in Dukes' stage III & IV

92. 3% present in Dukes' stage III-IV vs. 61. 5% in older patients

No significant difference in Dukes' stage with older group

**Characteristics** 

Poorly differentiated tumours tended to decrease with age, 16. 9% < 30 years. 6. 2% > 80 years. Similar trend in Mucin producing characteristics (36% vs7. 5%)

Poorly differentiated and / or mucinous or signet cell carcinoma (52% vs. 20. 5%)

Mucinous tumour 80% Poor differentiation 50%

Significantly higher poorly differentiated and mucinous carcinoma in the young.

Significant presence of mucinous (13. 3%) or signet ring type (5%) tumours.

**Country Sr** 

India

**No**

Taiwan 7 Chiang

**Reference, Period of Study** 

7 year period

<sup>8</sup>Nath 2003-2007

<sup>9</sup>Gupta 2000-2008

2002a

15 yr period

Nepal 10 Singh

SriLanka 11 de Silva


403 patients in two age groups, <40yrs and >40yrs

39 patients <30yrs, mean age 25yrs

> 523 Asian cohorts 19-50 years Of them <40 yrs:134; >40yrs:389

21 patients

406 patients, 203 in < 50 years

<50 years 90; 190 > 50 years

**Age profile Disease Stage Tumour** 

Older group: 53. 2% in stage II; younger group: 45%in stage III.

Advanced disease stage in 70% patients.

63% Advanced stage (III-IV) <40yr group: 89/134;66% >40yr group: 245/389;63%

<30yrs Extensive disease Mucinous

More advanced stage III-IV at diagnosis (56 vs. 41%) higher rate of N-2 disease (29 vs. 16%)

40% Advanced stage (III-IV) <40yrs:47/90;52% >40yrs:61/190;32% **Characteristics** 

Poorly differentiated tumours found in larger proportion in younger patients (22% vs. 5. 9%)

Mucinous histology: 18%; differentiation: moderate 61%, poor 36%

Predominantly poorly differentiated: (30% in <40 years 12% in > 40 years) mucinous, signet ring cell histological subtypes (16% vs. 9%).

histology

No difference in other features

Mucinous tumour in 11% in early onset group, 7% in late onset group

**Country Sr** 

**ASIA**

Singapore

Israel

**No**

Iran 1 Fazeli

**Reference, Period of Study** 

1995-2001

1989-2001

<sup>2</sup>Kam

<sup>3</sup>Chew 1997-2005

1990-94

Shemesh-Bar, 1997-2007

<sup>6</sup>Neufeld 1999-2005

Malaysia 4 Shahruddin

5


Colorectal Carcinoma in the Young 23

differentiation have also been linked to molecular genetic differences. Recent molecular biology studies have shown characteristic features of mucinous carcinoma, e. g., lower expression of p53, more frequent DNA replication errors expressed as microsatellite instability and specific codon 12 K-ras mutations and, when ploidy has been determined, a

**Tumour subsite:** The issue of subsite location is important in screening strategies and in choice of treatment protocols. In the literature (e. g. , Breivik et. al. 1997) preference for subsite location has been associated with molecular genetic roots of CRC. Molecular genetic findings classify CRC into two groups. The first class of tumours show microsatellite instability (MSI), occur more frequently in the right colon, have diploid DNA, behave indolently, of which Hereditary Non polyposis Colorectal Cancer Syndrome (HNPCC) is an example. The larger incidence of proximal colon cancer in patients with HNPCC syndrome highlights the importance of genetics in preference for subsite location in colon cancer. In the other group belong tumours which tend to be left sided, show aneuploid DNA, behave aggressively, of which Familial Adenomatous Polyposis (FAP) is an example. Each group

Breivik et. al. (1997)in a study of 282 patients from 7 hospitals in Norway in the period 1987- 9 concluded that proximal and distal CRC evolve by different genetic pathways and that these pathways are influenced by sex-related factors. Their results, analyzed by statistical models, pointed to hormonal mechanisms with important clinical implications. They found that presence of TP 53 mutations was dependent on tumour location only, with a positive association to cancers occurring distally (p=0. 002). Microsatellite instability was found

Stigliano et. al. (2008) compared a cohort of 40 HNPCC cases with 573 sporadic CRC cases in the period 1970-1993. Median age of diagnosis was 46. 8 years in HNPCC cases and 61 years in sporadic CRC cases. 85% had right sided lesion in HNPCC group as opposed to 57% in

Slattery et. al. (1996) studied age, sex and tumour sub-site distribution in 1709 CRC patients from three geographic areas in the USA. Approximately 50% of CRC in men and greater than 50% of CRC in women were in the proximal segment of the colon. Men who were diagnosed prior to age 50 and both men and women diagnosed at age 70 or older had predominantly proximal cancers. People with proximal cancers and those diagnosed prior to age 50 were likely to have more advanced disease. In general, both men and women had more proximal

Ionov et. al. (1993) showed that 12% of CRC patients carried ubiquitous somatic deletions in poly (dA. dT) sequences and other simple repeats. Tumours with these mutations showed distinctive genotypic and phenotypic features. Patients with these deletions showed a predominance of right sided tumours while those without deletions had a predominance of

Thibodeau et. al. (1993)studied the association of microsatellite alterations with preference for tumour subsite. All four sites of alteration studied showed a dramatic change in preference from distal to proximal colon in the mutated form (typical values:

Fancher et. al. (2011) studied 45 young patients, 20 males and 25 females,mean age 43. 6 years, in the USA and found preference for left sided lesions in females (16/8)and a preference for right sided lesions (12/10)in men (p=0. 35; small sample size);right sided

cancers with advancing age, which were associated with more advanced disease.

higher index if diploidy was found than for non-mucinous carcinoma (Negri 2005).

has its own characteristic gene mutations (Lynch and de la Chapelle, 1999).

almost exclusively in proximal colon cancers.

proximal/distal; (26,49), (11,1 in the mutated form)).

cancers had a higher stage at presentation.

sporadic cancer group.

left sided lesions.


\*Soliman et. al. (1997)

Table 3. Summary of references in the literature on stage and tumour characteristics from Asia and Africa.

Reports from Asia and Africa are listed in Table 3. Features of only the younger patients were assessed in four reports (Sr. No. 2-4,9). The younger groups were larger in several studies (523:Sr No. 3; 203:Sr No 5; 370: Sr No. 7and 576 : Sr No. 12) from Asia and Africa as compared to ones from the USA (Table 1) and Europe (Table 2). Higher incidence of CRC in the young in Asia and Africa was found to be consistent with these figures. In two studies (Sr No. 11,12) the disease in the young was assessed as less advanced at presentation and less aggressive. In one report (Sr No. 5),a more advanced disease stage was noted but no difference in tumour grade was found. A more advanced disease and tumour grade was reported in the young as compared to the older patients (which is usually the case in Table 1 & 2) in 5 of 12 reports (Sr No. 1,6-8,10). In a report by Chew et. al (2009, Sr No. 3) the same conclusion was reached; 'older' patients were however in the age group 40-50years. The frequency of advanced disease and high tumour grade in the young in these reports were similar to that in reports restricted to only the young patients (Sr No. 2,4,9).

Irrespective of the country, the size of the study group, time span and the year of study, the dominant result is the same. Young CRC patients present at a more advanced clinical stage, the tumours are mucinous and poorly differentiated, more so in comparison with the older patient group. The features in India and neighbouring Nepal and Sri Lanka are the same as in the rest of the world. We have noticed some difference in disease pattern in Asia and Africa as compared to the West in our discussions of the data in Tables 1-3. The issue of ethnic differences in determining the difference in disease characteristics is important. This issue, without specific reference to the disease in the young, received attention in several papers, e. g., Isbister (1992; New Zealand and Saudi Arabia), Soliman et. al. (2001; Egypt and the West), Fireman et. al. (2001;Arab and Jewish neighbours in Israel),Qing et. al. (2003;USA and China), Sung et. al. (2005;Asia and the West),Goh et. al. (2005;Asian patients of different races in Malaysia) and Fairley et. al. (2006;Blacks,Asians/Pacific Islanders and Whites).

The advanced stage at presentation of many colorectal cancers in young patients is not just a result of a delay in diagnosis. It may also be that the cancer in younger patients is more virulent by nature. This feature is rooted in subtleties of genetic differences. More aggressive' tumour characteristics, as evidenced by its mucinous nature and poor

1608 patients; 35. 6% <40 yrs

Table 3. Summary of references in the literature on stage and tumour characteristics from

Reports from Asia and Africa are listed in Table 3. Features of only the younger patients were assessed in four reports (Sr. No. 2-4,9). The younger groups were larger in several studies (523:Sr No. 3; 203:Sr No 5; 370: Sr No. 7and 576 : Sr No. 12) from Asia and Africa as compared to ones from the USA (Table 1) and Europe (Table 2). Higher incidence of CRC in the young in Asia and Africa was found to be consistent with these figures. In two studies (Sr No. 11,12) the disease in the young was assessed as less advanced at presentation and less aggressive. In one report (Sr No. 5),a more advanced disease stage was noted but no difference in tumour grade was found. A more advanced disease and tumour grade was reported in the young as compared to the older patients (which is usually the case in Table 1 & 2) in 5 of 12 reports (Sr No. 1,6-8,10). In a report by Chew et. al (2009, Sr No. 3) the same conclusion was reached; 'older' patients were however in the age group 40-50years. The frequency of advanced disease and high tumour grade in the young in these reports were similar to that in reports restricted to only the young patients

Irrespective of the country, the size of the study group, time span and the year of study, the dominant result is the same. Young CRC patients present at a more advanced clinical stage, the tumours are mucinous and poorly differentiated, more so in comparison with the older patient group. The features in India and neighbouring Nepal and Sri Lanka are the same as in the rest of the world. We have noticed some difference in disease pattern in Asia and Africa as compared to the West in our discussions of the data in Tables 1-3. The issue of ethnic differences in determining the difference in disease characteristics is important. This issue, without specific reference to the disease in the young, received attention in several papers, e. g., Isbister (1992; New Zealand and Saudi Arabia), Soliman et. al. (2001; Egypt and the West), Fireman et. al. (2001;Arab and Jewish neighbours in Israel),Qing et. al. (2003;USA and China), Sung et. al. (2005;Asia and the West),Goh et. al. (2005;Asian patients of different races in Malaysia) and Fairley et. al.

The advanced stage at presentation of many colorectal cancers in young patients is not just a result of a delay in diagnosis. It may also be that the cancer in younger patients is more virulent by nature. This feature is rooted in subtleties of genetic differences. More aggressive' tumour characteristics, as evidenced by its mucinous nature and poor

**Age profile Disease Stage Tumour** 

Dukes' stage is worse in > 40 years group (72% vs. 57%)

**Characteristics** 

Tumour grade comparable in two groups; mucin producing tumours: 31% in younger group, 14% in older group

**Country Sr** 

**AFRICA** 

\*Soliman et. al. (1997)

Asia and Africa.

(Sr No. 2,4,9).

**No**

Egypt 12 Soliman\*

**Reference, Period of Study** 

1982-1996

(2006;Blacks,Asians/Pacific Islanders and Whites).

differentiation have also been linked to molecular genetic differences. Recent molecular biology studies have shown characteristic features of mucinous carcinoma, e. g., lower expression of p53, more frequent DNA replication errors expressed as microsatellite instability and specific codon 12 K-ras mutations and, when ploidy has been determined, a higher index if diploidy was found than for non-mucinous carcinoma (Negri 2005).

**Tumour subsite:** The issue of subsite location is important in screening strategies and in choice of treatment protocols. In the literature (e. g. , Breivik et. al. 1997) preference for subsite location has been associated with molecular genetic roots of CRC. Molecular genetic findings classify CRC into two groups. The first class of tumours show microsatellite instability (MSI), occur more frequently in the right colon, have diploid DNA, behave indolently, of which Hereditary Non polyposis Colorectal Cancer Syndrome (HNPCC) is an example. The larger incidence of proximal colon cancer in patients with HNPCC syndrome highlights the importance of genetics in preference for subsite location in colon cancer. In the other group belong tumours which tend to be left sided, show aneuploid DNA, behave aggressively, of which Familial Adenomatous Polyposis (FAP) is an example. Each group has its own characteristic gene mutations (Lynch and de la Chapelle, 1999).

Breivik et. al. (1997)in a study of 282 patients from 7 hospitals in Norway in the period 1987- 9 concluded that proximal and distal CRC evolve by different genetic pathways and that these pathways are influenced by sex-related factors. Their results, analyzed by statistical models, pointed to hormonal mechanisms with important clinical implications. They found that presence of TP 53 mutations was dependent on tumour location only, with a positive association to cancers occurring distally (p=0. 002). Microsatellite instability was found almost exclusively in proximal colon cancers.

Stigliano et. al. (2008) compared a cohort of 40 HNPCC cases with 573 sporadic CRC cases in the period 1970-1993. Median age of diagnosis was 46. 8 years in HNPCC cases and 61 years in sporadic CRC cases. 85% had right sided lesion in HNPCC group as opposed to 57% in sporadic cancer group.

Slattery et. al. (1996) studied age, sex and tumour sub-site distribution in 1709 CRC patients from three geographic areas in the USA. Approximately 50% of CRC in men and greater than 50% of CRC in women were in the proximal segment of the colon. Men who were diagnosed prior to age 50 and both men and women diagnosed at age 70 or older had predominantly proximal cancers. People with proximal cancers and those diagnosed prior to age 50 were likely to have more advanced disease. In general, both men and women had more proximal cancers with advancing age, which were associated with more advanced disease.

Ionov et. al. (1993) showed that 12% of CRC patients carried ubiquitous somatic deletions in poly (dA. dT) sequences and other simple repeats. Tumours with these mutations showed distinctive genotypic and phenotypic features. Patients with these deletions showed a predominance of right sided tumours while those without deletions had a predominance of left sided lesions.

Thibodeau et. al. (1993)studied the association of microsatellite alterations with preference for tumour subsite. All four sites of alteration studied showed a dramatic change in preference from distal to proximal colon in the mutated form (typical values: proximal/distal; (26,49), (11,1 in the mutated form)).

Fancher et. al. (2011) studied 45 young patients, 20 males and 25 females,mean age 43. 6 years, in the USA and found preference for left sided lesions in females (16/8)and a preference for right sided lesions (12/10)in men (p=0. 35; small sample size);right sided cancers had a higher stage at presentation.

Colorectal Carcinoma in the Young 25

constitute 72%.

3 Singh, Nepal Rectum: most frequent site of tumor

and the old.

tumour:20%;

mean age 47 years

9 Shemesh-Bar, Israel Higher proportion of left side tumour in the young (82% vs. 71%)

no effect of age.

general population

11 Malekzadeh, Iran Predominantly right sided tumour,

> 80 years

vs. > 40 years group,

(~65%) in both age groups

older patient group.

69. 7% distal tumours in < 40 yr group (rectal 57. 9%,left colon 11. 8%).

(76. 9% vs. 36. 9% in older age group)

Rectosigmoid region:most common (29%)' Left colon 19%,Splenic flexure 4%,Transverse

No significant difference between <and >

46% rectal and rectosigmoid; right-sided

66. 7%rectal lesions; younger patients;

(61. 4% <50years, 36% <40 yrs,16% <30 yrs)

Predominantly left sided tumour (~80%)in <40

Rectosigmoid most common site in general population (553 patients),70. 7%;also in 32,<29

Subsite distribution nearly independent

No change in subsite preference in < 40 years

larger proportion of distal tumours

(~90%)of left sided lesion in both age groups.

No significant difference in tumour distribution between the young

colon 9%,Hepatic flexure 4%, Cecum 24% ;all patients<30yrs

65 years group; predominance

patient group, all young <30yrs

years and 41-50 years age group;

years younger group: 84. 4%

of age group (< & > 40 years), distal ~ 80% in both groups. 14 Chiang, Taiwan No change in subsite preference from < 30 years to

Of patients with colon cancer proximal tumours

Rectum: commonest (83%) site of the lesion in young patients (21-30 yrs). No comparison with

**No. Author, Country Sub-site** 

Asia & Africa

**Sr.** 

6

1 Gupta, India

2 Singh, South Asia\*

4 de Silva, Sri Lanka

<sup>5</sup>Shahruddin,

7 Kam, Singapore

8 Ashenafi, Ethiopia

<sup>10</sup>Chew, Singapore (Asian patients)

12 Ibrahim, Lebanon

13 Fazeli , Iran

15 Soliman\*\*, Egypt

Malaysia

Goh, Asian patients of different races in Malaysia

Kaw et. al. (2002) studied 1277 Filipino patients of whom 218 (17%)were <40 years, a mean age of 31. 3 years. Cancers of the right colon were noted to be more common in females (55%)and rectal tumours were seen more frequently in males (55%;p=0. 014),but when analysed in relation to age, right colon cancers were actually more common in men <40 years of age (p=0. 013);the incidence in women was higher only above the age of 50 years. The proportion of CRCs located on the right side was 28% for <40 years patients and 20% for the 40+ group. On the other hand, left colon cancers were seen in 30% of the older age group compared with 18% in the younger population (p=0. 001). For rectal cancer, there was no significant difference in proportion between the young and the old (p=0. 414).

Elsaleh et. al. (2000) in an older patient group (mean age 66. 7 ±12. 9 years) in Australia reported that MSI positive tumours were slightly more frequent in women than in men (10 vs 7%). Right sided tumours were more frequently MSI positive than left sided tumours (20 vs 1%). Men with right sided tumours benefited from chemotherapy (37 vs 12%) but men with left sided tumours did not.

Mahdavinia et. al. (2005), Fazeli et. al. (2007) and Malekzadeh et. al (2009) found that in Iranian patients with positive family history of CRC, the most frequently affected site of colon was the right side. Malekzadeh et. al. (2009) found that MSI was more frequent in early-onset patients and in proximal tumours. They reported that proximal and distal tumours harbor different p53 mutational spectra;distal CRCs showed a higher frequency of G to A transitions at CpG whereas G to A transitions at non-CpGs were more frequent in proximal tumours. Fazeli et. al. (2007) found that 62. 5% of patients with proximal colon tumours were males.

Nelson et. al. (1997) and Saltzstein et. al. (1997) showed that there was an increase in the relative proportion of proximal colon cancers with increasing age **'**a shift to the right**'**. Thus with increasing age, full length colonoscopy will be a better screening tool. The exact age at which the shift occurs will vary with gender and ethnicity. There is a predominance of African-Americans amongst those at risk for proximal colon carcinoma and predominance of white males amongst those at risk for distal CRC.

Goh et. al. (2005) in a study of different races in Malaysia observed that demographic differences between Asia and the West may exist. No difference in anatomic distribution was found in Malay, Chinese and Indian races. They noted that in general CRC tends to be located distally in areas with a lower incidence of disease (parts of Asia) and migrated proximally with increasing incidence, as in Japan or Korea. They suggested that this may be related to a decrease in rectal cancer and an increasing proportion of elderly patients in the population. Young patients had a higher probability of having distal lesions as compared to the older patients.

Qing et. al. (2003) in a comparative study on Chinese and American patients, noted that the proportion of left sided lesions in Oriental patients (74%) was significantly higher than that in Whites (63. 7%) and that rectal cancers were significantly more common among Orientals (p<0. 001).

O'Connell et. al. (2004a) in their review quoted average values of subsite location in <40 years young patients as follows: ascending 22%, transverse 11%, descending colon 13%, rectum and sigmoid (including rectosigmoid junction) 54%,a dominance of left sided tumour in the young.

We summarize reports on preference for tumour sub-site from different countries in Table 4. Some of these are cited in Table 1-3 where patient groups are detailed. The others are detailed in Table 4.

Kaw et. al. (2002) studied 1277 Filipino patients of whom 218 (17%)were <40 years, a mean age of 31. 3 years. Cancers of the right colon were noted to be more common in females (55%)and rectal tumours were seen more frequently in males (55%;p=0. 014),but when analysed in relation to age, right colon cancers were actually more common in men <40 years of age (p=0. 013);the incidence in women was higher only above the age of 50 years. The proportion of CRCs located on the right side was 28% for <40 years patients and 20% for the 40+ group. On the other hand, left colon cancers were seen in 30% of the older age group compared with 18% in the younger population (p=0. 001). For rectal cancer, there was

Elsaleh et. al. (2000) in an older patient group (mean age 66. 7 ±12. 9 years) in Australia reported that MSI positive tumours were slightly more frequent in women than in men (10 vs 7%). Right sided tumours were more frequently MSI positive than left sided tumours (20 vs 1%). Men with right sided tumours benefited from chemotherapy (37 vs 12%) but men

Mahdavinia et. al. (2005), Fazeli et. al. (2007) and Malekzadeh et. al (2009) found that in Iranian patients with positive family history of CRC, the most frequently affected site of colon was the right side. Malekzadeh et. al. (2009) found that MSI was more frequent in early-onset patients and in proximal tumours. They reported that proximal and distal tumours harbor different p53 mutational spectra;distal CRCs showed a higher frequency of G to A transitions at CpG whereas G to A transitions at non-CpGs were more frequent in proximal tumours. Fazeli et. al. (2007) found that 62. 5% of patients with proximal colon

Nelson et. al. (1997) and Saltzstein et. al. (1997) showed that there was an increase in the relative proportion of proximal colon cancers with increasing age **'**a shift to the right**'**. Thus with increasing age, full length colonoscopy will be a better screening tool. The exact age at which the shift occurs will vary with gender and ethnicity. There is a predominance of African-Americans amongst those at risk for proximal colon carcinoma and predominance

Goh et. al. (2005) in a study of different races in Malaysia observed that demographic differences between Asia and the West may exist. No difference in anatomic distribution was found in Malay, Chinese and Indian races. They noted that in general CRC tends to be located distally in areas with a lower incidence of disease (parts of Asia) and migrated proximally with increasing incidence, as in Japan or Korea. They suggested that this may be related to a decrease in rectal cancer and an increasing proportion of elderly patients in the population. Young patients had a higher probability of having distal lesions as compared to

Qing et. al. (2003) in a comparative study on Chinese and American patients, noted that the proportion of left sided lesions in Oriental patients (74%) was significantly higher than that in Whites (63. 7%) and that rectal cancers were significantly more common among Orientals

O'Connell et. al. (2004a) in their review quoted average values of subsite location in <40 years young patients as follows: ascending 22%, transverse 11%, descending colon 13%, rectum and sigmoid (including rectosigmoid junction) 54%,a dominance of left sided

We summarize reports on preference for tumour sub-site from different countries in Table 4. Some of these are cited in Table 1-3 where patient groups are detailed. The others are

no significant difference in proportion between the young and the old (p=0. 414).

with left sided tumours did not.

of white males amongst those at risk for distal CRC.

tumours were males.

the older patients.

tumour in the young.

detailed in Table 4.

(p<0. 001).


Colorectal Carcinoma in the Young 27

In some of these papers, (Sr. No. 1, 2, 5, 7, 8, 10, 17, 23-25, 27), a preference for distal lesions in the young patients were cited, but were not compared with the older patient groups. In some others (Sr. No. 3, 9, 12, 16, 18-21) this comparison was made and a change in preference for tumour sub-site with increasing age was noticed. Shemesh-Bar et.al. (2010, Sr. No. 9) and Ibrahim et. al. (1986, Sr. No. 12) found that although left sided lesions formed the majority of tumours, their proportion decreased in the older group. Singh et. al (2002a, Sr. No. 3) and Fairley et. al. (2006, Sr. No. 20)found that the proportion of rectal cancers decreased with increasing age. In several reports preference for right sided lesions showed an increase with increasing age (Sr. No. 16, 18, 20-21). Slattery et. al. (1996, Sr. No. 19)reported an increase in proportion of proximal tumours with increasing age for women, exceeding 50% (62. 3%), only in the age group 70-79 years. Amongst men, proportion of proximal tumours exceeded 50%in the <50 yr groups (62. 5%, 30-39 yrs; 51. 1%, 40-49 yrs), falls below 50%in the 50-59 and 60-69 yrs groups and then rises again to 54% in the 70-79 yrs group. A decrease in proximal tumours with increasing age was reported by Karsten et. al. (2008, Sr. No. 22) and Fante et. al. (1997, Sr. No. 28). Both studies reported a dominance of distal tumours in different age groups (two in Sr. No. 22, three in Sr. No. 28), but proximal tumours decreased with increasing age. In a few papers (Sr. No 4, 6, 10, 13-15, 26) sub-site preference was found not to depend on age. Fazeli et.al. (2007, Sr. No. 13) reported that~ 80% of the tumours were distal in the young (<40 years) and also in the older age group. In these reports which did not find any effect of age on subsite preference, distal tumours were >50% in the young and in the older group. A preference for proximal tumours in a population of colon cancer patients were reported in several papers (Sr. No. 11 and Mahdavinia et. al. (2005) in general population of colon cancer patients from Iran, where incidence is lower than in the West and in Sr. No. 1 in young colon cancer patients < 40 years in India). Cozart et. al. (1993, Sr. No. 17) found tumour sub-site preference for left colon (24/12) in a small population of colon cancer patients; Dozois et. al. (2005, Sr. No 25) found the same preference in a much larger (1025 patients) young (<50yrs)population. We cite several prospective reports on change in relative preference of tumour sub-site over a long time period. Fazeli et. al. (2007, Sr. No. 13) reported that the nearly equal preference for distal tumours (~80%) in the <40 years and in the >40 years group in Iran, remained unchanged for two decades (1970-80, 1990-2000). In contrast, it was reported in a study on patients from New Zealand, in the period 1974-83 (Jass 1991), that the incidence of right colon cancers remained stable in younger patients (<50 years), that in older patients showed an increase and a marked reduction in left colon and rectal colon cancer in <50 years group was observed. An increase in proximal CRC relative to distal tumours was reported in

another retrospective study in the period 1940-79 in the US (Beart et. al. 1983).

The problem arising from inability to detect cancer early because of hospital infrastructure and relative lack of awareness of the disease may not be the only problem peculiar to the developing world in Asia and Africa. Sung et. al. (2005) pointed out that non-polypoidal (flat or depressed) lesions and colorectal neoplasm arising without preceding adenoma (de novo cancer) seemed to be more common in Asian than in other populations. Although most cases of colorectal cancer are thought to arise from a sequence of adenoma to carcinoma, evidence from Asia, in particular Japan suggests another mechanism. Clinicopathological studies have shown that there are two groups of colorectal cancer,

**4. de novo CRC in Asia** 


\*Singh et. al. 1984; \*\*Soliman et. al. (1997)

Table 4. Tumour Sub-site in the young in different countries

Increase in primary lesions in the right colon with

(12 right colon, 24 left colon,11 rectum) and left colon amongst colon cancer patients; study group comprises of only young patients<30yrs. No

Proximal cancers more frequent (>50%) in men<50

Rectal cancers more frequent in <50yrs group (37% vs. 26%); proximal colon cancer more frequent in >50 age group (42. 6% vs. 32. 1%),remaining <50%in both groups.

Older patients: more transverse/right sided lesions (p=0. 003). 138 patients;mean age of patients with different sites of tumour: Right colon 72 yrs,left colon 66. 1 yrs,

compared to 21% in older group, p=0. 004.

Tumours evenly distributed in colon and rectum

Rectum and sigmoid colon most frequent sites (54%) in the young <40 yrs patient group.

Predominantly rectum (49. 1%) or left colon (29. 1%) than proximal colon (21. 9%).

No comparison with older patient group.

Colon: Right 21%,Transverse 21%, Left 14% Sigmoid & Rectum 44% in the <40 yrs group; these figures are 34%,4%,8%,54% respectively in the

Only 12% right-sided colon cancer,<40 yrs patients, no comparison with older group.

whole patient group <40 – 55 years.

14% in 51-55 years group.

Majority of tumours in left colon and rectum in the

Right colon: 37% in <40 years, 18% in 41-50 years,

increasing age at diagnosis; <40 yrs group compared with general population.

Significant shift to right sided lesion with

years and in both men and women >70

Dominance of left sided lesions

comparison with older group.

increasing age;<50 vs. >50yrs.

years(details in text).

rectum 61. 6 yrs 22 Karsten, USA Right sided lesion more frequent (44%)in young

(under-40 group).

older group.

Older group not compared.

All young patients <50yrs.

**No. Author, Country Sub-site**

U. S. A.

Europe

**Sr.** 

16 Bedekian, USA

17 Cozart, USA

<sup>18</sup>Nelson, USA & Saltzstein, USA

19 Slattery, USA

20 Fairley, USA

21 Lichtman, USA

23 Minardi, USA

24 O'Connella

25 Dozois, USA

26 Behbehani, USA

27 Leff, UK

28 Fante, Italy

Table 4. Tumour Sub-site in the young in different countries

\*Singh et. al. 1984; \*\*Soliman et. al. (1997)

(International Review)

In some of these papers, (Sr. No. 1, 2, 5, 7, 8, 10, 17, 23-25, 27), a preference for distal lesions in the young patients were cited, but were not compared with the older patient groups. In some others (Sr. No. 3, 9, 12, 16, 18-21) this comparison was made and a change in preference for tumour sub-site with increasing age was noticed. Shemesh-Bar et.al. (2010, Sr. No. 9) and Ibrahim et. al. (1986, Sr. No. 12) found that although left sided lesions formed the majority of tumours, their proportion decreased in the older group. Singh et. al (2002a, Sr. No. 3) and Fairley et. al. (2006, Sr. No. 20)found that the proportion of rectal cancers decreased with increasing age. In several reports preference for right sided lesions showed an increase with increasing age (Sr. No. 16, 18, 20-21). Slattery et. al. (1996, Sr. No. 19)reported an increase in proportion of proximal tumours with increasing age for women, exceeding 50% (62. 3%), only in the age group 70-79 years. Amongst men, proportion of proximal tumours exceeded 50%in the <50 yr groups (62. 5%, 30-39 yrs; 51. 1%, 40-49 yrs), falls below 50%in the 50-59 and 60-69 yrs groups and then rises again to 54% in the 70-79 yrs group. A decrease in proximal tumours with increasing age was reported by Karsten et. al. (2008, Sr. No. 22) and Fante et. al. (1997, Sr. No. 28). Both studies reported a dominance of distal tumours in different age groups (two in Sr. No. 22, three in Sr. No. 28), but proximal tumours decreased with increasing age. In a few papers (Sr. No 4, 6, 10, 13-15, 26) sub-site preference was found not to depend on age. Fazeli et.al. (2007, Sr. No. 13) reported that~ 80% of the tumours were distal in the young (<40 years) and also in the older age group. In these reports which did not find any effect of age on subsite preference, distal tumours were >50% in the young and in the older group. A preference for proximal tumours in a population of colon cancer patients were reported in several papers (Sr. No. 11 and Mahdavinia et. al. (2005) in general population of colon cancer patients from Iran, where incidence is lower than in the West and in Sr. No. 1 in young colon cancer patients < 40 years in India). Cozart et. al. (1993, Sr. No. 17) found tumour sub-site preference for left colon (24/12) in a small population of colon cancer patients; Dozois et. al. (2005, Sr. No 25) found the same preference in a much larger (1025 patients) young (<50yrs)population. We cite several prospective reports on change in relative preference of tumour sub-site over a long time period. Fazeli et. al. (2007, Sr. No. 13) reported that the nearly equal preference for distal tumours (~80%) in the <40 years and in the >40 years group in Iran, remained unchanged for two decades (1970-80, 1990-2000). In contrast, it was reported in a study on patients from New Zealand, in the period 1974-83 (Jass 1991), that the incidence of right colon cancers remained stable in younger patients (<50 years), that in older patients showed an increase and a marked reduction in left colon and rectal colon cancer in <50 years group was observed. An increase in proximal CRC relative to distal tumours was reported in another retrospective study in the period 1940-79 in the US (Beart et. al. 1983).

#### **4. de novo CRC in Asia**

The problem arising from inability to detect cancer early because of hospital infrastructure and relative lack of awareness of the disease may not be the only problem peculiar to the developing world in Asia and Africa. Sung et. al. (2005) pointed out that non-polypoidal (flat or depressed) lesions and colorectal neoplasm arising without preceding adenoma (de novo cancer) seemed to be more common in Asian than in other populations. Although most cases of colorectal cancer are thought to arise from a sequence of adenoma to carcinoma, evidence from Asia, in particular Japan suggests another mechanism. Clinicopathological studies have shown that there are two groups of colorectal cancer,

Colorectal Carcinoma in the Young 29

concluded that major age related differences in the clinical and molecular features of CRC

Farrington et. al. (1998) pointed out that germ-line mutations in DNA mismatch-repair (MMR) genes impart a markedly elevated cancer risk, often presenting as autosomal dominant HNPCC. Not all gene carriers have a family history. Young probands with early onset CRC irrespective of family history were genetically tested and it was found that an appreciable proportion of young colon cancer probands carry a germline mutation in a

Losi et. al. (2005) evaluated clinical features and molecular pathways, chromosomal instability (CNI) and MSI in early onset CRC. Of 71 patients (<45 years), 14 showed both MSI and altered expressions of MMR proteins. In the 57 MSI -negative (-) lesions, altered expression of APC, β-catenin and p53 genes were found more frequently than in MSIpositive (+) tumours. 7/14 MSI (+) tumours were associated with clinical features of HNPCC and in all but one, constitutional mutations in MLH-1 and MSH-2 genes could be detected. The same mutations were found in other family members. Involvement of

Chan et. al. (1999)studied 59 Chinese patients <45 yrs and 58, >45 yrs in Hong Kong. The incidence of MSI-H varied statistically significantly with age, being observed in >60% of those <31 years at diagnosis and in <15% of those ≥46 years. More than 80% of Chinese CRC patients <31 years had germline mutations in MMR genes. In a novel case, mutation in

Ho et. al. (2000) in a study on 124 young (<50yrs) Hong Kong Chinese CRC patients concluded that MSI occurs in a significant proportion of the subjects. Young age at CRC diagnosis, proximal tumour location, increasing number of first degree relatives with CRC and a personal history of metachronous cancer were independent predictors of MSI status in the patient group. In patients <30 years, MSI tumours were more likely to be located in distal large bowel. In a proportion of patients with MSI tumours, germline mutation in the two MMR genes hMSH2 and hMLH1 was identified. The authors opined that this observation suggests a differential activity of the MMR pathway in colorectal carcinogenesis in different age groups. They observed that the inconsistency between MSI-H and a family

Liang et. al. (2003) studied 138 below-40 CRC patients and 339 patients who were 60+. They found a higher percentage of normal p53 expression (61. 1 vs. 46. 8%, p=0. 023) and high frequency microsatellite instability (MSI-H) ( 29. 4 vs. 6. 3% p<0. 001) in the young. The

Durno et. al. (2005, 2006) found evidence of MSI in 73% cancers from individuals in 9-24 years of age, 50% of whom had features of HNPCC. Other reports found MSI in 46% of

Sanchez et. al. (2009) performed a molecular classification of CRC based on microsattellite instability (MSI), CpG-island methylator phenotype (CIMP) and mutations in the K-ras and BRAF oncogenes. There were four classes, combinations of MSI-H and MSS with CIMP–H or CIMP (-). 69. 8% of tumours (391 subjects) were MSS-CIMP(-) and less likely to be poorly differentiated (p=0. 009). CIMP-H tumours were more common in older patients (p<0. 001). MSI-H/CIMP-H tumours had a high frequency of BRAF mutation and a low rate of K-ras mutation, the opposite was true for MSS-CIMP(-) tumours (p<0. 001). The four molecular phenotypes tended towards divergent survival. MSI-H cancers were associated with better

chromosomal instability was demonstrated in a majority of early onset CRC.

exist.

DNA MMR gene.

hMSH-6 was present but MSI was absent.

family history of the two groups was similar.

disease free survival.

history in the early onset patients deserves further attention.

under- 21 patients with only 1/3 having a clear family history.

polypoid and non-polypoid (superficial) tumours. The latter are flat lesions with a raised or depressed surface. Since these tumours are small (<1cm in diameter) and there are no adenomatous elements in their vicinity, they were proposed not to have originated from any precursor lesion and were termed de-novo carcinomas. These non-polypoid tumours are less likely to have K-ras mutations than are CRC arising from the adenoma-carcinoma sequence. Non-polypoid tumours of the colorectal regions tend to reach deeper layers of the intestinal wall in the early stage of the disease and with a higher degree of dysplasia. They are therefore more invasive than the polypoid adenomas (Sung et. al., 2005). Reports on de novo cancer have been published from Japan (Goto et. al 2004) and from Taiwan (Chen et. al 2003). About one-third of CRC patients in both countries have de-novo cancer. One study from UK also reported this feature (Rembacken et. al., 2000). Whether this feature is unique to Asia or whether it shows any preference for the younger or the older group of patients is not reported. Because of their flat appearance they are harder to identify by conventional colonoscopy**.** Chromoendoscopy and the use of magnifying colonoscopy may be necessary. The absence of polypoid growth preceding malignancy has posed difficulties in screening for early CRC by radiological imaging or even endoscopic techniques.

#### **5. Early onset CRC and genetics**

Colorectal tumours provide an excellent model system for understanding the molecular events that control the process of initiation and progression of human tumours. Rate of random mutational events alone cannot account for the number of genetic alterations found in most human cancers and it has been suggested that destabilization of the genome may be a prerequisite early in carcinogenesis. In CRC there are two separate destabilizing pathways. The more common involves chromosomal instability (CNI). The second mutational pathway in CRC displays increased rate of intragenic mutation characterized by generalized instability in microsatellites (MSI). Defects in mismatch repair genes (MMR) lead to high frequency MSI in CRC. National Cancer Institute definitions of MSI-L (L=low), MSI-H (H=high) and MSS (microsatellite stable) in CRC are given in Boland et. al. (1998). A recently recognized molecular alteration found frequently in MSI cancers is the CpG island methylator phenotype (CIMP). Colon cancer is usually observed in one of three specific patterns: sporadic, inherited or familial. Fewer than 10% of patients have an inherited predisposition to colon cancer. Sporadic cancer is common in persons older than 50 years of age, probably as a result of dietary and environmental factors as well as normal aging. Patients with inherited disease have CRC at a younger age, 10-20 years earlier than general population and are of interest in this essay. The area of hereditary CRC has been reviewed by Lynch and de la Chapelle (2003)and earlier by Lynch et. al (1991). Different aspects of molecular genetics of CRC have been discussed in this series (Ewart Toland, 2012) and elsewhere (Fearon and Volgenstein 1990; Loeb 1994; Jass 1995; Lynch 1996; Baba 1997; Gryfe et. al. 1997; Lengauer et. al. 1998; Lynch and Smyrk 1998; Lynch and de la Chapelle 1999; Yang 1999; Potter 1999; Jass et. al. 2002; Calvert and Frucht 2002; Zbuk 2009). In this section we discuss several recent papers which highlight the difference in genetic characteristics of younger CRC patients and those of the older group.

Morris et. al. (2007) showed that the incidence of tumours with microsattelite instability was significantly higher in patients aged 40 years, 18. 3% compared to 6. 6% in those aged 41 – 60 yrs (p<0. 0001). TP53 mutations were also more frequent (p=0. 002). However K-ras mutations were less common (p=0. 0001) when comparing the same age groups. They

polypoid and non-polypoid (superficial) tumours. The latter are flat lesions with a raised or depressed surface. Since these tumours are small (<1cm in diameter) and there are no adenomatous elements in their vicinity, they were proposed not to have originated from any precursor lesion and were termed de-novo carcinomas. These non-polypoid tumours are less likely to have K-ras mutations than are CRC arising from the adenoma-carcinoma sequence. Non-polypoid tumours of the colorectal regions tend to reach deeper layers of the intestinal wall in the early stage of the disease and with a higher degree of dysplasia. They are therefore more invasive than the polypoid adenomas (Sung et. al., 2005). Reports on de novo cancer have been published from Japan (Goto et. al 2004) and from Taiwan (Chen et. al 2003). About one-third of CRC patients in both countries have de-novo cancer. One study from UK also reported this feature (Rembacken et. al., 2000). Whether this feature is unique to Asia or whether it shows any preference for the younger or the older group of patients is not reported. Because of their flat appearance they are harder to identify by conventional colonoscopy**.** Chromoendoscopy and the use of magnifying colonoscopy may be necessary. The absence of polypoid growth preceding malignancy has posed difficulties in screening

Colorectal tumours provide an excellent model system for understanding the molecular events that control the process of initiation and progression of human tumours. Rate of random mutational events alone cannot account for the number of genetic alterations found in most human cancers and it has been suggested that destabilization of the genome may be a prerequisite early in carcinogenesis. In CRC there are two separate destabilizing pathways. The more common involves chromosomal instability (CNI). The second mutational pathway in CRC displays increased rate of intragenic mutation characterized by generalized instability in microsatellites (MSI). Defects in mismatch repair genes (MMR) lead to high frequency MSI in CRC. National Cancer Institute definitions of MSI-L (L=low), MSI-H (H=high) and MSS (microsatellite stable) in CRC are given in Boland et. al. (1998). A recently recognized molecular alteration found frequently in MSI cancers is the CpG island methylator phenotype (CIMP). Colon cancer is usually observed in one of three specific patterns: sporadic, inherited or familial. Fewer than 10% of patients have an inherited predisposition to colon cancer. Sporadic cancer is common in persons older than 50 years of age, probably as a result of dietary and environmental factors as well as normal aging. Patients with inherited disease have CRC at a younger age, 10-20 years earlier than general population and are of interest in this essay. The area of hereditary CRC has been reviewed by Lynch and de la Chapelle (2003)and earlier by Lynch et. al (1991). Different aspects of molecular genetics of CRC have been discussed in this series (Ewart Toland, 2012) and elsewhere (Fearon and Volgenstein 1990; Loeb 1994; Jass 1995; Lynch 1996; Baba 1997; Gryfe et. al. 1997; Lengauer et. al. 1998; Lynch and Smyrk 1998; Lynch and de la Chapelle 1999; Yang 1999; Potter 1999; Jass et. al. 2002; Calvert and Frucht 2002; Zbuk 2009). In this section we discuss several recent papers which highlight the difference in genetic characteristics of

Morris et. al. (2007) showed that the incidence of tumours with microsattelite instability was significantly higher in patients aged 40 years, 18. 3% compared to 6. 6% in those aged 41 – 60 yrs (p<0. 0001). TP53 mutations were also more frequent (p=0. 002). However K-ras mutations were less common (p=0. 0001) when comparing the same age groups. They

for early CRC by radiological imaging or even endoscopic techniques.

**5. Early onset CRC and genetics** 

younger CRC patients and those of the older group.

concluded that major age related differences in the clinical and molecular features of CRC exist.

Farrington et. al. (1998) pointed out that germ-line mutations in DNA mismatch-repair (MMR) genes impart a markedly elevated cancer risk, often presenting as autosomal dominant HNPCC. Not all gene carriers have a family history. Young probands with early onset CRC irrespective of family history were genetically tested and it was found that an appreciable proportion of young colon cancer probands carry a germline mutation in a DNA MMR gene.

Losi et. al. (2005) evaluated clinical features and molecular pathways, chromosomal instability (CNI) and MSI in early onset CRC. Of 71 patients (<45 years), 14 showed both MSI and altered expressions of MMR proteins. In the 57 MSI -negative (-) lesions, altered expression of APC, β-catenin and p53 genes were found more frequently than in MSIpositive (+) tumours. 7/14 MSI (+) tumours were associated with clinical features of HNPCC and in all but one, constitutional mutations in MLH-1 and MSH-2 genes could be detected. The same mutations were found in other family members. Involvement of chromosomal instability was demonstrated in a majority of early onset CRC.

Chan et. al. (1999)studied 59 Chinese patients <45 yrs and 58, >45 yrs in Hong Kong. The incidence of MSI-H varied statistically significantly with age, being observed in >60% of those <31 years at diagnosis and in <15% of those ≥46 years. More than 80% of Chinese CRC patients <31 years had germline mutations in MMR genes. In a novel case, mutation in hMSH-6 was present but MSI was absent.

Ho et. al. (2000) in a study on 124 young (<50yrs) Hong Kong Chinese CRC patients concluded that MSI occurs in a significant proportion of the subjects. Young age at CRC diagnosis, proximal tumour location, increasing number of first degree relatives with CRC and a personal history of metachronous cancer were independent predictors of MSI status in the patient group. In patients <30 years, MSI tumours were more likely to be located in distal large bowel. In a proportion of patients with MSI tumours, germline mutation in the two MMR genes hMSH2 and hMLH1 was identified. The authors opined that this observation suggests a differential activity of the MMR pathway in colorectal carcinogenesis in different age groups. They observed that the inconsistency between MSI-H and a family history in the early onset patients deserves further attention.

Liang et. al. (2003) studied 138 below-40 CRC patients and 339 patients who were 60+. They found a higher percentage of normal p53 expression (61. 1 vs. 46. 8%, p=0. 023) and high frequency microsatellite instability (MSI-H) ( 29. 4 vs. 6. 3% p<0. 001) in the young. The family history of the two groups was similar.

Durno et. al. (2005, 2006) found evidence of MSI in 73% cancers from individuals in 9-24 years of age, 50% of whom had features of HNPCC. Other reports found MSI in 46% of under- 21 patients with only 1/3 having a clear family history.

Sanchez et. al. (2009) performed a molecular classification of CRC based on microsattellite instability (MSI), CpG-island methylator phenotype (CIMP) and mutations in the K-ras and BRAF oncogenes. There were four classes, combinations of MSI-H and MSS with CIMP–H or CIMP (-). 69. 8% of tumours (391 subjects) were MSS-CIMP(-) and less likely to be poorly differentiated (p=0. 009). CIMP-H tumours were more common in older patients (p<0. 001). MSI-H/CIMP-H tumours had a high frequency of BRAF mutation and a low rate of K-ras mutation, the opposite was true for MSS-CIMP(-) tumours (p<0. 001). The four molecular phenotypes tended towards divergent survival. MSI-H cancers were associated with better disease free survival.

Colorectal Carcinoma in the Young 31

cancers of several different types, particularly in HNPCC. CRC in majority of <35 years group (58% of 31 patients) exhibited instability whereas CRC in > 35 years group uncommonly did (12% of 158). In 12 of <35 years group, instability was evaluated for alterations of MMR genes and in 5, it was found to harbor germline mutations. These data suggested that the mechanisms underlying tumour development in young CRC patients

Lin et. al. (2010) showed in a study cohort of 950 patients (2000-2005) that carcinogenic effects of Western lifestyle might be mediated via insulin-like growth factor-1 (IGF-1). IGF-1 is a peptide growth factor that promotes cell proliferation and inhibits apoptosis. Both in vitro and in vivo studies suggested that IGF-1 could promote CRC growth. Further, circulating levels of IGF-1 were associated with various cancers including CRC. It was shown that genetic variation controls variability of circulating IGF-1. The expression of IGF-1 was reported to vary in different ethnicities. In turn it was speculated that polymorphisms of the genes involved in the IGF axis might affect IGF-1 expression and possible cancer risk. The age at onset of CRC varied considerably. Extreme age at the CRC onset, very young or very old seemed to be associated with different carcinogenesis. It was shown that some genetic polymorphism affects age of onset of cancers. For example IGF-1 polymorphism plays a significant role in affecting disease onset in Lynch syndrome. These authors showed that older patients have a higher frequency of AA genotype of IGF-1 (- 2995C/A), significantly higher (12. 7%) than that in younger patients (4. 2%). Mucinous differentiation, but not other clinicopathological factors was associated with the CA /AA genotype of IGF-1. The authors concluded that the genotype of the IGF-1 promoter was different in young CRC patients compared to older CRC patients and that IGF-1 SNP was

Yantis et. al. (2009) provided data to show that post translational regulation of mRNA and subsequent protein expression may be particularly important to the development of CRC in young patients. They compared 24 patients <40 yrs of age with 45 patients 40 yrs of age, who served as controls. Cases were evaluated for clinical risk factors of malignancy and pathologic feature predictive of outcome. More aggressive features in tumours of young patients, namely more frequent lymphovascular (81%) and venous (48%) invasion, an infiltrative growth pattern (81%) were reported. Significantly increased expression of miR-

Family history of CRC at a young age is a significant risk factor. Johns and Houlston (2001) performed a meta-analysis of 27 case-control and cohort studies of colorectal cancer risk and found that a family history of one affected first degree relative diagnosed before the age of 45 carried a 3. 87 fold (95% confidence interval 2. 40 – 6. 22) increased risk for the disease. Fuchs et. al. (1994) concluded that a family history of CRC is associated with an increased risk of disease, especially amongst the young. The relative risk factor of an under-45 yrs person with one or more affected first degree relative as compared with those without a family history was 3. St. John et. al. (1993) performed a case-control study of relatives of CRC patients and of matched control patients. They concluded that first degree relatives of patients with CRC have an increased risk of colorectal cancer. The risk was greater if diagnosed at an early age and when other first degree relatives were affected. Winawer et. al. (1996) observed that siblings and parents of patients with adenomatous polyps were at

21, miR-20a, miR-145, miR-181b, and miR-203 was noted in the younger group.

differ from those in most older patients.

associated with mucinous adenocarcinoma.

**6. Predisposing factors** 

Alsop et. al. (2006) investigated association of young age in below-45 patients with somatic mutation of K-ras gene, a common event in CRC tumorigenesis. The role of these mutations was found to be comparatively minor in the younger group, in contrast to its significant role in CRC of older age of onset.

Soliman et. al. (2001) compared molecular pathology of CRC in Egyptian (44% <40 years) and Western patients. They found MSI-H carcinoma in 17% (2/12) of under-40 and 46% (12/26) of 40+ Egyptian patients; K-ras gene mutation in 0% (0/18) of under-40 group and in 17% (5/29) of 40+ group; p-53 overexpression in 57% (13/23) of under-40 group and 39% (13/33) of 40+ group. These data show that molecular pathology of CRC in young Egyptians differed from that in the old; in particular, K-ras mutation played a distinctly minor role in the younger group. Unique differences in molecular pathology of CRC between the Egyptian and Western patients were also discussed.

Breivik et. al. (1997) found that the presence of K-ras mutation was dependent on age and gender of the patient, with an especially low frequency amongst young males. Microsatellite instability was rare in tumours with K-ras and TP53 mutations.

Berg et. al. (2010) focused on the somatic tumour development in young patients with no known inherited syndromes. They studied mutations in oncogenes K-ras, BRAF, PIK3CA and the tumour suppressor gene PTEN and in TP53, in three age groups in 181 patients (45, < 50 yrs; 67, 51-70 yrs; 69, >70 yrs). Distinct genetic differences were found in tumours in the young and the elderly patients, who were comparable for known clinical and pathological variables. This result indicated that young patients had a different genetic risk profile for CRC development than older patients. Clinical implications of these differences were discussed by the authors. The total gene mutation index was lowest in tumours from the younger patients. In contrast the genome complexity assessed as copy number aberrations was highest in tumours from the youngest patients.

Casper et. al. (1994) showed in a study on 225 FAP patients that deletion of 5 base pairs at codon 1309 within exon 15 (the most common mutation) was identified in 20 families. Other mutations within exons 7-15 were found in 49 families. The 1309 mutation leads to development of colonic polyps at a younger age thus giving rise to an earlier malignant transformation. In patients with 5 base pair deletion at codon 1309, gastrointestinal symptoms and death from CRC occurred about 10 years earlier than in patients with other mutations.

Khan et. al. (2008) studied 35 patients with CRC diagnosed at <30 years age. They found no mutations in exons 4-10 of the p53 gene. The frequencies of polymorphism in p53 and in MDM2SNP309 did not differ from rates previously reported for normal control populations and no polymorphism in either gene could be associated with early onset CRC.

Ahmed et. al. (2005) reported a study on 363 CRC patients of whom 18 were of Bangladeshi origin. 22% of Bangladeshi patients presented with a locally advanced or a metastatic CRC, whereas the same figure for non-Bangladeshi patients was 11%. Sixty one percent of the Bangladeshi patients were below 40 years of age and did not report any family history. Microarray profiling between these two groups demonstrated 1203 differentially expressed genes (p<0. 05). The patient groups studied by Nath et. al. (2005) and by our group, (Gupta et. al. 2010) (Table 3) and by Pal (2006) belong predominantly to West Bengal in India, which is adjacent to Bangladesh. These studies reported dominance of younger patients in their study groups, advanced disease stage and aggressive tumour characteristics.

Liu et. al. (1995) studied the prevalence of DNA replication errors (RER) associated with genetic instability in relation to age among patients without HNPCC. RER was found in

Alsop et. al. (2006) investigated association of young age in below-45 patients with somatic mutation of K-ras gene, a common event in CRC tumorigenesis. The role of these mutations was found to be comparatively minor in the younger group, in contrast to its significant role

Soliman et. al. (2001) compared molecular pathology of CRC in Egyptian (44% <40 years) and Western patients. They found MSI-H carcinoma in 17% (2/12) of under-40 and 46% (12/26) of 40+ Egyptian patients; K-ras gene mutation in 0% (0/18) of under-40 group and in 17% (5/29) of 40+ group; p-53 overexpression in 57% (13/23) of under-40 group and 39% (13/33) of 40+ group. These data show that molecular pathology of CRC in young Egyptians differed from that in the old; in particular, K-ras mutation played a distinctly minor role in the younger group. Unique differences in molecular pathology of CRC between the

Breivik et. al. (1997) found that the presence of K-ras mutation was dependent on age and gender of the patient, with an especially low frequency amongst young males. Microsatellite

Berg et. al. (2010) focused on the somatic tumour development in young patients with no known inherited syndromes. They studied mutations in oncogenes K-ras, BRAF, PIK3CA and the tumour suppressor gene PTEN and in TP53, in three age groups in 181 patients (45, < 50 yrs; 67, 51-70 yrs; 69, >70 yrs). Distinct genetic differences were found in tumours in the young and the elderly patients, who were comparable for known clinical and pathological variables. This result indicated that young patients had a different genetic risk profile for CRC development than older patients. Clinical implications of these differences were discussed by the authors. The total gene mutation index was lowest in tumours from the younger patients. In contrast the genome complexity assessed as copy number aberrations

Casper et. al. (1994) showed in a study on 225 FAP patients that deletion of 5 base pairs at codon 1309 within exon 15 (the most common mutation) was identified in 20 families. Other mutations within exons 7-15 were found in 49 families. The 1309 mutation leads to development of colonic polyps at a younger age thus giving rise to an earlier malignant transformation. In patients with 5 base pair deletion at codon 1309, gastrointestinal symptoms and death from CRC occurred about 10 years earlier than in patients with other

Khan et. al. (2008) studied 35 patients with CRC diagnosed at <30 years age. They found no mutations in exons 4-10 of the p53 gene. The frequencies of polymorphism in p53 and in MDM2SNP309 did not differ from rates previously reported for normal control populations

Ahmed et. al. (2005) reported a study on 363 CRC patients of whom 18 were of Bangladeshi origin. 22% of Bangladeshi patients presented with a locally advanced or a metastatic CRC, whereas the same figure for non-Bangladeshi patients was 11%. Sixty one percent of the Bangladeshi patients were below 40 years of age and did not report any family history. Microarray profiling between these two groups demonstrated 1203 differentially expressed genes (p<0. 05). The patient groups studied by Nath et. al. (2005) and by our group, (Gupta et. al. 2010) (Table 3) and by Pal (2006) belong predominantly to West Bengal in India, which is adjacent to Bangladesh. These studies reported dominance of younger patients in their

Liu et. al. (1995) studied the prevalence of DNA replication errors (RER) associated with genetic instability in relation to age among patients without HNPCC. RER was found in

and no polymorphism in either gene could be associated with early onset CRC.

study groups, advanced disease stage and aggressive tumour characteristics.

in CRC of older age of onset.

Egyptian and Western patients were also discussed.

was highest in tumours from the youngest patients.

mutations.

instability was rare in tumours with K-ras and TP53 mutations.

cancers of several different types, particularly in HNPCC. CRC in majority of <35 years group (58% of 31 patients) exhibited instability whereas CRC in > 35 years group uncommonly did (12% of 158). In 12 of <35 years group, instability was evaluated for alterations of MMR genes and in 5, it was found to harbor germline mutations. These data suggested that the mechanisms underlying tumour development in young CRC patients differ from those in most older patients.

Lin et. al. (2010) showed in a study cohort of 950 patients (2000-2005) that carcinogenic effects of Western lifestyle might be mediated via insulin-like growth factor-1 (IGF-1). IGF-1 is a peptide growth factor that promotes cell proliferation and inhibits apoptosis. Both in vitro and in vivo studies suggested that IGF-1 could promote CRC growth. Further, circulating levels of IGF-1 were associated with various cancers including CRC. It was shown that genetic variation controls variability of circulating IGF-1. The expression of IGF-1 was reported to vary in different ethnicities. In turn it was speculated that polymorphisms of the genes involved in the IGF axis might affect IGF-1 expression and possible cancer risk. The age at onset of CRC varied considerably. Extreme age at the CRC onset, very young or very old seemed to be associated with different carcinogenesis. It was shown that some genetic polymorphism affects age of onset of cancers. For example IGF-1 polymorphism plays a significant role in affecting disease onset in Lynch syndrome. These authors showed that older patients have a higher frequency of AA genotype of IGF-1 (- 2995C/A), significantly higher (12. 7%) than that in younger patients (4. 2%). Mucinous differentiation, but not other clinicopathological factors was associated with the CA /AA genotype of IGF-1. The authors concluded that the genotype of the IGF-1 promoter was different in young CRC patients compared to older CRC patients and that IGF-1 SNP was associated with mucinous adenocarcinoma.

Yantis et. al. (2009) provided data to show that post translational regulation of mRNA and subsequent protein expression may be particularly important to the development of CRC in young patients. They compared 24 patients <40 yrs of age with 45 patients 40 yrs of age, who served as controls. Cases were evaluated for clinical risk factors of malignancy and pathologic feature predictive of outcome. More aggressive features in tumours of young patients, namely more frequent lymphovascular (81%) and venous (48%) invasion, an infiltrative growth pattern (81%) were reported. Significantly increased expression of miR-21, miR-20a, miR-145, miR-181b, and miR-203 was noted in the younger group.

#### **6. Predisposing factors**

Family history of CRC at a young age is a significant risk factor. Johns and Houlston (2001) performed a meta-analysis of 27 case-control and cohort studies of colorectal cancer risk and found that a family history of one affected first degree relative diagnosed before the age of 45 carried a 3. 87 fold (95% confidence interval 2. 40 – 6. 22) increased risk for the disease. Fuchs et. al. (1994) concluded that a family history of CRC is associated with an increased risk of disease, especially amongst the young. The relative risk factor of an under-45 yrs person with one or more affected first degree relative as compared with those without a family history was 3. St. John et. al. (1993) performed a case-control study of relatives of CRC patients and of matched control patients. They concluded that first degree relatives of patients with CRC have an increased risk of colorectal cancer. The risk was greater if diagnosed at an early age and when other first degree relatives were affected. Winawer et. al. (1996) observed that siblings and parents of patients with adenomatous polyps were at

Colorectal Carcinoma in the Young 33

an average of 4. 5 years. Those few patients who presented early in the course of their disease responded well to radical resection. Okuno et. al. (1987) reported frequent occurrence of mucinous carcinoma, lymph node involvement and advanced stage according to Dukes' classification in the younger group (<39 years). The overall survival rate was poorer in the younger group (41% vs. 55. 9%), whereas the difference between the two

**Pre-2000, favourable prognosis:** Howard et. al. (1975) found that younger patients had a greater frequency of advanced signs, later stages of cancer and mucoid carcinoma, but when compared by clinical stage, they did as well or better than older patients. 5-year survival rates were 31% in <40 years group and 32% in>40 years group. Clinical staging was the most important prognostic factor irrespective of age. No inherent difference was found in the virulence of cancer in the young, survival rate being essentially the same. Adloff et. al. (1986) in a paper published much later reached identical conclusion. Walton et. al. (1976) in a study on 70 under-40 patients reported that survival time was shorter in patients with mucinous and anaplastic tumours and their incidences increased in this age group. Overall survival rates, however, did not significantly depend on age. Early diagnosis and prompt aggressive surgical treatment produced survival equivalent to that in patients of other age groups. Scarpa et. al. (1976) in a study on 47 adults in the age group 20-40 years found smaller tumours and depth of invasion as important prognostic factors but tumour grade had no correlation with survival. They concluded that there was no difference in survival rate between the young and the old. Bulow (1980) found, in an extensive study spread over 25 years (951 <40 patients, all <40 patients in Denmark in the period 1943-1967) that stage according to Dukes' classification and presence of intestinal obstruction and/or perforation and not age, determined prognosis. Ahlberg et. al. (1980) in a study group of 27 patients, aged <30 years, in 1969-70 in Scandinavia, concluded that prognosis was good, if predisposing factors were absent (9/15 survived 5 years), but not so otherwise. Ohman (1982) in a study group of 1061 patients, of whom 48 were below 40, in Sweden reported a five year survival rate in the overall population and in curable cases. Both rates were equal in the two age groups. Age factor had no impact. Five year survival was 100% in stage A, 50% in B, 33% in C. Proportion of Dukes' A lesion was equal in the two groups; there were fewer B and more C lesions amongst the young. Survival was not altered if ulcerative colitis was superimposed on carcinoma. Beckman et. al. (1984) studied 69 patients, 20-39 years and reported good prognosis. Neither age, sex, tumour size, location, mere presence of lymph node metastases, depth of tumour invasion nor predisposing disease of the colon was a strong prognostic factor. Metastases of six or more lymph nodes and distant spread of the tumour at the time of initial surgery were ominous findings; so was mucinous carcinoma, a relatively frequent occurrence. Jarvinen and Turunen (1984) in a study on 249 under-40 year patients between 1970 and 1979 found no difference in their 5 year survival rate from that of the general population. A premalignant condition was more common as age decreased. Family cancer syndrome, FAP and other predisposing diseases were observed in a significant proportion of study group. It was suggested that more emphasis should be placed on identification, family screening and treatment of conditions predisposing to colorectal cancer. LaQuaglia et. al. (1992) analysed their experience with 29 histologically verified cases of whom 20 were resected for cure. The predictors for survival were resectability, regional nodal involvement, depth of invasion, grade (Signet ring (45%) or anaplastic lesions (24%) were considered high grade) and interval from symptom onset to diagnosis. Median age at diagnosis was 19 years (10-21 years), median survival was 16

groups in rates of curative resection was not statistically significant.

an increased risk for CRC particularly when adenoma was diagnosed at < 60 yrs age. Despite limited accuracy and compliance, family history is still the most easily obtainable risk factor for colorectal cancer.

Deficiency in host response to carcinogenesis is less easily recognized and treated. A personal history of other cancers, especially chronic immunosensitive cancers such as melanoma, if occurring at a young age, may indicate an increased susceptibility to CRC. Chronic immune suppression or clinical suggestions of impaired immunity may also mean the same.

FAP and HNPCC patients have a lifetime risk of 100 and 80 percent respectively, of developing CRC. In FAP, the affected persons develop hundreds to thousands of colonic polyps. Although the rate of transition to cancer is slow, the vast number of polyps virtually assures colon cancer development at a young age. Average age of developing cancer is 39 years, with 7% diagnosed by the age of 20 and 15% by 25. In HNPCC, the affected persons have a very high risk for CRC but do not develop the hundreds of polyps seen in FAP. These polyps are very likely to make a transition to cancer. Although sporadic colon cancer usually arises in colon polyps after a 5-10 years period of growth and transformation, in HNPCC this progression can occur within 1 -2 years. HNPCC occurs at a relatively young age, median 42-45 years, with 35%-40% diagnosed before 40 years of age. The proportion of HNPCC or familial colorectal cancer among all CRC varies by country from 1-10% with a median of 2-5% (Mecklin and Ponz de Leon 1994). HNPCC has been reported from many different populations, Europeans, white and Indian Americans, Asians, Australasians, South Americans and Egyptians (Sarroca et. al. 1978; Bamezai et. al. 1984; Ushio 1985; Lynch et. al. 1985; Mecklin 1987; Vasen et. al. 1990; Mecklin and Jarvinen 1991; Jass and Stewart 1992; Soliman et. al. 1998).

Ulcerative colitis (UC) is another important predisposing factor. The most important risk factors for development of CRC in UC patients are prolonged duration of disease, pancolonic disease, continuously active disease and severity of inflammation. Eaden et. al. (2001) performed a meta-analysis of the risk of CRC in UC. 116 of 194 reported studies were included in this analysis. Overall prevalence of CRC in UC patients was 3. 7%. The risk of CRC in UC patients was determined by decade of disease and a non significant increase in risk over time was observed.

#### **7. Prognosis and survival of young patients**

Opinion on the issue of survival of younger patients is not unanimous. We have divided literature reports on this issue, pre-2000 and post-2000, in two separate sections. The reports in which prognosis for the young is shown to be poorer and the ones in which they are not so, are separately grouped.

**Pre-2000, poor prognosis**: Moore et. al. (1984) concluded that poorer survival in younger (<40 years) patients was a result of an inherently more virulent lesion, a conclusion supported by a greater incidence of mucinous tumours, an indicator of poor prognosis and a higher incidence of advanced disease, especially in the second and third decades. They did not find delay in diagnosis as an important factor in determination of survival. Adkins et. al. (1987) ascribed poorer prognosis in the young (<35 years) to unfavourable histological features of the tumours and advanced disease at the time of presentation in these patients. Of 45 under-35 patients, 19 patients with poorly differentiated tumours survived for an average of 1 year, whereas 19 with well or moderately differentiated tumours survived for

an increased risk for CRC particularly when adenoma was diagnosed at < 60 yrs age. Despite limited accuracy and compliance, family history is still the most easily obtainable

Deficiency in host response to carcinogenesis is less easily recognized and treated. A personal history of other cancers, especially chronic immunosensitive cancers such as melanoma, if occurring at a young age, may indicate an increased susceptibility to CRC. Chronic immune suppression or clinical suggestions of impaired immunity may also mean

FAP and HNPCC patients have a lifetime risk of 100 and 80 percent respectively, of developing CRC. In FAP, the affected persons develop hundreds to thousands of colonic polyps. Although the rate of transition to cancer is slow, the vast number of polyps virtually assures colon cancer development at a young age. Average age of developing cancer is 39 years, with 7% diagnosed by the age of 20 and 15% by 25. In HNPCC, the affected persons have a very high risk for CRC but do not develop the hundreds of polyps seen in FAP. These polyps are very likely to make a transition to cancer. Although sporadic colon cancer usually arises in colon polyps after a 5-10 years period of growth and transformation, in HNPCC this progression can occur within 1 -2 years. HNPCC occurs at a relatively young age, median 42-45 years, with 35%-40% diagnosed before 40 years of age. The proportion of HNPCC or familial colorectal cancer among all CRC varies by country from 1-10% with a median of 2-5% (Mecklin and Ponz de Leon 1994). HNPCC has been reported from many different populations, Europeans, white and Indian Americans, Asians, Australasians, South Americans and Egyptians (Sarroca et. al. 1978; Bamezai et. al. 1984; Ushio 1985; Lynch et. al. 1985; Mecklin 1987; Vasen et. al. 1990; Mecklin and Jarvinen 1991; Jass and Stewart 1992;

Ulcerative colitis (UC) is another important predisposing factor. The most important risk factors for development of CRC in UC patients are prolonged duration of disease, pancolonic disease, continuously active disease and severity of inflammation. Eaden et. al. (2001) performed a meta-analysis of the risk of CRC in UC. 116 of 194 reported studies were included in this analysis. Overall prevalence of CRC in UC patients was 3. 7%. The risk of CRC in UC patients was determined by decade of disease and a non significant increase in

Opinion on the issue of survival of younger patients is not unanimous. We have divided literature reports on this issue, pre-2000 and post-2000, in two separate sections. The reports in which prognosis for the young is shown to be poorer and the ones in which they are not

**Pre-2000, poor prognosis**: Moore et. al. (1984) concluded that poorer survival in younger (<40 years) patients was a result of an inherently more virulent lesion, a conclusion supported by a greater incidence of mucinous tumours, an indicator of poor prognosis and a higher incidence of advanced disease, especially in the second and third decades. They did not find delay in diagnosis as an important factor in determination of survival. Adkins et. al. (1987) ascribed poorer prognosis in the young (<35 years) to unfavourable histological features of the tumours and advanced disease at the time of presentation in these patients. Of 45 under-35 patients, 19 patients with poorly differentiated tumours survived for an average of 1 year, whereas 19 with well or moderately differentiated tumours survived for

risk factor for colorectal cancer.

the same.

Soliman et. al. 1998).

risk over time was observed.

so, are separately grouped.

**7. Prognosis and survival of young patients** 

an average of 4. 5 years. Those few patients who presented early in the course of their disease responded well to radical resection. Okuno et. al. (1987) reported frequent occurrence of mucinous carcinoma, lymph node involvement and advanced stage according to Dukes' classification in the younger group (<39 years). The overall survival rate was poorer in the younger group (41% vs. 55. 9%), whereas the difference between the two groups in rates of curative resection was not statistically significant.

**Pre-2000, favourable prognosis:** Howard et. al. (1975) found that younger patients had a greater frequency of advanced signs, later stages of cancer and mucoid carcinoma, but when compared by clinical stage, they did as well or better than older patients. 5-year survival rates were 31% in <40 years group and 32% in>40 years group. Clinical staging was the most important prognostic factor irrespective of age. No inherent difference was found in the virulence of cancer in the young, survival rate being essentially the same. Adloff et. al. (1986) in a paper published much later reached identical conclusion. Walton et. al. (1976) in a study on 70 under-40 patients reported that survival time was shorter in patients with mucinous and anaplastic tumours and their incidences increased in this age group. Overall survival rates, however, did not significantly depend on age. Early diagnosis and prompt aggressive surgical treatment produced survival equivalent to that in patients of other age groups. Scarpa et. al. (1976) in a study on 47 adults in the age group 20-40 years found smaller tumours and depth of invasion as important prognostic factors but tumour grade had no correlation with survival. They concluded that there was no difference in survival rate between the young and the old. Bulow (1980) found, in an extensive study spread over 25 years (951 <40 patients, all <40 patients in Denmark in the period 1943-1967) that stage according to Dukes' classification and presence of intestinal obstruction and/or perforation and not age, determined prognosis. Ahlberg et. al. (1980) in a study group of 27 patients, aged <30 years, in 1969-70 in Scandinavia, concluded that prognosis was good, if predisposing factors were absent (9/15 survived 5 years), but not so otherwise. Ohman (1982) in a study group of 1061 patients, of whom 48 were below 40, in Sweden reported a five year survival rate in the overall population and in curable cases. Both rates were equal in the two age groups. Age factor had no impact. Five year survival was 100% in stage A, 50% in B, 33% in C. Proportion of Dukes' A lesion was equal in the two groups; there were fewer B and more C lesions amongst the young. Survival was not altered if ulcerative colitis was superimposed on carcinoma. Beckman et. al. (1984) studied 69 patients, 20-39 years and reported good prognosis. Neither age, sex, tumour size, location, mere presence of lymph node metastases, depth of tumour invasion nor predisposing disease of the colon was a strong prognostic factor. Metastases of six or more lymph nodes and distant spread of the tumour at the time of initial surgery were ominous findings; so was mucinous carcinoma, a relatively frequent occurrence. Jarvinen and Turunen (1984) in a study on 249 under-40 year patients between 1970 and 1979 found no difference in their 5 year survival rate from that of the general population. A premalignant condition was more common as age decreased. Family cancer syndrome, FAP and other predisposing diseases were observed in a significant proportion of study group. It was suggested that more emphasis should be placed on identification, family screening and treatment of conditions predisposing to colorectal cancer. LaQuaglia et. al. (1992) analysed their experience with 29 histologically verified cases of whom 20 were resected for cure. The predictors for survival were resectability, regional nodal involvement, depth of invasion, grade (Signet ring (45%) or anaplastic lesions (24%) were considered high grade) and interval from symptom onset to diagnosis. Median age at diagnosis was 19 years (10-21 years), median survival was 16

Colorectal Carcinoma in the Young 35

(2008) in the USA performed a comparative study of two groups, < 40 years and > 60 years of age, ethnically diverse, between 1998 and 2005. Fifty one percent of 41 young patients were Hispanic. Young patients were more likely to have a family history. Aggressive nature of tumour in the young was noted, but operative intervention and survival was similar in the two groups. Tohme et. al. (2008) in a study of 325 patients, 13. 2% of whom were below 45 concluded that age by itself was not a significant prognostic factor. The independent prognostic factors were delay in consultation, which was more frequent in younger patients (29. 7 vs. 18. 6 weeks, p=0. 01), positive family history in the young (44. 1% vs. 18. 2%), right sided tumour and peritoneal carcinomatosis. Leff et. al. (2010) in a British study of 49, <40 patients reported a 5-year and overall survival of 58% and 46% respectively. They concluded that prognosis in the young was not worse than that for CRC in the population as a whole. Mitry et. al. (2001) reported, in both overall and stage for stage comparisons that patients below the age of 45 years had a better survival rate than older patients, mortality rate was lower in the younger group (2. 1% vs. 8. 4%) although advanced stage presentation was more frequent and predisposing conditions were significantly higher in the below 45 group (11. 7 vs. 0. 4%, p<0. 001). Lin et. al. (2005) studied 45 histologically confirmed under-40 patients, 90% of whom reported with advanced (C+D) stage, between 1992-2002 in Taiwan. They reported that disease stage was an important prognostic factor, 5 year survival in B, C and D stage patients being 25, 16, and 0% respectively. Karnofsky performance status (KPS 70%), lymph node involvement and preoperative LDH levels were major determinants of survival. Surgical resection and adjuvant chemotherapy improved survival of advanced stage patients, but the improvement achieved does reach the level of a patient who reports early. Liang et. al. (2003) reported that although the younger patients with colorectal cancer had more mucin producing (14. 7 vs. 4. 7%, p<0. 001) tumours and a more advanced tumour stage at presentation (p<0. 001) than older patients, the operative mortality rate was lower (0. 7 vs. 5%) and cancer specific survival was similar (p<0. 05) in stage I, II, III disease or better in stage IV disease (22-28 vs. 12-

**Post-2000, poor prognosis:** Endreseth et. al. (2006) in a study on 2283 rectal cancer patients found overall 5-year survival to be 54% for patients younger than 40 years compared to 71-88% for the older patients (p=0. 029). Among those treated for cure, 56% of <40 group developed distant metastasis compared to 20-26% in the older group. Age younger than 40 years was a significant prognostic factor in this group and increased the risk of metastasis and death. A study from Nepal by Singh et. al. (2002a) reported a more aggressive disease in the younger group (<40 years) and a significantly lower 2 years survival rate (4% vs. 55%). Singh et. al. (2002b)in a study on 18 under 40 patients in India found that the tumor was unresectable in 5 patients (28%). Fourteen patients (78%) had advanced cancer indicated by TNM stage III or IV disease. Among the 13 patients subjected to surgical treatment followed by adjuvant chemotherapy, only 3 had long term

**Prognosis and Genetics:** There have been reports in the literature that suggest that the survival of MSI-H CRC patients is longer than that of patients with MSS CRC. This latter group constitutes the majority. In some studies however no survival advantage was detected and a National Cancer Institute workshop held in 1998 (Boland et. al. 1998) concluded that MSI had not been shown to be an independent predictor of prognosis (Gryfe et. al., 2000). We cite a number of mostly post-2000 papers that link prognosis to MSI tumour

17 months, p< 0. 001).

pathway.

disease free survival beyond 2 years.

months, that for those undergoing complete resection was 33 months. In those undergoing resection for cure, tumour grade, regional nodal involvement and depth of invasion were the only factors that affected prognosis. Hidalgo (1995) in Spain studied 26, under-45 CRC patients (17. 2% of the whole group) whose potential risk factors were no different from those of the general population. Clinical presentation, tumour site and Dukes' stage were similar in the younger group and in the general population, but morbidity, mortality and post operative complications were lower. There were no differences in resection or survival rates. Chung et.al. (1998) in a study on 101 under 40 patients and 2064 older patients found no difference in tumour characteristics, Dukes' stage and overall 5 year survival, but reported a higher adjusted hazard ratio and adverse outcome in the <40 years group compared to 40-59 years group. They noted that a significant family history and predisposing conditions in the young warrants aggressive screening, surveillance and treatment. Heys et. al. (1994) in a review reported histological evaluations of the cancers in the younger age group patients and found that approximately four times as many tumours were of the mucinous type. This was associated with an increased risk of local recurrence. Dukes' staging and vascular invasion by tumour were prognostic indicators for overall patient survival. However survival rates for young patients with CRC were comparable to those of older patients, when equivalent Dukes' stage was considered.

**Post-2000, favourable prognosis**: O'Connell et. al. (2004b) used SEER (Surveillance, Epidemiology and End Results) database in the period 1991-1999 in the USA (1334 younger patients, 20-40 years; 46,457 older patients 60-80 years) to conclude that 5-year stage-specific survival was similar for stage I and III patients and better for younger patients in stage II and IV (p<0. 01). The same patient group showed later stage (more of stage III and IV) and higher grade tumours for younger patients. The authors noted that their population-based finding contradicts earlier single institution reports. Stigliano et. al. (2008) compared a cohort of 40 HNPCC cases with 573 sporadic CRC cases in the period 1970-1993. Median age of diagnosis was 46. 8 years in HNPCC cases and 61 years in sporadic CRC cases. Early stage cancer (Dukes' A & B) was 70% in HNPCC group and 61. 6% in sporadic group. The crude 5-year cumulative survival for primary CRC was 94. 2% in HNPCC vs. 75. 3% in sporadic cancer patients (p < 0. 0001). The influence of age on prognosis is apparent.

Berg et. al. (2010) studied 181 patients (45, < 50 yr; 67 (51-70 yr); 69, > 70 yr) and found no difference in survival while comparing age groups, even when adjustment for tumour stage at diagnosis had been made. Younger patients however presented at a more advanced disease stage (54, 46, and 36% in three groups). Tumour stage was the most powerful prognostic variable (p < 0. 001). Turkiewicz et.al. (2001) in a study spanning 29 years in Australia concluded that young patients with CRC had the potential to do just as well. The overall 5 years survival among younger patients in Stage A and B (53%) was found to be better than their counterparts in the older group. With influence of a family history of CRC being very apparent in this group, the authors conclude that emphasis must be on screening. Makela et. al. (2002) in a study of 102 under-50 patients in Finland over a 20 years period (1980-1999) concluded that young age is not a poor prognostic marker in colorectal cancer. Radical operation, venous invasion and tumour grade were good predictors of survival in patients below 50 years. Kam et.al. (2004) in a study in Singapore on 39 under-30 years patients inferred that age did not affect survival and recommended early endoscopy for all with persistent symptoms. They concluded that early diagnosis, radical resection and adjuvant therapy still form the cornerstone in management of colorectal cancer in this age group. Karsten et.al.

months, that for those undergoing complete resection was 33 months. In those undergoing resection for cure, tumour grade, regional nodal involvement and depth of invasion were the only factors that affected prognosis. Hidalgo (1995) in Spain studied 26, under-45 CRC patients (17. 2% of the whole group) whose potential risk factors were no different from those of the general population. Clinical presentation, tumour site and Dukes' stage were similar in the younger group and in the general population, but morbidity, mortality and post operative complications were lower. There were no differences in resection or survival rates. Chung et.al. (1998) in a study on 101 under 40 patients and 2064 older patients found no difference in tumour characteristics, Dukes' stage and overall 5 year survival, but reported a higher adjusted hazard ratio and adverse outcome in the <40 years group compared to 40-59 years group. They noted that a significant family history and predisposing conditions in the young warrants aggressive screening, surveillance and treatment. Heys et. al. (1994) in a review reported histological evaluations of the cancers in the younger age group patients and found that approximately four times as many tumours were of the mucinous type. This was associated with an increased risk of local recurrence. Dukes' staging and vascular invasion by tumour were prognostic indicators for overall patient survival. However survival rates for young patients with CRC were comparable to

**Post-2000, favourable prognosis**: O'Connell et. al. (2004b) used SEER (Surveillance, Epidemiology and End Results) database in the period 1991-1999 in the USA (1334 younger patients, 20-40 years; 46,457 older patients 60-80 years) to conclude that 5-year stage-specific survival was similar for stage I and III patients and better for younger patients in stage II and IV (p<0. 01). The same patient group showed later stage (more of stage III and IV) and higher grade tumours for younger patients. The authors noted that their population-based finding contradicts earlier single institution reports. Stigliano et. al. (2008) compared a cohort of 40 HNPCC cases with 573 sporadic CRC cases in the period 1970-1993. Median age of diagnosis was 46. 8 years in HNPCC cases and 61 years in sporadic CRC cases. Early stage cancer (Dukes' A & B) was 70% in HNPCC group and 61. 6% in sporadic group. The crude 5-year cumulative survival for primary CRC was 94. 2% in HNPCC vs. 75. 3% in sporadic cancer patients (p < 0. 0001). The influence of age

Berg et. al. (2010) studied 181 patients (45, < 50 yr; 67 (51-70 yr); 69, > 70 yr) and found no difference in survival while comparing age groups, even when adjustment for tumour stage at diagnosis had been made. Younger patients however presented at a more advanced disease stage (54, 46, and 36% in three groups). Tumour stage was the most powerful prognostic variable (p < 0. 001). Turkiewicz et.al. (2001) in a study spanning 29 years in Australia concluded that young patients with CRC had the potential to do just as well. The overall 5 years survival among younger patients in Stage A and B (53%) was found to be better than their counterparts in the older group. With influence of a family history of CRC being very apparent in this group, the authors conclude that emphasis must be on screening. Makela et. al. (2002) in a study of 102 under-50 patients in Finland over a 20 years period (1980-1999) concluded that young age is not a poor prognostic marker in colorectal cancer. Radical operation, venous invasion and tumour grade were good predictors of survival in patients below 50 years. Kam et.al. (2004) in a study in Singapore on 39 under-30 years patients inferred that age did not affect survival and recommended early endoscopy for all with persistent symptoms. They concluded that early diagnosis, radical resection and adjuvant therapy still form the cornerstone in management of colorectal cancer in this age group. Karsten et.al.

those of older patients, when equivalent Dukes' stage was considered.

on prognosis is apparent.

(2008) in the USA performed a comparative study of two groups, < 40 years and > 60 years of age, ethnically diverse, between 1998 and 2005. Fifty one percent of 41 young patients were Hispanic. Young patients were more likely to have a family history. Aggressive nature of tumour in the young was noted, but operative intervention and survival was similar in the two groups. Tohme et. al. (2008) in a study of 325 patients, 13. 2% of whom were below 45 concluded that age by itself was not a significant prognostic factor. The independent prognostic factors were delay in consultation, which was more frequent in younger patients (29. 7 vs. 18. 6 weeks, p=0. 01), positive family history in the young (44. 1% vs. 18. 2%), right sided tumour and peritoneal carcinomatosis. Leff et. al. (2010) in a British study of 49, <40 patients reported a 5-year and overall survival of 58% and 46% respectively. They concluded that prognosis in the young was not worse than that for CRC in the population as a whole. Mitry et. al. (2001) reported, in both overall and stage for stage comparisons that patients below the age of 45 years had a better survival rate than older patients, mortality rate was lower in the younger group (2. 1% vs. 8. 4%) although advanced stage presentation was more frequent and predisposing conditions were significantly higher in the below 45 group (11. 7 vs. 0. 4%, p<0. 001). Lin et. al. (2005) studied 45 histologically confirmed under-40 patients, 90% of whom reported with advanced (C+D) stage, between 1992-2002 in Taiwan. They reported that disease stage was an important prognostic factor, 5 year survival in B, C and D stage patients being 25, 16, and 0% respectively. Karnofsky performance status (KPS 70%), lymph node involvement and preoperative LDH levels were major determinants of survival. Surgical resection and adjuvant chemotherapy improved survival of advanced stage patients, but the improvement achieved does reach the level of a patient who reports early. Liang et. al. (2003) reported that although the younger patients with colorectal cancer had more mucin producing (14. 7 vs. 4. 7%, p<0. 001) tumours and a more advanced tumour stage at presentation (p<0. 001) than older patients, the operative mortality rate was lower (0. 7 vs. 5%) and cancer specific survival was similar (p<0. 05) in stage I, II, III disease or better in stage IV disease (22-28 vs. 12- 17 months, p< 0. 001).

**Post-2000, poor prognosis:** Endreseth et. al. (2006) in a study on 2283 rectal cancer patients found overall 5-year survival to be 54% for patients younger than 40 years compared to 71-88% for the older patients (p=0. 029). Among those treated for cure, 56% of <40 group developed distant metastasis compared to 20-26% in the older group. Age younger than 40 years was a significant prognostic factor in this group and increased the risk of metastasis and death. A study from Nepal by Singh et. al. (2002a) reported a more aggressive disease in the younger group (<40 years) and a significantly lower 2 years survival rate (4% vs. 55%). Singh et. al. (2002b)in a study on 18 under 40 patients in India found that the tumor was unresectable in 5 patients (28%). Fourteen patients (78%) had advanced cancer indicated by TNM stage III or IV disease. Among the 13 patients subjected to surgical treatment followed by adjuvant chemotherapy, only 3 had long term disease free survival beyond 2 years.

**Prognosis and Genetics:** There have been reports in the literature that suggest that the survival of MSI-H CRC patients is longer than that of patients with MSS CRC. This latter group constitutes the majority. In some studies however no survival advantage was detected and a National Cancer Institute workshop held in 1998 (Boland et. al. 1998) concluded that MSI had not been shown to be an independent predictor of prognosis (Gryfe et. al., 2000). We cite a number of mostly post-2000 papers that link prognosis to MSI tumour pathway.

Colorectal Carcinoma in the Young 37

Chemotherapeutic treatment protocol will progressively become more specific as the genetic basis of CRC gets better understood and the heterogeneity of the disease is better

Fallik et. al. (2003) studied response to irinotecan in 72 patients, of whom 1 responded completely and 11 partially. Among the 7 tumours that displayed MSI-H phenotype, 4 responded to irinotecan whereas only 7 out of 65 MSI-L tumours did (p=0. 009). A better response to irinotecan was observed in the patients whose tumours have lost BAX expression (p<0. 001). 7 of 72 tumours had inactivating mutations in the coding repeat of the target genes. Amongst these seven, five responded to irinotecan, whereas only 6 of the other 65 tumours did (p<0. 001) indicating that MSI-driven inactivation of target genes modifies tumour sensitivity. It has been shown that tumours with mucinous histology, a common feature of many tumours in the young(section 7) and whose molecular genetic signatures have been referred to earlier (section 3), show poor response to fluorouracil-based first line chemotherapy (Negri et. al. 2005) and first-line oxaliplatin /irinotecan based combination

Adkins RB Jr., DeLozier JB, McKnight WG, et. al. Carcinoma of the colon in patients 35 years

Adloff M, Arnaud JP, Schloegel M et. al. Colorectal cancer in patients under 40 years of age.

Ahlberg J, Bergstrand O, Holmstrom B, et. al. Malignant tumours of the colon and rectum in patients aged 30 and younger. Acta Chir Scand suppl 1980; 500:29-31. Ahmed S, Banerjea A., Hands RE et. al.; Microarray profiling of colorectal cancer in

Alsop K, Mead L, Smith LD, et. al.; Low somatic K-ras mutation frequency in colorectal

Ashenafi, S. The frequency of large bowel cancer as seen in Addis Ababa University,

Baba S. Recent advances in molecular genetics of colorectal cancer. World J. Surg. 1997; 21:

Bae JM, Jung KW, Won YJ. Estimation of cancer deaths in Korea for the upcoming years; J.

Bamezai R, Singh G, Khanna NN et. al. Genetics of site specific colon cancer: a family study.

Barnetson RA, Tenesa A, Farrington SM. Identification and survival of carriers of mutations in DNA mismatch repair genes in colon cancer. N. Engl. J. Med. 2006; 354: 2751-3. Beckman EN, Gathright JB, Ray JE. A potentially brighter prognosis for colon carcinoma in

cancer diagnosed under the age of 45 years. European Journal of Cancer 2006; 42:

characterized. There are several literature reports that show these connections.

**8. Molecular genetics and treatment protocol** 

chemotherapy (Catalano et. al. 2009).

We thank Mr. Anil Kumar Verma and Dr. Nilanjana Bose.

of age and younger. Am Surg 1987; 53: 141-5.

Bangladeshi patients. Colorectal Dis. 2005; 7:571-5.

Pathology Department. Ethiop. Med. J. 2000; 38:277-82.

the third and fourth decades. Cancer 1984; 54:1478-81.

Dis Colon Rectum 1986; 29:322-5.

Korean Med. Sci. 2002, 17:611-5.

Clin. Genet., 1984; 26: 129–32.

**9. Acknowledgement** 

1357-61.

678-87.

**10. References** 

Gryfe et. al. (2000), in a study on 607 under-50 patients found MSI in 17% of patients and concluded that MSI was associated with a significant survival advantage independent of all standard prognostic factors including tumour stage. Regardless of depth of tumour invasion, MSI-H CRC had a decreased likelihood of metastasis to regional lymph nodes. Elsaleh et. al. (2000) (mean age 66. 7±12. 9 years) in Australia report striking survival benefits for patients with MSI tumours (90 vs 35%, p=0. 0007) and also for patients with right sided lesions, who received adjuvant chemotherapy as compared to those who did not (48 vs 27% alive at end of study, p<0. 0001) and for women (53 vs 33%, p<0. 0001). Suh et. al. (2002) in a comparative study of MSI(+) and MSI(-) sporadic young (<40years) CRC patients showed that the former had better prognosis (p=0. 051). Their results suggested that sporadic MSI(+) CRC in the young had different histomorphologic features as compared to MSI(-) CRC and HNPCC cancers. Samowitz et. al. (2009)in a study of 990 rectal cancer patients in the US showed that even though MSI-H has been associated in many studies with improved prognosis of colon cancer, the effect of MSI-H and K-ras mutations posed significantly higher risk of death for rectal cancers. Liang et. al. (2003) reported that there was a higher percentage of normal p53 expression (61 vs. 48%) and high frequency microsatellite instability (MSI-H) (29. 4 vs. 6. 3%, p, 0. 001) in the young. Lukish et. al. (1998) in a study group of 36 patients in the <40 year age group determined their DNA replication error (RER) status (expressed as MSI) and compared the clinical and pathologic characteristics of RER(+) and RER(-) cases. They concluded that RER(+) tumours were common (47%) in young patients and patients with RER(+) tumours had a significantly improved prognosis:5 year survival probability 68% in RER(+), 32% in RER () tumours (p<0. 05). Knowledge of RER status therefore could affect initial therapy, postoperative chemotherapy and follow up.

The paradoxical good survival after surgery for patients with young age at diagnosis of CRC supports the idea that many cancers in the young are microsatellite unstable. A number of studies linked high frequency MSI to poor tumour differentiation or mucinous histology, a signature of many tumours in the young (Sanchez et. al. 2009; Kim et. al. 1994, Lin et. al 2010, Suh et. al., 2002). Ionov et. al. (1993) in their study of mutations involving poly (dA. dT) sequences (Section 5 for details) found that the presence of mutations was accompanied by an increase in the proportion of poorly differentiated lesions (6/9 vs 17/90, poor/well, moderate)and also in an increase in proportion of Stage A +B disease (2/14 vs 53/68; C+D/A+B). Crude survival was expected to be better than usual in young patients because of their youth and the improved tolerance to surgery and complications that youth confers (Liang and Church 2010).

Berg et. al. (2010) in a study of patients in different age groups (Section 5 for details), found that patients with TP53 mutated tumours had poorer survival rates than patients with wild type TP53 (938 vs. 1016 days, p =0. 04); however the difference was not significant when corrected for tumour stage. TP53 mutation were of higher prognostic significance in right sided tumours (883, 1051 days; mutated, wild type; p = 0. 005). Among patients in the younger age group, those with K-ras mutation had significantly shorter survival than patients with K-ras wild type samples (841, 1033 days, p=0. 02).

Barnetson et. al. (2006) studied a group of 870 below-55 years CRC patients for germline mutations in DNA MMR genes, proposed a model for prediction of the presence of mutations in these genes and validated the model in an independent group of 155 patients. Survival in carriers and non carriers was similar.

Gryfe et. al. (2000), in a study on 607 under-50 patients found MSI in 17% of patients and concluded that MSI was associated with a significant survival advantage independent of all standard prognostic factors including tumour stage. Regardless of depth of tumour invasion, MSI-H CRC had a decreased likelihood of metastasis to regional lymph nodes. Elsaleh et. al. (2000) (mean age 66. 7±12. 9 years) in Australia report striking survival benefits for patients with MSI tumours (90 vs 35%, p=0. 0007) and also for patients with right sided lesions, who received adjuvant chemotherapy as compared to those who did not (48 vs 27% alive at end of study, p<0. 0001) and for women (53 vs 33%, p<0. 0001). Suh et. al. (2002) in a comparative study of MSI(+) and MSI(-) sporadic young (<40years) CRC patients showed that the former had better prognosis (p=0. 051). Their results suggested that sporadic MSI(+) CRC in the young had different histomorphologic features as compared to MSI(-) CRC and HNPCC cancers. Samowitz et. al. (2009)in a study of 990 rectal cancer patients in the US showed that even though MSI-H has been associated in many studies with improved prognosis of colon cancer, the effect of MSI-H and K-ras mutations posed significantly higher risk of death for rectal cancers. Liang et. al. (2003) reported that there was a higher percentage of normal p53 expression (61 vs. 48%) and high frequency microsatellite instability (MSI-H) (29. 4 vs. 6. 3%, p, 0. 001) in the young. Lukish et. al. (1998) in a study group of 36 patients in the <40 year age group determined their DNA replication error (RER) status (expressed as MSI) and compared the clinical and pathologic characteristics of RER(+) and RER(-) cases. They concluded that RER(+) tumours were common (47%) in young patients and patients with RER(+) tumours had a significantly improved prognosis:5 year survival probability 68% in RER(+), 32% in RER () tumours (p<0. 05). Knowledge of RER status therefore could affect initial therapy, postoperative

The paradoxical good survival after surgery for patients with young age at diagnosis of CRC supports the idea that many cancers in the young are microsatellite unstable. A number of studies linked high frequency MSI to poor tumour differentiation or mucinous histology, a signature of many tumours in the young (Sanchez et. al. 2009; Kim et. al. 1994, Lin et. al 2010, Suh et. al., 2002). Ionov et. al. (1993) in their study of mutations involving poly (dA. dT) sequences (Section 5 for details) found that the presence of mutations was accompanied by an increase in the proportion of poorly differentiated lesions (6/9 vs 17/90, poor/well, moderate)and also in an increase in proportion of Stage A +B disease (2/14 vs 53/68; C+D/A+B). Crude survival was expected to be better than usual in young patients because of their youth and the improved tolerance to surgery and complications that youth confers

Berg et. al. (2010) in a study of patients in different age groups (Section 5 for details), found that patients with TP53 mutated tumours had poorer survival rates than patients with wild type TP53 (938 vs. 1016 days, p =0. 04); however the difference was not significant when corrected for tumour stage. TP53 mutation were of higher prognostic significance in right sided tumours (883, 1051 days; mutated, wild type; p = 0. 005). Among patients in the younger age group, those with K-ras mutation had significantly shorter survival than

Barnetson et. al. (2006) studied a group of 870 below-55 years CRC patients for germline mutations in DNA MMR genes, proposed a model for prediction of the presence of mutations in these genes and validated the model in an independent group of 155 patients.

patients with K-ras wild type samples (841, 1033 days, p=0. 02).

Survival in carriers and non carriers was similar.

chemotherapy and follow up.

(Liang and Church 2010).

#### **8. Molecular genetics and treatment protocol**

Chemotherapeutic treatment protocol will progressively become more specific as the genetic basis of CRC gets better understood and the heterogeneity of the disease is better characterized. There are several literature reports that show these connections.

Fallik et. al. (2003) studied response to irinotecan in 72 patients, of whom 1 responded completely and 11 partially. Among the 7 tumours that displayed MSI-H phenotype, 4 responded to irinotecan whereas only 7 out of 65 MSI-L tumours did (p=0. 009). A better response to irinotecan was observed in the patients whose tumours have lost BAX expression (p<0. 001). 7 of 72 tumours had inactivating mutations in the coding repeat of the target genes. Amongst these seven, five responded to irinotecan, whereas only 6 of the other 65 tumours did (p<0. 001) indicating that MSI-driven inactivation of target genes modifies tumour sensitivity. It has been shown that tumours with mucinous histology, a common feature of many tumours in the young(section 7) and whose molecular genetic signatures have been referred to earlier (section 3), show poor response to fluorouracil-based first line chemotherapy (Negri et. al. 2005) and first-line oxaliplatin /irinotecan based combination chemotherapy (Catalano et. al. 2009).

#### **9. Acknowledgement**

We thank Mr. Anil Kumar Verma and Dr. Nilanjana Bose.

#### **10. References**


Colorectal Carcinoma in the Young 39

deSilva MV, Fernando MS, Fernando D. Comparison of some clinical and histological

Dozois EJ, Boardman LA, Suwanthawna W. et. al. Young-Onset colorectal cancer in patients

Durno C, Arnoson M, Bapat B, et. al. ; Family history and molecular features of children adolescents, and young adults with colorectal carcinoma. Gut. 2005; 54: 1146-50. Durno CA, Gallinger S. Genetic predisposition to colorectal cancer: new pieces in the

Eaden JA, Abrams KR, Mayberry JT. The risk of colorectal cancer in ulcerative colitis: a

Elsaleh H, Joseph D, Grieu F. et. al. Association of tumour site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. The Lancet 2000; 355: 1745-50. Endreseth BH, Romundstad P. Myrvold HE, et. al. ; Rectal Cancer in the young patient. Dis.

Ewart Toland, A., Germline Genetics in Colorectal Cancer Susceptibility and Prognosis,

Fairley TL, Cardinez CJ, Martin J et. al. Colorectal cancer in US Adults younger than 50

Fallik D, Borrini F, Boige V, et. al.; Microsatellite instability is a predictive factor of the

Fancher TT,Palesty JA,Rashidi L et. al.; Is Gender Related to the Stage of Colorectal Cancer

Fante R, Benatti P, diGregorio C. , et. al.; Colorectal carcinoma in different age groups: a population Fancher tbased investigation. Am. J. Gastroenterol 1997; 92: 1505-9. Farrington SM, Lin-Goerke J, Ling J. et. al.; Systematic Analysis of hMSH2 and hMLH1 in young colon cancer patients and controls. Am. J. Hum. Genet. 1998; 63:749-59. Fazeli MS, Adel MG, Lebaschi AH. Colorectal carcinoma: A retrospective, descriptive study

Fearon ER and Vogelstein B. A Genetic Model for Colorectal Tumorigenesis. Cell 1990;

Fireman Z, Sandler E, Kopelman Y et. al. Ethnic differences in Colorectal Cancer among Arab and Jewish neighbours in Israel. Am. J. Gastroenterol, 2001;96:204-7. Frizis A, Papadopoulos A, Akriditis G et. al. Are there any differences in colorectal cancer between young and older patients. Tech. Colorectal, 2004; 8 Supplement 1: s147-8. Fry, R. D. , Mahmoud, N; Maron, D. J. et. al. (2008,18th Ed;vol 2) in Sabiston Textbook of Surgery, Townsend CM, Beauchamp RD, Evers BM et. al. ; Elsevier, p 1404-5. Fuchs CS, Guovannucci EL, Coldetz GA, et. al. A prospective study of family history and the risk of colorectal cancer. New England J. Of Medicine 1994; 331: 1669-74. Goh KL, Quek KF, Yeo GTS et. al. Colorectal cancer in Asians: a demographic and anatomic

observed in Tehran University. Dis. Colon Rectum, 2007; 50:990-5.

tumour response to irinotecan in patients with advanced colorectal cancer. Cancer

at Initial Presentation in Young Patients? Journal of Surgical Reserch 2011;165,15-18

of Age, Gender, sub site, stage and differentiation in Iran from 1995 to 2001 as

survey in Malaysian patients undergoing colonoscopy. Aliment Pharmacol Ther

Goto H, Oda Y, Tanaka T, et. al. ; Estimated incidence of colorectal de novo cancer in Japan.

with no known genetic predisposition. Medicine 2008; 87: 259-63.

pediatric puzzle. J Pediatr Gastroenterol Nutr 2006; 43: 5-15.

Colorectal Cancer Biology - From Genes to Tumor, 2012.

years of age, 1998-2001. Cancer 2006; 107: 1153-61.

Ceylon Med J 2000;45:166–8.

metaanalysis. Gut 2001; 48: 526-35.

Colon Rectum 2006; 49:993-1001.

Res. 2003; 63: 5738-44.

61:759-67.

2005; 22: 859–64.

Gut 2004: 53 (suppl 36): A30.

features of colorectal carcinoma occurring in patients below and above 40 years.


Bedikian AY, Kantarjian H, Nelson RS, et al. Colorectal cancer in young adults. South Med J

Behbehani A, Sakwa M, Ehrlichman R, et. al. Colorectal carcinoma in patients under age 40.

Berg M, Danielsen SA, Ahlquist T, et. al. ; DNA sequence profiles of the colorectal cancer

Beart RW Jr, Melton LJ, Maruta et. al. ; Trends in right and left sided colon cancer. Dis Colon

Boland CR, Thibodeau SN , Hamilton SR et. al. A National Cancer Institute Workshop on

Breivik J, Ragnhild AL, Gunn IM et. al. ; Different genetic pathways to proximal and distal

Bulow S. Colorectal cancer in patients less than 40 years of age in Denmark, 1943-1967. Dis

Calvert PM, Frucht H. The Genetics of Colorectal Cancer. Annals of Internal Medicine 2002;

Caspari R, Friedi W, Mandil M et. al. ; Familial adenomatous polyposis: mutation at codon

Catalano V, Loupakis F, Graziano F. , et. al. Mucinous histology predicts for poor response

Chan TL, Yuen ST, Chung LP, et. al. Frequent Microsatellite Instability and mismatch repair

Chen CD, Yen MF, Wang WM et. al. A case-control study for the disease natural history of

Chew MH, Koh PK, Ng KH. et. al.; Improved survival in an Asian cohort of young

Chiang JM, Chan MC, Changchien CR, et. al. Favorable influence of age on tumour

Chung YFA, Eu KW, Machin D et. al. Young age is not a poor prognostic marker in

Cozart DT, Lang NP, Hauer-Jensen M. ; Colorectal cancer in patients under 30 years of age.

Cusack JC, Giacco GG, Cleary K, et al. ; Survival factors in 186 patients younger than 40 years old with colorectal adenocarcinoma. J Am Coll Surg 1996;183:105–12.

colorectal cancer patients. Ins. J. Colorectal Dis. 2009; 24: 1075-83.

colorectal cancer. British J. Surg. 1998; 85:1255-9.

Instability in colorectal cancer. Cancer Res 1998;58:5248-57.

critical gene set KRAS-BRAF-PIK3CA-PTEN-TP53 Related to Age at Disease Onset.

Microsatellite Instability for cancer detection and familial predisposition: Development of International Criteria for the determination of Microsatellite

colorectal cancer influenced by sex-related factors. Int. J. Cancer (Pred. Oncol. ),

1309 and early onset of colon cancer. Lancet 1994; 343: 629-32 (Erratum in Lancet

rate and overall survival of patients with colorectal cancer and treated with firstline oxaliplatin- and/or irinotecan-based chemotherapy. Br J Cancer 2009; 100: 881-

gene mutations in Young Chinese patients with Colorectal cancer. J. Natl. Cancer

adenoma-carcinoma and de novo carcinoma and surveillance of colon and rectum after polypectomy: implication for efficacy of colonoscopy. Br J Cancer 2003;

characteristics of sporadx colorectal adenocarcinoma. Dis. Col. Rectum 2003; 46:

Contributors to the Southwestern Surgical Congress Unusual Case Registry. Am J

1981;74:920–4.

Ann. Surg. , 1985; 202: 610-4.

PLoS ONE 2010; 5:e13978.

Colon Rectum 1980; 23: 327-36.

Rectum 1983;26: 393-8.

1997; 74: 664-9.

137: 603-12.

1994; 343: 863).

Inst. 1999; 91:1221-6.

Surg 1993;166:764–7.

88:1866-73.

904-10.

7.


Colorectal Carcinoma in the Young 41

Kaw LL,Punzalan CK,Crisostomo AC et. al. Surgical Pathology of Colorectal Cancer in Filipinos:Implications for Clinical Practice. J. Am. Coll Surg 2002;195:188-95. Khan SA, Idrees K, Forslund A, et. al. ; Genetic variants in germline TP53 and MDM2 SNP309 are not associated with early onset colorectal cancer. J Surg Oncol 2008; 97:621-5. Kim H,Jen J,Vogelstein B et. al. Clinical and pathological characteristics of sporadic

LaQuaglia MP, Heller G, Filippa DA, et. al. ; Prognostic factors and outcome in patients 21 years and under with colorectal carcinoma. J. Pediatr. Surg. 1992; 27: 1085-90. Lee PY, Fletcher WS, Sullivan ES, et al. Colorectal cancer in young patients: characteristics

Leff D. R. , Chen, A. , Roberts D. et. al. ; Colorectal cancer in the young patient. Am. Surg

Lengauer C,Kinzler KW,Vogelstein B . Genetic instabilities in human cancers. Nature

Liang JT, Huang KC, Chen AL et. al. ; Clinicopathological and molecular biological features of colorectal cancer in patients less than 40 years of age. Br J Surg 2003; 19:205-14. Liang J and Church J; How to Manage the patient with Early-Age-of-Onset (<50 years)

Lichtman SM, Mandel F, Hoexter B et. al. Prospective Analysis of colorectal carcinoma. Dis.

Lin JT, Wang WS, Yen CC, et. al. ; Outcome of colorectal carcinoma in patients under-40

Lin J-K, Shen M-Y, Lin T-C, et. al. Distribution of a single nucleotide polymorphism of

Liu B,Farrington SM,Peterson GM et. al. Genetic instability occurs in the majority of young

Loeb LA. Microsatellite Instability:Marker of a mutated phenotype in cancer. Cancer

Losi L, Di Gregorio C, Pedroni M, et. al. ; Molecular genetic alterations and clinical features

Lukish JR,Muro K,De Nobile J et. al. Prognostic significance of DNA Replication Errors in Young patients with Colorectal cancer. Annals of Surgery,1998,227:51-6. Lynch, H. T. , Drouhard, T. J. , Schuelke, G. S. et. al. Hereditary nonpolyposis colorectal cancer in Navajo Indian family. Cancer Genet. Cytogenet. , 1985;15: 209–13. Lynch, H. T. , Smyrk, T. C. ,Watson, P. et. al. ; Hereditary colorectal cancer. Semin. Oncol. ,

Lynch, H. T. Desmoid tumours:Genotype-Phenotype differences in familial adenomatous polyposis-A nosological dilemma. Am. J. Hum. Genet. ,1996;59:1184-5. Lynch, H. T. , Smyrk, T. C. Classification of familial adenomatous polyposis:A diagnostic

Lynch HT, and de la Chapelle A; Genetic susceptibility to non-polyposis colorectal cancer. J.

Lynch HT and de la Chapelle; Hereditary Colorectal Cancer. N. Engl J Med 2003;348:919-932.

insulin-like growth factor-1 in colorectal cancer patients and its association with

in early-onset colorectal carcinomas and their role for the recognition of hereditary

Colorectal Cancer? Surg. Oncol Clin N Am 2010; 19:725-31.

mucinous adeno carcinoma. Int. J. Biol. Markers 2010; 25: 195-9.

patients with colorectal cancer. Nature Medicine 1995,1:348-52.

cancer syndromes. Am J. Gastroenterol,2005;100:2280-7.

nightmare. Am. J. Hum. Genet. , 1998; 62: 1288-9.

years of age. J. Gastroenterol-Hepatol 2005; 20: 900-5.

J Pathol. 1994;145:148-56.

2007; 73:42-7.

1998;396:643-9.

and outcome. Am Surg 1994; 60: 607–12.

Colon. Rectum, 1994; 37:1286-90.

Research 1994;54:5059-63.

1991; 18:337–66.

Med. Genet; 1999; 36: 801-18.

colorectal carcinomas with DNA replication errors in microsatellite sequences. Am.


Gryfe R, Swallow C, Bapat B et. al. Molecular biology of colorectal cancer. Curr Probl

Gryfe R, Kim H, Hsieh ETK et. al. Tumour microsatellite instability and clinical outcome in

Gupta S, Bhattacharya D, Acharya AN et. al. Colorectal carcinoma in young adults: a

Guraya SY, Eltinay OE. Higher prevalence in young population and rightward shift of

Heys SD, O'Hanrahan TJ, Briltenden J, et. al. ; Colorectal cancer in young patients: a review

Hidalgo PM, Moreno SC, Moreno Gonzale E et. al. The incidence, prognostic factors and

Ho JWC,Yuen S,Chung L et. al. ,Distinct Clinical Features Associated With Microsattelite

Hosseini SV, Izadpanah A, Yarmohammadi H. Epidemiological changes in colorectal cancer

Howard E. W, Cavallo C, Hovey LM et. al. Colon Rectal cancer in the young adult; 1975;

Huang J, Seow A, Shi CY et. al. Colorectal Carcinoma among Ethnic Chinese in Singapore.

Ibrahim NK and Abdul Karim FW. Colorectal Adenocarcinoma in Young Lebanese adults.

Isbister WH, Fraser J. Large-Bowel cancer in the young: a national survival study. Dis. Col.

Isbister WH. Colorectal cancer below age 40 in the Kingdom of Saudi Arabia. Aust NZJ.

Ionov Y,Peinado MA,Malkhosyan S. Ubiquitous somatic mutations in simple repeated

Jarvinen HJ, Turunen MJ. Colorectal carcinoma before 40 years of age: prognosis and

Jass JR. Subsite distribution and incidence of colorectal cancer in New Zealand, 1974-1983.

Jass, JR and Stewart, S. M. , Evolution of hereditary non-polyposis colorectal cancer. Gut,

Jass JR. Colorectal adenoma progression and genetic change: is there a link? Ann Med. 1995;

Johns LE & Houlston RS; A systematic review and metaanalysis of familial colorectal cancer

Kam MH, Eu KW, Barben CP et. al. Colorectal Cancer in the young: a 12 year review of

Karsten B, Kim J, King J et. al. Characteristics of colorectal cancer in young patients at an

predisposing conditions . Scand J Gastroenterol 1984; 19: 634–8.

sequences reveal a new mechanism for colonic carcinogenesis. Nature 1993;

colorectal carcinoma. Saudi Med J 2006; 27:1391-3. Hamilton W. Letter to the Editor, Am. J. Surg. 2005; 189: 504.

of the literature. Eur. J. Surg. Oncol. 1994; 20: 225-31.

in Shiraz, Iran: 1980-2000. ANZ J Surg, 2004; 74: 547-9.

young patients with colorectal cancer. New England J. of Medicine 2000; 342:69-77.

retrospective study on Indian patients: 2000-2008. Colorectal Disease 2010; 12, e182-9.

survival in young adults with colorectal adenocarcinoma. Rev. Esp Enfrm Dig.

Instability In Clorectal Cancers Of Young Patients. Int. J. Cancer(Pred

Trends in Incidence Rate by Anatomic Sub site from 1968 to 1992. Cancer. 1999;85:

Cancer. 1997; 21: 23-300.

1995; 87: 431-6.

Oncol):2000;89:356-60

Am. Surg. 41:260-5.

Cancer 1986; 58:816-20.

Rectum 1990; 33: 363-6.

Surg. 1992; 62: 468-72.

Dis Colon Rectum 1991; 34:56-9.

risk. Am. J. Gastroenterol 2001, 96, 2992-3003.

urban county hospital. Am Surg. 2008; 74: 973-6.

patients 30 years or less. Colorectal Disease 2004; 6: 191-4.

2519-25.

363:558-61.

1992; **33**, 783–6.

27: 301-6.


Colorectal Carcinoma in the Young 43

O'Connell J. B. , Maggard M. A. , Liu J. H. et. al. , Do young colon cancer patients have

Ohman U. Colorectal carcinoma in patients less than 40 years of age. Dis Colon Rectum

Okuno M, Ikehara T, Nagayama M, et. al. Colorectal carcinoma in young adults. Am J Surg

Pal, M. Proportionate increased in incidence of colorectal cancer at an age below 40 years:

Paraf F, Jothy S. Colorectal cancer before the age of 40: a case-control study. Dis. Col.

Popat S, Hubner R, Houlston RS. Systematic review of microsatellite instability and

Rembacken BJ, Fujii T, Cairns A, et. al. Flat and depressed colonic neoplasms: a prospective

Saltzstein SL, Behling CA, Savides TJ. The relation of age, race, and gender to the subsite

Samowitz WS, Curtin K, Wolff RK et. al. Microsatellite instability and survival in rectal

Sanchez JA, Krumroy L, Plummer S. et. al. Genetic and epigentic classifications define

Sarroca, C. , Quadrelli, R. and Praderi, R. , Cancer colique familial. (Article in French) Nouv.

Scarpa JF, Hartmann WH, Sawyers JL. Adenocarcinoma of the colon and rectum in young

Shahrudin MD, Noori SM. Cancer of the colon and rectum in the first three decades of life.

Shemesh-Bar L, Kundel Y, Idelevich E. et. al. Colorectal cancer in young patients in Israel.

Singh JP, Maini VK and Bhatnagar A. Large Bowel Malignancy; Epidemiology and gut

Singh Y, Vaidya P, Hemandas AK et. al. Colorectal carcinoma in Nepalese young adults: presentation and outcome. Gan To Kagaku Ryoho. 2002a; 29 Suppl 1:223-9. Singh LJ, Moirangthem GS, Debnath K. Colorectal cancer in younger patients. Trop

Slattery ML, Friedman GD, Potter JD et. al. Description of age, sex and site distributions of colon carcinoma in three geographic areas. Cancer, 1996; 78: 1666-70. Soliman AS, Bondy ML, Levin B. et. al. Colorectal cancer in Egyptian patients under 40 years

Soliman AS, Bondy ML, El-Badawy SA et. al. Contrasting molecular pathology of colorectal

carcinoma in Egyptian and Western patients **,**British Journal of Cancer 2001; **85**(7):

Motility Studies in South Asia. Dis. Colon. Rectum 1984; 27: 10-15.

clinical pheotypes and determine patient outcomes in colorectal cancer. Br J Surg

colorectal cancer: a study of differences between American and Chinese patients.

Potter JD. Colorectal Cancer:Molecules and population. J Natl Cancer Inst 1999;91:916-32. Qing SH, Rao KY, Jiang HY et. al. Racial differences in the anatomical distribution of

worse outcome? World J. Surg. 2004b; 28: 558-62.

An observation. J. Can. Res. Ther. 2006; 2: 97-9.

colorectal cancer prognosis. J Clin Oncol 2005; 23: 609-18.

study of 1000 colonoscopies in the UK. Lancet 2000; 355:1211-4.

location of colorectal carcinoma. Cancer 1998; 82:1408–10.

cancer. Cancer Causes Control. 2009; 20: 1763-8.

1982; 25:209-14.

1987; 154:264-8.

Rectum 2000, 43: 1222-6.

2009; 96:1196-204.

Presse Med. , 1978;7: 1412.

adults. South Med J 1976; 69: 24-7.

World J. Surg. 2010; 34:2701-9.

Gastroenterol 2002b; 23:144-5.

1037–1046.

of age. Int. J. Cancer 1997; 71: 26-30.

Hepatogastroenterology 1997;44:441–4.

World J Gastroenterol 2003; 9:721-5.


Mahadavinia M, Bishehsari F, Ansari R, et. al. ; Family history of colorectal cancer in Iran.

Makela J, Kiviniemi H, Laitinen S. Prognostic factors after surgery in patients younger than

Malekzadeh R, Bishehswin F, Mahdavinia M. et. al. Epidemiology and Molecular Genetics of Colorectal Cancer in Iran :A review. Arch. Iranian Mediane 2009; 12: 161-9. Mecklin, J. P. , Frequency of hereditary colorectal carcinoma. Gastroenterology, 1987; 93,

Mecklin, J. P. and Jarvinen, H. J. , Tumor spectrum in cancer family syndrome (hereditary

Mecklin, J. P. and Ponz de Leon, M. , Epidemiology of HNPCC. Anticancer Res. , 1994; 14,

Minardi AJ Jr, Sittig KM, Zibari GB, et al. Colorectal cancer in the young patient. Am Surg

Mitry E, Benhamiche AM, Jouve JL. Colorectal adenocarcinoma in patients under 45 years of

Mohandas KM, Desai DC. Epidemiology of digestive tract cancer in India. V. Large and

Moore PA, Dilawari RA, Fidler WJ. Adenocarcinoma of the colon and rectum in patients less

Morris M, Platell C, Iacopetta B. A population-based study of age-related variation in

Murday V, and Slack J. , Inherited disorders associated with colorectal cancer. Cancer Surv. ,

Nath J. , Wigley C. , Keighley MRB et. al. Rectal cancer in young adults: a series of 102 patients at a tertiary care centre in India. Colorectal Disease, 2009; 11:475-9. National Cancer Registry Programme, Indian Council of Medical Research, NewDelhi. aConsolidated report of population based cancer registries(PBCR), 2001-4.

Negri FV, Wotherspoon A, Cunnigham D, et. al. ; Mucinous histology predicts for reduced

Nelson RL, Dollear T, Freels S et. al. The relation of age, race, and gender to the subsite

Neufeld D, Shpitz B. Bugaev N. et. al. Young age onset of colorectal cancer in Israel. Tech.

O'Connell JB, Maggard MA, Liu JH et. al. Rates of Colon and Rectal Cancers are increasing

O'Connell, J. B. , Maggard, M. A. , Livingston, E. H. et. al. Colorectal cancer in the young.

location of colorectal carcinoma. Cancer 1997;80:193–7.

in young adults. Am. Surg. 2003; 69: 866-72.

age: comparison with older patients in a well-defined French population. Dis.

clinicopathological features, molecular markers and outcome from colorectal

December 2006, (supplement to Consolidated Report of Population Based Cancer Registries 1990-96:August 2001) www. icmr. nic. in bAtlas of Cancer in India (Chapter 7 of First All India Report 2001-02) National Cancer Registry Programme

fluorouracil responsiveness and survival in advanced colorectal cancer. Annals of

nonpolyposis colorectal cancer). Cancer, 1991; 68, 1109–12.

small bowel. Indian J of Gastroenterol 1999; 18: 118-21.

than 40 years of age. Am Surg 1984; 50: 10-4.

cancer. Anticancer Research 2007; 27:2833-8.

ICMR, www. canceratlasindia. org

oncology 2005, 16: 1305-10.

Coloproctal 2009; 13: 201-4.

The Am J Surg. 2004a;187:343-8.

50 years old with colorectal adenocarcinoma. Hepatogastroenterology. 2002; 49:

BMC Cancer 2005; 5:112.

971-5.

1021–5.

1625–9.

1998;64:849–53.

1989; **8**, 139–57.

Colon. Rectum 2001; 44: 380-87.


**3** 

*Greece* 

**Early Detection of Colorectal Cancer** 

**Issues of early detection of colorectal cancer with reference to the value of screening** 

Colorectal cancer (CRC) is the most common newly-diagnosed cancer, one of the leading causes of illness and death in the Western world, and the second most common cause of cancer morbidity in Europe. Yet, CRC is a preventable disease and, if detected early, highly treatable. Early detection and prevention are health care strategies of critical importance for the reduction of CRC morbidity and mortality. In a number of countries, screening programmes have been implemented on nationwide scale since the 1960s for other forms of cancer. The early detection of cancer increases the likelihood of successful outcomes, but in order to have early detection, education and training promoting early diagnosis and resulting in increased screening, participation is needed. Additionally, the effectiveness of screening can be measured by the reduction on mortality, but it greatly depends upon tangible and sometimes intangible factors, contingent on setting and target population; it is essential, for example, to identify and screen the appropriate target population and to overcome implementation and uptake barriers. All of these issues, with emphasis on obstacles encountered at the level of general and family practice are highlighted in a recent

Although the screening is performed in the context of public health, and for the benefit of the community, the rights and welfare of the individual should also be respected. The role of the General Practitioner/Family Practitioner (GP/FP) and generally of the Primary Care Provider (PCP) is challenging yet instrumental in achieving this balance, as it is at that level screening is initiated (Viguier et al, 2011). The involvement and the role of GPs and PCPs in convincing patients to participate and initiate CRC screening should be further explored and elucidated, as it is of key importance in cultural and organisational context and health policy issues (Sarfaty, 2006). CRC screening of asymptomatic population groups is currently recommended in the USA and many European countries, and a number of pilot and nationwide programmes have been developed for this purpose. More specifically, mass screening programmes are currently established in 13 of 39 European countries (Pox et al, 2007; Manfredi et al, 2011) with feasibility studies undertaken as pilot actions in many more.

**programmes and the role of the primary care practitioner** 

editorial in Family Practice (Lionis and Petelos, 2011).

**1. Introduction** 

**and Population Screening Tests** 

Christos Lionis and Elena Petelos *Clinic of Social and Family Medicine, Faculty of Medicine, University of Crete,* 


### **Early Detection of Colorectal Cancer and Population Screening Tests**

Christos Lionis and Elena Petelos *Clinic of Social and Family Medicine, Faculty of Medicine, University of Crete,* 

#### *Greece*

#### **1. Introduction**

44 Colorectal Cancer – From Prevention to Patient Care

St. John DJB, McDermott FT, Hopper JL et. al. Cancer Risk in Relatives of Patients with Common colorectal cancer. Annals of internal Medicine 1993; 118: 785-90. Stigliano V, Assisi D, Cosimelli M, et. al. ; Survival of hereditary non-polyposis colorectal

Suh JH, Lim SD, Kim JC, et. al. ; Comparison of clinicopathologic characteristics and genetic

Sung JJY, Lau JY, Goh KL et. al. Increasing incidence of colorectal cancer in Asia:

Tamura K, Ishiguro S, Munakata A et. al. Annual changes in colorectal carcinoma incidence

Thibodeau SN, Bren G, Schaid D . Microsatellite Instability in Cancer of the Proximal Colon.

Tohme C, Labaki M, Hajj G, et. al. ; Colorectal cancer in young patients: presentation,

Turkiewich D, Miller B, Schache D et. al. Young patients with colorectal cancer: how to they

Ushio, K., Genetic and familial factors in colorectal cancer. J. clin. Oncol. , 1985; 15(Suppl. 1),

Varma JR, Sample L. Colorectal cancer in patients aged less than 40 years. J Am Board Fam

Vasen, H. F., Offerhaus, G. J. , Den Hartog Jager, F. C. et. al. The tumor spectrum in

Walton WW, Hagihara PG, Griffen WO. Colorectal Adenocarcinoma in Patients Less than 40

Winawer SJ, Zauber AG, Gerdes H, et. al. ; Risk of colorectal cancer in the families of

Yang VW. The Molecular Genetics of Colorectal Cancer; Current Gastroenterology

Yantiss RK, Goodarzi M, Zhou XK et. al. ; Clinical pathologic and molecular features of early

Yilmazlar T, Zorluoglu A, Ozguc H, et. al. ; Colorectal cancer in young adults, 1995; 81:230-3. Yiu HY, Whittemore AS, Shibata A. Increasing colorectal cancer incidence rates in Japan.

Yuen ST, Chung LP and Leung SY. Colorectal carcinoma in Hong Kong, epidemiology and

Zbuk K, Sidebotham EL, Bleyer A. et. al. , Colorectal cancer in young adults. Seminars in

onset colorectal carcinoma. Am J Surg Pathol 2009; 33: 572-82.

hereditary non-polyposis colorectal cancer: a study of 24 kindreds in the

patients with adenomatous polyps. National Polyp Study Workgroup. The N.

implications for screening. Lancet Oncol. 2005;6: 871-6.

Cancer Res 2008; 27: 39.

SCIENCE 1993;260:816-9

fare. ANZ J. Surg 2001; 71: 707-10.

Netherlands. Int. J. Cancer, 1990; 46, 31–4.

England J. of Med 1996; 334:82-7.

Years Old; Dis Colon and Rectum, 1976; 19: 529-34.

International Journal of Cancer, 2004; 109: 777-81.

genetic mutation. Brit. J. Cancer; 1997; 76: 1610-6.

(Article in French).

Pract 1990; 3: 54–9.

Reports1999;1:449-54

Oncology 2009; 36: 439-50.

78: 1187-94.

281–98.

www. Cancer. org

Colon Rectum 2002; 45: 219-28.

cancer patients compared with sporadic colorectal cancer patients. J. Exp. Clin.

alterations between microsatellite instability-positive and microsatellite instabilitynegative sporadic colorectal carcinomas in patients younger than 40 years old. Dis.

in Japan. Analysis of survey data on incidence in Aomori Prefecture. Cancer 1996;

clinicopathological characteristics and outcome. J. Med. Liban, 2008; 56: 208-14

#### **Issues of early detection of colorectal cancer with reference to the value of screening programmes and the role of the primary care practitioner**

Colorectal cancer (CRC) is the most common newly-diagnosed cancer, one of the leading causes of illness and death in the Western world, and the second most common cause of cancer morbidity in Europe. Yet, CRC is a preventable disease and, if detected early, highly treatable. Early detection and prevention are health care strategies of critical importance for the reduction of CRC morbidity and mortality. In a number of countries, screening programmes have been implemented on nationwide scale since the 1960s for other forms of cancer. The early detection of cancer increases the likelihood of successful outcomes, but in order to have early detection, education and training promoting early diagnosis and resulting in increased screening, participation is needed. Additionally, the effectiveness of screening can be measured by the reduction on mortality, but it greatly depends upon tangible and sometimes intangible factors, contingent on setting and target population; it is essential, for example, to identify and screen the appropriate target population and to overcome implementation and uptake barriers. All of these issues, with emphasis on obstacles encountered at the level of general and family practice are highlighted in a recent editorial in Family Practice (Lionis and Petelos, 2011).

Although the screening is performed in the context of public health, and for the benefit of the community, the rights and welfare of the individual should also be respected. The role of the General Practitioner/Family Practitioner (GP/FP) and generally of the Primary Care Provider (PCP) is challenging yet instrumental in achieving this balance, as it is at that level screening is initiated (Viguier et al, 2011). The involvement and the role of GPs and PCPs in convincing patients to participate and initiate CRC screening should be further explored and elucidated, as it is of key importance in cultural and organisational context and health policy issues (Sarfaty, 2006). CRC screening of asymptomatic population groups is currently recommended in the USA and many European countries, and a number of pilot and nationwide programmes have been developed for this purpose. More specifically, mass screening programmes are currently established in 13 of 39 European countries (Pox et al, 2007; Manfredi et al, 2011) with feasibility studies undertaken as pilot actions in many more.

Early Detection of Colorectal Cancer and Population Screening Tests 47

processes. The chapter starts with concepts and definitions, proceeds with the recommended screening tests and concludes by outlining main points of interest and corresponding recommended tasks and actions for PCPs, for the purpose of increasing

Population screening is the systematic application of a suitable test with the aim of identifying individuals at a risk of a specific condition or disorder, but who have not sought medical attention on account of symptoms for that particular condition or disorder, and who can benefit from further investigation or direct preventive action (Wald, 1994). The notion differs from opportunistic screening, and it is a systematic process that includes certain steps from call or recall to screening, feedback of the results and follow-up in welldefined intervals. For population screening, the organised framework in which it takes place provides opportunities for more effective management, quality assurance and evaluation. In our empirical view, understanding of the notion of screening, population or opportunistic, greatly varies between health care practitioners with the result of adversely impacting the effective implementation of the early detection programmes for CRC. It is for this reason we have decided to provide an extensive review on the existing literature and

As stated by Wilson and Jungner in their seminal paper (Wilson and Jungner, 1968) "*the central idea of early disease detection and treatment is essentially simple. However, the path to its successful achievement (on the one hand bringing to treatment those with previously undetected disease, and, on the other, avoiding harm to those persons not in need of treatment) is far from simple though sometimes it may appear deceptively easy*". On the basis of whether early detection is possible at an early stage of the disease and taking into consideration whether an appropriate treatment is available, they attempted to formulate criteria that could help guide the selection of conditions and population groups suitable for screening. They also noted case-finding differences, depending on whether it is performed by a public health agency or by a general practitioner, and, almost four decades ago, emphasised the aspect of cost by underlining the importance of assessing effectiveness not only from an individual,

The fast pace of genetic research and the advent of new therapies has resulted in the generation of many other lists of screening criteria; most of them based to a greater or lesser degree on the Wilson-Jungner criteria. Additionally, even when consensus at the national or regional level is reached on which set of criteria to apply, there are other social, ethical and even logistical considerations to be examined. More recent trends on patient-centric and evidence-based health care, as well as cost-effectiveness and quality assurance, have resulted through a series of consultations to the modified Wilson and Jungner criteria (Andermann et al, 2008). In these amended criteria, opportunistic screening, essentially casefinding performed outside a framework of an organised programme as the one required to ensure such criteria are met, is, therefore, not a valid alternative; additionally to being less efficient it is also more costly, and, most importantly, quality assurance mechanisms cannot

The definition of an "organised" screening programme according to the International Agency for Research on Cancer (IARC) includes: 1) an explicit policy with specified age categories, method and interval for screening; 2) a defined target population; 3) a

uptake and facilitating implementation of CRC screening programmes.

consensus criteria to define screening, focusing on CRC screening.

**2. Concepts and definitions** 

but also from a public health perspective.

be embedded in a standardised fashion in such a process.

Although many of these screening programmes, both opportunistic and population-based are already implemented, and screening and early detection of adenomatous polyps has been shown to be effective in the reduction of CRC morbidity and mortality, the rate of screening participation remains low in many population groups at risk for the disease. Good news have recently arrived from across the Atlantic, where decision analysis tools were employed to inform recommendation updates and "microsimulation modelling demonstrated that declines in CRC death rates are consistent with a relatively large contribution from screening" (Edwards et al, 2010), nevertheless, similar efforts are lacking in some European countries, an issue that is given its due importance in this chapter. The success in the US can be attributed to the efforts of international organisations and national task forces, as they have resulted in a level of high awareness of CRC screening among US primary care providers (PCPs) in the US (Klabunde et al, 2003; Levin et al, 2008), but also in certain European countries. However, there is a variation in the evidence that explains the low rate of CRC screening, especially in younger patients (Walsh et al, 2009), while, few physicians recommend screening for the majority of their patients (McGregor et al, 2004). Compounding this effect is evidence that close to a quarter of physicians report not following national screening guidelines, and only half reported the adoption of recommendations that was consistent with the guidelines (Meissner at al, 2006), another key issue that requires special attention.

Additionally, very few PCPs use chart reminders or outreach programmes to contact patient populations most likely to benefit from screening (Klabunde et al, 2009). There is limited research focusing on obstacles and barriers, and the role of the physician-patient relationship plays in determining participation in screening programmes, especially when it comes to ethnic and culturally diverse groups (Lionis and Petelos, 2011). The importance of culturally relevant strategies for designing and implementing screening programmes has been already highlighted (Tu et al, 2006). Additionally, the role of socioeconomic disparities in CRC screening has been highlighted and documented (Meissner et al, 2011) if not explored in detail (Aubin-Auger et al, 2011), thus indicating a need for a close collaboration between medical and social care scientists in order to improve the requisite understanding for increased compliance to CRC screening recommendations. To compound the increasing complexity of national guidelines and the sensitivity of implementing them to culturally and linguistically varied patients, support through interventions focusing on organizational changes and further education and training for PCPs on early diagnosis, prevention and health promotion is needed. These are all issues that this chapter attempts to address. All of these factors are relevant for and have an impact on the ongoing debate about the role of GPs/FPs and PCPs, as well as the contribution these have on the effective implementation of screening programmes, opportunistic and population-based.

From all of the above one can surmise that the early detection of CRC is an issue of complexity requiring clear messages to increase the awareness and performance of the health care actors. This is another objective of the present chapter. Thus, the particular aims of this chapter are: (a) to provide information about the recommendations issued by certain large national and international organizations, including those issued by the U.S. Preventive Services Task Force (US PSTF) on the use of the available screening tests for the early detection of CRC and adenomas for average-risk subjects, (b) to critically review the role of clinical physicians and mainly PCPs in the early detection of CRC, (c) to explore issues with an impact on CRC screening, and, finally, (d) to highlight some quality issues relevant to CRC screening and relevant guidelines for quality assurance mechanisms in the relevant processes. The chapter starts with concepts and definitions, proceeds with the recommended screening tests and concludes by outlining main points of interest and corresponding recommended tasks and actions for PCPs, for the purpose of increasing uptake and facilitating implementation of CRC screening programmes.

#### **2. Concepts and definitions**

46 Colorectal Cancer – From Prevention to Patient Care

Although many of these screening programmes, both opportunistic and population-based are already implemented, and screening and early detection of adenomatous polyps has been shown to be effective in the reduction of CRC morbidity and mortality, the rate of screening participation remains low in many population groups at risk for the disease. Good news have recently arrived from across the Atlantic, where decision analysis tools were employed to inform recommendation updates and "microsimulation modelling demonstrated that declines in CRC death rates are consistent with a relatively large contribution from screening" (Edwards et al, 2010), nevertheless, similar efforts are lacking in some European countries, an issue that is given its due importance in this chapter. The success in the US can be attributed to the efforts of international organisations and national task forces, as they have resulted in a level of high awareness of CRC screening among US primary care providers (PCPs) in the US (Klabunde et al, 2003; Levin et al, 2008), but also in certain European countries. However, there is a variation in the evidence that explains the low rate of CRC screening, especially in younger patients (Walsh et al, 2009), while, few physicians recommend screening for the majority of their patients (McGregor et al, 2004). Compounding this effect is evidence that close to a quarter of physicians report not following national screening guidelines, and only half reported the adoption of recommendations that was consistent with the guidelines (Meissner at al, 2006), another key

Additionally, very few PCPs use chart reminders or outreach programmes to contact patient populations most likely to benefit from screening (Klabunde et al, 2009). There is limited research focusing on obstacles and barriers, and the role of the physician-patient relationship plays in determining participation in screening programmes, especially when it comes to ethnic and culturally diverse groups (Lionis and Petelos, 2011). The importance of culturally relevant strategies for designing and implementing screening programmes has been already highlighted (Tu et al, 2006). Additionally, the role of socioeconomic disparities in CRC screening has been highlighted and documented (Meissner et al, 2011) if not explored in detail (Aubin-Auger et al, 2011), thus indicating a need for a close collaboration between medical and social care scientists in order to improve the requisite understanding for increased compliance to CRC screening recommendations. To compound the increasing complexity of national guidelines and the sensitivity of implementing them to culturally and linguistically varied patients, support through interventions focusing on organizational changes and further education and training for PCPs on early diagnosis, prevention and health promotion is needed. These are all issues that this chapter attempts to address. All of these factors are relevant for and have an impact on the ongoing debate about the role of GPs/FPs and PCPs, as well as the contribution these have on the effective implementation

From all of the above one can surmise that the early detection of CRC is an issue of complexity requiring clear messages to increase the awareness and performance of the health care actors. This is another objective of the present chapter. Thus, the particular aims of this chapter are: (a) to provide information about the recommendations issued by certain large national and international organizations, including those issued by the U.S. Preventive Services Task Force (US PSTF) on the use of the available screening tests for the early detection of CRC and adenomas for average-risk subjects, (b) to critically review the role of clinical physicians and mainly PCPs in the early detection of CRC, (c) to explore issues with an impact on CRC screening, and, finally, (d) to highlight some quality issues relevant to CRC screening and relevant guidelines for quality assurance mechanisms in the relevant

issue that requires special attention.

of screening programmes, opportunistic and population-based.

Population screening is the systematic application of a suitable test with the aim of identifying individuals at a risk of a specific condition or disorder, but who have not sought medical attention on account of symptoms for that particular condition or disorder, and who can benefit from further investigation or direct preventive action (Wald, 1994). The notion differs from opportunistic screening, and it is a systematic process that includes certain steps from call or recall to screening, feedback of the results and follow-up in welldefined intervals. For population screening, the organised framework in which it takes place provides opportunities for more effective management, quality assurance and evaluation.

In our empirical view, understanding of the notion of screening, population or opportunistic, greatly varies between health care practitioners with the result of adversely impacting the effective implementation of the early detection programmes for CRC. It is for this reason we have decided to provide an extensive review on the existing literature and consensus criteria to define screening, focusing on CRC screening.

As stated by Wilson and Jungner in their seminal paper (Wilson and Jungner, 1968) "*the central idea of early disease detection and treatment is essentially simple. However, the path to its successful achievement (on the one hand bringing to treatment those with previously undetected disease, and, on the other, avoiding harm to those persons not in need of treatment) is far from simple though sometimes it may appear deceptively easy*". On the basis of whether early detection is possible at an early stage of the disease and taking into consideration whether an appropriate treatment is available, they attempted to formulate criteria that could help guide the selection of conditions and population groups suitable for screening. They also noted case-finding differences, depending on whether it is performed by a public health agency or by a general practitioner, and, almost four decades ago, emphasised the aspect of cost by underlining the importance of assessing effectiveness not only from an individual, but also from a public health perspective.

The fast pace of genetic research and the advent of new therapies has resulted in the generation of many other lists of screening criteria; most of them based to a greater or lesser degree on the Wilson-Jungner criteria. Additionally, even when consensus at the national or regional level is reached on which set of criteria to apply, there are other social, ethical and even logistical considerations to be examined. More recent trends on patient-centric and evidence-based health care, as well as cost-effectiveness and quality assurance, have resulted through a series of consultations to the modified Wilson and Jungner criteria (Andermann et al, 2008). In these amended criteria, opportunistic screening, essentially casefinding performed outside a framework of an organised programme as the one required to ensure such criteria are met, is, therefore, not a valid alternative; additionally to being less efficient it is also more costly, and, most importantly, quality assurance mechanisms cannot be embedded in a standardised fashion in such a process.

The definition of an "organised" screening programme according to the International Agency for Research on Cancer (IARC) includes: 1) an explicit policy with specified age categories, method and interval for screening; 2) a defined target population; 3) a

Early Detection of Colorectal Cancer and Population Screening Tests 49

remains, the assessment of a screening program based on FIT for a one-year period in France seemed to be the most cost-effective approach (Hassan, et al, 2011). More research is necessary, as for example indicating what the best cut-off levels for colonoscopy referral are,

**3. Screening tests, guidelines for CRC and the importance of early detection**  As previously mentioned, CRC can be curable when diagnosed at an early stage. Also, CRC mostly develops from colorectal non-malignant precursor lesions, thus, rendering it a preventable disease through the removal of premalignant lesions. Systematic early detection and removal at the "adenoma-phase" can prevent the occurrence of CRC and markedly decrease overall population incidence (Winawer et al, 1993), as human colon carcinogenesis

The readers of this chapter are aware from previous chapters of this book that there are various tumour staging systems, the ones mainly used in Europe being Duke's classification and TNM (Tumour, Node, Metastasis) classification of malignant tumours, introduced by the Union Internationale Contra le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Despite the fact TNM yields greater information, there are several major issues due to the reclassification of the system. Most importantly, there seems to be great disparity between the therapeutic decision-making and the TNM staging, with multiple TNM staging versions being used in different countries and great variance in reporting. It has been argued that changes should only occur after extensive discussion within the scientific community (Quirke et al, 2010), and it is essential to note that the reporting on a nationwide scale for any given CRC screening programme should be performed on the basis of the same staging system. Lesion reporting within the frame of the screening programme should be standardized to allow for better evaluation and reporting, and, consequently, improved outcomes. TNM stages and version, frequency of CRC and distribution of TNM stages should be reported along with the presence of non-neoplastic lesions. According to the report of the EU on CRC quality guidelines (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation, 2011), without explicit criteria for the diagnosis and staging of early adenocarcinoma unnecessary radical resection would result in severe overtreatment, raising

Various screening technologies are currently available, from the more established Guaiac faecal occult blood tests (gFOBT) and immunochemical FOBT to sigmoidoscopy and colonoscopy, as well as combinations (i.e. combined FOBT with sigmoidoscopy) to the new screening technologies, as CT colonography, stool DNA and capsule endoscopy. Early reports during the previous decade suggested that biennial screening by FOBT reduces CRC mortality, as for example in the French (Faivre et al, 2004) and Danish populations (Jørgensen et al, 2002). This fact lead WHO (World Health Organization) and OMED (World Organization for Digestive Endoscopy) to suggest a choice of FOBT that should take into account dietary compliance to recommendations, but also colonoscopy resources (Young et al, 2002). Prior to any recommendation for screening tests for CRC, the PCPs should be able to recognise whether the individual visiting the practice/office is at average, increased or at high risk for CRC. In other words, to be able to identify whether the particular individual truly belongs to the target population of the screening programme or –should there not be one available– whether there is reason for referral in the context of opportunistic screening.

without compromising sensitivity, to determine optimal public health approaches.

progresses to the carcinoma pathway via the dysplasia-adenoma phase.

the morbidity and mortality in the context of the programmes.

management team responsible for implementation; 4) a health-care team for decisions and care; 5) a quality assurance structure; and 6) a method for identifying cancer occurrence and death in the population (IARC 2005). Such organised population-based screening programmes have a predefined specific population, according to epidemiological data and on the basis of target age and geographical area, and during all the stages, from the invitation of the eligible individuals to the assessment procedures following testing, a specific protocol is followed. As mentioned, quality aspects of the process can be better addressed, as for example during follow-up (Miles et al, 2004). Furthermore, such screening programmes usually do not incur any costs for the participants.

It is important to have also a concrete idea regarding more abstract terms determining the usefulness and, even, the effectiveness of screening tests, and to keep such definitions in mind. Although these terms have been widely used, there is also great variation of their understanding and usage in clinical decision-making. For example, the ability of a measure or test to predict a subsequent event is a form of validity. On the basis of such a criterion we determine the predictive value of a test. The positive predictive value (PPV) in terms of detection through FOBT screening is defined as "*the percentage of people with detection of at least one lesion/adenoma/advanced adenoma/cancer at follow-up colorectal screening among those with positive tests who have attended follow-up colorectal screening*" [Adapted from the European guidelines for quality assurance in colorectal cancer screening and diagnosis. 2011. European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation], whereas a positive test is, effectively, an abnormal result leading to further investigation (i.e. colonoscopy) or the removal of a lesion, for example, according to the protocol of the organized screening programme. By alluding to a false-positive result, we effectively mean that although the test indicated disease is present this is not the case. In a true-positive test, the result is correct and the disease is really present. Similarly a false-negative test indicates a disease-free subject has been tested, but the disease is present and might remain undetected if there is no further testing, symptoms, etc., whereas a true-negative test indicates a disease-free subject has been tested. Prevalence of the disease affects not only the positive predictive value of a screening test but also its negative predictive value, i.e. the probability that the person subjected to the screening test is truly free from the disease when a negative (normal) test result is obtained.

Two concepts often discussed in relation to true- and false- positive and negative results are sensitivity and specificity and both terms have an important impact on the PCP decisions on which of the available tests for the early detection of CRC should be recommended. The sensitivity refers to the number of cases the test can identify or in more simple words the probability of one diagnostic test telling the truth when the disease exists. It gives us a certainty that true positives will not be missed. Specificity refers to the accuracy of the finding or in simple words the probability in telling the truth when the disease is absent. Ideally, a test should be both highly sensitive and specific. To that direction, the US Preventive Services Task Force and the Institute of Medicine (IoM) recommend the fecal occult blood test (FOBT) test, and more specifically the guaiac test (gFOBT) for screening programmes. Nevertheless, when used on its own, it has relatively low sensitivity, whereas the a combination with a more sensitive test, such as the fecal immunochemical test (FIT) could help to render screening programmes more effective (Allison, et al, 2007). Another interesting consideration, especially given the public health context of mass screening, is cost-effectiveness and how it correlates to specificity and sensitivity. Although uncertainty

management team responsible for implementation; 4) a health-care team for decisions and care; 5) a quality assurance structure; and 6) a method for identifying cancer occurrence and death in the population (IARC 2005). Such organised population-based screening programmes have a predefined specific population, according to epidemiological data and on the basis of target age and geographical area, and during all the stages, from the invitation of the eligible individuals to the assessment procedures following testing, a specific protocol is followed. As mentioned, quality aspects of the process can be better addressed, as for example during follow-up (Miles et al, 2004). Furthermore, such screening

It is important to have also a concrete idea regarding more abstract terms determining the usefulness and, even, the effectiveness of screening tests, and to keep such definitions in mind. Although these terms have been widely used, there is also great variation of their understanding and usage in clinical decision-making. For example, the ability of a measure or test to predict a subsequent event is a form of validity. On the basis of such a criterion we determine the predictive value of a test. The positive predictive value (PPV) in terms of detection through FOBT screening is defined as "*the percentage of people with detection of at least one lesion/adenoma/advanced adenoma/cancer at follow-up colorectal screening among those with positive tests who have attended follow-up colorectal screening*" [Adapted from the European guidelines for quality assurance in colorectal cancer screening and diagnosis. 2011. European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation], whereas a positive test is, effectively, an abnormal result leading to further investigation (i.e. colonoscopy) or the removal of a lesion, for example, according to the protocol of the organized screening programme. By alluding to a false-positive result, we effectively mean that although the test indicated disease is present this is not the case. In a true-positive test, the result is correct and the disease is really present. Similarly a false-negative test indicates a disease-free subject has been tested, but the disease is present and might remain undetected if there is no further testing, symptoms, etc., whereas a true-negative test indicates a disease-free subject has been tested. Prevalence of the disease affects not only the positive predictive value of a screening test but also its negative predictive value, i.e. the probability that the person subjected to the screening test is truly free from the disease when a negative (normal) test

Two concepts often discussed in relation to true- and false- positive and negative results are sensitivity and specificity and both terms have an important impact on the PCP decisions on which of the available tests for the early detection of CRC should be recommended. The sensitivity refers to the number of cases the test can identify or in more simple words the probability of one diagnostic test telling the truth when the disease exists. It gives us a certainty that true positives will not be missed. Specificity refers to the accuracy of the finding or in simple words the probability in telling the truth when the disease is absent. Ideally, a test should be both highly sensitive and specific. To that direction, the US Preventive Services Task Force and the Institute of Medicine (IoM) recommend the fecal occult blood test (FOBT) test, and more specifically the guaiac test (gFOBT) for screening programmes. Nevertheless, when used on its own, it has relatively low sensitivity, whereas the a combination with a more sensitive test, such as the fecal immunochemical test (FIT) could help to render screening programmes more effective (Allison, et al, 2007). Another interesting consideration, especially given the public health context of mass screening, is cost-effectiveness and how it correlates to specificity and sensitivity. Although uncertainty

programmes usually do not incur any costs for the participants.

result is obtained.

remains, the assessment of a screening program based on FIT for a one-year period in France seemed to be the most cost-effective approach (Hassan, et al, 2011). More research is necessary, as for example indicating what the best cut-off levels for colonoscopy referral are, without compromising sensitivity, to determine optimal public health approaches.

#### **3. Screening tests, guidelines for CRC and the importance of early detection**

As previously mentioned, CRC can be curable when diagnosed at an early stage. Also, CRC mostly develops from colorectal non-malignant precursor lesions, thus, rendering it a preventable disease through the removal of premalignant lesions. Systematic early detection and removal at the "adenoma-phase" can prevent the occurrence of CRC and markedly decrease overall population incidence (Winawer et al, 1993), as human colon carcinogenesis progresses to the carcinoma pathway via the dysplasia-adenoma phase.

The readers of this chapter are aware from previous chapters of this book that there are various tumour staging systems, the ones mainly used in Europe being Duke's classification and TNM (Tumour, Node, Metastasis) classification of malignant tumours, introduced by the Union Internationale Contra le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Despite the fact TNM yields greater information, there are several major issues due to the reclassification of the system. Most importantly, there seems to be great disparity between the therapeutic decision-making and the TNM staging, with multiple TNM staging versions being used in different countries and great variance in reporting. It has been argued that changes should only occur after extensive discussion within the scientific community (Quirke et al, 2010), and it is essential to note that the reporting on a nationwide scale for any given CRC screening programme should be performed on the basis of the same staging system. Lesion reporting within the frame of the screening programme should be standardized to allow for better evaluation and reporting, and, consequently, improved outcomes. TNM stages and version, frequency of CRC and distribution of TNM stages should be reported along with the presence of non-neoplastic lesions. According to the report of the EU on CRC quality guidelines (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation, 2011), without explicit criteria for the diagnosis and staging of early adenocarcinoma unnecessary radical resection would result in severe overtreatment, raising the morbidity and mortality in the context of the programmes.

Various screening technologies are currently available, from the more established Guaiac faecal occult blood tests (gFOBT) and immunochemical FOBT to sigmoidoscopy and colonoscopy, as well as combinations (i.e. combined FOBT with sigmoidoscopy) to the new screening technologies, as CT colonography, stool DNA and capsule endoscopy. Early reports during the previous decade suggested that biennial screening by FOBT reduces CRC mortality, as for example in the French (Faivre et al, 2004) and Danish populations (Jørgensen et al, 2002). This fact lead WHO (World Health Organization) and OMED (World Organization for Digestive Endoscopy) to suggest a choice of FOBT that should take into account dietary compliance to recommendations, but also colonoscopy resources (Young et al, 2002). Prior to any recommendation for screening tests for CRC, the PCPs should be able to recognise whether the individual visiting the practice/office is at average, increased or at high risk for CRC. In other words, to be able to identify whether the particular individual truly belongs to the target population of the screening programme or –should there not be one available– whether there is reason for referral in the context of opportunistic screening.

Early Detection of Colorectal Cancer and Population Screening Tests 51

summarises its recommendations and recommends screening for CRC using an FOBT, sigmoidoscopy, or colonoscopy in adults beginning at the age of 50 years and continuing until the age of 75 (Grade: A Recommendation). However, US PSTF recommends against screening for CRC in adults over 85 (Grade: D Recommendation), while it concludes that the evidence is insufficient to assess the benefits of CT colonography and faecal DNA testing for CRC as screening modalities. Judging the benefits against harms, the American Task Force discusses among the benefits of the less invasive CRS screening the number of colonoscopies that may be reduced. However, it recommends that for any positive test there is a follow-up with colonoscopy. The Task Force Recommendation statement underlines that the benefits of CRC detection and early intervention decline at the age of 75 years, thus it leaves the decision for a routine screening at individual level. There is, as described, a lot of information, but slightly conflicting evidence and advice. Nevertheless, participation of the 50-75 years age group increased by 13.1% reaching 65.4%, whereas a significant CRC incidence decline was noted in 35 states and mortality declined in 49 states and DC (CDC, 2011). Further efforts are currently being made in the field of patient engagement and patient-reported outcomes, and in the context of comparative effectiveness research (PCORI,

In the United Kingdom, Cancer Research UK underlines the importance of screening for the reduction of CRC mortality and has elaborated upon the role of FOBT and flexible sigmoidoscopy (Cancer Research UK, 2011). This institute refers to evidence provided by four RCTs where the use of FOBT every two years reduced CRC mortality by 15% to 18% in people aged 45-74 years. Centralised systems, such as the Australian and, to a certain extent, the UK system, remove pressure from the individual GP and the organisational capacity at practice level, but could potentially result in a loss of involvement and a lowered feeling of

An individual is at increased risk when s/he has a personal and family history of CRC or adenomatous polyps but without reporting any of the high-risk familiar syndromes. Those hereditary syndromes include: the hereditary non-polyposis CRC (HNPCC), the familiar adenomatous polyposis (FPP) and the attenuated PAP (APAP). In this group, the clinical physicians and the PCP should change their strategy from screening to regular surveillance, and the tests that primarily detect cancer should be replaced by more sensitive diagnostic approaches and particularly colonoscopy, which should start at age 40 or younger (Sarfaty, 2008). The National Institute for Health and Clinical Excellence (UK) has recently published a new guideline on colonoscopic surveillance for the prevention of people with ulcerative colitis, Crohn's disease or adenomas (NICE, 2011). At the third category where the probability of developing CRC is high, the PCPs should be more cautious when recommending screening and surveillance. A family history of an adenomatous polypus or CRC in a relative under the age of 50 is suggesting a high probability of the presence of any of the above high-risk hereditary syndromes and the clinical physician requires genetic testing; a close collaboration with hospital specialists at a centre with expertise should be

In Europe, the high degree of heterogeneity in health care systems, policy, roles, screening programme resources and very different values in local and regional settings had previously created a rather fragmented picture. There are ongoing efforts toward harmonisation, for example, recently developed guidelines (2011) in an effort under the auspices of the European Commission, focus on quality assurance and provide clear and concise information to facilitate decision-making at the GP/FP and PCP levels. As illustrated by

2011).

responsibility.

established (Sarfaty, 2008).

In Australia, the Department of Health and Aging has issued clinical practice guidelines for the prevention, early detection and management, and national population-based screening programs are in place for various cancers, from breast cervix and to bowel (CRC) (Australian DoHA, 2011). Also, the National Bowel Cancer Screening Program Register plays an important role in the programme, as it assists participants through the screening pathway, allows for reminders and follow-ups without taxing local resources and GPs. There are online tools and decision-aids available to GPs and information in twenty languages targeted at patients. Pre-invitation, invitation, follow-up letters, FOBT kit instructions and an information booklet are all provided in all of these languages and are also available online. Additionally, a qualitative evaluation of opinions, attitudes and behaviours influencing CRC were examined in the pilot phase of the national screening programme and the report was published and integrated in future planning [A Qualitative Evaluation of Opinions, Attitudes and Behaviours Influencing the Bowel Cancer Screening Pilot Program: Final Report August 2005]. Most interestingly, the invitation is sent directly to the candidate participant and it is not necessary to nominate a physician in the forms submitted, although participants are encouraged to nominate a doctor in the context of follow-up if the FOBT is positive:


The American Cancer Society (ACS) and the National Colorectal Cancer (NCC) in cooperation with the Thomas Jefferson University have edited a Primary Care Clinician's Evidence-Based Toolbox and Guide (Sarfaty, 2008). According to this guide, an individual is at an average risk when s/he has no first-degree relatives with a history of either CRC or adenomatous polyps and no illness or past health problems have been reported (Sarfaty, 2008). For individuals at average risk, GPs are recommended to initially take a medical history, including age, symptoms, family medical history, and also individual history with a focus on bowel diseases and dietary habits, and to perform a clinical examination including a digital rectal examination. Also, various CRC screening guidelines and recommendations have been issued, both by national and international organisations and institutions. In 2008, a joint effort of the ACS and the American Gastroenterology Association was released regarding certain modalities including stool tests, flexible sigmoidoscopy (FS), colonoscopy (CS), double-contrast barium enema (DCBE), computer tomography, colonography (CTC) (McFarland, et al 2008). Those joint guidelines also stressed the importance of prevention of CRC important tasks for PCPs. The US Preventive Services Task Force (US PSTF) recommends routine asymptomatic screening for three cancer sites, including that of breast, CRC and cervix, mainly because they are asymptomatic to a high degree in early staging, have a high 5-year survival rate when the cancer is localised, and as there is a strong evidence on the screening effectiveness (Cardarelli, 2010).

In terms of a recommended start and stop age for screening, the ACS has issued guidelines for the early detection of CRC and polyps with recommended screening beginning at age 50 for both men and women (ACS, 2011). The US PSTF recommends a screening for averagerisk men and women 50 years of age and older, with colonoscopy every 10 years, flexible sigmoidoscopy or DCBE every five years and faecal occult blood test every year (U.S. Preventive Services Task Force, 2011). In a supporting document, this Task Force

In Australia, the Department of Health and Aging has issued clinical practice guidelines for the prevention, early detection and management, and national population-based screening programs are in place for various cancers, from breast cervix and to bowel (CRC) (Australian DoHA, 2011). Also, the National Bowel Cancer Screening Program Register plays an important role in the programme, as it assists participants through the screening pathway, allows for reminders and follow-ups without taxing local resources and GPs. There are online tools and decision-aids available to GPs and information in twenty languages targeted at patients. Pre-invitation, invitation, follow-up letters, FOBT kit instructions and an information booklet are all provided in all of these languages and are also available online. Additionally, a qualitative evaluation of opinions, attitudes and behaviours influencing CRC were examined in the pilot phase of the national screening programme and the report was published and integrated in future planning [A Qualitative Evaluation of Opinions, Attitudes and Behaviours Influencing the Bowel Cancer Screening Pilot Program: Final Report August 2005]. Most interestingly, the invitation is sent directly to the candidate participant and it is not necessary to nominate a physician in the forms submitted, although participants are encouraged to nominate a doctor in the context of

a. If no doctor is nominated, the FOBT results will only be sent to the participant.

b. If a doctor is nominated, the results of the FOBT will be sent to the participant and their

c. If the FOBT result is positive it is explained that it will be necessary to discuss the result

The American Cancer Society (ACS) and the National Colorectal Cancer (NCC) in cooperation with the Thomas Jefferson University have edited a Primary Care Clinician's Evidence-Based Toolbox and Guide (Sarfaty, 2008). According to this guide, an individual is at an average risk when s/he has no first-degree relatives with a history of either CRC or adenomatous polyps and no illness or past health problems have been reported (Sarfaty, 2008). For individuals at average risk, GPs are recommended to initially take a medical history, including age, symptoms, family medical history, and also individual history with a focus on bowel diseases and dietary habits, and to perform a clinical examination including a digital rectal examination. Also, various CRC screening guidelines and recommendations have been issued, both by national and international organisations and institutions. In 2008, a joint effort of the ACS and the American Gastroenterology Association was released regarding certain modalities including stool tests, flexible sigmoidoscopy (FS), colonoscopy (CS), double-contrast barium enema (DCBE), computer tomography, colonography (CTC) (McFarland, et al 2008). Those joint guidelines also stressed the importance of prevention of CRC important tasks for PCPs. The US Preventive Services Task Force (US PSTF) recommends routine asymptomatic screening for three cancer sites, including that of breast, CRC and cervix, mainly because they are asymptomatic to a high degree in early staging, have a high 5-year survival rate when the cancer is localised, and as there is a strong

In terms of a recommended start and stop age for screening, the ACS has issued guidelines for the early detection of CRC and polyps with recommended screening beginning at age 50 for both men and women (ACS, 2011). The US PSTF recommends a screening for averagerisk men and women 50 years of age and older, with colonoscopy every 10 years, flexible sigmoidoscopy or DCBE every five years and faecal occult blood test every year (U.S. Preventive Services Task Force, 2011). In a supporting document, this Task Force

follow-up if the FOBT is positive:

evidence on the screening effectiveness (Cardarelli, 2010).

doctor.

with a doctor.

summarises its recommendations and recommends screening for CRC using an FOBT, sigmoidoscopy, or colonoscopy in adults beginning at the age of 50 years and continuing until the age of 75 (Grade: A Recommendation). However, US PSTF recommends against screening for CRC in adults over 85 (Grade: D Recommendation), while it concludes that the evidence is insufficient to assess the benefits of CT colonography and faecal DNA testing for CRC as screening modalities. Judging the benefits against harms, the American Task Force discusses among the benefits of the less invasive CRS screening the number of colonoscopies that may be reduced. However, it recommends that for any positive test there is a follow-up with colonoscopy. The Task Force Recommendation statement underlines that the benefits of CRC detection and early intervention decline at the age of 75 years, thus it leaves the decision for a routine screening at individual level. There is, as described, a lot of information, but slightly conflicting evidence and advice. Nevertheless, participation of the 50-75 years age group increased by 13.1% reaching 65.4%, whereas a significant CRC incidence decline was noted in 35 states and mortality declined in 49 states and DC (CDC, 2011). Further efforts are currently being made in the field of patient engagement and patient-reported outcomes, and in the context of comparative effectiveness research (PCORI, 2011).

In the United Kingdom, Cancer Research UK underlines the importance of screening for the reduction of CRC mortality and has elaborated upon the role of FOBT and flexible sigmoidoscopy (Cancer Research UK, 2011). This institute refers to evidence provided by four RCTs where the use of FOBT every two years reduced CRC mortality by 15% to 18% in people aged 45-74 years. Centralised systems, such as the Australian and, to a certain extent, the UK system, remove pressure from the individual GP and the organisational capacity at practice level, but could potentially result in a loss of involvement and a lowered feeling of responsibility.

An individual is at increased risk when s/he has a personal and family history of CRC or adenomatous polyps but without reporting any of the high-risk familiar syndromes. Those hereditary syndromes include: the hereditary non-polyposis CRC (HNPCC), the familiar adenomatous polyposis (FPP) and the attenuated PAP (APAP). In this group, the clinical physicians and the PCP should change their strategy from screening to regular surveillance, and the tests that primarily detect cancer should be replaced by more sensitive diagnostic approaches and particularly colonoscopy, which should start at age 40 or younger (Sarfaty, 2008). The National Institute for Health and Clinical Excellence (UK) has recently published a new guideline on colonoscopic surveillance for the prevention of people with ulcerative colitis, Crohn's disease or adenomas (NICE, 2011). At the third category where the probability of developing CRC is high, the PCPs should be more cautious when recommending screening and surveillance. A family history of an adenomatous polypus or CRC in a relative under the age of 50 is suggesting a high probability of the presence of any of the above high-risk hereditary syndromes and the clinical physician requires genetic testing; a close collaboration with hospital specialists at a centre with expertise should be established (Sarfaty, 2008).

In Europe, the high degree of heterogeneity in health care systems, policy, roles, screening programme resources and very different values in local and regional settings had previously created a rather fragmented picture. There are ongoing efforts toward harmonisation, for example, recently developed guidelines (2011) in an effort under the auspices of the European Commission, focus on quality assurance and provide clear and concise information to facilitate decision-making at the GP/FP and PCP levels. As illustrated by

Early Detection of Colorectal Cancer and Population Screening Tests 53

evidence indicates that iFOBT is superior to gFOBT with respect to detection rate and

Given the lack of additional evidence, the interval for iFOBT screening can best be set at

In the absence of additional evidence, the age range for a screening programme with iFOBT can be based on the limited evidence for the optimal age range in gFOBT trials

There is reasonable evidence from one large RCT that flexible sigmoidoscopy (FS) screening reduces CRC incidence and mortality if performed in an organised screening programme with careful monitoring of the quality and systematic evaluation of the

The available evidence suggests that the optimal interval for FS screening should not be

There is limited evidence suggesting that the best age range for FS screening should be between 55 and 64 years. After age 74, average-risk FS screening should be discontinued,

Limited evidence exists on the efficacy of colonoscopy screening in reducing CRC incidence and mortality. However, recent studies suggest that colonoscopy screening

Limited available evidence suggests that the optimal interval for colonoscopy screening

Indirect evidence suggests that the prevalence of neoplastic lesions in the population below 50 years of age is too low to justify colonoscopic screening, while in the elderly population (75 years and above) lack of benefit could be a major issue. The optimal age for a single colonoscopy appears to be around 55 years. Average risk colonoscopy screening

The impact on CRC incidence and mortality of combining sigmoidoscopy screening with annual or biennial FOBT has not yet been evaluated in trials. There is currently no evidence for extra benefit from adding a once-only FOBT to sigmoidoscopy screening

There currently is no evidence on the effect new screening tests under evaluation on CRC incidence and mortality. New screening technologies such as CT colonography, stool DNA testing and capsule endoscopy should therefore not be used for screening the average-risk

Costs per life-year gained for both FOBT and endoscopy screening strategies are well

Available studies differ with respect to what screening strategies are most cost-effective. No recommendation of one screening strategy over the others can be made based on the

below the commonly-used threshold of US\$ 50 000 per life-year gained (LYG) There is some evidence that iFOBT is a cost-effective alternative to gFOBT

might not be as effective in the right colon as in other segments of the colorectum

should not be performed before age 50 and should be discontinued after age 74

should not be less than 10 years and may even extend up to 20 years

positive predictive value for adenomas and cancer

that of gFOBT, and should not exceed three years

less than 10 years and may even be extended to 20 years

given the increasing co-morbidity in this age range

outcomes, adverse effects and costs

*Combination of FOBT and sigmoidoscopy*

*New screening technologies under evaluation*

available evidence of cost-effectiveness Table 1. Recommendations and conclusions

*Sigmoidoscopy* 

*Colonoscopy* 

population *Cost-effectiveness*

Table 1, an evidenced-based brief overview of various conventional screening methods is given, although new technologies resulting in more modern forms of screening are not assessed for lack of evidence; some information regarding cost-effectiveness is provided along with the recommendations and examined in more details in the report (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation, 2011)

Finally, it is important to underline that early detection is directly dependent on acceptance of the screening test by both provider and patient, as well as the uptake of the screening programme. For example, early versions of stool DNA (sDNA) testing lacked the requisite sensitivity and markers, but improved sDNA tests are now available. It is important to understand patient preferences regarding screening options for selecting the right tool for a given population; for example, whether a non-invasive test is preferred to colonoscopy or whether accuracy is considered much more important than discomfort. In a study by Schroy et al, (2002), those preferring colonoscopy to sDNA or FOBT rated accuracy as the most important factor, whereas those rating concerns about discomfort or frequency of testing as the most important parameter preferred sDNA. Most subjects preferred a shared (54%) or patient-dominant (34%) decision-making process.

As previously highlighted, removal of all adenomas, without accurately distinguishing between those which will become malignant and those which will not, will effectively result in excessive overtreatment, and it is for this reason that newer screening tests, such as the sDNA, focusing on genomic changes affecting associated biological and metabolic processes should not be overlooked as options necessitating further research -particularly because of their potential to avoid iatrogenic care, but also because they might better reflect patient preferences for certain population groups (Sillars-Hardebol, et al, 2012).

[Adapted from the European guidelines for quality assurance in colorectal cancer screening and diagnosis. 2011. European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation.]

*Guaiac FOBT*

There is good evidence that invitation to screening with FOBT using the guaiac test reduces mortality from colorectal cancer (CRC) by approximately 15% in average risk populations of appropriate age

RCTs have only investigated annual and biennial screening with guaiac FOBT (gFOBT) (II). To ensure effectiveness of gFOBT screening, the screening interval in a national screening programme should not exceed two years

Circumstantial evidence suggests that mortality reduction from gFOBT is similar in different age ranges between 45 and 80 years. The age range for a national screening programme should at least include 60 to 64 years in which CRC incidence and mortality are high and life expectancy is still considerable. From there the age range could be expanded to include younger and older individuals, taking into account the balance between risk and benefit and the available resources

*Immunochemical FOBT*

There is reasonable evidence from an RCT that iFOBT screening reduces rectal cancer mortality, and from case control studies that it reduces overall CRC mortality; Additional evidence indicates that iFOBT is superior to gFOBT with respect to detection rate and positive predictive value for adenomas and cancer

Given the lack of additional evidence, the interval for iFOBT screening can best be set at that of gFOBT, and should not exceed three years

In the absence of additional evidence, the age range for a screening programme with iFOBT can be based on the limited evidence for the optimal age range in gFOBT trials

*Sigmoidoscopy* 

52 Colorectal Cancer – From Prevention to Patient Care

Table 1, an evidenced-based brief overview of various conventional screening methods is given, although new technologies resulting in more modern forms of screening are not assessed for lack of evidence; some information regarding cost-effectiveness is provided along with the recommendations and examined in more details in the report (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency

Finally, it is important to underline that early detection is directly dependent on acceptance of the screening test by both provider and patient, as well as the uptake of the screening programme. For example, early versions of stool DNA (sDNA) testing lacked the requisite sensitivity and markers, but improved sDNA tests are now available. It is important to understand patient preferences regarding screening options for selecting the right tool for a given population; for example, whether a non-invasive test is preferred to colonoscopy or whether accuracy is considered much more important than discomfort. In a study by Schroy et al, (2002), those preferring colonoscopy to sDNA or FOBT rated accuracy as the most important factor, whereas those rating concerns about discomfort or frequency of testing as the most important parameter preferred sDNA. Most subjects preferred a shared (54%) or

As previously highlighted, removal of all adenomas, without accurately distinguishing between those which will become malignant and those which will not, will effectively result in excessive overtreatment, and it is for this reason that newer screening tests, such as the sDNA, focusing on genomic changes affecting associated biological and metabolic processes should not be overlooked as options necessitating further research -particularly because of their potential to avoid iatrogenic care, but also because they might better reflect patient

[Adapted from the European guidelines for quality assurance in colorectal cancer screening and diagnosis. 2011. European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health

There is good evidence that invitation to screening with FOBT using the guaiac test reduces mortality from colorectal cancer (CRC) by approximately 15% in average risk

RCTs have only investigated annual and biennial screening with guaiac FOBT (gFOBT) (II). To ensure effectiveness of gFOBT screening, the screening interval in a national

Circumstantial evidence suggests that mortality reduction from gFOBT is similar in different age ranges between 45 and 80 years. The age range for a national screening programme should at least include 60 to 64 years in which CRC incidence and mortality are high and life expectancy is still considerable. From there the age range could be expanded to include younger and older individuals, taking into account the balance

There is reasonable evidence from an RCT that iFOBT screening reduces rectal cancer mortality, and from case control studies that it reduces overall CRC mortality; Additional

preferences for certain population groups (Sillars-Hardebol, et al, 2012).

for Health and Consumers, World Health Organisation, 2011)

patient-dominant (34%) decision-making process.

Organisation.] *Guaiac FOBT*

populations of appropriate age

*Immunochemical FOBT*

screening programme should not exceed two years

between risk and benefit and the available resources

There is reasonable evidence from one large RCT that flexible sigmoidoscopy (FS) screening reduces CRC incidence and mortality if performed in an organised screening programme with careful monitoring of the quality and systematic evaluation of the outcomes, adverse effects and costs

The available evidence suggests that the optimal interval for FS screening should not be less than 10 years and may even be extended to 20 years

There is limited evidence suggesting that the best age range for FS screening should be between 55 and 64 years. After age 74, average-risk FS screening should be discontinued, given the increasing co-morbidity in this age range

*Colonoscopy* 

Limited evidence exists on the efficacy of colonoscopy screening in reducing CRC incidence and mortality. However, recent studies suggest that colonoscopy screening might not be as effective in the right colon as in other segments of the colorectum

Limited available evidence suggests that the optimal interval for colonoscopy screening should not be less than 10 years and may even extend up to 20 years

Indirect evidence suggests that the prevalence of neoplastic lesions in the population below 50 years of age is too low to justify colonoscopic screening, while in the elderly population (75 years and above) lack of benefit could be a major issue. The optimal age for a single colonoscopy appears to be around 55 years. Average risk colonoscopy screening should not be performed before age 50 and should be discontinued after age 74

*Combination of FOBT and sigmoidoscopy*

The impact on CRC incidence and mortality of combining sigmoidoscopy screening with annual or biennial FOBT has not yet been evaluated in trials. There is currently no evidence for extra benefit from adding a once-only FOBT to sigmoidoscopy screening

*New screening technologies under evaluation*

There currently is no evidence on the effect new screening tests under evaluation on CRC incidence and mortality. New screening technologies such as CT colonography, stool DNA testing and capsule endoscopy should therefore not be used for screening the average-risk population

*Cost-effectiveness*

Costs per life-year gained for both FOBT and endoscopy screening strategies are well below the commonly-used threshold of US\$ 50 000 per life-year gained (LYG)

There is some evidence that iFOBT is a cost-effective alternative to gFOBT

Available studies differ with respect to what screening strategies are most cost-effective. No recommendation of one screening strategy over the others can be made based on the available evidence of cost-effectiveness

Table 1. Recommendations and conclusions

Early Detection of Colorectal Cancer and Population Screening Tests 55

Another important task for PCPs and other practitioners is to educate their patients/clients to contact the PHC services when some warning signs are experienced and among them are

Prior to the decision of the PCP to refer the subjects to either CS/FT or CT should be explored the access to that screening method and consider the existing diagnostic capacity

Finally, another essential consideration in the PCP decision to implement screening tests for early detection of CRC is that of quality of life. Quality of life in evidence-based medicine should always reflect the preferences of patients, as patient-centeredness is its cornerstone. Despite the fact everyone values particular aspects of life differently, all aspects of life that may be affected adversely or in a beneficial manner by aspects of health and illness should be taken into consideration. For screening programmes, it is important to understand the cultural context in which it is performed or is to be performed and to ensure the values of the patients are taken into consideration when determining and/or

As mentioned above, the CRC screening rate increase does not seem to apply in many countries and regions and the associated obstacles and barriers that have already been

 Obstacles at doctor level: Obstacles reported by the GPs were relevant to the difficulties in being convinced especially when signs and symptoms were lacking. There was, in other words, confusion in addressing difficulties stemming from conflict between personal experiences and public health implications (Aubin-Auger et al, 2011). Also, there is research indicating that even in countries with established screening programmes only 50% of the GPs considered themselves to be sufficiently trained, as

 Obstacles at patient level: Researchers examined obstacles at patient level and how these were linked to the physician-patient interaction and communication. For example, cancer screening did not fall in with the perception of some patients regarding health care, and they failed to identify benefits outside the context of familiar high-risk groups. Potentially inadvertently reflecting specificity and sensitivity issues, participants were afraid of poor technical skills, and taking ownership of the risk for performing the test, resulting in false positive or false negative results (Aubin-Auger et al. 2011). Mirroring the high number of GPs who do not feel they are sufficiently trained, patients cited the absence of recommendation as one of the most important reasons for not undergoing

 Obstacles at doctor-patient level: GPs and patients agreed the lack of symptoms and lack of familial risk were two of the main reasons for doubting how useful such a test could be, the GPs thought that the patients misunderstood the process and were

**5. Obstacles to implementing CRC screening in primary care** 

for example in France (Viguier et al, 2011).

screening (Viguier et al, 2011).

reported in the literature (Lionis and Petelos, 2011) could be classified as follows:

(Rudy and Zdon, 2000): - Hematochezia - Melaena


assessing outcomes.

**Obstacles in primary care** 


resources (Sarfaty and Wender, 2007).

Finally, the concepts of colonoscopic surveillance and screening for recurrent CRC should receive attention by PCPs. The adenomatous precursors of CRC are present in over 30% of individuals over 55 (Eide, 1991), placing them at higher risk of developing CRC, but the removal of these lesions reduces risk to that of the general population (Citarda et al, 2001). Recurrent CRC, as for example following resection, also necessitates an intensive surveillance programme, as the detection at an asymptomatic stage can result in survival benefit (Renehan, et al, 2002). This means that surveillance and follow-up programmes should also be combined or evaluated along with a screening programme.

#### **4. Primary care and CRC: Tasks and steps for screening implementation in primary care**

One of the most important factors for the effective implementation of a CRC screening programme is the involvement of a PCP, particularly of the GP or the FP, in convincing targeted individuals to participate and to initiate the screening. The PCPs have multiple and varying tasks, more specifically to (Sarfaty, 2008):


One of the most challenging issues that the PCPs encounter is to convince the average risk individual to use a simple and inexpensive test to initially detect if any hidden blood is present in stools, constituting a strong indication of the presence of an adenomatous polyp or CRC. To achieve it, an effective doctor-patient communication should be established, and the purpose of the GPs/PCPs might also need to be re-assessed by further education or training on early diagnosis, prevention and promotion. The role of a multidisciplinary team is also essential. FOB Testing serves this role, although it has received criticism because of the lack of specificity, particularly when the test is dehydrated, and because of the subsequent increase of the associated costs of screening programmes (WHO, Rudy and Zdon, 2000). FIT, also, fulfils this purpose; it is a simple procedure: the stool sample is collected by the individual/patient at home, and the completed test is sent to a laboratory or to the PCPs office. Usually two samples from different bowel movements are required and the instructions on sampling procedures on how the water sample should be transferred by the brush onto the test card are clear and readily understood.

Another important task for PCPs and other practitioners is to educate their patients/clients to contact the PHC services when some warning signs are experienced and among them are (Rudy and Zdon, 2000):


54 Colorectal Cancer – From Prevention to Patient Care

Finally, the concepts of colonoscopic surveillance and screening for recurrent CRC should receive attention by PCPs. The adenomatous precursors of CRC are present in over 30% of individuals over 55 (Eide, 1991), placing them at higher risk of developing CRC, but the removal of these lesions reduces risk to that of the general population (Citarda et al, 2001). Recurrent CRC, as for example following resection, also necessitates an intensive surveillance programme, as the detection at an asymptomatic stage can result in survival benefit (Renehan, et al, 2002). This means that surveillance and follow-up programmes

**4. Primary care and CRC: Tasks and steps for screening implementation in** 

One of the most important factors for the effective implementation of a CRC screening programme is the involvement of a PCP, particularly of the GP or the FP, in convincing targeted individuals to participate and to initiate the screening. The PCPs have multiple and

1. Assess the risk of developing CRC and increase the risk awareness, as described

2. Discuss options with patients/individuals and effectively engage in shared decisionmaking (SDM) – this would ensure patient perspectives and preferences are consistent

3. Convince to participate – this task requires communication and consultation skills, as

4. Implement the initial tests: those primarily used to detect cancer, including the annual Guaiac-based occult blood test (gFOBT), the annual faecal immunochemical test (FIT),

5. Consider and assess the available screening resources and capacity: it is an important task for PCPs, who should be aware of the available resources in their district or health region capacity, as well as patient limitations (e.g. socioeconomic, mobility, etc.), to determine the optimal referral pathway for the test(s) that detect adenomatous polyps

One of the most challenging issues that the PCPs encounter is to convince the average risk individual to use a simple and inexpensive test to initially detect if any hidden blood is present in stools, constituting a strong indication of the presence of an adenomatous polyp or CRC. To achieve it, an effective doctor-patient communication should be established, and the purpose of the GPs/PCPs might also need to be re-assessed by further education or training on early diagnosis, prevention and promotion. The role of a multidisciplinary team is also essential. FOB Testing serves this role, although it has received criticism because of the lack of specificity, particularly when the test is dehydrated, and because of the subsequent increase of the associated costs of screening programmes (WHO, Rudy and Zdon, 2000). FIT, also, fulfils this purpose; it is a simple procedure: the stool sample is collected by the individual/patient at home, and the completed test is sent to a laboratory or to the PCPs office. Usually two samples from different bowel movements are required and the instructions on sampling procedures on how the water sample should be transferred by

6. To make the necessary arrangement to complete the CRC screening.

the brush onto the test card are clear and readily understood.

should also be combined or evaluated along with a screening programme.

varying tasks, more specifically to (Sarfaty, 2008):

well as an established continuity of care.

**primary care** 

above.

to decisions made.

or stool DNA test (sDNA).

and CRC (FS, CS, DCBE, CRC).


Prior to the decision of the PCP to refer the subjects to either CS/FT or CT should be explored the access to that screening method and consider the existing diagnostic capacity resources (Sarfaty and Wender, 2007).

Finally, another essential consideration in the PCP decision to implement screening tests for early detection of CRC is that of quality of life. Quality of life in evidence-based medicine should always reflect the preferences of patients, as patient-centeredness is its cornerstone. Despite the fact everyone values particular aspects of life differently, all aspects of life that may be affected adversely or in a beneficial manner by aspects of health and illness should be taken into consideration. For screening programmes, it is important to understand the cultural context in which it is performed or is to be performed and to ensure the values of the patients are taken into consideration when determining and/or assessing outcomes.

#### **5. Obstacles to implementing CRC screening in primary care**

#### **Obstacles in primary care**

As mentioned above, the CRC screening rate increase does not seem to apply in many countries and regions and the associated obstacles and barriers that have already been reported in the literature (Lionis and Petelos, 2011) could be classified as follows:


Early Detection of Colorectal Cancer and Population Screening Tests 57

context in which health care delivery and provision are taking place, the practice itself, and the structural and operational characteristics of given settings, and also examined the provider and patient levels. The researchers concluded that although a given intervention may be implemented at one or multiple levels, the factors determining uptake and participation are, indeed, correlated with all of these levels in an interconnected and

In concluding this section, the role of the PCPs is extremely complex, and although research on obstacles, barriers and limitations is starting to create a more robust evidence base, further qualitative and translational research is required to identify best practices and intervention transferability. Additionally, policy measures for the purpose of supporting screening delivery mechanisms are required, and, similarly, policy should aim to facilitate the organisational changes necessary for creating and supporting the operational features of

Although messages about the effectiveness of CRC screening have been widely available, there are still concerns in terms of both physician involvement and PC user participation in CRC screening. This is not a message that concerns CRC screening per se, but prevention and health promotion activities undertaken by GPs in Europe. There are significant gaps between GP knowledge and practice in Europe, already reported upon (Brotons et al, 2005). Evidence from the literature indicates that less than one third of the PC physicians use chart reminders and 15% use outreach mechanisms to contact patients needing screening (Klabunde et al, 2009). Investment has been made on efforts and research programs to assess the impact of quality improvement intervention programs. One of them combined diverse components, such as performance activities, delivery system design, electronic medical record tools and patient activation (Ornstein et al, 2010), and reported promising results in the Evidence-Based Toolbox and Guide we currently have (Sarfaty, 2008). Thus, the implementation of educational programmes for PCPs and patients in addition to the development of shared-decision making tools, given the differing perspectives between doctor and patient, seem imperative, as otherwise lack of consensus could adversely impact

A page invitation to the health practitioners to avoid certain errors has been made and



A toolkit for a systematic approach in tracking and increasing screening for public health improvement of CRC intervention was prepared for the Agency for Health Care Research and Quality (AHRQ). It delivers tools, process guidelines, tips and evidence of the intervention effectiveness (Harris et al, 2010). It is strongly recommended for PCPs, health care planners and managers. However, the role of PCPs in increasing the CRC screening rate remains a key component. According to the ACS and NCS, the positive impact of its

interdependent manner.

preventive services.

**6. Increasing the CRC screening rate** 

the CRC screening rate (Schroy et al, 2011).

among them the following:

blood test

years

afraid of reactions to false negative results, whereas the patients complained about time, as well as the constipation effect from repeating the test, and did not express fears about such results (Aubin-Auger et al, 2011). Further evidence (Schroy et al, 2011) indicates that screening intentions and test ordering are adversely affected when patient and provider preferences differ. Interestingly, compounding previously reported data (Serra et al, 2008), having a screening habit (e.g. mammography) proved to be a positive factor for women, whereas increased participation was reported for those with a higher educational level, particularly for male patients. Without diminishing the importance of facilitators, a patient having a relative having already performed gFOBT was more likely to accept the test, but friends and family were not identified as obstacles.

#### **Further barriers:**

Cultural and linguistic barriers were also touched upon by these researchers, but not explored in detail; it is highlighted that even the wording a doctor uses has an effect and that further research is necessary (Lionis and Petelos, 2011). There is evidence that by employing culturally and linguistically relevant approaches for FOBT promotion, screening participation increases in target populations of low-income and/or less acculturated minority patients (Tu et al, 2006). Indeed, a challenge of equal significance to guideline adherence and compliance in screening is ensuring equity of access to screening. Part of ensuring equity of access is to ensure awareness issues have been addressed for all ethnic and culturally diverse groups. A study of all the patients aged 50-60 registered in general practices for a UK region (West Midlands), with a total number of over eleven thousand respondents, examined factors that contributed positively or negatively on behaviour toward screening (Taskila et al, 2009). People without a screening habit (men), older people, and those with Indian ethnic backgrounds were more likely to have negative attitudes, whereas Black-Caribbean ethnic background people reporting abdominal pain, bleeding or tiredness were more likely to have a positive attitude. This great variation in attitudes indicates that there are different needs to be addressed for increasing awareness and highlight the importance of culturally relevant strategies for designing and implementing screening programmes (Taskila et al, 2009). Evidence amasses from various countries, with a study focusing specifically on FOBT use, along with the subsequent investigation of a positive result (Bampton et al, 2005). Researchers established that both indications for use and follow-up of a positive result varied according to the ethnicity of the GP and independently of the medical training received (Koo et al, 2011). Additionally, it was indicated that the ethnicity of the patient and, similarly to results of other research, associated linguistic and cultural barriers affect screening uptake and was noted that this may adversely affect the health of immigrant populations.

To address all the obstacles and barriers previously mentioned, it is necessary to embrace the perspective of the users of screening programmes, and also to examine screening under the prism of public health perspective. A recently conducted review highlights the need for policy supporting both screening delivery and organisational transformation in a manner that promotes improvement of operational features for preventive services (Senore et al, 2010). The researchers examined recently proposed conceptual frameworks that were aimed at identifying key elements and, thus, potential targets for interventions aiming to improve screening (Cole et al, 2009 and Federici et al, 2005). The models developed conceptualised these potential targets at various levels: the organizational

Cultural and linguistic barriers were also touched upon by these researchers, but not explored in detail; it is highlighted that even the wording a doctor uses has an effect and that further research is necessary (Lionis and Petelos, 2011). There is evidence that by employing culturally and linguistically relevant approaches for FOBT promotion, screening participation increases in target populations of low-income and/or less acculturated minority patients (Tu et al, 2006). Indeed, a challenge of equal significance to guideline adherence and compliance in screening is ensuring equity of access to screening. Part of ensuring equity of access is to ensure awareness issues have been addressed for all ethnic and culturally diverse groups. A study of all the patients aged 50-60 registered in general practices for a UK region (West Midlands), with a total number of over eleven thousand respondents, examined factors that contributed positively or negatively on behaviour toward screening (Taskila et al, 2009). People without a screening habit (men), older people, and those with Indian ethnic backgrounds were more likely to have negative attitudes, whereas Black-Caribbean ethnic background people reporting abdominal pain, bleeding or tiredness were more likely to have a positive attitude. This great variation in attitudes indicates that there are different needs to be addressed for increasing awareness and highlight the importance of culturally relevant strategies for designing and implementing screening programmes (Taskila et al, 2009). Evidence amasses from various countries, with a study focusing specifically on FOBT use, along with the subsequent investigation of a positive result (Bampton et al, 2005). Researchers established that both indications for use and follow-up of a positive result varied according to the ethnicity of the GP and independently of the medical training received (Koo et al, 2011). Additionally, it was indicated that the ethnicity of the patient and, similarly to results of other research, associated linguistic and cultural barriers affect screening uptake and was noted that this

To address all the obstacles and barriers previously mentioned, it is necessary to embrace the perspective of the users of screening programmes, and also to examine screening under the prism of public health perspective. A recently conducted review highlights the need for policy supporting both screening delivery and organisational transformation in a manner that promotes improvement of operational features for preventive services (Senore et al, 2010). The researchers examined recently proposed conceptual frameworks that were aimed at identifying key elements and, thus, potential targets for interventions aiming to improve screening (Cole et al, 2009 and Federici et al, 2005). The models developed conceptualised these potential targets at various levels: the organizational

were not identified as obstacles.

may adversely affect the health of immigrant populations.

**Further barriers:** 

afraid of reactions to false negative results, whereas the patients complained about time, as well as the constipation effect from repeating the test, and did not express fears about such results (Aubin-Auger et al, 2011). Further evidence (Schroy et al, 2011) indicates that screening intentions and test ordering are adversely affected when patient and provider preferences differ. Interestingly, compounding previously reported data (Serra et al, 2008), having a screening habit (e.g. mammography) proved to be a positive factor for women, whereas increased participation was reported for those with a higher educational level, particularly for male patients. Without diminishing the importance of facilitators, a patient having a relative having already performed gFOBT was more likely to accept the test, but friends and family context in which health care delivery and provision are taking place, the practice itself, and the structural and operational characteristics of given settings, and also examined the provider and patient levels. The researchers concluded that although a given intervention may be implemented at one or multiple levels, the factors determining uptake and participation are, indeed, correlated with all of these levels in an interconnected and interdependent manner.

In concluding this section, the role of the PCPs is extremely complex, and although research on obstacles, barriers and limitations is starting to create a more robust evidence base, further qualitative and translational research is required to identify best practices and intervention transferability. Additionally, policy measures for the purpose of supporting screening delivery mechanisms are required, and, similarly, policy should aim to facilitate the organisational changes necessary for creating and supporting the operational features of preventive services.

#### **6. Increasing the CRC screening rate**

Although messages about the effectiveness of CRC screening have been widely available, there are still concerns in terms of both physician involvement and PC user participation in CRC screening. This is not a message that concerns CRC screening per se, but prevention and health promotion activities undertaken by GPs in Europe. There are significant gaps between GP knowledge and practice in Europe, already reported upon (Brotons et al, 2005). Evidence from the literature indicates that less than one third of the PC physicians use chart reminders and 15% use outreach mechanisms to contact patients needing screening (Klabunde et al, 2009). Investment has been made on efforts and research programs to assess the impact of quality improvement intervention programs. One of them combined diverse components, such as performance activities, delivery system design, electronic medical record tools and patient activation (Ornstein et al, 2010), and reported promising results in the Evidence-Based Toolbox and Guide we currently have (Sarfaty, 2008). Thus, the implementation of educational programmes for PCPs and patients in addition to the development of shared-decision making tools, given the differing perspectives between doctor and patient, seem imperative, as otherwise lack of consensus could adversely impact the CRC screening rate (Schroy et al, 2011).

A page invitation to the health practitioners to avoid certain errors has been made and among them the following:


A toolkit for a systematic approach in tracking and increasing screening for public health improvement of CRC intervention was prepared for the Agency for Health Care Research and Quality (AHRQ). It delivers tools, process guidelines, tips and evidence of the intervention effectiveness (Harris et al, 2010). It is strongly recommended for PCPs, health care planners and managers. However, the role of PCPs in increasing the CRC screening rate remains a key component. According to the ACS and NCS, the positive impact of its

Early Detection of Colorectal Cancer and Population Screening Tests 59

through engagement of local actors and community leaders, strong community orientation and high level of awareness of the PHPs, including most importantly nursing and all other

Additionally, elements of the care pathway ought to be assessed in context and real settings, and, especially where pragmatic trials are impossible, difficult or unethical. Participation in and completion of the screening test is not the only necessary part to ensure successful outcomes, as there has to be a follow up with the appropriate diagnostic testing. (Christou

Screening utilisation is also influenced by behavioural factors and health economics parameters, as well as by the organizational and cultural settings (Senore et al, 2010). A theoretical framework to explain the adoption of health-related behaviours is needed to underpin any given implementation effort. Models conceptualizing elements of the health care provision have been proposed, providing targets for intervention at patient and provider levels (Stone et al, 2002; Bastani et al, 2004, Senore et al, 2010). However, it is important to note that although any given intervention component may act upon more than

one levels, screening uptake is interdependent on all these factors (Senore et al, 2010).

**8. CRC screening: An issue of quality assurance in modern health care** 

Quality issues in colorectal cancer screening have been previously discussed in editorials in the journals of Quality in Primary Care and Family Practice (Lionis, 2007, Lionis and Petelos, 2011). These editorials address issues relevant to improvement of uptake of CRC screening with the use of cognitive methods and the translation of the Health Belief Model into education and training programs for health care providers. The authors call for a closer collaboration between medical and social care scientists, and reveals another important challenge that PCPs face: addressing health inequalities in a changing and financially restrained world, where for example, minority groups showcase low adoption rates of

As previously mentioned, the Directorate General for Health and Consumers, with tasks funded by the EU Health programme (CRC screening grant No 2005317), led an effort

Consideration of multidisciplinary teams is also essential. These teams can help direct resources along a predefined, according to the evidence-base, care pathway ensuring effective, efficient and sustainable implementation and, thus, better results. As we have previously discussed, CRC screening is complex and comprises of different stages involving the participation of different health care actors. For example, all abnormal results should be followed-up and after-care service following treatment should be available. Nevertheless, only a small portion of health plans monitor follow-up care (Klabunde et al, 2003). Multidisciplinary participation can help implement interventions that have added value at patient, provider and even public health-health care system level. For example same day follow-up for abnormal FS, offering on-site colonoscopy seems to lead to better compliance (Stern et al, 2000; Senore et al, 2010). Another benefit in the involvement of nursing and clerical staff is the integration of quality indicators and quality assurance mechanisms in delivery processes and their monitoring as part of standardised care delivery without

available healthcare personnel –especially in rural or remote areas.

and Thompson, 2010).

draining on valuable resources.

preventive measures and screening tests.

**systems** 

advice is well documented, and the magnitude of the doctor's impact is considerable (Sarfaty, 2008).

#### **7. Designing a national CRC screening programme/framework**

We have seen the importance and potential of CRC screening in detail. The importance of screening taking place in an organised framework for optimal results, as for example in nationwide programmes, has also been examined and noted. As many European countries are still in the process of designing such a programme and many other countries globally are far from implementing such interventions, it is important to see how to best learn from other experiences and how to use lessons already learned to help us morph a flexible and robust model that can be adapted according to regional and local needs to ensure high acceptance, uptake and, indeed, equal access and reduced disparities. Thus, we decided to include in this section some key issues that health planners, health policy makers and public health decision makers should take into account when considering the design and implementation of a CRC screening programme.

Many of the countries in which a nationwide programme is implemented have extensively reported on the outcomes and evaluation of such programmes. Australia has introduced such a programme, but reporting indicated the needs and beliefs of minority groups, as for example indigenous Australians, were not always taken into consideration, with stronger drives, as for example economic benefit at country level, determining the approach undertaken and the strategy selected (Christou and Thompson, 2010).

To start, the need of conducting a feasibility study should be evaluated. According to Bowen et al, (2009) performing a feasibility study may be indicated when *"(a) community partnerships need to be established, increased, or sustained, (b) there are few previously published studies or existing data using a specific intervention technique, (c) prior studies of a specific intervention technique in a specific population were not guided by in-depth research or knowledge of the population's socio-cultural health beliefs, by members of diverse research teams, or by researchers familiar with the target population and in partnership with the targeted communities, (d) the population or intervention target has been shown empirically to need unique consideration of the topic, method, or outcome in other research or (e) previous interventions that employed a similar method have not been successful, but improved versions may be successful; or previous interventions had positive outcomes but in different settings than the one of interest"*. By quickly reviewing these grounds it becomes apparent CRC programmes, independently of whether they are still being designed or already implemented, are prime candidates for qualitative research via feasibility studies. These pilot actions can help elicit patient preferences and elucidate obstacles adversely affecting participation. In terms of results, even small modifications to existing programmes or design can greatly affect outcomes. For example, a sound review being the starting point for deploying such a pilot action, recent systematic reviews of interventions in Australia, indicated that organisational level changes were the most effective in terms of screening behaviour enhancement. It is important to note necessary modifications for increased effectiveness were those that included non-physicians in the screening process (Christie et al, 2008, Wardle et al, 2003, Vernon 1997). Language and literacy barriers are, of course, the most difficult to overcome, as illustrated by the Alaskan and Australian Aboriginal examples, and can only be adequately researched and addressed

advice is well documented, and the magnitude of the doctor's impact is considerable

We have seen the importance and potential of CRC screening in detail. The importance of screening taking place in an organised framework for optimal results, as for example in nationwide programmes, has also been examined and noted. As many European countries are still in the process of designing such a programme and many other countries globally are far from implementing such interventions, it is important to see how to best learn from other experiences and how to use lessons already learned to help us morph a flexible and robust model that can be adapted according to regional and local needs to ensure high acceptance, uptake and, indeed, equal access and reduced disparities. Thus, we decided to include in this section some key issues that health planners, health policy makers and public health decision makers should take into account when considering the design and

Many of the countries in which a nationwide programme is implemented have extensively reported on the outcomes and evaluation of such programmes. Australia has introduced such a programme, but reporting indicated the needs and beliefs of minority groups, as for example indigenous Australians, were not always taken into consideration, with stronger drives, as for example economic benefit at country level, determining the approach

To start, the need of conducting a feasibility study should be evaluated. According to Bowen et al, (2009) performing a feasibility study may be indicated when *"(a) community partnerships need to be established, increased, or sustained, (b) there are few previously published studies or existing data using a specific intervention technique, (c) prior studies of a specific intervention technique in a specific population were not guided by in-depth research or knowledge of the population's socio-cultural health beliefs, by members of diverse research teams, or by researchers familiar with the target population and in partnership with the targeted communities, (d) the population or intervention target has been shown empirically to need unique consideration of the topic, method, or outcome in other research or (e) previous interventions that employed a similar method have not been successful, but improved versions may be successful; or previous interventions had positive outcomes but in different settings than the one of interest"*. By quickly reviewing these grounds it becomes apparent CRC programmes, independently of whether they are still being designed or already implemented, are prime candidates for qualitative research via feasibility studies. These pilot actions can help elicit patient preferences and elucidate obstacles adversely affecting participation. In terms of results, even small modifications to existing programmes or design can greatly affect outcomes. For example, a sound review being the starting point for deploying such a pilot action, recent systematic reviews of interventions in Australia, indicated that organisational level changes were the most effective in terms of screening behaviour enhancement. It is important to note necessary modifications for increased effectiveness were those that included non-physicians in the screening process (Christie et al, 2008, Wardle et al, 2003, Vernon 1997). Language and literacy barriers are, of course, the most difficult to overcome, as illustrated by the Alaskan and Australian Aboriginal examples, and can only be adequately researched and addressed

**7. Designing a national CRC screening programme/framework** 

undertaken and the strategy selected (Christou and Thompson, 2010).

implementation of a CRC screening programme.

(Sarfaty, 2008).

through engagement of local actors and community leaders, strong community orientation and high level of awareness of the PHPs, including most importantly nursing and all other available healthcare personnel –especially in rural or remote areas.

Additionally, elements of the care pathway ought to be assessed in context and real settings, and, especially where pragmatic trials are impossible, difficult or unethical. Participation in and completion of the screening test is not the only necessary part to ensure successful outcomes, as there has to be a follow up with the appropriate diagnostic testing. (Christou and Thompson, 2010).

Screening utilisation is also influenced by behavioural factors and health economics parameters, as well as by the organizational and cultural settings (Senore et al, 2010). A theoretical framework to explain the adoption of health-related behaviours is needed to underpin any given implementation effort. Models conceptualizing elements of the health care provision have been proposed, providing targets for intervention at patient and provider levels (Stone et al, 2002; Bastani et al, 2004, Senore et al, 2010). However, it is important to note that although any given intervention component may act upon more than one levels, screening uptake is interdependent on all these factors (Senore et al, 2010).

Consideration of multidisciplinary teams is also essential. These teams can help direct resources along a predefined, according to the evidence-base, care pathway ensuring effective, efficient and sustainable implementation and, thus, better results. As we have previously discussed, CRC screening is complex and comprises of different stages involving the participation of different health care actors. For example, all abnormal results should be followed-up and after-care service following treatment should be available. Nevertheless, only a small portion of health plans monitor follow-up care (Klabunde et al, 2003). Multidisciplinary participation can help implement interventions that have added value at patient, provider and even public health-health care system level. For example same day follow-up for abnormal FS, offering on-site colonoscopy seems to lead to better compliance (Stern et al, 2000; Senore et al, 2010). Another benefit in the involvement of nursing and clerical staff is the integration of quality indicators and quality assurance mechanisms in delivery processes and their monitoring as part of standardised care delivery without draining on valuable resources.

#### **8. CRC screening: An issue of quality assurance in modern health care systems**

Quality issues in colorectal cancer screening have been previously discussed in editorials in the journals of Quality in Primary Care and Family Practice (Lionis, 2007, Lionis and Petelos, 2011). These editorials address issues relevant to improvement of uptake of CRC screening with the use of cognitive methods and the translation of the Health Belief Model into education and training programs for health care providers. The authors call for a closer collaboration between medical and social care scientists, and reveals another important challenge that PCPs face: addressing health inequalities in a changing and financially restrained world, where for example, minority groups showcase low adoption rates of preventive measures and screening tests.

As previously mentioned, the Directorate General for Health and Consumers, with tasks funded by the EU Health programme (CRC screening grant No 2005317), led an effort

Early Detection of Colorectal Cancer and Population Screening Tests 61

Bowen D.J., Kreuter M., Spring B., Cofta-Woerpel L., Linnan L., Weiner D., Bakken S., (...),

Brotons, C., Björkelund, C., Bulc, M., Ciurana, R., Godycki-Cwirko, M., Jurgova, E., Kloppe,

Cardarelli, R. (2010). The Role of Primary Care Providers in Cancer Screening. Primary Care

http://www.centerforcommunityhealth.org/Portals/14/Reports/PCPCancerBrief

Christie, J., Itzkowitz, S., Lihau-Nkanza, I., Castillo, A., Redd, W., Jandorf, L. (2008). A

Christou, A., and Thompson, S.C. (2010). "How could the National Bowel Cancer Screening

Western Australia and Combined Universities Centre for Rural Health. Citarda, F., Tomaselli, G., Capocaccia, R., Barcherini, S., Crespi, M. (2001). The Italian

polypectomy in reducing colorectal cancer incidence. GUT. *Vol. 48:*812–815 Cole, S.R., Young G.P., Byrne, D., Guy, J.R., and Morcom, J. (2002). Participation in screening

endorsement by the primary care practitioner. *J Med Screen. Vol (9):*147**–**152. Edwards, B.K., Ward, E., Kohler, B.A., Eheman, C., Zauber, A.G., Anderson, R.N., Jemal,

European Commission, Directorate General for Health and Consumers, EAHC — Executive

 http://bookshop.europa.eu/is-bin/INTERSHOP.enfinity/WFS/EU-Bookshop-Site/en\_GB/-/EUR/ViewPublication-Start?PublicationKey=ND3210390). Faivre, J., Dancourt, V., Lejeune, C., Tazi, M.A., Lamour, J., Gerard, D., Dassonville, F.,

Eide, T.J. (1991). Natural history of adenomas. *World J Surgery. Vol. 15:3*–6.

(Accessed online September 10, 2011:

randomized controlled trial using patient navigation to increase colonoscopy screening among low-income minorities. *Journal of the National Medical Association.* 

Program for Aboriginal people in Western Australia be improved?" Report to the WA Bowel Cancer Screening Implementation Committee, Department of Health,

Multicentre Study Group: Efficacy in standard clinical practice of colonoscopic

for colorectal cancer based on a faecal occult blood test is improved by

A., Schymura, M.J., Lansdorp-Vogelaar, I., Seeff, L.C., van Ballegooijen, M., Goede, S.L., Ries, L.A. (2010). Annual report to the nation on the status of cancer, 1975-2006, featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. *Cancer. Vol. 1;116(3)*:544-

Agency for Health and Consumers, World Health Organisation. (2011). European guidelines for quality assurance in colorectal cancer screening and diagnosis.

Bonithon-Kopp, C. (2004). Reduction in colorectal cancer mortality by fecal occult blood screening in a French controlled study. *Gastroenterology. Vol. 126(7):*1674-

*Preventive Medicine. Vol. 36(5):*452-457.

Europe. *Prev Med. Vol. 40(5)*:595-601.

Final%282%29.pdf).

*Vol. 100(3):*278-284.

73.

80.

Institute. (Accessed online November 10, 2011:

Fernandez M. (2009) How We Design Feasibility Studies. *American Journal of* 

P., Lionis, C., Mierzecki, A., Piñeiro, R., Pullerits, L., Sammut, M.R., Sheehan, M., Tataradze, R., Thireos, E.A., Vuchak, J.; EUROPREV network. (2005) Prevention and health promotion in clinical practice: the views of general practitioners in

aiming to develop EU guidelines on best practice in CRC screening, which resulted in the publication of the first edition of the European guidelines for quality assurance in colorectal cancer screening and diagnosis in February of 2011 (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation). The guidelines systematically examine the evidence for efficacy and effectiveness of CRC screening and outline the guiding principles for organising CRC screening programmes. Most importantly, the authors underline the importance of the availability of comprehensive, evidence-based quality assurance guidelines that address all the steps of a screening programme, including invitation, information, surveillance and any other subsequent care, as a key factor to the success of any cancer screening programme. Finally, the authors advocate the widespread application of standardised indicators, as recommended and elaborated upon in the guidelines, to facilitate quality management and promote information exchange in the context of continuous quality improvement.

#### **9. Epilogue**

This chapter serves as an overview of the guidance available for CRC screening in the US, UK, Europe and Australia and briefly discussed the important role of GPs/FPs and PCPs, in general, in increasing the CRC screening rate. Although the literature is rich in information, guidelines and recommendation for CRC screening, there is room for improvement. It is to important invest in translating primary research into practice and combine qualitative and quantitative evidence for relevant, contextualised training and educational interventions, both at patient and provider levels.

#### **10. References and internet resources**

#### **10.1 References**


aiming to develop EU guidelines on best practice in CRC screening, which resulted in the publication of the first edition of the European guidelines for quality assurance in colorectal cancer screening and diagnosis in February of 2011 (European Commission, Directorate General for Health and Consumers, EAHC — Executive Agency for Health and Consumers, World Health Organisation). The guidelines systematically examine the evidence for efficacy and effectiveness of CRC screening and outline the guiding principles for organising CRC screening programmes. Most importantly, the authors underline the importance of the availability of comprehensive, evidence-based quality assurance guidelines that address all the steps of a screening programme, including invitation, information, surveillance and any other subsequent care, as a key factor to the success of any cancer screening programme. Finally, the authors advocate the widespread application of standardised indicators, as recommended and elaborated upon in the guidelines, to facilitate quality management and promote information exchange in the context of

This chapter serves as an overview of the guidance available for CRC screening in the US, UK, Europe and Australia and briefly discussed the important role of GPs/FPs and PCPs, in general, in increasing the CRC screening rate. Although the literature is rich in information, guidelines and recommendation for CRC screening, there is room for improvement. It is to important invest in translating primary research into practice and combine qualitative and quantitative evidence for relevant, contextualised training and educational interventions,

Allison JE, Sakoda LC, Levin TR, Tucker JP, Tekawa IS, Cuff T, Pauly MP, Shlager L, Palitz

characteristics. *J Natl Cancer Inst. Vol. 3;99(19):*1462-70. Epub 2007, Sep 25. Andermann, A., Blancquaert, I., Beauchamp, S., Déry, V.. (2008). Revisiting Wilson and

Aubin-Auger, I., Mercier, A., Lebeau. J.P., Baumann, L., Peremans, L., Van Royen, P. (2011).

Bampton, P.A., Sandford, J.J., Cole, S.R., Smith, A., Morcom, J., Cadd, B., Young, G.P. (2005).

Bastani, R., Yabroff, K.R., Myers, R.E. and Glenn, B. (2004). Interventions to improve followup of abnormal findings in cancer screening*. Cancer. Vol. 101:*1188**–**1200.

AM, Zhao WK, Schwartz JS, Ransohoff DF, Selby JV. (2007). Screening for colorectal neoplasms with new fecal occult blood tests: update on performance

Jungner in the genomic age: a review of screening criteria over the past 40 years.

Obstacles to colorectal screening in general practice: a qualitative study of GPs and patients. *Family Practice*. Advance Access: cmr020 first published online May 6,

Interval faecal occult blood testing in a colonoscopy based screening programme

continuous quality improvement.

both at patient and provider levels.

**10. References and internet resources** 

*Bull World Health Organ. Vol. 86(4):*317-9.

detects additional pathology*. Gut. Vol. 54(6):*803-6.

2011 doi:10.1093/fampra/cmr020.

**9. Epilogue** 

**10.1 References** 


 http://bookshop.europa.eu/is-bin/INTERSHOP.enfinity/WFS/EU-Bookshop-Site/en\_GB/-/EUR/ViewPublication-Start?PublicationKey=ND3210390).

Faivre, J., Dancourt, V., Lejeune, C., Tazi, M.A., Lamour, J., Gerard, D., Dassonville, F., Bonithon-Kopp, C. (2004). Reduction in colorectal cancer mortality by fecal occult blood screening in a French controlled study. *Gastroenterology. Vol. 126(7):*1674- 80.

Early Detection of Colorectal Cancer and Population Screening Tests 63

Ornstein, M., Nemeth, L.S., Jenkins, P.G., Nietert, P.J. (2010). Colorectal cancer

Pox, C., Schmiegel, W., Classen, M. (2007). Current status of screening colonoscopy in

Quirke, P., Cuvelier, C., Ensari, A., Glimelius, B., Laurberg, S., Ortiz, H., Piard, F., Punt, C.J.,

Renehan AG, Egger M, Saunders MP, O'Dwyer ST. (2002). Impact on survival of intensive

Rudy, D.R., Zdon, M.J. (2000). Update on Colorectal Cancer. *Am Fam Physician* Vol 61:1759-

Sarfaty, M. (2006) How to Increase Colorectal Cancer Screening Rates in Practice: a Primary

Sarfaty, M., Wender, R. (2007). How to increase colorectal cancer screening rates in practice.

Schroy, P.C. 3rd, Lal, S., Glick, J.T., Robinson, P.A., Zamor, P., Heeren, T.C. (2007). Patient

Schroy, P.C., Mylvaganam, S., Davidson, P. (2011). Provider perspectives on the utility of a

Senore, C., Malila, N., Minozzi, S., Armaroli, P. (2010). How to enhance physician and public

Sillars-Hardebol, A.H., Carvalho, B., van Engeland, M., Fijneman, R.J., Meijer, G.A. (2012).

Stern, M.A., Fendrick, A.M., McDonnell, W.M., Gunaratnam, N., Moseley, R., Chey, W.D. A

Stone, E.G., Morton, S.C., Hulscher, M.E., Maglione, M.A., Roth, E.A., Grimshaw, J.M.,

*Pathology. Jan;226(1):*1-6. doi: 10.1002/path.3012. Epub 2011 Nov.

http://www.cancer.org/acs/groups/content/documents/document/acspc-

Europe and in the United States. *Endoscopy 39:*168–173.

analysis of randomised trials. *BMJ. Vol. 6;324(7341):*813.

University. (Accessed online November 10, 2011:

CA *Cancer J Clin. Vol. 57(6):*354-66.

*Res Clin Gastroenterol. Vol 24(4):*509-20.

*Manag Care. Vol. 13(7):*393-400.

*Journal of Pathology. Vol. 221(4):*357-360.

48: 900-906.

70,1773-4.

024588.pdf).

7625.2011.00730.x.

*95:*2074–2079.

screening in Primary Care: Translating Research into Practice. *Medical Care* Vol

Glenthoj, A., Pennickx, F., Seymour, M., Valentini, V., Williams, G., Nagtegaal, J.D. (2010). Evidence-based medicine: the time has come to set standards for staging,

follow up after curative resection for colorectal cancer: systematic review and meta-

Care Clinician's Evidence-Based Toolbox and Guide. Atlanta, GA: The American Cancer Society, National Colorectal Cancer Roundtable and Thomas Jefferson

preferences for colorectal cancer screening: how does stool DNA testing fare? *Am J* 

colorectal cancer screening decision aid for facilitating shared decision-making. Health Expect September 8 [Epub ahead of print], doi:10.1111/j.1369-

acceptance and utilisation of colon cancer screening recommendations. *Best Pract* 

The adenoma hunt in colorectal cancer screening: defining the target. *Journal of* 

randomized, controlled trial to assess a novel colorectal cancer screening strategy: the conversion strategy–a comparison of sequential sigmoidoscopy andcolonoscopy with immediate conversion from sigmoidoscopy to colonoscopy in patients with an abnormal screening sigmoidoscopy. *Am J Gastroenterol. Vol.* 

Mittman, B.S., Rubenstein, L.V., Rubenstein, L.Z., Shekelle, P.G. (2002).


Federici A, Giorgi Rossi P, Bartolozzi F, Farchi S, Borgia P & Guasticchi G. (2005). Survey on

Harris, D.M., Borsky, A.E., Stello, B. et al. (2010*). Toolkit for the System Approach to Tracking* 

Hassan, C., Benamouzig, R., Spada, C., Ponchon, T., Zullo, A., Saurin, J.C., Costamagna, G.

Klabunde, C.N., Frame, P.S., Meadow, A., Jones, E., Nadel, M., Vernon, S.W. (2003). A

Klabunde, C.N., Lanier, D., Nadel M.R, et al. (2009*).* Colorectal Cancer Screening by primary

Koo J.H., You B., Liu K., Athureliya M.D., Tang C.W., Redmond D.M., Connor S.J., Leong R.

Levin, B., Lieberman, D.A., McFarland, et al. (2008). Screening and surveillance for the

Lionis C, and Petelos E. (2011). Early detection of colorectal cancer: barriers to screening in

Lionis, C. (2007). Colorectal cancer screening and the challenging role of general

McFarland, E.G., Levin, B., Lieberman, D.A., Pickhart, P., Johnson, C.D., Glick, S.N., Brooks,

McGregor, S.E., Hilsden, R.J., Murray, A., Bryant, H.E. (2004) Colorectal cancer screening: practices and opinions of primary care physicians. *Prev Me*d. *Vol 39(2):*279-85. Meissner, H.I., Breen, N., Klabunde, C.N., Vernon, S.W. (2006). Patterns of colorectal cancer

Miles, A., Cockburn, J., Smith, R.A., Wardle J. (2004). A perspective from countries using

the primary care setting. *Family Practice. Vol 28(6):*589-91.

American Cancer Society. *CA Cancer J Clin* Vol 2008;58:160.

organized screening programs. *Cancer. Vol. 101(5):*201-1213.

screening in France. *Endoscopy. Vol. 43(9):*780-93. Epub 2011 May 27. IARC (2005). *Cervix Cancer Screening*. IARC Handbooks of Cancer Prevention. Volume 10. Jørgensen, O.D., Kronborg, O., Fenger, C. (2002). A randomised study of screening for

physicians in Lazio, Italy. *Prev Med. Vol(41):*30**–**35.

*Intervention*. AHRQ Publication No 11-0016.

biennial screening rounds. *Gut. Vol. 50(1)*:29-32.

1746.2011.06872.x. [Epub ahead of print]

*Biomarkers Prev*. *Vol 15(2):*389-94.

*Vol.(37)*:8-16.

*58:*130-160.

*15:*129-131.

recommendations and practices*. Prev Med. Vol. 36(3):*352-62.

colorectal cancer screening knowledge, attitudes, and practices of general practice

*and Increasing Screening for Public Health Improvement of Colorectal Cancer* 

(2011). Cost effectiveness and projected national impact of colorectal cancer

colorectal cancer using faecal occult blood testing: results after 13 years and seven

national survey of primary care physicians' colorectal cancer screening

care physicians: recommendations and practices; 2006-2007. *Am J Prev Med.*

(2011). Colorectal cancer screening practices is influenced by ethnicity of medical practitioner and patient. *J Gastroenterol Hepatol.* doi: 10.1111/j.1440-

early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. *CA Cancer J. Clin*. *Vol* 

practitioner/family physician: an issue of quality. *Quality in Primary Care. Vol.* 

D., Smith, R.A. (2008). Revised colorectal screening guidelines: Joint effort of the

screening uptake among men and women in the United States. *Cancer Epidemiol* 


<sup>024588.</sup>pdf).

Early Detection of Colorectal Cancer and Population Screening Tests 65

Australian Government Department of Health and Ageing. (2008). *Clinical Practice Guidelines* 

http://www.health.gov.au/internet/screening/publishing.nsf/Content/bw-gp-

http://www.health.gov.au/internet/screening/publishing.nsf/Content/qual-

*Program*. Final Report. Screening Monograph: 2/2005.

Australian Government Department of Health and Ageing. (2005) *A Qualitative Evaluation of* 

Australian Population Health Development Principal Committee. (2008). *Population-Based* 

http://www.cancerscreening.gov.au/internet/screening/publishing.nsf/Content

http://info.cancerresearchuk.org/cancerstats/types/bowel/screeningandpreventi

Centers for Diseases Control and Prevention (US) MMWR. Vital Signs: Colorectal Cancer

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6026a4.htm?s\_cid=mm602

National Institute for Health and Clinical Excellence, (2011). NICE publishes new guidelines

http://www.nice.org.uk/newsroom/pressreleases/2011045colonoscopicsurveillan

International Agency for Research on Cancer. Early detection and Prevention - Quality

Patient-Centered Outcomes Research Institute. (2011-ongoing) Review and Synthesis of

Evidence for Eliciting the Patient's Perspective in Patient-Centered Outcomes

http://www.iarc.fr/en/researchs-groups/QAS/current-topics.php).

http://www.pcori.org/patient-centered-outcomes-research/ and

U.S. Preventive Services Task Force. (2008). *Screening for Colorectal Cancer.* 

on colonoscopic surveillance for the prevention of colorectal cancer in people with

Screening, Incidence, and Mortality: United States, 2002—2010.

*General Practitioners*. (3rd ed.). (Accessed online August 11, 2011:

(Accessed online September 16, 2011:

(Accessed online September 16, 2011:

(Accessed online September 15, 2011:

(Accessed online September 15, 2011:

(Accessed online August 9, 2011:

Assurance Group Resources. (Accessed online August 11, 2011:

(Accessed online September 19, 2011:

(Accessed online August 8, 2011:

/other-pop-health#framework).

Cancer Research UK. Bowel cancer – screening and prevention.

ulcerative colitis, Crohn's disease or adenomas.

Research. Literature review and interviews.

http://www.pcori.org/committee-charters/).

crc-guide).

eval-cnt).

*Screening*.

on/).

6a4\_w).

ce.jsp).

*for the Prevention, Early Detection and Management of Colorectal Cancer – A Guide for* 

*Opinions, Attitudes and Behaviours Influencing the Bowel Cancer Screening Pilot* 

Interventions that increase use of adult immunization and cancer screening services: a meta-analysis. *Ann Intern Med. Vol. 136:*641**–**651.


#### **10.2 Internet resources**

American Cancer Society (ACS) Guidelines for the early detection of cancer: Colorectal cancer and polyps.

(Accessed online September 19, 2011:

 http://www.cancer.org/Healthy/FindCancerEarly/CancerScreeningGuidelines/a merican-cancer-society-guidelines-for-the-early-detection-of-cancer).

Australian Government Department of Health and Ageing. (2008). *Clinical Practice Guidelines for the Prevention, Early Detection and Management of Colorectal Cancer – A Guide for General Practitioners*. (3rd ed.).

(Accessed online August 11, 2011:

64 Colorectal Cancer – From Prevention to Patient Care

Sung, J.J., Lau, J.Y., Goh, K.L., Leung, W.K.; Asia Pacific Working Group on Colorectal

Taskila T., Wilson S., Damery S., Roalfe A., Redman V., Ismail T., Hobbs R. (2009). Factors

Tu S.P., Taylor V., Yasui Y, Chun A., Yip M.P., Acorda E. et al. (2006). Promoting culturally

Vernon SW. (2003). A national survey of primary care physicians' colorectal cancer screening recommendations and practices. *Prev Med. Vol. 36(3)*:352-62. Viguier J, Calazel-Benque A, Eisinger F, Pivot X. (2011). Organized colorectal cancer

Walsh, J.M., Posner, S.F., Perez-Stable, E.J. (2002) Colon cancer screening in the ambulatory

Wardle, J., Williamson, S., McCaffery, K. (2003). Increasing attendance at colorectal cancer

Winawer, S.J., Zauber, A.G., Ho, M.N., O'Brien, M.J., Gottlieb, L.S., Sternberg, S.S., Waye,

Young, G.P., St John, D.J., Winawer, S.J., Rozen, P. WHO (World Health Organization) &

American Cancer Society (ACS) Guidelines for the early detection of cancer: Colorectal

 http://www.cancer.org/Healthy/FindCancerEarly/CancerScreeningGuidelines/a merican-cancer-society-guidelines-for-the-early-detection-of-cancer).

community sample of older adults. *Health Psychology. Vol. 22:*99-105. Wilson, J.M.G., and Jungner, G. (1968). Principles and practice of screening for disease. Geneva: World Health Organization.(Accessed online October 25, 2011 from:

http://www.who.int/bulletin/volumes/86/4/07-050112BP.pdf).

services: a meta-analysis. *Ann Intern Med. Vol. 136:*641**–**651.

screening. *Lancet Oncol*. Vol 6(11):871-6.

population. *Br J Cancer. Vol 21;101(2):*250.

controlled trial. *Cancer. Vol 107:*959**–**966.

*Eur J Cancer Prev. Vol. 20 Suppl 1:S*26-32. Wald N.J. (1994). Guidance on terminology. *J Med Screen. 1:* 76.

*Journal of Medicine. Vol. 329(27):*1977-1981.

*Am J Gastroenterol. Vol 97(10):*2499-507.

**10.2 Internet resources** 

cancer and polyps.

(Accessed online September 19, 2011:

setting. *Prev Med*. *Vol 35(3):*209-18.

Interventions that increase use of adult immunization and cancer screening

Cancer. (2005) Increasing incidence of colorectal cancer in Asia: implications for

affecting attitudes toward colorectal cancer screening in the primary care

appropriate colorectal cancer screening through a health educator: a randomized

screening programmes: how to optimize efficiency among general practitioners.

screening: testing the efficacy of a mailed, psychoeducational intervention in a

J.D., Schapiro, M., Bond, J.H., and Panish,J.F. (1993). Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup, *New England* 

OMED (World Organization for Digestive Endoscopy). (2002). Choice of faecal occult blood tests for colorectal cancer screening: recommendations based on performance characteristics in population studies: a WHO (World Health Organization) and OMED (World Organization for Digestive Endoscopy) report.  http://www.health.gov.au/internet/screening/publishing.nsf/Content/bw-gpcrc-guide).

Australian Government Department of Health and Ageing. (2005) *A Qualitative Evaluation of Opinions, Attitudes and Behaviours Influencing the Bowel Cancer Screening Pilot Program*. Final Report. Screening Monograph: 2/2005.

(Accessed online September 16, 2011:

 http://www.health.gov.au/internet/screening/publishing.nsf/Content/qualeval-cnt).

Australian Population Health Development Principal Committee. (2008). *Population-Based Screening*.

(Accessed online September 16, 2011:

 http://www.cancerscreening.gov.au/internet/screening/publishing.nsf/Content /other-pop-health#framework).

Cancer Research UK. Bowel cancer – screening and prevention.

(Accessed online September 15, 2011:

 http://info.cancerresearchuk.org/cancerstats/types/bowel/screeningandpreventi on/).

Centers for Diseases Control and Prevention (US) MMWR. Vital Signs: Colorectal Cancer Screening, Incidence, and Mortality: United States, 2002—2010. (Accessed online September 15, 2011:

 http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6026a4.htm?s\_cid=mm602 6a4\_w).

National Institute for Health and Clinical Excellence, (2011). NICE publishes new guidelines on colonoscopic surveillance for the prevention of colorectal cancer in people with ulcerative colitis, Crohn's disease or adenomas.

(Accessed online August 9, 2011:

 http://www.nice.org.uk/newsroom/pressreleases/2011045colonoscopicsurveillan ce.jsp).

International Agency for Research on Cancer. Early detection and Prevention - Quality Assurance Group Resources.

(Accessed online August 11, 2011:

http://www.iarc.fr/en/researchs-groups/QAS/current-topics.php).

Patient-Centered Outcomes Research Institute. (2011-ongoing) Review and Synthesis of Evidence for Eliciting the Patient's Perspective in Patient-Centered Outcomes Research. Literature review and interviews.

(Accessed online September 19, 2011:

http://www.pcori.org/patient-centered-outcomes-research/ and

http://www.pcori.org/committee-charters/).

U.S. Preventive Services Task Force. (2008). *Screening for Colorectal Cancer.*  (Accessed online August 8, 2011:

**4** 

*Australia* 

**Turning Intention Into Behaviour: The Effect of** 

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in males and second in females; throughout the world over 1.2 million new CRC cases and 608,7000 deaths are estimated to have occurred in 2008 (Jemal et al., 2011). The only developed country to have demonstrated a significantly decreasing incidence in both males and females is the United States, and this is largely due to the early detection and removal of pre-cancerous lesions through CRC screening (Jemal et al., 2011). Thus, an understanding of the variables that encourage people to participate in CRC screening is important because early detection and treatment of precancerous lesions and adenomas results in a significantly higher survival rate than if treatment is delayed until physical symptoms of the condition are apparent. Population screening using a Faecal Occult Blood Test (FOBT) can facilitate the detection of CRC at its early stages. FOBT is the collective term for a guiaic FOBT (gFOBT) or a faecal immunochemical test (FIT). Both are home-based tests which, although differing in the technology utilised, involve a stool sample being sent to a laboratory to be analysed for occult blood, ideally followed by colonoscopy for those with a positive result. The cost effectiveness of FOBTs for the screening of CRC, measured as Quality Adjusted Life Years gained, is comparable to other screening procedures (Frazier et al., 2000) and more costeffective than treatment after physical symptoms are evident (Fisher et al., 2006). Randomised clinical trials have shown that both biennial and annual screening using FOBT screening reduces CRC incidence (Mandel et al., 2000) and mortality (Hardcastle et al., 1996; Kronborg et al., 2004; Mandel et al., 1993), and a systematic review concluded that FOBT screening is likely to avoid 1 in 6 colorectal cancer deaths (Hewitson et al., 2007). Effectiveness, however, depends upon yield and is critically dependent upon participation rates, which for population-based screening programs have been low, often despite high levels of intention to participate. For example, in Australia the National Bowel Cancer Screening Program, which provides people turning 50, 55 and 60 years with a free FOBT, had a participation rate in 2008 of 41% of the eligible population (AIHW, 2010). In England, the second round (2003–2005) of the pilot bowel cancer screening program had a

**1. Introduction** 

**Providing Cues to Action on Participation** 

**Rates for Colorectal Cancer Screening** 

Ingrid Flight1, Carlene Wilson2 and Jane McGillivray3

*2Cancer Council South Australia and Flinders University,* 

*1CSIRO Preventative Health Flagship,* 

*Flinders Centre for Innovation in Cancer, 3School of Psychology, Deakin University,* 

 http://www.uspreventiveservicestaskforce.org/uspstf/uspscolo.htm). World Health Organization. (2011). *Screening for Colorectal Cancer.*  (Accessed online August 1, 2011:

http://www.who.int/cancer/detection/colorectalcancer/en/).

### **Turning Intention Into Behaviour: The Effect of Providing Cues to Action on Participation Rates for Colorectal Cancer Screening**

Ingrid Flight1, Carlene Wilson2 and Jane McGillivray3 *1CSIRO Preventative Health Flagship, 2Cancer Council South Australia and Flinders University, Flinders Centre for Innovation in Cancer, 3School of Psychology, Deakin University, Australia* 

#### **1. Introduction**

66 Colorectal Cancer – From Prevention to Patient Care

http://www.uspreventiveservicestaskforce.org/uspstf/uspscolo.htm).

World Health Organization. (2011). *Screening for Colorectal Cancer.* 

http://www.who.int/cancer/detection/colorectalcancer/en/).

(Accessed online August 1, 2011:

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in males and second in females; throughout the world over 1.2 million new CRC cases and 608,7000 deaths are estimated to have occurred in 2008 (Jemal et al., 2011). The only developed country to have demonstrated a significantly decreasing incidence in both males and females is the United States, and this is largely due to the early detection and removal of pre-cancerous lesions through CRC screening (Jemal et al., 2011). Thus, an understanding of the variables that encourage people to participate in CRC screening is important because early detection and treatment of precancerous lesions and adenomas results in a significantly higher survival rate than if treatment is delayed until physical symptoms of the condition are apparent. Population screening using a Faecal Occult Blood Test (FOBT) can facilitate the detection of CRC at its early stages. FOBT is the collective term for a guiaic FOBT (gFOBT) or a faecal immunochemical test (FIT). Both are home-based tests which, although differing in the technology utilised, involve a stool sample being sent to a laboratory to be analysed for occult blood, ideally followed by colonoscopy for those with a positive result. The cost effectiveness of FOBTs for the screening of CRC, measured as Quality Adjusted Life Years gained, is comparable to other screening procedures (Frazier et al., 2000) and more costeffective than treatment after physical symptoms are evident (Fisher et al., 2006). Randomised clinical trials have shown that both biennial and annual screening using FOBT screening reduces CRC incidence (Mandel et al., 2000) and mortality (Hardcastle et al., 1996; Kronborg et al., 2004; Mandel et al., 1993), and a systematic review concluded that FOBT screening is likely to avoid 1 in 6 colorectal cancer deaths (Hewitson et al., 2007). Effectiveness, however, depends upon yield and is critically dependent upon participation rates, which for population-based screening programs have been low, often despite high levels of intention to participate. For example, in Australia the National Bowel Cancer Screening Program, which provides people turning 50, 55 and 60 years with a free FOBT, had a participation rate in 2008 of 41% of the eligible population (AIHW, 2010). In England, the second round (2003–2005) of the pilot bowel cancer screening program had a

Turning Intention Into Behaviour: The Effect of Providing

this by providing the plan for goal achievement.

resources necessary for achieving their goal.

FOBT (DeVellis et al., 1990).

Orbell et al. (1997) cited improved rates of breast self-examination.

Cues to Action on Participation Rates for Colorectal Cancer Screening 69

behavioural intention and behaviour (Gollwitzer, 1993). This model suggests that individuals achieve volitional control of their intention to act by the development of implementation intentions; the plans made to achieve a specific behavioural target (e.g., a statement describing when, where and how a specific behaviour will be carried out). These plans serve to provide the cue to action identified by the Health Belief Model but go beyond

Recent empirical work suggests that the approach of providing cues to action in the form of a specific implementation intention improves prediction of behaviour over and above the intention to act alone. Thus, Milne, Orbell and Sheeran (2002) reported improved exercise participation; Sheeran and Orbell (2000) reported beneficial effects on the uptake of cervical cancer screening; Verplanken and Faes (1999) described improved dietary regimens; and

A study examining uptake in the National Health Service Breast Screening Program (NHSBSP) in the UK (Rutter et al., 2006) has highlighted the importance of providing guidance on how to plan for a behaviour in order to ensure that people move from intention to actual behaviour (i.e., from the preparation to the action stage of the Transtheoretical Model, TTM). In this study, women invited to screen for breast cancer were asked to make specific plans for attending. The plans consisted of organising their travel, arranging to take time off work if necessary and changing the appointment if it was inconvenient. The results indicated that when women produced a *written* plan, actual rate of compliance with the screening appointment was 15% greater than in the control condition (no intervention) and 7% greater than women who failed to write down a plan although instructed to do so. Moreover, the influence from the production of cues to action in the form of a written plan was greatest for those who initially had a high intention to comply but a weak sense of control over making the necessary arrangements to put that intention into effect. This research suggests that uptake of FOBT might be significantly improved by providing a cue to action that seeks to stimulate people to do more than simply express their intention to screen. An effective informational intervention that results in the development of implementation intentions in the form of a plan describing the when, where, and how of faecal occult blood testing, and which enables the individual to deal with their own personal and environmental constraints, should provide those with the intention to act the further

One possible mechanism for explaining the effectiveness in previous studies of asking participants to form implementation intentions is that doing so forces people to think through the steps necessary for actually completing the screening. This 'thinking through', in turn, may serve to raise people's confidence about their ability to successfully carry out the screening behaviour. Confidence in one's own capacity to act is known in the literature as 'self efficacy' and is widely reported as predicting health behaviour participation (Schwarzer & Fuchs, 1995). People's feelings of self efficacy are likely to be a particular consideration in using the FOBT because the test is performed by the individual and not administered, like mammography or Pap smear, by a health care professional. Previous studies looking at consumer-initiated screening behaviours have shown that feelings of confidence in one's ability to correctly perform the behaviour bear a strong relationship to people's performance of these behaviours. This includes performance of breast selfexamination (Luszczynska, 2004), testicular self-examination (Lechner et al., 2002), and

significantly lower uptake than in the first round (52% vs 58%) (Weller et al., 2006) and reported participation rates in 2008 in other countries with an established or pilot population FOBT screening program ranged mostly from a moderate 45–50% (Italy and Denmark, respectively) to a low 16–18% (Korean Republic and Japan, respectively) level (International Cancer Screening Network, 2008). Understanding motivators to intention to participate and motivators to test completion are critical issues that need to be addressed.

The central question in research within health psychology is identifying and understanding the range of influences that prompt an individual to take up healthy behaviours or reject patterns of behaviour which compromise their health. Many social cognitive health behaviour models include a measure of intention to behave in a specific way as a precursor to action (e.g., Theory of Planned Behaviour; (Ajzen, 1985). Stage models focus specifically on the importance of addressing intention as a core component of public health interventions. For example, the Transtheoretical, or Stages of Change, Model (Prochaska, 2008; Prochaska et al., 1988) suggests that people can be characterised in terms of their readiness to make a change. Stages include precontemplation (benefits of lifestyle change are not being considered), contemplation (starting to consider change but not yet begun to act on this intention), preparation (ready to change the behaviour and preparing to act), action (making the initial steps toward behaviour change), and maintenance of the behaviour over time; with both contemplation and preparation measuring aspects of intention.

One of the most difficult questions for researchers examining screening participation has been the question of how to move people along these stages to the performance of the actual behaviour and, ideally, maintenance of the behaviour. A range of social cognitive models of health behaviour have proven effective in describing individual motivation to perform a variety of health behaviours, including screening, by identifying a range of attitudinal predictors (Conner & Norman, 2005). Each of these deliberative models can successfully map variables that describe individual differences in the intention to perform a behaviour. However, the relationship between behavioural intention and actual behaviour is less than perfect; it has been shown that around 50% of people with positive intentions to engage in health behaviours successfully translate those intentions into action (Sheeran, 2002), and a medium-to-large change in intention leads to only a small-to-medium change in behaviour (Webb & Sheeran, 2006).

This 'gap', the difference between an individual's commitment to act and initiation of the necessary processes to actually carry out the behaviour, needs to be bridged—in other words, research that influences 'intention to try' (Bagozzi & Warshaw, 1990) needs to also identify cues that will enable people to link to the means for achieving the intended behaviour. Some health behaviour models incorporate a stimulus to action in their operationalisations in an attempt to capture this intervening, or additive, influence that prompts individuals to actually implement behaviour. For example, Becker and colleagues (1977) incorporated 'cues to action' as additional, independent predictors of health behaviour, over and above attitudinal variables. Although incorporated in the earliest descriptions of the Health Belief Model, a cue to action, or strategy to initiate "readiness", is a variable that has received limited attention in the empirical literature. Nevertheless, research does suggest that certain acts may serve to stimulate health behaviour including physician advice, advertising campaigns, and postcard reminders (Sheeran & Orbell, 2000).

Research originating outside the health area has examined the notion of volitional control and how it might be used to explain the problematic nature of the relationship between

significantly lower uptake than in the first round (52% vs 58%) (Weller et al., 2006) and reported participation rates in 2008 in other countries with an established or pilot population FOBT screening program ranged mostly from a moderate 45–50% (Italy and Denmark, respectively) to a low 16–18% (Korean Republic and Japan, respectively) level (International Cancer Screening Network, 2008). Understanding motivators to intention to participate and motivators to test completion are critical issues that need to be addressed. The central question in research within health psychology is identifying and understanding the range of influences that prompt an individual to take up healthy behaviours or reject patterns of behaviour which compromise their health. Many social cognitive health behaviour models include a measure of intention to behave in a specific way as a precursor to action (e.g., Theory of Planned Behaviour; (Ajzen, 1985). Stage models focus specifically on the importance of addressing intention as a core component of public health interventions. For example, the Transtheoretical, or Stages of Change, Model (Prochaska, 2008; Prochaska et al., 1988) suggests that people can be characterised in terms of their readiness to make a change. Stages include precontemplation (benefits of lifestyle change are not being considered), contemplation (starting to consider change but not yet begun to act on this intention), preparation (ready to change the behaviour and preparing to act), action (making the initial steps toward behaviour change), and maintenance of the behaviour over time; with both contemplation and preparation measuring aspects of

One of the most difficult questions for researchers examining screening participation has been the question of how to move people along these stages to the performance of the actual behaviour and, ideally, maintenance of the behaviour. A range of social cognitive models of health behaviour have proven effective in describing individual motivation to perform a variety of health behaviours, including screening, by identifying a range of attitudinal predictors (Conner & Norman, 2005). Each of these deliberative models can successfully map variables that describe individual differences in the intention to perform a behaviour. However, the relationship between behavioural intention and actual behaviour is less than perfect; it has been shown that around 50% of people with positive intentions to engage in health behaviours successfully translate those intentions into action (Sheeran, 2002), and a medium-to-large change in intention leads to only a small-to-medium change in behaviour

This 'gap', the difference between an individual's commitment to act and initiation of the necessary processes to actually carry out the behaviour, needs to be bridged—in other words, research that influences 'intention to try' (Bagozzi & Warshaw, 1990) needs to also identify cues that will enable people to link to the means for achieving the intended behaviour. Some health behaviour models incorporate a stimulus to action in their operationalisations in an attempt to capture this intervening, or additive, influence that prompts individuals to actually implement behaviour. For example, Becker and colleagues (1977) incorporated 'cues to action' as additional, independent predictors of health behaviour, over and above attitudinal variables. Although incorporated in the earliest descriptions of the Health Belief Model, a cue to action, or strategy to initiate "readiness", is a variable that has received limited attention in the empirical literature. Nevertheless, research does suggest that certain acts may serve to stimulate health behaviour including physician advice, advertising campaigns, and postcard reminders (Sheeran & Orbell, 2000). Research originating outside the health area has examined the notion of volitional control and how it might be used to explain the problematic nature of the relationship between

intention.

(Webb & Sheeran, 2006).

behavioural intention and behaviour (Gollwitzer, 1993). This model suggests that individuals achieve volitional control of their intention to act by the development of implementation intentions; the plans made to achieve a specific behavioural target (e.g., a statement describing when, where and how a specific behaviour will be carried out). These plans serve to provide the cue to action identified by the Health Belief Model but go beyond this by providing the plan for goal achievement.

Recent empirical work suggests that the approach of providing cues to action in the form of a specific implementation intention improves prediction of behaviour over and above the intention to act alone. Thus, Milne, Orbell and Sheeran (2002) reported improved exercise participation; Sheeran and Orbell (2000) reported beneficial effects on the uptake of cervical cancer screening; Verplanken and Faes (1999) described improved dietary regimens; and Orbell et al. (1997) cited improved rates of breast self-examination.

A study examining uptake in the National Health Service Breast Screening Program (NHSBSP) in the UK (Rutter et al., 2006) has highlighted the importance of providing guidance on how to plan for a behaviour in order to ensure that people move from intention to actual behaviour (i.e., from the preparation to the action stage of the Transtheoretical Model, TTM). In this study, women invited to screen for breast cancer were asked to make specific plans for attending. The plans consisted of organising their travel, arranging to take time off work if necessary and changing the appointment if it was inconvenient. The results indicated that when women produced a *written* plan, actual rate of compliance with the screening appointment was 15% greater than in the control condition (no intervention) and 7% greater than women who failed to write down a plan although instructed to do so. Moreover, the influence from the production of cues to action in the form of a written plan was greatest for those who initially had a high intention to comply but a weak sense of control over making the necessary arrangements to put that intention into effect. This research suggests that uptake of FOBT might be significantly improved by providing a cue to action that seeks to stimulate people to do more than simply express their intention to screen. An effective informational intervention that results in the development of implementation intentions in the form of a plan describing the when, where, and how of faecal occult blood testing, and which enables the individual to deal with their own personal and environmental constraints, should provide those with the intention to act the further resources necessary for achieving their goal.

One possible mechanism for explaining the effectiveness in previous studies of asking participants to form implementation intentions is that doing so forces people to think through the steps necessary for actually completing the screening. This 'thinking through', in turn, may serve to raise people's confidence about their ability to successfully carry out the screening behaviour. Confidence in one's own capacity to act is known in the literature as 'self efficacy' and is widely reported as predicting health behaviour participation (Schwarzer & Fuchs, 1995). People's feelings of self efficacy are likely to be a particular consideration in using the FOBT because the test is performed by the individual and not administered, like mammography or Pap smear, by a health care professional. Previous studies looking at consumer-initiated screening behaviours have shown that feelings of confidence in one's ability to correctly perform the behaviour bear a strong relationship to people's performance of these behaviours. This includes performance of breast selfexamination (Luszczynska, 2004), testicular self-examination (Lechner et al., 2002), and FOBT (DeVellis et al., 1990).

Turning Intention Into Behaviour: The Effect of Providing

320 (160 men and 160 women).

provided.

**3.1.3 Study conduct** 

were ineligible.

Cues to Action on Participation Rates for Colorectal Cancer Screening 71

allowing for non-contactability by telephone, a subsequent rejection rate of 30% and ineligibility, we needed to recruit at least 1600 participants to achieve a final sample size of

A random sample of 6000 (3000 males, 3000 females) potential invitees aged between 50 and 76 years and residing in southern urban Adelaide, South Australia, was provided by the Australian Electoral Commission (AEC). The Australian Government was conducting a pilot National Bowel Cancer Screening Program (NBCSP) at the same time (2004) so individuals with postcodes within the Federal screening program were deleted from the sample

Telephone contact numbers for the remaining sample were obtained by comparing the list against information contained in the electronic White Pages telephone directory. Those persons for whom telephone contact details were not indicated were excluded from the list, as were those whose address indicated that they resided in a hostel or nursing home; such individuals were unlikely to be in the position of deciding for themselves whether they should screen for CRC. The remaining sample was randomized separately by sex using a random number generator (Microsoft ® Office Excel 2003) and 400 (200 m; 200 f) names

The trial proceeded through a number of phases, as described below and illustrated in Table 1. Phase 1: All potential participants were mailed an advance notification letter and accompanying information, to the effect that an attempt would be made to contact them by telephone to invite them to participate in a study on how best to encourage people to participate in screening for colorectal cancer. Potential participants were advised that they were ineligible to participate if they had ever participated in CRC screening or been diagnosed with CRC or polyps. This exclusion criterion was because in Australia such diagnoses normally follow a positive FOBT and subsequent colonoscopy, and we wanted to target those who had not displayed overt symptoms but were of average risk (that is, based solely on the fact that they were aged 50 years or more) of developing CRC. An opportunity was provided at this point for individuals to decline participation or to indicate that they

Phase 2: One week after the advance notification letter, attempts were made (to a maximum of 3 occasions) to telephone individuals and recruit them to the study. A Computer Assisted Telephone Interview (CATI) format was used by trained interviewers to collect interview responses (Microsoft ® Office Access 2003). For those who were contactable and agreed to participate, informed consent was formally requested and recorded before commencement of the CATI. The recruiting interviewers were blinded to an individual's group allocation until they reached that part of the CATI (after having determined eligibility) that, as part of obtaining informed consent, provided details of the particular intervention to which the participant had been assigned. To those that agreed to participate, the interviewer briefly described what an FOBT was and asked whether they had heard of it: "*Before we contacted you, had you ever heard of a screening test for colorectal cancer, where you are given a set of cards to take home and asked to smear a part of your stool on the cards on two separate occasions, and then return the cards to be tested for blood? This is called a Faecal Occult Blood Test, or FOBT. This is the type of screening test we will be sending you*". Baseline measures were obtained: background

demographics, level of commitment to using an FOBT, and confidence to use the kit.

were assigned sequentially to one of 4 groups. In total 1642 names were allocated.

### **2. Aims**

This study was designed to investigate the effect of the formulation of implementation intentions upon people's participation in screening using FOBT. We chose to examine uptake of FOBT rather than colonoscopy because, in comparison to the United States, usual CRC screening practice in Australia is by FOBT followed by colonoscopy for those with a positive result—in other words, colonoscopy is regarded as a diagnostic test rather than a screening test.

An additional aim was to monitor the impact upon participation of differing levels of directedness in formulating these intentions and to determine the impact of self efficacy and prior levels of generalised intention upon both implementation intention formation and participation.

Consistent with prior research, it was anticipated that the formulation of implementation intentions (regardless of level of directedness) would increase participation in FOBT over levels of participation in the control group. Furthermore, previous work in the area of preventive health behaviour suggests that people's feelings of self efficacy, or confidence to use the test (the terms 'self efficacy' [SE] and 'confidence' will hereinafter be used interchangeably) can be increased in response to appropriate cues to action, and it was anticipated that the provision of directions for the formulation of implementation intentions would increase people's feelings of self efficacy. It was further hypothesised that those who were already strongly intending to use an FOBT were expected to differ in implementation intention formation and participation from those whose intentions to test were initially weaker.

We conducted two randomised controlled trials to test these hypotheses. Study 1 was a trial conducted amongst a group of eligible, randomly selected males and females who were approached and agreed to participate in the trial. Study 2 was also a randomised controlled trial to examine the generalisability of results to population settings and which differed from Study 1 in that prior commitment to trial participation was not obtained and eligibility was unknown.

### **3. Study 1**

#### **3.1 Methods**

#### **3.1.1 Study design**

The study was a parallel, randomised, controlled trial, stratified by sex, comparing return of FOBT between three intervention groups and one control group. People in the intervention groups received an FOBT of the immunochemical type (FIT) in the mail together with instructions on how to construct a (1) participant-determined and retained plan, (2) participant-determined and shared plan, or (3) researcher-directed and shared plan. The control group received the FOBT only.

#### **3.1.2 Sample size and selection**

Previous studies of implementation intentions have demonstrated that the effect of their formation upon behaviour is medium to large (Gollwitzer & Sheeran, 2006). To achieve statistical power of .80 to detect a medium-sized effect (allowing for the possibility of self efficacy and generalised intention as co-variants) and an alpha of 0.05, we aimed to recruit a minimum of 80 participants in each of the four groups described above. Accordingly, allowing for non-contactability by telephone, a subsequent rejection rate of 30% and ineligibility, we needed to recruit at least 1600 participants to achieve a final sample size of 320 (160 men and 160 women).

A random sample of 6000 (3000 males, 3000 females) potential invitees aged between 50 and 76 years and residing in southern urban Adelaide, South Australia, was provided by the Australian Electoral Commission (AEC). The Australian Government was conducting a pilot National Bowel Cancer Screening Program (NBCSP) at the same time (2004) so individuals with postcodes within the Federal screening program were deleted from the sample provided.

Telephone contact numbers for the remaining sample were obtained by comparing the list against information contained in the electronic White Pages telephone directory. Those persons for whom telephone contact details were not indicated were excluded from the list, as were those whose address indicated that they resided in a hostel or nursing home; such individuals were unlikely to be in the position of deciding for themselves whether they should screen for CRC. The remaining sample was randomized separately by sex using a random number generator (Microsoft ® Office Excel 2003) and 400 (200 m; 200 f) names were assigned sequentially to one of 4 groups. In total 1642 names were allocated.

#### **3.1.3 Study conduct**

70 Colorectal Cancer – From Prevention to Patient Care

This study was designed to investigate the effect of the formulation of implementation intentions upon people's participation in screening using FOBT. We chose to examine uptake of FOBT rather than colonoscopy because, in comparison to the United States, usual CRC screening practice in Australia is by FOBT followed by colonoscopy for those with a positive result—in other words, colonoscopy is regarded as a diagnostic test rather than a

An additional aim was to monitor the impact upon participation of differing levels of directedness in formulating these intentions and to determine the impact of self efficacy and prior levels of generalised intention upon both implementation intention formation and

Consistent with prior research, it was anticipated that the formulation of implementation intentions (regardless of level of directedness) would increase participation in FOBT over levels of participation in the control group. Furthermore, previous work in the area of preventive health behaviour suggests that people's feelings of self efficacy, or confidence to use the test (the terms 'self efficacy' [SE] and 'confidence' will hereinafter be used interchangeably) can be increased in response to appropriate cues to action, and it was anticipated that the provision of directions for the formulation of implementation intentions would increase people's feelings of self efficacy. It was further hypothesised that those who were already strongly intending to use an FOBT were expected to differ in implementation intention formation and participation from those whose intentions to test were initially

We conducted two randomised controlled trials to test these hypotheses. Study 1 was a trial conducted amongst a group of eligible, randomly selected males and females who were approached and agreed to participate in the trial. Study 2 was also a randomised controlled trial to examine the generalisability of results to population settings and which differed from Study 1 in that prior commitment to trial participation was not obtained and eligibility

The study was a parallel, randomised, controlled trial, stratified by sex, comparing return of FOBT between three intervention groups and one control group. People in the intervention groups received an FOBT of the immunochemical type (FIT) in the mail together with instructions on how to construct a (1) participant-determined and retained plan, (2) participant-determined and shared plan, or (3) researcher-directed and shared plan. The

Previous studies of implementation intentions have demonstrated that the effect of their formation upon behaviour is medium to large (Gollwitzer & Sheeran, 2006). To achieve statistical power of .80 to detect a medium-sized effect (allowing for the possibility of self efficacy and generalised intention as co-variants) and an alpha of 0.05, we aimed to recruit a minimum of 80 participants in each of the four groups described above. Accordingly,

**2. Aims** 

screening test.

participation.

weaker.

was unknown.

control group received the FOBT only.

**3.1.2 Sample size and selection** 

**3. Study 1 3.1 Methods 3.1.1 Study design**  The trial proceeded through a number of phases, as described below and illustrated in Table 1. Phase 1: All potential participants were mailed an advance notification letter and accompanying information, to the effect that an attempt would be made to contact them by telephone to invite them to participate in a study on how best to encourage people to participate in screening for colorectal cancer. Potential participants were advised that they were ineligible to participate if they had ever participated in CRC screening or been diagnosed with CRC or polyps. This exclusion criterion was because in Australia such diagnoses normally follow a positive FOBT and subsequent colonoscopy, and we wanted to target those who had not displayed overt symptoms but were of average risk (that is, based solely on the fact that they were aged 50 years or more) of developing CRC. An opportunity was provided at this point for individuals to decline participation or to indicate that they were ineligible.

Phase 2: One week after the advance notification letter, attempts were made (to a maximum of 3 occasions) to telephone individuals and recruit them to the study. A Computer Assisted Telephone Interview (CATI) format was used by trained interviewers to collect interview responses (Microsoft ® Office Access 2003). For those who were contactable and agreed to participate, informed consent was formally requested and recorded before commencement of the CATI. The recruiting interviewers were blinded to an individual's group allocation until they reached that part of the CATI (after having determined eligibility) that, as part of obtaining informed consent, provided details of the particular intervention to which the participant had been assigned. To those that agreed to participate, the interviewer briefly described what an FOBT was and asked whether they had heard of it: "*Before we contacted you, had you ever heard of a screening test for colorectal cancer, where you are given a set of cards to take home and asked to smear a part of your stool on the cards on two separate occasions, and then return the cards to be tested for blood? This is called a Faecal Occult Blood Test, or FOBT. This is the type of screening test we will be sending you*". Baseline measures were obtained: background demographics, level of commitment to using an FOBT, and confidence to use the kit.

Turning Intention Into Behaviour: The Effect of Providing

screening package without any accompanying plan.

time of interview (data not included in these analyses).

**3.2.1 Development of implementation plans** 

(before reminder) or after 6 weeks.

**3.2 Materials** 

author on request.

Cues to Action on Participation Rates for Colorectal Cancer Screening 73

return one copy of the completed plan to CSIRO ('*Aide to return'*). The third intervention group received a plan in the form of a researcher-directed 'Checklist' ('*Checklist to return'*) which *directed* participants to think about how they were going to deal with potential barriers. This group was also provided with two copies of the checklist and asked to return one completed checklist to CSIRO. Thus, those in the intervention groups were invited to formulate implementation plans at differing levels of directedness, and the researchers, through their requirement that two of the intervention groups return a completed plan, were able to verify that in fact a plan had been completed. The control group received a

Phase 4: Receipt of completed FOBTs was recorded by the Bowel Health Service (Repatriation General Hospital, Bedford Park, South Australia) and participation data relayed to the researchers. People who did not return their test after six weeks were sent a reminder letter. Participation in screening was defined as receipt of kit within 6 weeks

Phase 5: Approximately 7 weeks following FOBT despatch, participants were contacted by telephone. Confidence to use the FOBT was again measured, as was (for those who had returned their FOBT) commitment to screen every two years in the future, following recommended screening guidelines. Additionally, participants' reasons for screening or not screening were elicited, depending on whether a completed FOBT had been returned at the

Two versions of implementation plan were designed; one as an 'aide' and the other as a more prescriptive 'checklist'. Each version was introduced to the participant with the words "*Many people find that they intend to complete the FOBT but then forget or 'never get around to it'. It has been found that if you form a definite plan of exactly when and where you will carry out an intended behaviour you are more likely to actually do so and less likely to forget or find that you don't get around to doing it. It would be useful for you to plan when, where and how you will complete the FOBT. To help you do this, we would like you to use the attached sheets we have provided"* (adapted from Milne et al., 2002). Both plans were designed to support confidence and addressed practical aspects of completing the test (reading the instructions; deciding the most convenient time to use the FOBT; deciding the most convenient location to use the FOBT; preparing for the test; using the FOBT; remembering to use the FOBT; sending the FOBT for analysis). Both versions commenced with the instruction: "*Using this plan, decide when you will use the screening kit, where you will use the kit, and the procedure you will use to carry out the screening test and obtain your result from the Bowel Health Service*". They thereafter differed in their level of directedness in covering the practical aspects. For example, for 'remembering to use the kit' the aides contained the following instruction: "*It is easy to forget to do things unless we have a way to remind us. Decide now how you can make it easier for you to remember—for example, by leaving the kit or this plan in a prominent location, or writing yourself a note. Write below how you will remind yourself to use the kit on two separate occasions".* In contrast, for the same instruction the checklist stated "*Place a reminder in a prominent place so that you do not forget to use the kit*" with two check boxes (1st sample done; 2nd sample done) to indicate that this instruction had been carried out. The complete documents are available from the first


Table 1. Study 1 interventions by phase and arm, with attrition rates

Phase 3: The day following the recruitment interview, all participants were mailed a screening package which included an immunochemical FOBT. Accompanying the package, intervention groups also received an implementation plan to serve as a 'cue to action' to provide a strategy for goal achievement (completion and return of the FOBT). Two intervention groups received a participant-directed plan in the form of an 'Aide' that *invited* participants to think about, and write down, how they were going to deal with potential barriers to using the FOBT. Suggestions were made as to how these barriers could be addressed. Participants in one of these two groups were asked to retain their completed plan ('*Aide to retain'*); the other group were sent two copies of the plan and requested to return one copy of the completed plan to CSIRO ('*Aide to return'*). The third intervention group received a plan in the form of a researcher-directed 'Checklist' ('*Checklist to return'*) which *directed* participants to think about how they were going to deal with potential barriers. This group was also provided with two copies of the checklist and asked to return one completed checklist to CSIRO. Thus, those in the intervention groups were invited to formulate implementation plans at differing levels of directedness, and the researchers, through their requirement that two of the intervention groups return a completed plan, were able to verify that in fact a plan had been completed. The control group received a screening package without any accompanying plan.

Phase 4: Receipt of completed FOBTs was recorded by the Bowel Health Service (Repatriation General Hospital, Bedford Park, South Australia) and participation data relayed to the researchers. People who did not return their test after six weeks were sent a reminder letter. Participation in screening was defined as receipt of kit within 6 weeks (before reminder) or after 6 weeks.

Phase 5: Approximately 7 weeks following FOBT despatch, participants were contacted by telephone. Confidence to use the FOBT was again measured, as was (for those who had returned their FOBT) commitment to screen every two years in the future, following recommended screening guidelines. Additionally, participants' reasons for screening or not screening were elicited, depending on whether a completed FOBT had been returned at the time of interview (data not included in these analyses).

#### **3.2 Materials**

72 Colorectal Cancer – From Prevention to Patient Care

**Interventions Phase 3** 

FOBT screening package only (n=91)

*Aide to retain*  FOBT screening package + implementation plan to be formulated and retained by participant (n=81)

*Aide to return*  FOBT screening package + implementation plan to be formulated and returned to researcher (n=95)

*Checklist to return*  FOBT screening package + implementation plan devised by researcher completed and returned to researcher (n=97)

Phase 3: The day following the recruitment interview, all participants were mailed a screening package which included an immunochemical FOBT. Accompanying the package, intervention groups also received an implementation plan to serve as a 'cue to action' to provide a strategy for goal achievement (completion and return of the FOBT). Two intervention groups received a participant-directed plan in the form of an 'Aide' that *invited* participants to think about, and write down, how they were going to deal with potential barriers to using the FOBT. Suggestions were made as to how these barriers could be addressed. Participants in one of these two groups were asked to retain their completed plan ('*Aide to retain'*); the other group were sent two copies of the plan and requested to

Table 1. Study 1 interventions by phase and arm, with attrition rates

N=364 *Control*  **Measures Phase 4** 

N=350 after exclusions (All groups)

*Measures:*  Return of kit within 6 weeks Return of kit after 6 weeks

**Measures Phase 5** 

N=328 after exclusions (All groups)

*CATI interview measures:*  Commitment to rescreen Self efficacy to use FOBT Reasons for not screening (if applicable) Reasons for screening (if applicable)

**Phase 1 Recruitment** 

N=1642 Potentially eligible participants randomised to study arm then notification of intention to contact by telephone + information sheet mailed

**Phase 2** 

N=994 contactable. N=364 agreed to participate in CATI interview:

screen

*Baseline measures:*  Demographics Commitment to

Self efficacy to use FOBT

#### **3.2.1 Development of implementation plans**

Two versions of implementation plan were designed; one as an 'aide' and the other as a more prescriptive 'checklist'. Each version was introduced to the participant with the words "*Many people find that they intend to complete the FOBT but then forget or 'never get around to it'. It has been found that if you form a definite plan of exactly when and where you will carry out an intended behaviour you are more likely to actually do so and less likely to forget or find that you don't get around to doing it. It would be useful for you to plan when, where and how you will complete the FOBT. To help you do this, we would like you to use the attached sheets we have provided"* (adapted from Milne et al., 2002). Both plans were designed to support confidence and addressed practical aspects of completing the test (reading the instructions; deciding the most convenient time to use the FOBT; deciding the most convenient location to use the FOBT; preparing for the test; using the FOBT; remembering to use the FOBT; sending the FOBT for analysis). Both versions commenced with the instruction: "*Using this plan, decide when you will use the screening kit, where you will use the kit, and the procedure you will use to carry out the screening test and obtain your result from the Bowel Health Service*". They thereafter differed in their level of directedness in covering the practical aspects. For example, for 'remembering to use the kit' the aides contained the following instruction: "*It is easy to forget to do things unless we have a way to remind us. Decide now how you can make it easier for you to remember—for example, by leaving the kit or this plan in a prominent location, or writing yourself a note. Write below how you will remind yourself to use the kit on two separate occasions".* In contrast, for the same instruction the checklist stated "*Place a reminder in a prominent place so that you do not forget to use the kit*" with two check boxes (1st sample done; 2nd sample done) to indicate that this instruction had been carried out. The complete documents are available from the first author on request.

Turning Intention Into Behaviour: The Effect of Providing

Male Female

Highest level of education

Some high school Completed high school/trade

Country of birth:

<sup>≠</sup>n=2 missing values

Australia

University qualification

Never heard of FOBT prior to participation

\*\* n=3 missing values for age group

contemplation stage (Table 2).

**3.4.1 FOBT participation** 

Control N=90 (%)

48 (53) 42

46 (51)

32 (36) 12 (13)

\*percentages have been rounded so may not be equivalent to 100%

Table 2. Study 1 Participant demographic characteristics\*

therefore available for n=328/994 (33%) participants.

Cues to Action on Participation Rates for Colorectal Cancer Screening 75

be contacted and invited to participate. Of n=994 able to be contacted and eligible, n=364 individuals (36.6%) agreed to participate in the study. Subsequently n=14 were excluded from analysis because they didn't receive an FOBT (n=3); had undergone screening since joining the study (n=4); reported symptoms that precluded them from using the FOBT (n=4), or were unable to participate due to barriers unrelated to the study (n=3). Baseline and screening participation data were therefore available for n=350/994 participants (35%).

> Aide to retain n=79 (%)

41 (52) 38

35 (45)

27 (35) 16 (20)

At follow-up (post intervention and mailing of FOBT), n=13 participants declined or were unable to be interviewed and n=9 were unable to be contacted; follow-up data were

At recruitment, the groups (n=350 participants) were balanced for gender, mean age, age group, level of education and Australian birth, and awareness of FOBT. The majority of participants had never heard of an FOBT before they were approached, i.e. they were in pre-

Completed FOBTs were returned by n=286/350 (81.7%) of eligible participants over a period of 15 weeks (mean = 3.12 weeks). Contrary to the hypothesis that formation of implementation plans would improve FOBT uptake, there was no significant difference

Age group\*\* *X*2 (6)=2.236,

Age, mean 61.1 60.5 61.2 61.7 NS

Age 50–59 43 (48) 38 (48) 45 (49) 37 (41) Age 60–69 31 (34) 29 (37) 32 (35) 39 (43) Age 70–76 15 (17) 11 (14) 14 (15) 13 (14)

Aide to return n=91 (%)

44 (48) 47

*X*2 (6)=5.894

67 (74) 57 (72) 71 (78) 61 (68) *X*2 (3)=2.539,

64 (71) 65 (82) 59 (65)≠ 65 (72) *X*2 (3)=5.618

39 (43)

36 (40) 16 (18) Checklist to return n=90 (%)

34 (38) 56

52 (58)

26 (29) 12 (13) Test of difference

p=.897

p=.435

p=.468

p=.132

*X*2 (3)=5.270, p=.153

#### **3.2.2 Development of self efficacy scale**

Self efficacy was measured using 4 items derived from terms developed by Vernon et al. (1997) and our clinical experience of the challenges and impediments surrounding FOBT use. Participants were asked to rate their degree of confidence in surmounting the barriers described. The items were scored on a 5-point Likert scale ranging from *strongly disagree* (1) to *strongly agree* (5). The items were: "I feel confident that I would be able to carry out an FOBT"; "I feel confident that the test will not be overly distasteful or embarrassing"; I feel confident that I would be able to find time in the day to complete the test"; "I feel confident that I could complete the test correctly". The scale had good internal consistency, with a Cronbach alpha coefficient of .86.

#### **3.2.3 Commitment to screen**

Commitment to screen was measured in Phase 2 by asking "*Right now, how strongly committed are you to doing this test, where 1 is undecided and 5 is very committed?".* The followup interview measured commitment to screen again (for those who had returned their FOBT): "*Now that you have done this screening test once, do you think you'll go on doing it every two years?"* (yes/no answer) and "*Right now, how strongly committed are you to doing this test again, where 1 is undecided and 5 is very committed?"* 

#### **3.2.4 Screening offer**

The screening package, or kit, included (a) a bowel cancer screening information pamphlet; (b) an immunochemical FOBT ((iFOBT also known as a faecal immunochemical test for haemoglobin [FIT], InSure™, Enterix Australia) that does not require dietary or drug restrictions; (c) a combined Participant Details and Consent Form confirming personal details, nominating a preferred doctor for follow-up, and consent to obtain clinical followup reports if required; and (d) a reply-paid return envelope.

#### **3.3 Data analysis**

Random missing values on pre- and post self efficacy (SE) variables (17/2800, 0.61%) were imputed using the expectation maximisation method, so that as many observations as possible were available for computing self efficacy total scores. The scores were split at the median baseline SE score of 17; scores ≤16 were designated 'low' and scores ≥17 'high' SE. Participation rates were viewed as 'early' or 'late' at a cut-off point of 6 weeks following despatch of FOBT, at which time a reminder was sent to non-responders. Chi-square analysis was conducted to assess FOBT awareness, FOBT participation and return of implementation plans between groups; Fishers exact test was utilised where cells contained <5. Paired samples t-tests and one-way ANOVAs compared score means for self efficacy and commitment to screen. A median split was not performed for commitment to screen as the majority of people had high intention to screen. Binary logistic regression was used to examine the ability of self efficacy and commitment to screen to predict return of FOBT, and Generalised linear models (GLM) were used to assess interactions between variables. All tests were conducted using a two-sided alpha level of 0.05.

#### **3.4 Results**

Recruitment and participation attrition rates are shown at Table 1. From a sampling frame of potential participants (3,000 men and 3,000 women), n=1642 were notified that they would be contacted and invited to participate. Of n=994 able to be contacted and eligible, n=364 individuals (36.6%) agreed to participate in the study. Subsequently n=14 were excluded from analysis because they didn't receive an FOBT (n=3); had undergone screening since joining the study (n=4); reported symptoms that precluded them from using the FOBT (n=4), or were unable to participate due to barriers unrelated to the study (n=3). Baseline and screening participation data were therefore available for n=350/994 participants (35%).


\*percentages have been rounded so may not be equivalent to 100%

\*\* n=3 missing values for age group

<sup>≠</sup>n=2 missing values

74 Colorectal Cancer – From Prevention to Patient Care

Self efficacy was measured using 4 items derived from terms developed by Vernon et al. (1997) and our clinical experience of the challenges and impediments surrounding FOBT use. Participants were asked to rate their degree of confidence in surmounting the barriers described. The items were scored on a 5-point Likert scale ranging from *strongly disagree* (1) to *strongly agree* (5). The items were: "I feel confident that I would be able to carry out an FOBT"; "I feel confident that the test will not be overly distasteful or embarrassing"; I feel confident that I would be able to find time in the day to complete the test"; "I feel confident that I could complete the test correctly". The scale had good internal consistency, with a

Commitment to screen was measured in Phase 2 by asking "*Right now, how strongly committed are you to doing this test, where 1 is undecided and 5 is very committed?".* The followup interview measured commitment to screen again (for those who had returned their FOBT): "*Now that you have done this screening test once, do you think you'll go on doing it every two years?"* (yes/no answer) and "*Right now, how strongly committed are you to doing this test* 

The screening package, or kit, included (a) a bowel cancer screening information pamphlet; (b) an immunochemical FOBT ((iFOBT also known as a faecal immunochemical test for haemoglobin [FIT], InSure™, Enterix Australia) that does not require dietary or drug restrictions; (c) a combined Participant Details and Consent Form confirming personal details, nominating a preferred doctor for follow-up, and consent to obtain clinical follow-

Random missing values on pre- and post self efficacy (SE) variables (17/2800, 0.61%) were imputed using the expectation maximisation method, so that as many observations as possible were available for computing self efficacy total scores. The scores were split at the median baseline SE score of 17; scores ≤16 were designated 'low' and scores ≥17 'high' SE. Participation rates were viewed as 'early' or 'late' at a cut-off point of 6 weeks following despatch of FOBT, at which time a reminder was sent to non-responders. Chi-square analysis was conducted to assess FOBT awareness, FOBT participation and return of implementation plans between groups; Fishers exact test was utilised where cells contained <5. Paired samples t-tests and one-way ANOVAs compared score means for self efficacy and commitment to screen. A median split was not performed for commitment to screen as the majority of people had high intention to screen. Binary logistic regression was used to examine the ability of self efficacy and commitment to screen to predict return of FOBT, and Generalised linear models (GLM) were used to assess interactions between variables. All

Recruitment and participation attrition rates are shown at Table 1. From a sampling frame of potential participants (3,000 men and 3,000 women), n=1642 were notified that they would

**3.2.2 Development of self efficacy scale** 

Cronbach alpha coefficient of .86.

*again, where 1 is undecided and 5 is very committed?"* 

up reports if required; and (d) a reply-paid return envelope.

tests were conducted using a two-sided alpha level of 0.05.

**3.2.3 Commitment to screen** 

**3.2.4 Screening offer** 

**3.3 Data analysis** 

**3.4 Results** 

Table 2. Study 1 Participant demographic characteristics\*

At follow-up (post intervention and mailing of FOBT), n=13 participants declined or were unable to be interviewed and n=9 were unable to be contacted; follow-up data were therefore available for n=328/994 (33%) participants.

At recruitment, the groups (n=350 participants) were balanced for gender, mean age, age group, level of education and Australian birth, and awareness of FOBT. The majority of participants had never heard of an FOBT before they were approached, i.e. they were in precontemplation stage (Table 2).

#### **3.4.1 FOBT participation**

Completed FOBTs were returned by n=286/350 (81.7%) of eligible participants over a period of 15 weeks (mean = 3.12 weeks). Contrary to the hypothesis that formation of implementation plans would improve FOBT uptake, there was no significant difference

Turning Intention Into Behaviour: The Effect of Providing

\*\* p<.01 \*\*\*p<.001

return/non return of FOBTs

complete initial screening, regardless of initial level of confidence.

Cues to Action on Participation Rates for Colorectal Cancer Screening 77

increased significantly, regardless of group assignment. This result suggests that, in general, confidence to complete the test in the future is likely to decrease for those people who don't

SE score non-returners (full sample) Time 1 16.77 1.893 47 1.758 Time 2 16.15 2.278

Low baseline SE score non-returners Time 1 15.46 1.208 25 0.220 Time 2 15.35 2.279

SE score returners (full sample) Time 1 17.57 1.944 279 8.674\*\*\* Time 2 18.71 1.561

Low baseline SE score returners Time 1 15.69 .978 116 15.388\*\*\* Time 2 18.32 1.711

High baseline SE score non-returners Time 1 18.32 1.287 21 2.752\*\* Time 2 17.09 1.925

High baseline SE score returners Time 1 18.92 1.202 162 0.489 Time 2 18.99 1.383

Table 5. Study 1 mean self efficacy scores pre- and post-intervention, overall and by

was decreased when the initial level was high but the test was not completed.

**3.4.4 Commitment to screen and maintain screening** 

from those that had a high initial SE score (Table 7).

In order to determine whether confidence at baseline influenced reaction to the various interventions, participants were characterised as having a low or high SE score at baseline (determined by a median-split between 16 and 17), and change in confidence over time compared (See Table 5). Low SE non-returners did not significantly change their SE scores post intervention, whereas low SE returners' scores significantly increased post intervention. Similarly, for those with a high SE score at baseline, non-returners' scores significantly decreased post intervention but did not significantly change if they returned an FOBT. This latter result is likely to reflect ceiling effects given that the maximum score possible for SE was 20. These results suggest that self efficacy was increased when the test was completed but the initial level of confidence to complete the test was low, and conversely confidence

At baseline, the majority of people (n=217/343, 63%) were committed or very committed to doing the test (M=4.39, SD=.924; median=5) and there were no group differences (Table 6). Those who returned an FOBT were asked their level of commitment to maintain screening, and just over half (n=137/239, 57.3%) were "very committed" to screening again (M=4.38, SD=.840, median=5, n=47 missing values), regardless of intervention assignment. For those that did return an FOBT, a paired-sample t-test indicated that for the sample as a whole there was a statistically significant decrease in commitment to screen from baseline, ie after exposure to the intervention and FOBT (Table 7). When we examined the relationship between commitment and self efficacy by comparing commitment level between those who had a low or high SE baseline score, it was apparent that the decrease in commitment came

M SD df t


between the groups in FOBT participation or return within 6 weeks (i.e., before and after reminder) (Table 3).

\*These numbers do not correspond with participants who returned FOBTs within 6 weeks

Table 3. Study 1 return of kits and implementation plans by group

#### **3.4.2 Return of implementation plans**

Most participants who returned a completed FOBT and were also required to return a completed implementation plan did so. There was no significant difference in rate of return between aide and checklist (Table 3), suggesting that differing levels of directedness had no impact on whether the plans were completed. There were no cases of a plan being returned without an accompanying completed kit.

#### **3.4.3 Self Efficacy (SE)**

A mixed between-within subjects analysis of variance was conducted to assess the impact of the different interventions on follow-up SE scores. There was no significant interaction between intervention group and time [F(3, 324) = .874, p=.455]. There was a substantial main effect for time [F(1,324) = 46.424, p=<.005), η2 = .125] with groups showing an increase in self efficacy (Time 1, M = 17.45, SD = 1.95; Time 2, M = 18.3, SD = 1.91). The main effect comparing the groups was not significant [F(3,324) = .156, p=.93], suggesting that provision of assistance with planning did not influence SE (Table 4).


Table 4. Study 1 group mean self efficacy scores pre- and post intervention

Subsequent analyses compared self-efficacy between those who returned FOBTs and those who did not. Table 5 shows that when we compared SE over time for FOBT non-returners using a paired samples t-test there was a decrease in confidence that approached significance (*p*=.08). In other words, the confidence of non-participants to screen was impacted negatively by the provision of the FOBT. By contrast, confidence among those who returned an FOBT


increased significantly, regardless of group assignment. This result suggests that, in general, confidence to complete the test in the future is likely to decrease for those people who don't complete initial screening, regardless of initial level of confidence.

\*\* p<.01

76 Colorectal Cancer – From Prevention to Patient Care

between the groups in FOBT participation or return within 6 weeks (i.e., before and after

FOBTs returned 76 (84) 66 (84) 70 (77) 74 (82) X2 (3)=1.980,

Plans returned\* 62 66 X2 (1)=.367,

Most participants who returned a completed FOBT and were also required to return a completed implementation plan did so. There was no significant difference in rate of return between aide and checklist (Table 3), suggesting that differing levels of directedness had no impact on whether the plans were completed. There were no cases of a plan being returned

A mixed between-within subjects analysis of variance was conducted to assess the impact of the different interventions on follow-up SE scores. There was no significant interaction between intervention group and time [F(3, 324) = .874, p=.455]. There was a substantial main effect for time [F(1,324) = 46.424, p=<.005), η2 = .125] with groups showing an increase in self efficacy (Time 1, M = 17.45, SD = 1.95; Time 2, M = 18.3, SD = 1.91). The main effect comparing the groups was not significant [F(3,324) = .156, p=.93], suggesting that provision

> Aide to return mean (SD)

\*These numbers do not correspond with participants who returned FOBTs within 6 weeks

Table 3. Study 1 return of kits and implementation plans by group

Aide to return n=91 (%)

67 (74) 61 (77) 62 (68) 66 (73) X2 (3)=.869,

Checklist to return n=90 (%)

Test of difference

p=.577

p=.833

p=.545

Checklist to return

mean (SD)

Aide to retain n=79 (%)

reminder) (Table 3).

Return of kits within 6 weeks Control N=90 (%)

**3.4.2 Return of implementation plans** 

without an accompanying completed kit.

of assistance with planning did not influence SE (Table 4).

Aide to retain mean (SD)

Table 4. Study 1 group mean self efficacy scores pre- and post intervention

Time 1 17.21 (1.81) 17.67 (2.03) 17.50 (1.73) 17.45 (2.22) Time 2 18.39 (2.04) 18.26 (1.98) 18.32 (1.93) 18.36 (1.73)

Subsequent analyses compared self-efficacy between those who returned FOBTs and those who did not. Table 5 shows that when we compared SE over time for FOBT non-returners using a paired samples t-test there was a decrease in confidence that approached significance (*p*=.08). In other words, the confidence of non-participants to screen was impacted negatively by the provision of the FOBT. By contrast, confidence among those who returned an FOBT

**3.4.3 Self Efficacy (SE)** 

Control

mean (SD)

\*\*\*p<.001

Table 5. Study 1 mean self efficacy scores pre- and post-intervention, overall and by return/non return of FOBTs

In order to determine whether confidence at baseline influenced reaction to the various interventions, participants were characterised as having a low or high SE score at baseline (determined by a median-split between 16 and 17), and change in confidence over time compared (See Table 5). Low SE non-returners did not significantly change their SE scores post intervention, whereas low SE returners' scores significantly increased post intervention. Similarly, for those with a high SE score at baseline, non-returners' scores significantly decreased post intervention but did not significantly change if they returned an FOBT. This latter result is likely to reflect ceiling effects given that the maximum score possible for SE was 20. These results suggest that self efficacy was increased when the test was completed but the initial level of confidence to complete the test was low, and conversely confidence was decreased when the initial level was high but the test was not completed.

#### **3.4.4 Commitment to screen and maintain screening**

At baseline, the majority of people (n=217/343, 63%) were committed or very committed to doing the test (M=4.39, SD=.924; median=5) and there were no group differences (Table 6). Those who returned an FOBT were asked their level of commitment to maintain screening, and just over half (n=137/239, 57.3%) were "very committed" to screening again (M=4.38, SD=.840, median=5, n=47 missing values), regardless of intervention assignment. For those that did return an FOBT, a paired-sample t-test indicated that for the sample as a whole there was a statistically significant decrease in commitment to screen from baseline, ie after exposure to the intervention and FOBT (Table 7). When we examined the relationship between commitment and self efficacy by comparing commitment level between those who had a low or high SE baseline score, it was apparent that the decrease in commitment came from those that had a high initial SE score (Table 7).

Turning Intention Into Behaviour: The Effect of Providing

**4.1.1 Sample size and selection** 

**4.1 Methods** 

names were allocated.

**4.1.2 Study conduct** 

deliberately ascertained.

**4.1.3 Data analysis** 

**4.2 Results** 

efficacy and commitment to screening.

participation data relayed to the researchers.

conducted to assess FOBT participation between groups.

Cues to Action on Participation Rates for Colorectal Cancer Screening 79

Sample selection proceeded as described for Study 1. A separate sample of 6000 men and women aged between 50 and 76 years, randomly selected from four South Australian electoral divisions, was obtained from the AEC. People residing in postcodes included in the pilot NBCSP were omitted from the sample, as were those whose address indicated they resided in a hostel or nursing home. The remaining sample was randomised separately by sex and 400 men and women were assigned sequentially to one of 4 groups. In total 1600

Phase 1: All potential participants were mailed an advance notification letter (which aligns with the protocol adopted by the NBCSP) and accompanying information as for Study 1, and were informed that they would shortly be receiving a screening package in the mail. Exclusion due to ineligibility was dependent upon self-identification and communication of this fact to the researchers before despatch of FOBT. Willingness to participate was not

Phase 2: Three weeks after the advance notification letter, a screening kit including an immunochemical FOBT was sent to individuals. As for Study 1, intervention groups also received a discrete implementation plan. The nature of this approach precluded us from ascertaining willingness to participate and from obtaining pre- and post measures of self

Phase 3: Receipt of completed FOBTs was recorded by the Bowel Health Service and

Participation rates were viewed as 'early' or 'late' at a cut-off point of 6 weeks following despatch of FOBT, when a reminder was sent to non-responders. Chi-square analysis was

N=1600 men and women were sent an advance warning letter. Those who did not identify themselves as ineligible or not wishing to participate were then mailed a screening kit and accompanying material according to intervention group. In total, n=225 were excluded from the study (n=118 identified themselves as ineligible; n=83 didn't wish to participate; n=24 packages were undeliverable). Analyses were therefore conducted for n=1375 men and

At baseline, the groups were balanced for gender (Table 9). It wasn't possible to ascertain age group breakdowns because the AEC supplied a random sample within an age range (50–74 years) which wasn't broken down into groups (for Study 1 we ascertained age from the participant). The study design also precluded us obtaining other demographic information (mean age, education, country of birth) as we did for Study 1. However, given that the underlying sampling mechanism was identical (i.e., supplied by the AEC), there is

women. Recruitment and participation attrition rates are shown at Table 8.

some confidence that the groups were balanced on these other factors.


\*Includes non-returners and returners; 7/350 missing values

\*\*Includes only those who returned an FOBT; 47/286 missing values

Table 6. Study 1 mean commitment to screen by group pre- and post intervention


\*<.05 \*\*\*<.001

Table 7. Study 1 FOBT returners' commitment to screen pre- and post-intervention, overall and by SE level at baseline

#### **3.4.5 Effect of self efficacy and commitment to screen on use of FOBT**

Logistic regression was used to assess the independent and joint effects of baseline SE and baseline commitment to screen on return of FOBT. SE alone made a statistically significant contribution, X2 (1, n=350)=11.535, p<.001, OR=1.27, CI 1.10-1.47), predicting 5.3% of the variance (Nagelkerke R squared) in screening uptake. Commitment to screen alone also made a statistically significant contribution X2 (1, n=343)=13.837, p<001, OR=1.67, CI 1.28- 2.18), and explained 6.4% of the variance. When these predictors were entered together into the logistic regression model, there was a statistically significant effect, X2(2, n=343)=17.487, p<.001), but only commitment to screen displayed a unique and statistically significant contribution (p=.012, OR=1.46, CI 1.07-1.97); baseline self efficacy was marginally significant (p=.06, OR=1.17, CI.993-1.37). This suggests that those who are committed to using the FOBT will do so regardless of their level of confidence. The total variance explained by the combined model was R2=8.0%, indicating that factors other than these also contribute to the likelihood of completing an FOBT.

#### **4. Study 2**

Study 2 was conducted to examine the generalisability of Study 1's results to the broader population. This approach more closely approximated that undertaken in current population screening programs utilising FOBTs.

#### **4.1 Methods**

78 Colorectal Cancer – From Prevention to Patient Care

Checklist to return mean (SD)

ANOVA

(3, 339)

(3,235)

M SD df t

Time 1 4.20 0.947 98 -1.522 Time 2 4.36 0.814

133 4.485\*\*\* Time 2 4.39 0.866

Time 1 4.76 0.578

.087

.294

Aide to return mean (SD)

*F*, (df) p

Table 6. Study 1 mean commitment to screen by group pre- and post intervention

Commitment to screen (full sample, n=233) Time 1 4.52 0.804 232 2.15\* Time 2 4.38 0.843

Table 7. Study 1 FOBT returners' commitment to screen pre- and post-intervention, overall

Logistic regression was used to assess the independent and joint effects of baseline SE and baseline commitment to screen on return of FOBT. SE alone made a statistically significant contribution, X2 (1, n=350)=11.535, p<.001, OR=1.27, CI 1.10-1.47), predicting 5.3% of the variance (Nagelkerke R squared) in screening uptake. Commitment to screen alone also made a statistically significant contribution X2 (1, n=343)=13.837, p<001, OR=1.67, CI 1.28- 2.18), and explained 6.4% of the variance. When these predictors were entered together into the logistic regression model, there was a statistically significant effect, X2(2, n=343)=17.487, p<.001), but only commitment to screen displayed a unique and statistically significant contribution (p=.012, OR=1.46, CI 1.07-1.97); baseline self efficacy was marginally significant (p=.06, OR=1.17, CI.993-1.37). This suggests that those who are committed to using the FOBT will do so regardless of their level of confidence. The total variance explained by the combined model was R2=8.0%, indicating that factors other than these also contribute to the

Study 2 was conducted to examine the generalisability of Study 1's results to the broader population. This approach more closely approximated that undertaken in current

**3.4.5 Effect of self efficacy and commitment to screen on use of FOBT** 

4.48 (.844) 4.57 (.854) 4.33 (.974) 4.23 (.984) 2.21

4.41 (.938) 4.33 (.816) 4.53 (.704) 4.24 (.878) 1.25

Control

Time 1\* (n=343)

Time 2\*\* (n=239)

(n=99)

(n=134)

and by SE level at baseline

likelihood of completing an FOBT.

population screening programs utilising FOBTs.

**4. Study 2** 

\*<.05 \*\*\*<.001 mean (SD)

Aide to retain mean (SD)

\*Includes non-returners and returners; 7/350 missing values \*\*Includes only those who returned an FOBT; 47/286 missing values

Low baseline SE score commitment to screen

High baseline SE score commitment to screen

#### **4.1.1 Sample size and selection**

Sample selection proceeded as described for Study 1. A separate sample of 6000 men and women aged between 50 and 76 years, randomly selected from four South Australian electoral divisions, was obtained from the AEC. People residing in postcodes included in the pilot NBCSP were omitted from the sample, as were those whose address indicated they resided in a hostel or nursing home. The remaining sample was randomised separately by sex and 400 men and women were assigned sequentially to one of 4 groups. In total 1600 names were allocated.

#### **4.1.2 Study conduct**

Phase 1: All potential participants were mailed an advance notification letter (which aligns with the protocol adopted by the NBCSP) and accompanying information as for Study 1, and were informed that they would shortly be receiving a screening package in the mail. Exclusion due to ineligibility was dependent upon self-identification and communication of this fact to the researchers before despatch of FOBT. Willingness to participate was not deliberately ascertained.

Phase 2: Three weeks after the advance notification letter, a screening kit including an immunochemical FOBT was sent to individuals. As for Study 1, intervention groups also received a discrete implementation plan. The nature of this approach precluded us from ascertaining willingness to participate and from obtaining pre- and post measures of self efficacy and commitment to screening.

Phase 3: Receipt of completed FOBTs was recorded by the Bowel Health Service and participation data relayed to the researchers.

#### **4.1.3 Data analysis**

Participation rates were viewed as 'early' or 'late' at a cut-off point of 6 weeks following despatch of FOBT, when a reminder was sent to non-responders. Chi-square analysis was conducted to assess FOBT participation between groups.

#### **4.2 Results**

N=1600 men and women were sent an advance warning letter. Those who did not identify themselves as ineligible or not wishing to participate were then mailed a screening kit and accompanying material according to intervention group. In total, n=225 were excluded from the study (n=118 identified themselves as ineligible; n=83 didn't wish to participate; n=24 packages were undeliverable). Analyses were therefore conducted for n=1375 men and women. Recruitment and participation attrition rates are shown at Table 8.

At baseline, the groups were balanced for gender (Table 9). It wasn't possible to ascertain age group breakdowns because the AEC supplied a random sample within an age range (50–74 years) which wasn't broken down into groups (for Study 1 we ascertained age from the participant). The study design also precluded us obtaining other demographic information (mean age, education, country of birth) as we did for Study 1. However, given that the underlying sampling mechanism was identical (i.e., supplied by the AEC), there is some confidence that the groups were balanced on these other factors.

Turning Intention Into Behaviour: The Effect of Providing

**4.2.2 Return of implementation plans** 

**5. Discussion** 

evidence of the use of these plans.

Cues to Action on Participation Rates for Colorectal Cancer Screening 81

A considerable proportion of those who returned an FOBT and were also required to return a completed implementation plan did not do so, and significantly fewer people returned the prescriptive plan (i.e., checklist) than the aide (Table 10), suggesting that level of directedness may have an effect on whether the plans were completed—those who were required to formulate their own plan based on suggestions for action were more likely to return a plan compared to those given a prescriptive checklist. Notwithstanding this result, given that the requirement for return was to act as an indicator of whether plans had actually been formulated, it appears that around half the participants used the FOBT without adhering to

We hypothesised that the formation of implementation plans would assist return of FOBT kits by providing a physical cue to action. In addition we hypothesised that the process of completing a plan would increase confidence in ability to complete the test and that those who were strongly committed to screening at baseline would differ in formation of implementation plans and participation to those with a less strong initial commitment. Notwithstanding the difference in overall participation figures between Study 1 (81.7%) and Study 2 (39.9%), we found that for both studies provision of assistance with planning, regardless of directedness, had no influence on completion of an FOBT. The lack of influence of an implementation plan concurs with the conclusions of other researchers who have also found no effect of implementation planning on subsequent behaviour (Jackson et al., 2005; Michie et al., 2004; Rutter et al., 2006; Skar et al., 2011). Even so, this result goes against the large body of evidence suggesting that formulating action plans has a positive effect on the intention-behaviour gap. It has been suggested, however, that there exists sparse evidence for a positive effect of implementation intentions on behaviours outside student samples, who are more likely to comply with task demands (actually formulating the plan) (Jackson et al., 2005; Schweiger Gallo & Gollwitzer, 2007). It has also been argued that implementation plans are only effective where there is motivation to achieve a goal (Sniehotta, 2009) and that where goal intentions are positive, so will be the effects of implementation intentions (Gollwitzer, 1993; Oettingen et al., 2000). The majority of FOBT returners in Study 1 already had a high intention to screen, which may be attributable to the fact that that they had made a conscious decision to participate and were presumably more motivated to act, but in any case there was no evidence of a differential effect of combining high commitment with formation of implementation plans on FOBT return. Indeed, the high proportion of implementation plans returned by Study 1 participants (82%) may be indicative only of compliance with the study requirements (i.e., to return plans) rather than

However, and in contrast to Study 1, it is evident that nearly half the FOBTs returned in Study 2 were completed without making a plan, a result which could reasonably be extrapolated to the group that was asked to retain their formulated plan. It has been suggested that non-completion may reflect ambiguity of study instructions (Michie et al., 2004) but, given that nearly all Study 1 participants returned identically-constructed implementation plans with a completed FOBT, this was not the case in our population. Rather, this outcome suggests that some felt they had no need to complete plans, perhaps because their intentions were sufficiently strong to make the use of plans unnecessary.

planning instructions, particularly those who received a prescriptive plan.


Table 8. Study 2 interventions by phase and arm, with attrition rates


Table 9. Study 2 participant demographic characteristics

#### **4.2.1 FOBT participation**

Completed FOBTs were returned by 548/1375 (39.9%) of participants over a period of 26 weeks (mean = 5.51 weeks). This rate is similar to that achieved in the NBCSP in 2008 (i.e., 41% (AIHW, 2010). As for Study 1, contrary to our hypothesis that the formation of implementation plans would improve FOBT uptake, there was no significant difference between the groups in FOBT participation or return within 6 weeks (before and after reminder) (Table 10).


Table 10. Study 2 overall return of kits and within 6 weeks (i.e. before reminder) by group

#### **4.2.2 Return of implementation plans**

A considerable proportion of those who returned an FOBT and were also required to return a completed implementation plan did not do so, and significantly fewer people returned the prescriptive plan (i.e., checklist) than the aide (Table 10), suggesting that level of directedness may have an effect on whether the plans were completed—those who were required to formulate their own plan based on suggestions for action were more likely to return a plan compared to those given a prescriptive checklist. Notwithstanding this result, given that the requirement for return was to act as an indicator of whether plans had actually been formulated, it appears that around half the participants used the FOBT without adhering to planning instructions, particularly those who received a prescriptive plan.

#### **5. Discussion**

80 Colorectal Cancer – From Prevention to Patient Care

FOBT screening package + implementation plan to be formulated and retained by participant (n=400)

FOBT screening package + implementation plan to be formulated and returned to researcher (n=400)

FOBT screening package + implementation plan devised by researcher to be completed and

> Checklist to return n=346 (%)

178 (51.4) 168

Test of difference

*X*2 (3)=0.96, p=.992

> Test of difference

p=.652

p=.352

p=.001

62/142 (43.6) X2 (1)=9.389,

FOBT screening package only (n=400)

**Measures Phase 3** 

(All groups, n=1375) Return of kit within and after 6 weeks

**Interventions Phase 2** 

*Aide to retain* 

*Aide to return* 

*Checklist to return* 

Table 8. Study 2 interventions by phase and arm, with attrition rates

Aide to retain n=350 (%)

176 (50.3) 174

Table 9. Study 2 participant demographic characteristics

n=345 (%)

returned to researcher (n=400)

Aide to return n=334 (%)

170 (50.9) 164

Aide to retain n=350 (%)

83/131

Completed FOBTs were returned by 548/1375 (39.9%) of participants over a period of 26 weeks (mean = 5.51 weeks). This rate is similar to that achieved in the NBCSP in 2008 (i.e., 41% (AIHW, 2010). As for Study 1, contrary to our hypothesis that the formation of implementation plans would improve FOBT uptake, there was no significant difference between the groups in FOBT participation or return within 6 weeks (before and after

FOBTs returned 144 (41.7) 131 (37.4) 131 (39.2) 142 (41.0) X2(3)=1.633,

Table 10. Study 2 overall return of kits and within 6 weeks (i.e. before reminder) by group

Aide to return n=334 (%)

(58.4)

106 (30.7) 98 (28.0) 97 (29.0) 94 (27.2) X2 (3)=3.269,

Checklist to return n=346 (%)

*Control* 

**Recruitment Phase 1** 

Potentially eligible participants randomised to study arm then mailed information

(Control)

Male Female n=345 (%)

176 (51.0) 169

**4.2.1 FOBT participation** 

reminder) (Table 10).

Return of kits within 6 weeks

Return of plans with FOBT

Control

N=1600

sheet and notification that they would shortly receive an FOBT kit. Ineligibility was defined and dependent upon self-report

We hypothesised that the formation of implementation plans would assist return of FOBT kits by providing a physical cue to action. In addition we hypothesised that the process of completing a plan would increase confidence in ability to complete the test and that those who were strongly committed to screening at baseline would differ in formation of implementation plans and participation to those with a less strong initial commitment.

Notwithstanding the difference in overall participation figures between Study 1 (81.7%) and Study 2 (39.9%), we found that for both studies provision of assistance with planning, regardless of directedness, had no influence on completion of an FOBT. The lack of influence of an implementation plan concurs with the conclusions of other researchers who have also found no effect of implementation planning on subsequent behaviour (Jackson et al., 2005; Michie et al., 2004; Rutter et al., 2006; Skar et al., 2011). Even so, this result goes against the large body of evidence suggesting that formulating action plans has a positive effect on the intention-behaviour gap. It has been suggested, however, that there exists sparse evidence for a positive effect of implementation intentions on behaviours outside student samples, who are more likely to comply with task demands (actually formulating the plan) (Jackson et al., 2005; Schweiger Gallo & Gollwitzer, 2007). It has also been argued that implementation plans are only effective where there is motivation to achieve a goal (Sniehotta, 2009) and that where goal intentions are positive, so will be the effects of implementation intentions (Gollwitzer, 1993; Oettingen et al., 2000). The majority of FOBT returners in Study 1 already had a high intention to screen, which may be attributable to the fact that that they had made a conscious decision to participate and were presumably more motivated to act, but in any case there was no evidence of a differential effect of combining high commitment with formation of implementation plans on FOBT return. Indeed, the high proportion of implementation plans returned by Study 1 participants (82%) may be indicative only of compliance with the study requirements (i.e., to return plans) rather than evidence of the use of these plans.

However, and in contrast to Study 1, it is evident that nearly half the FOBTs returned in Study 2 were completed without making a plan, a result which could reasonably be extrapolated to the group that was asked to retain their formulated plan. It has been suggested that non-completion may reflect ambiguity of study instructions (Michie et al., 2004) but, given that nearly all Study 1 participants returned identically-constructed implementation plans with a completed FOBT, this was not the case in our population. Rather, this outcome suggests that some felt they had no need to complete plans, perhaps because their intentions were sufficiently strong to make the use of plans unnecessary.

Turning Intention Into Behaviour: The Effect of Providing

with the pre-contemplation stage.

**7. Acknowledgements** 

**8. References** 

**6. Conclusion** 

Cues to Action on Participation Rates for Colorectal Cancer Screening 83

The low rate of participation in Study 2 may reflect a dissonance of messages appropriate to an individual's stage of readiness to screen (Prochaska, 2008). The differences in study design, particularly recruitment strategy, between studies 1 and 2 may have resulted in basic sample differences in stage of readiness to screen at baseline. Specifically, including only participants prepared to complete questionnaires in Study 1 resulted in a highly committed sample, likely to be in contemplation or preparation to act stage, characterised by a high participation rate. By contrast, Study 2 invitees were a population sample, most of who were probably in pre-contemplation on receipt of the FOBT, with participation rates comparable with those achieved by the national screening program (i.e. ~ 40%). Precontemplation is a stage where it could be argued that a person's knowledge, attitudes and intentions are in a more unstable state. People in this stage have been shown to have higher barriers, higher chance health locus of control, low powerful others health locus of control, lower perceived susceptibility and lower CRC knowledge (Gregory et al., 2011). It follows that these factors should be addressed to facilitate movement through contemplation to the action stage. However, our implementation plans as formulated were aimed at those with an intention to act and focused on the where, when and how of successful completion of the FOBT. It could be daunting for those who had never heard of FOBT screening to receive a test and accompanying material designed to assist with completing the test without first being given information aimed at overcoming barriers and lack of knowledge associated

The provision of assistance with the preparation of implementation plans, regardless of their level of directedness, had no influence on FOBT participation in the 2 studies conducted. One reason for their lack of effect may be that the majority of participants were likely to be in pre-contemplation stage in Study 2 and in the action stage in Study 1. Thus ceiling effects limited the potential for cues to impact behaviour among participants in Study 1, and Study 2 participants may have benefited from an intervention that tackled Contemplation as an intermediary to Action. This stage mismatch has implications for population-based screening programs and may contribute toward less than optimal screening uptake rates. Future research could usefully address the potential for the communication within a population setting of material targeted to specific decision stages, designed to progressively move an individual toward action and maintenance of action. Our research indicated that confidence to screen and commitment to screen separately and together exerted a greater influence on actual FOBT participation; however, these factors accounted for a small amount

of variance and future research should address the contribution of other factors.

AIHW. (2010). *National Bowel Cancer Screening Program: Annual monitoring report 2009; Data* 

*supplement 2010*. Canberra: Australian Institute of Health and Welfare & Australian

We would like to thank Ian Zajac, CSIRO, for timely statistical advice.

Department of Health and Ageing

Indeed, we found from Study 1 that commitment had the most significant influence on FOBT use—because the majority of participants were strongly committed, we were unable to determine if having a weak level of commitment would influence formulation of an implementation plan or use of the FOBT. Of those Study 2 participants that did formulate and return plans, significantly fewer used the prescriptive 'checklist' format. Participants may have been "turned off" by the directedness of the checklist, particularly since they were a population sample and had not made a mindful decision to participate in a study. Study 1 demonstrated that provision of directions did not increase people's self efficacy. These results accord with a meta-analysis of 66 randomised controlled studies that concluded that forming implementation intentions had negligible effects on self efficacy and goal intentions (Webb & Sheeran, 2008).

For the group as a whole, baseline self efficacy did not have a strong influence on whether people used the test; rather, the act itself of completing the FOBT determined confidence—self efficacy was increased when the initial level of confidence to complete the FOBT was low, and conversely confidence was decreased when the initial level was high but the test was not completed. Rather than confidence to use the FOBT, from Study 1 it appears that being initially committed to screening had a more significant influence on whether people actually did use the FOBT, confirming the general consensus that intention to perform a behaviour is a necessary precursor of action. Even so, we found in Study 1 that commitment to screening, while a significant predictor of FOBT use, in conjunction with self efficacy explained only 8.0% of the variance, indicating that other factors exist which contribute to the likelihood of completing an FOBT. For example, Gregory et al. (2011) found that social-cognitive predictors of intention to screen for CRC and actual screening behaviour, although overlapping, were not the same, and Power and colleagues (2008) in their study of CRC screening found that life difficulty variables were better predictors of action than intention.

It is puzzling to note that there was a significant decrease in commitment to repeat screening by those that did use the FOBT and had a high initial level of confidence, in contrast to those with low confidence whose level of commitment to screening did not change. It may be that initial commitment was high for most because the participants were an 'interested' sample, and that those with high SE who screened reinforced their view that they were capable of completing an FOBT without necessarily moving from that conclusion and forming a commitment to rescreen. Conversely, those with low confidence but who did complete their test, thereby increasing their confidence, could have felt 'motivated' to repeat the experience again and so not changed their level of commitment. Interestingly, the same lessening of intention by those with high self efficacy was noted in a study examining the role of self efficacy in testicular self-examination (Umeh & Chadwick, 2010). The researchers found that those with high self efficacy appeared to have worsened attitudes toward self examination when both vulnerability and severity estimates were low. The same situation could well apply to CRC screening, particularly as perceived susceptibility is a Preventive Health Model (PHM) construct demonstrated to be associated with CRC screening ((Flight et al., 2010; Tiro et al., 2005). Commitment to future CRC screening in one or 2 years would perhaps, as Umeh and Chadwick (2010) have suggested, be temporarily rejected if the penalties of inaction are deemed insignificant, a viewpoint which may stem from a defensive reaction activated by anxiety. This view suggests that an emphasis on the development of messages designed to increase perceptions of personal risk of CRC without raising anxiety are warranted.

The low rate of participation in Study 2 may reflect a dissonance of messages appropriate to an individual's stage of readiness to screen (Prochaska, 2008). The differences in study design, particularly recruitment strategy, between studies 1 and 2 may have resulted in basic sample differences in stage of readiness to screen at baseline. Specifically, including only participants prepared to complete questionnaires in Study 1 resulted in a highly committed sample, likely to be in contemplation or preparation to act stage, characterised by a high participation rate. By contrast, Study 2 invitees were a population sample, most of who were probably in pre-contemplation on receipt of the FOBT, with participation rates comparable with those achieved by the national screening program (i.e. ~ 40%). Precontemplation is a stage where it could be argued that a person's knowledge, attitudes and intentions are in a more unstable state. People in this stage have been shown to have higher barriers, higher chance health locus of control, low powerful others health locus of control, lower perceived susceptibility and lower CRC knowledge (Gregory et al., 2011). It follows that these factors should be addressed to facilitate movement through contemplation to the action stage. However, our implementation plans as formulated were aimed at those with an intention to act and focused on the where, when and how of successful completion of the FOBT. It could be daunting for those who had never heard of FOBT screening to receive a test and accompanying material designed to assist with completing the test without first being given information aimed at overcoming barriers and lack of knowledge associated with the pre-contemplation stage.

#### **6. Conclusion**

82 Colorectal Cancer – From Prevention to Patient Care

Indeed, we found from Study 1 that commitment had the most significant influence on FOBT use—because the majority of participants were strongly committed, we were unable to determine if having a weak level of commitment would influence formulation of an implementation plan or use of the FOBT. Of those Study 2 participants that did formulate and return plans, significantly fewer used the prescriptive 'checklist' format. Participants may have been "turned off" by the directedness of the checklist, particularly since they were a population sample and had not made a mindful decision to participate in a study. Study 1 demonstrated that provision of directions did not increase people's self efficacy. These results accord with a meta-analysis of 66 randomised controlled studies that concluded that forming implementation intentions had negligible effects on self efficacy and goal intentions

For the group as a whole, baseline self efficacy did not have a strong influence on whether people used the test; rather, the act itself of completing the FOBT determined confidence—self efficacy was increased when the initial level of confidence to complete the FOBT was low, and conversely confidence was decreased when the initial level was high but the test was not completed. Rather than confidence to use the FOBT, from Study 1 it appears that being initially committed to screening had a more significant influence on whether people actually did use the FOBT, confirming the general consensus that intention to perform a behaviour is a necessary precursor of action. Even so, we found in Study 1 that commitment to screening, while a significant predictor of FOBT use, in conjunction with self efficacy explained only 8.0% of the variance, indicating that other factors exist which contribute to the likelihood of completing an FOBT. For example, Gregory et al. (2011) found that social-cognitive predictors of intention to screen for CRC and actual screening behaviour, although overlapping, were not the same, and Power and colleagues (2008) in their study of CRC screening found that life difficulty variables were

It is puzzling to note that there was a significant decrease in commitment to repeat screening by those that did use the FOBT and had a high initial level of confidence, in contrast to those with low confidence whose level of commitment to screening did not change. It may be that initial commitment was high for most because the participants were an 'interested' sample, and that those with high SE who screened reinforced their view that they were capable of completing an FOBT without necessarily moving from that conclusion and forming a commitment to rescreen. Conversely, those with low confidence but who did complete their test, thereby increasing their confidence, could have felt 'motivated' to repeat the experience again and so not changed their level of commitment. Interestingly, the same lessening of intention by those with high self efficacy was noted in a study examining the role of self efficacy in testicular self-examination (Umeh & Chadwick, 2010). The researchers found that those with high self efficacy appeared to have worsened attitudes toward self examination when both vulnerability and severity estimates were low. The same situation could well apply to CRC screening, particularly as perceived susceptibility is a Preventive Health Model (PHM) construct demonstrated to be associated with CRC screening ((Flight et al., 2010; Tiro et al., 2005). Commitment to future CRC screening in one or 2 years would perhaps, as Umeh and Chadwick (2010) have suggested, be temporarily rejected if the penalties of inaction are deemed insignificant, a viewpoint which may stem from a defensive reaction activated by anxiety. This view suggests that an emphasis on the development of messages designed to increase perceptions of personal risk of CRC without

(Webb & Sheeran, 2008).

better predictors of action than intention.

raising anxiety are warranted.

The provision of assistance with the preparation of implementation plans, regardless of their level of directedness, had no influence on FOBT participation in the 2 studies conducted. One reason for their lack of effect may be that the majority of participants were likely to be in pre-contemplation stage in Study 2 and in the action stage in Study 1. Thus ceiling effects limited the potential for cues to impact behaviour among participants in Study 1, and Study 2 participants may have benefited from an intervention that tackled Contemplation as an intermediary to Action. This stage mismatch has implications for population-based screening programs and may contribute toward less than optimal screening uptake rates. Future research could usefully address the potential for the communication within a population setting of material targeted to specific decision stages, designed to progressively move an individual toward action and maintenance of action. Our research indicated that confidence to screen and commitment to screen separately and together exerted a greater influence on actual FOBT participation; however, these factors accounted for a small amount of variance and future research should address the contribution of other factors.

#### **7. Acknowledgements**

We would like to thank Ian Zajac, CSIRO, for timely statistical advice.

#### **8. References**

AIHW. (2010). *National Bowel Cancer Screening Program: Annual monitoring report 2009; Data supplement 2010*. Canberra: Australian Institute of Health and Welfare & Australian Department of Health and Ageing

Turning Intention Into Behaviour: The Effect of Providing

pp. 95-103, ISSN 1070-5503

No. 19, pp. 1365-1371, ISSN 0028-4793

Vol. 55, pp. 218-222, ISSN 0738-3991

ISSN 0146-1672

134, ISSN 0883-6612

0036-5521

Cues to Action on Participation Rates for Colorectal Cancer Screening 85

Lechner, L.; Oenema, A. & Nooijer, J. (2002). Testicular self-examination (TSE) among Dutch

Luszczynska, A. (2004). Change in breast self-examination behavior: Effects of intervention

Mandel, J.; Bond, J.; Church, T.; Snover, D.; Bradley, G.; Schuman, L. & Ederer, F. (1993).

Mandel, J.; Church, T.; Bond, J.; Ederer, F.; Geisser, M.; Mongin, S.; Snover, D. & Schuman, L.

Milne, S.; Orbell, S. & Sheeran, P. (2002). Combining motivational and volitional interventions

Power, E., Van Jaarsveld, C.H.M., McCaffery, K., Miles, A., Atkin, W. & Wardle, J. (2008).

Prochaska, J. (2008). Decision making in the Transtheoretical Model of Behavior Change.

Prochaska, J.; Velicer, W.; DiClemente, C. & Fava, J. (1988). Measuring processes of change:

Rutter, D.; Steadman, L. & Quine, L. (2006). An implementations intervention to increase

Schwarzer, R. & Fuchs, R. (1995). Self-efficacy and health behaviours. In:*Predicting health* 

*Behavioral Medicine,* Vol. 35, pp. 285-294, ISSN 0883-6612

*Psychology,* Vol. 56, No. 4, pp. 520-528, ISSN 0022-006X

*Medical Decision Making,* Vol. 28, pp. 845-849, ISSN 0272-989X

*Education Research,* Vol. 17, pp. 73-84, ISSN 0268-1153

rounds. *Scandinavian Journal of Gastroenterology,* Vol. 39, No. 9, pp. 846-851, ISSN

young men aged 15-19: Determinants of the intention to practice TSE. *Health* 

on enhancing self-efficacy. *International Journal of Behavioral Medicine,* Vol. 11, No. 2,

Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study. *New England Journal of Medicine,* Vol. 328,

(2000). The effect of fecal occult-blood screening on the incidence of colorectal cancer. *New England Journal of Medicine,* Vol. 343, No. 22, pp. 1603-1607, ISSN 0028-4793 Michie, S.; Dormandy, E. & Marteau, T. (2004). Increasing screening uptake amongst those

intending to be screened: the use of action plans. *Patient Education and Counseling,* 

to promote exercise participation: Protection motivation theory and implementation intentions. *British Journal of Health Psychology,* Vol. 7, pp. 183-184, ISSN 1359-107X Oettingen, G.; Honig, G. & Gollwitzer, P. (2000). Effective self-regulation of goal attainment. *International Journal of Educational Research,* Vol. 33, pp. 705-732, ISSN:0883-0355 Orbell, S.; Hodgins, S. & Sheeran, P. (1997). Implementation intentions and the theory of

planned behaviour. *Personality and Social Psyschology Bulletin,* Vol. 23, pp. 945-954,

Understanding intentions and action in colorectal cancer screening. *Annals of* 

applications to the cessation of smoking. *Journal of Consulting and Clinical* 

uptake of mammography. *Annals of Behavioural Medicine,* Vol. 32, No. 2, pp. 127-

*behavior: Research and practice with social cognition models*, M. Conner & P. Norman (Eds.), (pp. 163-196). ISBN 033519320X, Buckingham, UK: Open University Press Schweiger Gallo, I. & Gollwitzer, P. (2007). Implementation intentions: A look back at fifteen years of progress. *Psicothema,* Vol. 19, No. 1, pp. 37-42, ISSN 0214-9915 Sheeran, P. (2002). Intention-behavior relations: A conceptual and empirical review. *European Review of Social Psychology,* Vol. 12, No. 1, pp. 1-36, ISSN 1046-3283 Sheeran, P. & Orbell, S. (2000). Using implementation intentions to increase attendance for cervical cancer screening. *Health Psychology,* Vol. 19, pp. 283-289, ISSN 0278-6133 Skar, S.; Sniehotta, F.; Molloy, G.; Prestwich, A. & Araujo-Soares, V. (2011). Do brief online

planning intereventions increase physical activity amongst university students? A


Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In: *Action-Control:* 

Bagozzi, R. & Warshaw, P. (1990). Trying to consume. *Journal of Consumer Research,* Vol. 17,

Becker, M.; Haefner, D. & Maiman, L. (1977). The health belief model in the prediction of

Conner, M. & Norman, P. (Eds.). (2005). *Predicting Health Behaviour: Research and Practice with* 

DeVellis, B., Blalock, S.J. & Sandler, R. (1990). Predicting participation in cancer screening:

Fisher, J.; Fikry, C. & Troxel, A. (2006). Cutting Cost and Increasing Access to Colorectal

*Health Education & Behavior,* Vol. 37, No. 5, pp. 724-736, ISSN 1090-1981 Frazier, A.; Colditz, G.; Fuchs, C. & Kuntz, K. (2000). Cost-effectiveness of screening for

Gollwitzer, P. (1993). Goal achievement: The role of intentions. *European Review of Social* 

Gollwitzer, P. & Sheeran, P. (2006). Implementation intentions and goal achievement: A

Gregory, T.; Wilson, C.; Duncan, A.; Turnbull, D.; Cole, S. & Young, G. (2011). Demographic,

for colorectal cancer. *Lancet,* Vol. 348, pp. 1472-1477, ISSN 0099-5355 Hewitson, P.; Glasziou, P.; Irwig, L.; Towler, B. & Watson, E. (2007). *Screening for colorectal* 

7134-451, Heidelberg, Germany: Springer

pp. 127-140, ISSN 0093-5301

University Press

0098-7484

18, pp. 348-366, ISSN 0022-1465

20, pp. 639-660, ISSN 0021-9010

*Psychology,* Vol. 4, pp. 141-185, ISSN 1046-3283

*Medicine,* Vol. 60, pp. 2382-2391, ISSN 0037-7856

Vol. 38, pp. 69-119, ISSN 0065-2601

*From cognition to behavior*, J. Kuhl & J. Beckman (Eds.), (pp. 11-39), ISBN 978-038-

dietary compliance: A field experiment. *Journal of Health and Social Behaviour,* Vol.

*Social Cognition Models* (2nd ed.). ISBN 13 978 0335 21176 0, Maidenhead, UK: Open

The role of perceived behavioural control. *Journal of Applied Social Psychology,* Vol.

Cancer Screening: Another Approach to Following the Guidelines. *Cancer Epidemiology Biomarkers & Prevention,* Vol. 15, No. 1, pp. 108-113, ISSN 1055-9965 Flight, I.; Wilson, C.; McGillivray, J. & Myers, R. (2010). Cross-cultural validation of the

Preventive Health Model for colorectal cancer screening: An Australian study.

colorctal cancer in the general population. *JAMA,* Vol. 284, pp. 1954-1961, ISSN

meta-analysis of effects and processes. *Advances in Experimental Social Psychology,* 

social cognitive and social ecological predictors of intention and participation in screening for colorectal cancer. *BMC Public Health,* Vol. 11, pp. 38, ISSN 1471-2458 Hardcastle, J.; Chamberlain, J.; Robinson, M.; Moss, S.; Amar, S.; Balfour, T.; James, P. &

Mangham, C. (1996). Randomised controlled trial of faecal-occult-blood screening

*cancer using the faecal occult blood test, Hemoccult*: Cochrane Database of Systematic Reviews Issue 1. Art. No.: CD001216. DOI: 10.1002/14651858.CD001216.pub2 International Cancer Screening Network. (2008). Inventory of Colorectal Cancer Screening

Activities in ICSN Countries, May 2008. Retrieved 22 June 2011, from

Beyond intention: do specific plans increase health behaviours in patients in primary care? A study of fruit and vegetable consumption. *Social Science &* 

biennial screening with a faecal occult blood test: results after nine screening

http://www.appliedresearch.cancer.gov/icsn/colorectal/screening.html Jackson, C.; Lawton, R.; Knapp, P.; Ranor, D.; Conner, M.; Lowe, C. & Closs, S. (2005).

Jemal, A.; Bray, F.; Center, M.; Ferlay, J.; Ward, E. & Forman, D. (2011). Global Cancer Statistics. *CA: A Cancer Journal for Clinicians,* Vol. 61, pp. 69-90, ISSN 0007-9235 Kronborg, D.; Jorgensen, O.; Fenger, C. & Rasmussen, M. (2004). Randomized study of rounds. *Scandinavian Journal of Gastroenterology,* Vol. 39, No. 9, pp. 846-851, ISSN 0036-5521


**5** 

Hitoshi Okamura *Hiroshima University,* 

*Japan* 

**Psychological Impact and Associated** 

**Results Concerning Hereditary** 

**Nonpolyposis Colorectal Cancer**

**Factors After Disclosure of Genetic Test** 

Advances in genetics in recent years have made major contributions to the development of medical genetics. The existence of "familial tumors" has been recognized, and genetic testing, with a potentially incalculable benefit to humanity, is being attempted (Offit, 1998). Numerous gene analyses related to the genesis and development of colorectal cancer have been conducted, and the existence of hereditary colorectal tumors in the form of hereditary nonpolyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP) has

HNPCC is caused by inherited germline mutations in mismatch repair genes and accounts for 2 -5% of colorectal cancers. The condition is characterized by young-onset, synchronous and metachronous tumors, and a predisposition to gynecologic, urinary tract, and extracolonic gastrointestinal cancers. Genetic testing usually begins with a family member who has been diagnosed with an HNPCC syndrome-related cancer (proband). If a deleterious mutation is identified, testing can be offered to the proband's family members, since they are at risk of carrying the mutation. Knowing one's genetic risk for hereditary

However, in contrast to the advances in scientific techniques, a great deal of apprehension exists with regard to the psychological or ethical, legal, and social issues (ELSI) associated with the application of these techniques. Since important personal genetic information that does not change throughout one's lifetime is handled during genetic diagnosis and an individual's genetic information is partly shared with blood relatives, with the impact of such genetic information not being limited to the individual, we find ourselves in a situation where new life health-care norms that also take psychosocial aspects into consideration are required. For this reason, a variety of studies have been conducted regarding the psychosocial aspects involved in the screening-test-taking behavior of high-risk people, the psychological aspects of high-risk people, interest in genetic counseling and genetic testing, and the psychosocial effects of genetic counseling. Studies on psychosocial aspects after being informed of the test results have also been reported recently, but many of these studies are concerning hereditary breast and ovarian cancer, and very few studies

cancers may facilitate the early detection or prevention of cancer.

**1. Introduction** 

been identified.

randomised controlled trial. *Psychology and Health,* Vol. 26, No. 4, pp. 399-417, ISSN 0887-0446


### **Psychological Impact and Associated Factors After Disclosure of Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer**

Hitoshi Okamura *Hiroshima University, Japan* 

#### **1. Introduction**

86 Colorectal Cancer – From Prevention to Patient Care

Sniehotta, F. (2009). towards a theory of intentional behaviour change: Plans, planning, and

Tiro, J.; Vernon, S.; Hyslop, T. & Myers, R. (2005). Factorial validity and invariance of a

Umeh, K. & Chadwick, R. (2010). Early detection of testicular cancer: revsiting the role of

Vernon, S.; Myers, R. & Tilley, B. (1997). Development and validation of an instrument to

*Epidemiology Biomarkers & Prevention,* Vol. 12, pp. 339-349, ISSN 1573-3521 Verplanken, B. & Faes, S. (1999). Good intentions, bad habits, and effects of forming

Webb, T. & Sheeran, P. (2006). Does changing behavioral intentions engender behavior

Webb, T. & Sheeran, P. (2008). Mechanisms of implementation intention effects: The role of

Weller, D.; Moss, S.; Butler, P.; Campbell, C.; Coleman, D.; Melia, J. & Robertson, R. (2006).

*the Department of Health*. Edinburgh: University of Edinburgh. UK.

0887-0446

1359-107X

ISSN 1573-3521

Vol. 14, pp. 2855-286, ISSN 1573-3521

Vol. 29, pp. 592-604, ISSN 1046-3283

132, No. 2, pp. 249-268, ISSN 0033-2909

*Social Psychology,* Vol. 47, pp. 373-395, ISSN 2044-8309

randomised controlled trial. *Psychology and Health,* Vol. 26, No. 4, pp. 399-417, ISSN

self-regulation. *British Journal of Health Psychology,* Vol. 14, No., pp. 261-273, ISSN

survey measuring psychosocial correlates of colorectal cancer screening among African Americans and Caucasians. *Cancer Epidemiology Biomarkers & Prevention,* 

self-efficacy in testicular self-examination among young asymptomatic males [Epub ahead of print]. *Journal of Behavioral Medicine,* published online 22 April 2010,

measure factors related to colorectal cancer screening adherence. *Cancer* 

implementation intentions on healthy eating. *European Review of Social Psychology,* 

change? A meta-analysis of the experimental evidence. *Psychological Bulletin,* Vol.

goal intentions, self-efficacy, and accessibility of plan components. *British Journal of* 

*English Pilot of Bowel Cancer Screening: an evaluation of the second round. Final Report to* 

Advances in genetics in recent years have made major contributions to the development of medical genetics. The existence of "familial tumors" has been recognized, and genetic testing, with a potentially incalculable benefit to humanity, is being attempted (Offit, 1998). Numerous gene analyses related to the genesis and development of colorectal cancer have been conducted, and the existence of hereditary colorectal tumors in the form of hereditary nonpolyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP) has been identified.

HNPCC is caused by inherited germline mutations in mismatch repair genes and accounts for 2 -5% of colorectal cancers. The condition is characterized by young-onset, synchronous and metachronous tumors, and a predisposition to gynecologic, urinary tract, and extracolonic gastrointestinal cancers. Genetic testing usually begins with a family member who has been diagnosed with an HNPCC syndrome-related cancer (proband). If a deleterious mutation is identified, testing can be offered to the proband's family members, since they are at risk of carrying the mutation. Knowing one's genetic risk for hereditary cancers may facilitate the early detection or prevention of cancer.

However, in contrast to the advances in scientific techniques, a great deal of apprehension exists with regard to the psychological or ethical, legal, and social issues (ELSI) associated with the application of these techniques. Since important personal genetic information that does not change throughout one's lifetime is handled during genetic diagnosis and an individual's genetic information is partly shared with blood relatives, with the impact of such genetic information not being limited to the individual, we find ourselves in a situation where new life health-care norms that also take psychosocial aspects into consideration are required. For this reason, a variety of studies have been conducted regarding the psychosocial aspects involved in the screening-test-taking behavior of high-risk people, the psychological aspects of high-risk people, interest in genetic counseling and genetic testing, and the psychosocial effects of genetic counseling. Studies on psychosocial aspects after being informed of the test results have also been reported recently, but many of these studies are concerning hereditary breast and ovarian cancer, and very few studies

Psychological Impact and Associated Factors After Disclosure of

aid, is needed.

HNPCC.

**3. Risk perception** 

perceived risk was unchanged.

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 89

however, concerns about psychological and psychosocial issues may present barriers to undergoing genetic testing. The development of patient education tools, such as the decision

HNPCC mutation carriers have a life-time risk of colorectal cancer of about 80%, while female carriers have a 40-60% risk of endometrial cancer and a 10-15% risk of ovarian cancer. Communicating cancer risk and assessing the perceived risk is very important for genetic counseling because of subsequent cancer prevention behavior or cancer-related distress. Four reports were extracted regarding risk perception among individuals at risk for

Codori et al. (2005) assessed the effect of genetic counseling on perceived lifetime risk and cancer-distress among 101 adult first-degree-relatives of colorectal cancer patients from families with known or suspected HNPCC. Most persons overestimated their cancer risk, and a higher perceived risk was associated with believing that colorectal cancer cannot be prevented. The individual perceived risk changed after counseling, although the mean

Domanska et al. (2007) investigated the perceived cancer risk among 47 HNPCC mutation carriers and correlated the findings with individual characteristics. A perceived risk of colorectal cancer above 60% was reported by 49% individuals, and only one reported a perceived risk > 80%. Female mutation carriers, individuals under the age of 50 years, and individuals who received their counseling within 1 year prior to the study reported a higher perceived risk of colorectal cancer. Individuals who had lost a parent to HNPCC-related cancer at an early age also reported a higher perceived risk. Regarding gynecological cancer, 33% of the women reported a perceived risk of 40-60% for endometrial cancer, whereas the

van Oostrom et al. (2007) studied the difference in cancer risk perception among 271 individuals who opted for genetic cancer susceptibility testing for a known familial BRCA1/2 or HNPCC related germline mutation. The assessment was conducted before, 1 week after, and 6 months after disclosure of the test results. Individuals from BRCA1/2 and HNPCC mutation families did not differ with regard to their risk perceptions over time. Individuals from BRCA1/2 families perceived hereditary cancer as being more serious. Grover et al. (2009) examined colorectal cancer risk perception among individuals tested for mismatch repair genes mutation and identified factors associated with an appropriate interpretation of their cancer risk. In this study, in particular, the authors paid attention to individuals with an indeterminate genetic test result. Pathogenic mutations in *MLH1* and *MSH2* have been identified in only 30% to 64% of families who meet the clinical criteria for HNPCC and have undergone testing. Genetic testing may not yield a definitive result because of the lack of an identifiable mutation in one of the known genes or a mutation of unclear pathogenic significance. In the absence of an identified family mutation, these results are considered indeterminate or uninformative. Patients remain at an increased risk for colorectal cancer, and intensive cancer screening recommendations are made based on their personal and family cancer histories. A total of 159 individuals who met the Revised Bethesda Guidelines and had previously undergone genetic testing participated in this study. Ninety individuals with a pathogenic mutation (true positive) correctly estimated their cancer risk. However, only 62% of individuals with an indeterminate genetic test result

remaining 67% either underestimated or overestimated their risk.

examining the impact of genetic testing for hereditary colorectal tumors have been performed.

In this article, the psychological consequences related to HNPCC are reviewed with regard to the following four points: (1) attitude toward genetic testing, (2) risk perception, (3) psychosocial effects of genetic counseling, and (4) psychosocial aspects after undergoing genetic testing and being informed of the test results. I have reviewed and selected nearly all the articles regarding these themes using the PubMed database.

#### **2. Attitude toward genetic testing**

Many subjects who undergo genetic counseling for HNPCC also wish to undergo genetic testing. However, some subjects refuse to undergo genetic testing, despite its potential benefits. Some previous studies investigated the relationships between the intention to undergo genetic testing and psychosocial variables.

Hadley et al. (2003) investigated attitudes, intention, and the completion of genetic testing among 111 newly identified family members (first-degree relatives) of individuals with HNPCC. Most (97%) stated their intention to pursue testing. Fifty-one percent reported that learning about their children's risks was the most important reason to consider testing. The participants' intentions to pursue genetic testing were significantly affected by concerns regarding their ability to handle the emotional aspects of testing and the psychosocial effect on family members. On the other hand, 39% identified the potential effect on their health insurance as the most important reason not to undergo testing.

Wakefield et al. (2007a) qualitatively assessed 22 individuals' attitudes toward genetic testing for HNPCC. The most frequently reported pros were "to help manage my risk of developing cancer", "to help my family", and "to know my cancer risk." The participants expressed concern about the potential psychological impact of genetic testing. The authors also found that some affected individuals may not fully comprehend the meaning of their potential test results.

Wakefield et al. (2008) conducted a randomized trial to measure the effectiveness of a tailored decision aid designed specifically to assist individuals to make informed decisions regarding genetic testing for HNPCC. The decision aid explains the evidence available regarding HNPCC-related cancer risks, the differences between a mutation search and predictive testing, and the potential benefits, risks, and limitations of testing (Wakefield et al., 2007b). One hundred and fifty-three individuals were randomly assigned to a group who received the decision aid or a group who received a control pamphlet. Evaluations were conducted 1 week after consultation and 6 months after the completion of the intervention using a questionnaire, and 95 subjects completed the 6-month follow-up questionnaire. Although the decision aid had no significant effect on the actual genetic testing decision, the participants who received the decision aid had significantly lower levels of decisional conflict regarding genetic testing and were more likely to be classified as having made an informed choice concerning genetic testing than participants who received a control pamphlet. Furthermore, men who received the decision aid had significantly higher knowledge levels regarding genetic testing than men who received a control pamphlet.

These reports suggest that most individuals pursue genetic testing to help manage their own risk of developing cancer and to learn about their children's risks. On the other hand, however, concerns about psychological and psychosocial issues may present barriers to undergoing genetic testing. The development of patient education tools, such as the decision aid, is needed.

#### **3. Risk perception**

88 Colorectal Cancer – From Prevention to Patient Care

examining the impact of genetic testing for hereditary colorectal tumors have been

In this article, the psychological consequences related to HNPCC are reviewed with regard to the following four points: (1) attitude toward genetic testing, (2) risk perception, (3) psychosocial effects of genetic counseling, and (4) psychosocial aspects after undergoing genetic testing and being informed of the test results. I have reviewed and selected nearly all

Many subjects who undergo genetic counseling for HNPCC also wish to undergo genetic testing. However, some subjects refuse to undergo genetic testing, despite its potential benefits. Some previous studies investigated the relationships between the intention to

Hadley et al. (2003) investigated attitudes, intention, and the completion of genetic testing among 111 newly identified family members (first-degree relatives) of individuals with HNPCC. Most (97%) stated their intention to pursue testing. Fifty-one percent reported that learning about their children's risks was the most important reason to consider testing. The participants' intentions to pursue genetic testing were significantly affected by concerns regarding their ability to handle the emotional aspects of testing and the psychosocial effect on family members. On the other hand, 39% identified the potential effect on their health

Wakefield et al. (2007a) qualitatively assessed 22 individuals' attitudes toward genetic testing for HNPCC. The most frequently reported pros were "to help manage my risk of developing cancer", "to help my family", and "to know my cancer risk." The participants expressed concern about the potential psychological impact of genetic testing. The authors also found that some affected individuals may not fully comprehend the meaning of their

Wakefield et al. (2008) conducted a randomized trial to measure the effectiveness of a tailored decision aid designed specifically to assist individuals to make informed decisions regarding genetic testing for HNPCC. The decision aid explains the evidence available regarding HNPCC-related cancer risks, the differences between a mutation search and predictive testing, and the potential benefits, risks, and limitations of testing (Wakefield et al., 2007b). One hundred and fifty-three individuals were randomly assigned to a group who received the decision aid or a group who received a control pamphlet. Evaluations were conducted 1 week after consultation and 6 months after the completion of the intervention using a questionnaire, and 95 subjects completed the 6-month follow-up questionnaire. Although the decision aid had no significant effect on the actual genetic testing decision, the participants who received the decision aid had significantly lower levels of decisional conflict regarding genetic testing and were more likely to be classified as having made an informed choice concerning genetic testing than participants who received a control pamphlet. Furthermore, men who received the decision aid had significantly higher knowledge levels regarding genetic testing than men who received a control

These reports suggest that most individuals pursue genetic testing to help manage their own risk of developing cancer and to learn about their children's risks. On the other hand,

the articles regarding these themes using the PubMed database.

insurance as the most important reason not to undergo testing.

**2. Attitude toward genetic testing** 

undergo genetic testing and psychosocial variables.

performed.

potential test results.

pamphlet.

HNPCC mutation carriers have a life-time risk of colorectal cancer of about 80%, while female carriers have a 40-60% risk of endometrial cancer and a 10-15% risk of ovarian cancer. Communicating cancer risk and assessing the perceived risk is very important for genetic counseling because of subsequent cancer prevention behavior or cancer-related distress. Four reports were extracted regarding risk perception among individuals at risk for HNPCC.

Codori et al. (2005) assessed the effect of genetic counseling on perceived lifetime risk and cancer-distress among 101 adult first-degree-relatives of colorectal cancer patients from families with known or suspected HNPCC. Most persons overestimated their cancer risk, and a higher perceived risk was associated with believing that colorectal cancer cannot be prevented. The individual perceived risk changed after counseling, although the mean perceived risk was unchanged.

Domanska et al. (2007) investigated the perceived cancer risk among 47 HNPCC mutation carriers and correlated the findings with individual characteristics. A perceived risk of colorectal cancer above 60% was reported by 49% individuals, and only one reported a perceived risk > 80%. Female mutation carriers, individuals under the age of 50 years, and individuals who received their counseling within 1 year prior to the study reported a higher perceived risk of colorectal cancer. Individuals who had lost a parent to HNPCC-related cancer at an early age also reported a higher perceived risk. Regarding gynecological cancer, 33% of the women reported a perceived risk of 40-60% for endometrial cancer, whereas the remaining 67% either underestimated or overestimated their risk.

van Oostrom et al. (2007) studied the difference in cancer risk perception among 271 individuals who opted for genetic cancer susceptibility testing for a known familial BRCA1/2 or HNPCC related germline mutation. The assessment was conducted before, 1 week after, and 6 months after disclosure of the test results. Individuals from BRCA1/2 and HNPCC mutation families did not differ with regard to their risk perceptions over time. Individuals from BRCA1/2 families perceived hereditary cancer as being more serious.

Grover et al. (2009) examined colorectal cancer risk perception among individuals tested for mismatch repair genes mutation and identified factors associated with an appropriate interpretation of their cancer risk. In this study, in particular, the authors paid attention to individuals with an indeterminate genetic test result. Pathogenic mutations in *MLH1* and *MSH2* have been identified in only 30% to 64% of families who meet the clinical criteria for HNPCC and have undergone testing. Genetic testing may not yield a definitive result because of the lack of an identifiable mutation in one of the known genes or a mutation of unclear pathogenic significance. In the absence of an identified family mutation, these results are considered indeterminate or uninformative. Patients remain at an increased risk for colorectal cancer, and intensive cancer screening recommendations are made based on their personal and family cancer histories. A total of 159 individuals who met the Revised Bethesda Guidelines and had previously undergone genetic testing participated in this study. Ninety individuals with a pathogenic mutation (true positive) correctly estimated their cancer risk. However, only 62% of individuals with an indeterminate genetic test result

Psychological Impact and Associated Factors After Disclosure of

experienced increased worry and physical symptoms after counseling.

insurance, choice or change of jobs, and obtaining a mortgage.

Event Scale) with a correct classification of 86%.

accumulation is needed.

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 91

with those for patients with cancer. A substantial minority, however, said that they

Bleiker et al. (2007) examined: 1) levels of cancer-specific distress more than one year after genetic counseling for HNPCC; 2) associations between sociodemographic, clinical and psychosocial factors and levels of distress; 3) the impact of genetic counseling on family relationships; and 4) the social consequences of genetic counseling. One hundred and sixteen individuals who participated in this study completed a self-report questionnaire by mail an average of 4 years after the last counseling session. Among all the subjects, 6% had clinically significant levels of cancer-specific distress (Impact of Event Scale). Having had contact with a professional psychosocial worker for cancer risk in the past 10 years was significantly associated with higher levels of current cancer specific distress. Only a minority of the subjects reported any adverse effects of genetic counseling on communication regarding genetic counseling with their children, family relationships, obtaining life

Keller et al. (2008) conducted a prospective study that examined the impact of multidisciplinary risk counseling on the psychosocial outcome of 139 affected cancer patients and 233 family members without cancer but at risk for HNPCC. Participants completed questionnaires specific to HNPCC before and 8 weeks after attending the cancer clinic. The levels of distress among affected patients exceeded those of unaffected individuals, as did worry regarding their relatives' risk. A significant reduction in general anxiety (Hospital Anxiety and Depression Scale), distress specific to familial colorectal cancer (Impact of Events Scale), and general cancer worry (Distress due to Hereditary Disorder) was demonstrated after counseling among both the affected patients and unaffected individuals. The reduction in distress was more pronounced among affected

patients given a high risk of HNPCC than among those with an intermediate risk.

Hasenbring et al. (2011) prospectively examined the impact of an initial interdisciplinary genetic counseling on feelings of anxiety with a special focus on subgroups related to personal cancer history, sex, age, and education. A significant interaction between time, sex, and age was identified for change in anxiety. While women in general and men older than 50 years revealed a significant reduction in anxiety, younger men did not show any change over time. A logistic regression analysis indicated that clinical Hospital Anxiety and Depression Scale-A cases could be predicted based on general distress (Brief Symptom Inventory) as well as by HNPCC-related cognitions of intrusion and avoidance (Impact of

These studies indicate that anxiety and cancer-specific distress are reduced after genetic counseling, suggesting an overall beneficial impact of comprehensive counseling. On the other hand, a minority of individuals, such as cancer-affected younger men, exhibited adverse effects of genetic counseling on psychosocial variables. Thus, healthcare providers (genetic counselors, human geneticists, oncologists, and psycho-oncologists) should always be aware of psychosocial issues after genetic counseling. However, as little data is available on the psychosocial effects of genetic counseling regarding HNPCC, further data

Since 1991, when a gene for hereditary cancer was first identified, studies expressing concern about the psychosocial aspects of gene diagnosis began in Western countries, with

**5. Psychosocial aspects after being informed of genetic test results** 

correctly estimated their risk. Individuals with a history of HNPCC-associated cancer or indeterminate genetic test results were significantly less likely to estimate their cancer risk as being increased.

These reports suggest that despite educational efforts and an increasing amount of data on the risk of cancer associated with HNPCC, few individuals report a perceived risk that is actually correct. In particular, individuals at risk for HNPCC who receive an indeterminate genetic test result may be falsely reassured. It is important that health care providers continue to device a counseling approach for promoting a correct understanding of cancer risk and for discussing the implications of uninformative results on the lifetime cancer risk.

#### **4. Psychosocial effects of genetic counseling**

Cancer genetic counseling has become popular as a result of the recent development of genetic tests that pinpoint familial cancer risk. Such counseling is composed of presymptomatic risk assessment and management (cancer risk counseling) and reproductive risk counseling. The former has two components: risk assessment and counseling regarding behavioral, medical, and surgical options to decrease risk. A basic goal of cancer risk counseling is to derive and explain an individual's cancer risk in clear terms, and the counselor's role is to educate and enumerate options for patients and clinicians, answer questions regarding what is known, and suggest appropriate referrals to help individuals reach difficult decisions.

A cancer risk counseling session is comprised of the following components: 1) baseline risk perception; 2) medical history and exposure history; 3) pedigree construction and pedigree documentation; 4) empiric risk assessment and genetic risk assessment; 5) options for early detection and prevention; 6) options, risks, and benefits of genetic testing; and 7) response to questions, support, and plans for follow-up. Throughout these discussions, a sensitivity to the psychological and ethical aspects of counseling is essential. Therefore, continued followup by the counselor after the session is the best way to limit the potential for adverse effects as a result of the knowledge of an inherited cancer risk, and ready access to liaison mental health professionals with experience in cancer genetics is thought to be a valued asset of cancer risk counseling.

Psychological research on aspects of cancer genetic counseling has focused on three broad areas: factors predicting interest in cancer genetic testing (Lerman et al., 1996), the psychological impact and effect of genetic counseling and testing for inherited cancer risk (Lerman et al., 1997), and the relationship between psychological distress and preventive behaviors (Kash et al., 1992). In each of these areas, the results have implications for the management of at-risk individuals. However, such data is unlikely to be applicable to every case because of cultural differences among study populations and the complexity of the instruments used in research studies, in addition to the fact that most of these studies have been performed for hereditary breast cancer. In this section, four studies on the psychological impacts of genetic counseling regarding HNPCC are reviewed.

Keller et al. (2002) explored distress before and after comprehensive interdisciplinary counseling in families at risk for HNPCC. Sixty-five individuals (31 patients with colorectal cancer and 34 unaffected at-risk persons) participated in this study. Data were collected from semi-structured questionnaires before, as well as 4-6 weeks after counseling. Distress declined after counseling, as did worries related to HNPCC. A trend toward a greater anticipated ability to cope with a positive gene test was also observed after counseling. Changes after counseling were generally more pronounced for persons at risk, compared

correctly estimated their risk. Individuals with a history of HNPCC-associated cancer or indeterminate genetic test results were significantly less likely to estimate their cancer risk

These reports suggest that despite educational efforts and an increasing amount of data on the risk of cancer associated with HNPCC, few individuals report a perceived risk that is actually correct. In particular, individuals at risk for HNPCC who receive an indeterminate genetic test result may be falsely reassured. It is important that health care providers continue to device a counseling approach for promoting a correct understanding of cancer risk and for discussing the implications of uninformative results on the lifetime cancer risk.

Cancer genetic counseling has become popular as a result of the recent development of genetic tests that pinpoint familial cancer risk. Such counseling is composed of presymptomatic risk assessment and management (cancer risk counseling) and reproductive risk counseling. The former has two components: risk assessment and counseling regarding behavioral, medical, and surgical options to decrease risk. A basic goal of cancer risk counseling is to derive and explain an individual's cancer risk in clear terms, and the counselor's role is to educate and enumerate options for patients and clinicians, answer questions regarding what is known, and suggest appropriate referrals to help

A cancer risk counseling session is comprised of the following components: 1) baseline risk perception; 2) medical history and exposure history; 3) pedigree construction and pedigree documentation; 4) empiric risk assessment and genetic risk assessment; 5) options for early detection and prevention; 6) options, risks, and benefits of genetic testing; and 7) response to questions, support, and plans for follow-up. Throughout these discussions, a sensitivity to the psychological and ethical aspects of counseling is essential. Therefore, continued followup by the counselor after the session is the best way to limit the potential for adverse effects as a result of the knowledge of an inherited cancer risk, and ready access to liaison mental health professionals with experience in cancer genetics is thought to be a valued asset of

Psychological research on aspects of cancer genetic counseling has focused on three broad areas: factors predicting interest in cancer genetic testing (Lerman et al., 1996), the psychological impact and effect of genetic counseling and testing for inherited cancer risk (Lerman et al., 1997), and the relationship between psychological distress and preventive behaviors (Kash et al., 1992). In each of these areas, the results have implications for the management of at-risk individuals. However, such data is unlikely to be applicable to every case because of cultural differences among study populations and the complexity of the instruments used in research studies, in addition to the fact that most of these studies have been performed for hereditary breast cancer. In this section, four studies on the

Keller et al. (2002) explored distress before and after comprehensive interdisciplinary counseling in families at risk for HNPCC. Sixty-five individuals (31 patients with colorectal cancer and 34 unaffected at-risk persons) participated in this study. Data were collected from semi-structured questionnaires before, as well as 4-6 weeks after counseling. Distress declined after counseling, as did worries related to HNPCC. A trend toward a greater anticipated ability to cope with a positive gene test was also observed after counseling. Changes after counseling were generally more pronounced for persons at risk, compared

psychological impacts of genetic counseling regarding HNPCC are reviewed.

as being increased.

**4. Psychosocial effects of genetic counseling** 

individuals reach difficult decisions.

cancer risk counseling.

with those for patients with cancer. A substantial minority, however, said that they experienced increased worry and physical symptoms after counseling.

Bleiker et al. (2007) examined: 1) levels of cancer-specific distress more than one year after genetic counseling for HNPCC; 2) associations between sociodemographic, clinical and psychosocial factors and levels of distress; 3) the impact of genetic counseling on family relationships; and 4) the social consequences of genetic counseling. One hundred and sixteen individuals who participated in this study completed a self-report questionnaire by mail an average of 4 years after the last counseling session. Among all the subjects, 6% had clinically significant levels of cancer-specific distress (Impact of Event Scale). Having had contact with a professional psychosocial worker for cancer risk in the past 10 years was significantly associated with higher levels of current cancer specific distress. Only a minority of the subjects reported any adverse effects of genetic counseling on communication regarding genetic counseling with their children, family relationships, obtaining life insurance, choice or change of jobs, and obtaining a mortgage.

Keller et al. (2008) conducted a prospective study that examined the impact of multidisciplinary risk counseling on the psychosocial outcome of 139 affected cancer patients and 233 family members without cancer but at risk for HNPCC. Participants completed questionnaires specific to HNPCC before and 8 weeks after attending the cancer clinic. The levels of distress among affected patients exceeded those of unaffected individuals, as did worry regarding their relatives' risk. A significant reduction in general anxiety (Hospital Anxiety and Depression Scale), distress specific to familial colorectal cancer (Impact of Events Scale), and general cancer worry (Distress due to Hereditary Disorder) was demonstrated after counseling among both the affected patients and unaffected individuals. The reduction in distress was more pronounced among affected patients given a high risk of HNPCC than among those with an intermediate risk.

Hasenbring et al. (2011) prospectively examined the impact of an initial interdisciplinary genetic counseling on feelings of anxiety with a special focus on subgroups related to personal cancer history, sex, age, and education. A significant interaction between time, sex, and age was identified for change in anxiety. While women in general and men older than 50 years revealed a significant reduction in anxiety, younger men did not show any change over time. A logistic regression analysis indicated that clinical Hospital Anxiety and Depression Scale-A cases could be predicted based on general distress (Brief Symptom Inventory) as well as by HNPCC-related cognitions of intrusion and avoidance (Impact of Event Scale) with a correct classification of 86%.

These studies indicate that anxiety and cancer-specific distress are reduced after genetic counseling, suggesting an overall beneficial impact of comprehensive counseling. On the other hand, a minority of individuals, such as cancer-affected younger men, exhibited adverse effects of genetic counseling on psychosocial variables. Thus, healthcare providers (genetic counselors, human geneticists, oncologists, and psycho-oncologists) should always be aware of psychosocial issues after genetic counseling. However, as little data is available on the psychosocial effects of genetic counseling regarding HNPCC, further data accumulation is needed.

#### **5. Psychosocial aspects after being informed of genetic test results**

Since 1991, when a gene for hereditary cancer was first identified, studies expressing concern about the psychosocial aspects of gene diagnosis began in Western countries, with

Psychological Impact and Associated Factors After Disclosure of

Table 1. (continued)

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 93

the results starting to be reported in 1993. Although studies investigating psychosocial aspects after the subjects had undergone actual genetic testing and had been informed of the test results have been reported, many of these studies have concerned hereditary breast and ovarian cancer, and only a few studies have been performed for HNPCC. Furthermore, little is known about the factors associated with psychosocial aspects. However, HNPCC testing might offer more benefit than hereditary breast and ovarian cancer testing because of the differences in the risk management options available to mutation carriers. In HNPCC, a colonoscopy every 1–2 years is more effective for detecting and preventing adverse health outcomes than measures available to carriers of hereditary breast and ovarian cancer mutations. Therefore, identifying the psychosocial situations in which individuals at risk for colorectal cancer have lived after the disclosure of genetic information or the way in which healthcare providers are able to support the mental states of these individuals are important.

Ten original articles (review articles were not included) assessing psychosocial aspects after individuals had been informed of genetic test results regarding HNPCC were extracted. In this chapter, cross-sectional studies that assessed psychosocial aspects at one time point after disclosure and prospective studies that followed-up psychosocial aspects for 1 year or longer after disclosure are described separately. A summary is shown in Table 1.

#### **5.1 Cross-sectional studies assessing psychosocial aspects after the subjects had been informed of the test results**

Four articles were extracted. Esplen et al. (2001) investigated psychosocial function in 50 individuals who were engaged in the genetic test process for HNPCC (the period between the psychosocial assessment and the disclosure of the test results was 1 – 48 months). Twenty-three individuals were identified as carriers (13 had a previous history of CRC), seven were non-carriers and 20 individuals were still awaiting their test results. The psychosocial scores demonstrated that a subgroup of individuals exhibited distress, with greater distress for those individuals awaiting results or testing positive. A high level of satisfaction was associated with the experience of testing.

Claes et al. (2004) assessed the short-term impact (1month after test result disclosure) of genetic testing using a semi-structured interview and self-reported questionnaires. The subjects were 40 cancer-unaffected relatives who had undergone predictive testing for HNPCC. Distress was within the normal ranges. Distress decreased significantly from preto post-test in non-carriers but not in carriers.

Murakami et al. (2004) identified the prevalence rates and predictors of psychological distress and evaluated the feelings of guilt at one month after the disclosure of test results in Japanese probands and unaffected relatives. The prevalence of major and minor depression, acute stress disorder (ASD), posttraumatic stress disorder (PTSD), and posttraumatic stress symptoms (PTSS) were assessed using the Structured Clinical Interview based on the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition revised (DSM-III-R) or the DSM-IV; feelings of guilt were investigated using a numeric scale and a semi-structured interview. Forty-two participants completed the 1-month follow-up interview. Although none of the participants met the criteria for major depression, ASD, or PTSD at the time of the follow-up interview, 7% of the participants met the criteria for minor depression and 5% had PTSS. The only predictor of psychological distress was the presence of a history of major or minor depression. Twelve percent of the participants had feelings of guilt.


Table 1. (continued)

92 Colorectal Cancer – From Prevention to Patient Care

the results starting to be reported in 1993. Although studies investigating psychosocial aspects after the subjects had undergone actual genetic testing and had been informed of the test results have been reported, many of these studies have concerned hereditary breast and ovarian cancer, and only a few studies have been performed for HNPCC. Furthermore, little is known about the factors associated with psychosocial aspects. However, HNPCC testing might offer more benefit than hereditary breast and ovarian cancer testing because of the differences in the risk management options available to mutation carriers. In HNPCC, a colonoscopy every 1–2 years is more effective for detecting and preventing adverse health outcomes than measures available to carriers of hereditary breast and ovarian cancer mutations. Therefore, identifying the psychosocial situations in which individuals at risk for colorectal cancer have lived after the disclosure of genetic information or the way in which healthcare providers are able to support the mental states of these individuals are

Ten original articles (review articles were not included) assessing psychosocial aspects after individuals had been informed of genetic test results regarding HNPCC were extracted. In this chapter, cross-sectional studies that assessed psychosocial aspects at one time point after disclosure and prospective studies that followed-up psychosocial aspects for 1 year or

longer after disclosure are described separately. A summary is shown in Table 1.

**5.1 Cross-sectional studies assessing psychosocial aspects after the subjects had** 

Four articles were extracted. Esplen et al. (2001) investigated psychosocial function in 50 individuals who were engaged in the genetic test process for HNPCC (the period between the psychosocial assessment and the disclosure of the test results was 1 – 48 months). Twenty-three individuals were identified as carriers (13 had a previous history of CRC), seven were non-carriers and 20 individuals were still awaiting their test results. The psychosocial scores demonstrated that a subgroup of individuals exhibited distress, with greater distress for those individuals awaiting results or testing positive. A high level of

Claes et al. (2004) assessed the short-term impact (1month after test result disclosure) of genetic testing using a semi-structured interview and self-reported questionnaires. The subjects were 40 cancer-unaffected relatives who had undergone predictive testing for HNPCC. Distress was within the normal ranges. Distress decreased significantly from pre-

Murakami et al. (2004) identified the prevalence rates and predictors of psychological distress and evaluated the feelings of guilt at one month after the disclosure of test results in Japanese probands and unaffected relatives. The prevalence of major and minor depression, acute stress disorder (ASD), posttraumatic stress disorder (PTSD), and posttraumatic stress symptoms (PTSS) were assessed using the Structured Clinical Interview based on the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition revised (DSM-III-R) or the DSM-IV; feelings of guilt were investigated using a numeric scale and a semi-structured interview. Forty-two participants completed the 1-month follow-up interview. Although none of the participants met the criteria for major depression, ASD, or PTSD at the time of the follow-up interview, 7% of the participants met the criteria for minor depression and 5% had PTSS. The only predictor of psychological distress was the presence of a history of

major or minor depression. Twelve percent of the participants had feelings of guilt.

important.

**been informed of the test results** 

satisfaction was associated with the experience of testing.

to post-test in non-carriers but not in carriers.


Table 1. (continued)

Psychological Impact and Associated Factors After Disclosure of

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 95

CES-D: Center for Epidemiological Studies-Depression, IES: Impact of Event Scale (IES-R: Impact of

ASD: acute stress disorder, PTSD: post-traumatic stress disorder, PTSS: post-traumatic stress symptoms Table 1. Characteristics of studies on psychosocial aspects and associated factors after being

QLI: Quality of Life Index, SCL-90: Symptom Checklist, STAI: State-Trait Anxiety Inventory

Event Scale-Revised), HADS: Hospital Anxiety and Depression Scale,

informed of genetic test results regarding HNPCC


CES-D: Center for Epidemiological Studies-Depression, IES: Impact of Event Scale (IES-R: Impact of Event Scale-Revised), HADS: Hospital Anxiety and Depression Scale,

QLI: Quality of Life Index, SCL-90: Symptom Checklist, STAI: State-Trait Anxiety Inventory ASD: acute stress disorder, PTSD: post-traumatic stress disorder, PTSS: post-traumatic stress symptoms

Table 1. Characteristics of studies on psychosocial aspects and associated factors after being informed of genetic test results regarding HNPCC

94 Colorectal Cancer – From Prevention to Patient Care

Table 1. (continued)

Psychological Impact and Associated Factors After Disclosure of

and long-term increases in distress.

cancer-related distress and state anxiety.

years, were similar to the baseline scores.

positive results.

**5.3 Summary** 

life, or lower social support.

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 97

the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at 2 weeks after test result disclosure, compared with non-carriers, in their respective groups; the scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months after test result disclosure. Higher levels of baseline mood disturbance, a lower quality of life, and lower social support were associated with a risk for both short-

Claes et al. (2005) evaluated distress one year after the disclosure of a predictive genetic test result for HNPCC in 72 cancer-unaffected relatives (36 carriers and 36 non-carriers). The mean levels of distress (cancer-specific distress, state anxiety, and psychoneuroticism) were within the normal ranges and none of the participants had an overall pattern (on all scales) of clinically elevated levels of distress. Carriers had significantly higher cancer-related distress one year after test result disclosure than non-carriers. In both groups, colorectal cancer-related distress decreased. Non-carriers additionally showed decreased endometrial

Collins et al. (2007) conducted a 3-year study of individuals who received predictive genetic test results for previously identified familial mutations regarding HNPCC. Questionnaires were sent before attendance and 2 weeks, 4 months, 1 year, and 3 years after receiving the test results. Psychological measures were included each time. The study included 73 individuals with no personal cancer history (19 carriers and 54 non-carriers). The results showed an increase in mean cancer-specific distress in carriers at 2 weeks with a return to baseline levels by 12 months. This level was maintained until 3 years. Non-carriers showed sustained decreases after testing with a significantly lower level at 3 years compared with at baseline. These scores tended to be lower than those for carriers at 3 years. The mean depression and anxiety scores did not differ between carriers and non-carriers and, at 3

Shiloh et al. (2008) assessed the emotional effects of genetic testing for HNPCC at baseline before testing and again at 6 and 12 months after testing. The subjects were 253 canceraffected and -unaffected individuals. Negative emotional reactions were evaluated using the Revised Impact of Event Scale and the Center for Epidemiological Studies-Depression Scale. Monitoring coping style was assessed at baseline using the Miller Behavioral Style Scale. Mean reductions were indicated in distress and depression levels within the first 6 months after testing. High monitors (individuals who vigilantly attended to threatening cues in their environment in an attempt to emotionally process the situation and who actively engaged in information seeking and cognitive problem solving with the intention of taking precautions) were generally more distressed than low monitors, specifically if they had indeterminate or

Many studies have shown that genetic testing does not result in short- or long-term significant adverse psychological outcomes, including depression, anxiety, and posttraumatic stress disorder (PTSD), in either carriers or non-carriers or in either canceraffected or cancer–unaffected individuals. However, healthcare providers should assess psychological responses, such as minor depression, posttraumatic stress symptoms (PTSS), and feelings of guilt, particularly in individuals who have a history of major or minor depression, nervous personality tendencies, baseline mood disturbances, a lower quality of

Lastly, two cases that showed adverse psychological reactions after being informed of genetic test results will be presented. The first case is a man who was diagnosed as having

Yamashita et al. (2008) elucidated the psychological impact at one month after the disclosure of genetic test results regarding HNPCC and assessed the associated factors, focusing on memory function in particular. The subjects were persons who were suspected of having HNPCC and had been given the choice of undergoing genetic testing. The post-genetic testing psychological impact was evaluated using the Impact of Event Scale-Revised (IES-R), and personality tendencies and memory function were evaluated. Final data were obtained from 46 Japanese probands and unaffected relatives (mutation-positive in 18 subjects, uninformative in 18 subjects, and mutation-negative in 10 subjects). A comparison of the IES-R scores showed that they tended to be higher in the mutation-positive group, but the differences were not statistically significant. The personality tendency "nervousness" and the verbal memory assessed prior to disclosure were significantly associated with the total IES-R score.

#### **5.2 Prospective studies assessing psychosocial aspects after the subjects had been informed of the test results**

Six articles were extracted. Aktan-Collan et al. (2001) assessed general anxiety, fear of cancer and death, satisfaction with life, and attitude regarding the future using a questionnaire survey in 271 individuals with no personal cancer history who were tested for HNPCC. Measurements were made before the first counseling (baseline), at the test disclosure session, and 1 and 12 months after disclosure. Although the mutation-positive individuals were more afraid of cancer than those who were mutation negative at every measurement point, the fear of cancer decreased significantly from the baseline until after disclosure in both groups. The mutation-positive subjects were more anxious than their counterparts immediately after the test disclosure, but the differences had disappeared at the follow-up examinations. Regarding the other variables, no differences among the groups defined according to mutation status or changes over time were detected.

Meiser et al. (2004) assessed the psychological impact of predictive genetic testing for HNPCC in 114 individuals with no personal cancer history (32 carriers and 82 non-carriers) using mailed self-administered questionnaires prior to and 2 weeks, 4 months and 12 months after the disclosure of the test results. Compared with the baseline results, carriers showed a significant increase in the mean scores for intrusive and avoidant thoughts regarding colorectal cancer at 2 weeks after test result disclosure and a significant decrease in the mean depression scores at 2 weeks and 4 months after test result disclosure. For noncarriers, significant decreases in the mean scores for intrusive and avoidant thoughts regarding colorectal cancer were observed at all follow-up assessment time points relative to the baseline. Non-carriers also showed significant decreases from the baseline in the mean depression scores at 2 weeks, 4 months and 12 months after test result disclosure. Significant decreases in the mean state anxiety scores from the baseline were also observed for noncarriers at 2 weeks after test result disclosure.

Gritz et al. (2005) examined the impact of HNPCC genetic test results on the psychological outcomes of cancer-affected and -unaffected participants up to 1 year after test result disclosure. A total of 155 persons completed the study measures before HNPCC genetic testing and at 2 weeks and 6 and 12 months after the disclosure of the test results. The mean scores for all the outcome measures remained stable and within the normal limits for canceraffected participants, regardless of the mutation status. Among unaffected carriers of HNPCC-predisposing mutations, the mean depression, state anxiety, and cancer worry scores increased from baseline to 2 weeks after test result disclosure and decreased from 2 weeks to 6 months after test result disclosure. Among unaffected non-carriers, the mean depression and anxiety scores did not differ, but the cancer worry scores decreased during the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at 2 weeks after test result disclosure, compared with non-carriers, in their respective groups; the scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months after test result disclosure. Higher levels of baseline mood disturbance, a lower quality of life, and lower social support were associated with a risk for both shortand long-term increases in distress.

Claes et al. (2005) evaluated distress one year after the disclosure of a predictive genetic test result for HNPCC in 72 cancer-unaffected relatives (36 carriers and 36 non-carriers). The mean levels of distress (cancer-specific distress, state anxiety, and psychoneuroticism) were within the normal ranges and none of the participants had an overall pattern (on all scales) of clinically elevated levels of distress. Carriers had significantly higher cancer-related distress one year after test result disclosure than non-carriers. In both groups, colorectal cancer-related distress decreased. Non-carriers additionally showed decreased endometrial cancer-related distress and state anxiety.

Collins et al. (2007) conducted a 3-year study of individuals who received predictive genetic test results for previously identified familial mutations regarding HNPCC. Questionnaires were sent before attendance and 2 weeks, 4 months, 1 year, and 3 years after receiving the test results. Psychological measures were included each time. The study included 73 individuals with no personal cancer history (19 carriers and 54 non-carriers). The results showed an increase in mean cancer-specific distress in carriers at 2 weeks with a return to baseline levels by 12 months. This level was maintained until 3 years. Non-carriers showed sustained decreases after testing with a significantly lower level at 3 years compared with at baseline. These scores tended to be lower than those for carriers at 3 years. The mean depression and anxiety scores did not differ between carriers and non-carriers and, at 3 years, were similar to the baseline scores.

Shiloh et al. (2008) assessed the emotional effects of genetic testing for HNPCC at baseline before testing and again at 6 and 12 months after testing. The subjects were 253 canceraffected and -unaffected individuals. Negative emotional reactions were evaluated using the Revised Impact of Event Scale and the Center for Epidemiological Studies-Depression Scale. Monitoring coping style was assessed at baseline using the Miller Behavioral Style Scale. Mean reductions were indicated in distress and depression levels within the first 6 months after testing. High monitors (individuals who vigilantly attended to threatening cues in their environment in an attempt to emotionally process the situation and who actively engaged in information seeking and cognitive problem solving with the intention of taking precautions) were generally more distressed than low monitors, specifically if they had indeterminate or positive results.

#### **5.3 Summary**

96 Colorectal Cancer – From Prevention to Patient Care

Yamashita et al. (2008) elucidated the psychological impact at one month after the disclosure of genetic test results regarding HNPCC and assessed the associated factors, focusing on memory function in particular. The subjects were persons who were suspected of having HNPCC and had been given the choice of undergoing genetic testing. The post-genetic testing psychological impact was evaluated using the Impact of Event Scale-Revised (IES-R), and personality tendencies and memory function were evaluated. Final data were obtained from 46 Japanese probands and unaffected relatives (mutation-positive in 18 subjects, uninformative in 18 subjects, and mutation-negative in 10 subjects). A comparison of the IES-R scores showed that they tended to be higher in the mutation-positive group, but the differences were not statistically significant. The personality tendency "nervousness" and the verbal memory

assessed prior to disclosure were significantly associated with the total IES-R score.

according to mutation status or changes over time were detected.

carriers at 2 weeks after test result disclosure.

**informed of the test results** 

**5.2 Prospective studies assessing psychosocial aspects after the subjects had been** 

Six articles were extracted. Aktan-Collan et al. (2001) assessed general anxiety, fear of cancer and death, satisfaction with life, and attitude regarding the future using a questionnaire survey in 271 individuals with no personal cancer history who were tested for HNPCC. Measurements were made before the first counseling (baseline), at the test disclosure session, and 1 and 12 months after disclosure. Although the mutation-positive individuals were more afraid of cancer than those who were mutation negative at every measurement point, the fear of cancer decreased significantly from the baseline until after disclosure in both groups. The mutation-positive subjects were more anxious than their counterparts immediately after the test disclosure, but the differences had disappeared at the follow-up examinations. Regarding the other variables, no differences among the groups defined

Meiser et al. (2004) assessed the psychological impact of predictive genetic testing for HNPCC in 114 individuals with no personal cancer history (32 carriers and 82 non-carriers) using mailed self-administered questionnaires prior to and 2 weeks, 4 months and 12 months after the disclosure of the test results. Compared with the baseline results, carriers showed a significant increase in the mean scores for intrusive and avoidant thoughts regarding colorectal cancer at 2 weeks after test result disclosure and a significant decrease in the mean depression scores at 2 weeks and 4 months after test result disclosure. For noncarriers, significant decreases in the mean scores for intrusive and avoidant thoughts regarding colorectal cancer were observed at all follow-up assessment time points relative to the baseline. Non-carriers also showed significant decreases from the baseline in the mean depression scores at 2 weeks, 4 months and 12 months after test result disclosure. Significant decreases in the mean state anxiety scores from the baseline were also observed for non-

Gritz et al. (2005) examined the impact of HNPCC genetic test results on the psychological outcomes of cancer-affected and -unaffected participants up to 1 year after test result disclosure. A total of 155 persons completed the study measures before HNPCC genetic testing and at 2 weeks and 6 and 12 months after the disclosure of the test results. The mean scores for all the outcome measures remained stable and within the normal limits for canceraffected participants, regardless of the mutation status. Among unaffected carriers of HNPCC-predisposing mutations, the mean depression, state anxiety, and cancer worry scores increased from baseline to 2 weeks after test result disclosure and decreased from 2 weeks to 6 months after test result disclosure. Among unaffected non-carriers, the mean depression and anxiety scores did not differ, but the cancer worry scores decreased during Many studies have shown that genetic testing does not result in short- or long-term significant adverse psychological outcomes, including depression, anxiety, and posttraumatic stress disorder (PTSD), in either carriers or non-carriers or in either canceraffected or cancer–unaffected individuals. However, healthcare providers should assess psychological responses, such as minor depression, posttraumatic stress symptoms (PTSS), and feelings of guilt, particularly in individuals who have a history of major or minor depression, nervous personality tendencies, baseline mood disturbances, a lower quality of life, or lower social support.

Lastly, two cases that showed adverse psychological reactions after being informed of genetic test results will be presented. The first case is a man who was diagnosed as having

Psychological Impact and Associated Factors After Disclosure of

area in collaboration with one another.

(4), 608-611

**7. References** 

Genetic Test Results Concerning Hereditary Nonpolyposis Colorectal Cancer 99

model, including psychosocial support; 2) the education of cancer genetic counselors; 3) the availability of appropriate information concerning cancer genetics; 4) the recruitment of subjects at risk for cancer susceptibility; and 5) the accumulation of further psycho-oncology research results. While it is by no means easy to deal with these problems, it is essential that medical oncologists, surgical oncologists, psycho-oncologists, medical geneticists, nurses, and all other health care providers involved in cancer care vigorously approach this new

Aktan-Collan K, Haukkala A, Mecklin JP, Uutela A, Kääriäinen H. (2001). Psychological

Bleiker EM, Menko FH, Kluijt I, Taal BG, Gerritsma MA, Wever LD, Aaronson NK. (2007).

health-related behavior at one year post-test. *Genetic Testing,* 9 (1), 54-65 Collins VR, Meiser B, Ukoumunne OC, Gaff C, St John DJ, Halliday JL. (2007). The impact of

Codori AM, Waldeck T, Petersen GM, Miglioretti D, Trimbath JD, Tillery MA. (2005).

Esplen MJ, Madlensky L, Butler K, McKinnon W, Bapat B, Wong J, Aronson M, Gallinger

Gritz ER, Peterson SK, Vernon SW, Marani SK, Baile WF, Watts BG, Amos CI, Frazier ML,

nonpolyposis colorectal cancer. *Journal of Clinical Oncology,* 23 (9), 1902-1910 Grover S, Stoffel EM, Mercado RC, Ford BM, Kohlman WK, Shannon KM, Conrad PG,

Hadley DW, Jenkins J, Dimond E, Nakahara K, Grogan L, Liewehr DJ, Steinberg SM, Kirsch

Hasenbring MI, Kreddig N, Deges G, Epplen JT, Kunstmann E, Stemmler S, Schulmann K,

hereditary colorectal cancer. *Journal of Genetic Counseling,* 14 (2), 119-132 Domanska K, Nilbert M, Soller M, Silfverberg B, Carlsson C. (2007). Discrepancies between

nonpolyposis colorectal cancer. *Genetic Testing,* 11 (2), 183-186

after testing. *Genetics in Medicine,* 9 (5), 290-297

*American Journal of Medical Genetics,* 103 (1), 9-15

syndrome. *Journal of Clinical Oncology,* 27 (24), 3981-3986

colorectal cancer. *Archives of Internal Medicine,* 163 (5), 573-582

consequences of predictive genetic testing for hereditary non-polyposis colorectal cancer (HNPCC): a prospective follow-up study. *International Journal of Cancer,* 93

Colorectal cancer in the family: psychosocial distress and social issues in the years following genetic counselling. *Hereditary Cancer in Clinical Practice,* 5 (2), 59-66 Claes E, Denayer L, Evers-Kiebooms G, Boogaerts A, Legius E. (2004). Predictive testing for

hereditary non-polyposis colorectal cancer: motivation, illness representations and short-term psychological impact. *Patient Education and Counseling,* 55 (2), 265-274 Claes E, Denayer L, Evers-Kiebooms G, Boogaerts A, Philippe K, Tejpar S, Devriendt K,

Legius E. (2005). Predictive testing for hereditary nonpolyposis colorectal cancer: subjective perception regarding colorectal and endometrial cancer, distress, and

predictive genetic testing for hereditary nonpolyposis colorectal cancer: three years

Genetic counseling outcomes: perceived risk and distress after counseling for

estimated and perceived risk of cancer among individuals with hereditary

S.(2001). Motivations and psychosocial impact of genetic testing for HNPCC.

Lynch PM. (2005). Psychological impact of genetic testing for hereditary

Blanco AM, Terdiman JP, Gruber SB, Chung DC, Syngal S. (2009). Colorectal cancer risk perception on the basis of genetic test results in individuals at risk for Lynch

I. (2003). Genetic counseling and testing in families with hereditary nonpolyposis

Willert J, Schmiegel W. (2011). Psychological impact of genetic counseling for

acute stress disorder at a 1-month follow-up examination after the disclosure of a genetic test result, despite the fact that the test result had been negative. The second case is a man who felt guilty after hearing of the positive test results of family members of individuals belonging to his support group.

#### **5.4 Cases exhibiting adverse psychological reactions**

[Case 1] Mr. A was a 39-year-old married man without children who came for genetic counseling and testing because of a family history of colon cancer. He had no history of cancer, but his father had a history of colon cancer and his sister had died of the disease at an early age. To confirm the diagnosis of HNPCC, a blood sample was obtained and mutations in the hMSH2 and hMLH1 genes were analyzed. He then consented in writing to participate in our study, and a baseline interview was conducted. He did not meet any of the criteria for any psychiatric disorders.

Approximately two months after the blood test, he underwent post-test counseling and was informed that no mutations had been detected in either the hMSH2 or hMLH1 gene. Four weeks after the disclosure of the test result, at a 1-month follow-up examination, he was diagnosed as having acute stress disorder according to a structured clinical interview based on the DSM-IV. The total score of the Impact of Event Scale-Revised was high. The score for Total Mood Disturbance in the Profile of Mood States was higher than that at the baseline interview. He reported that although he felt emotional relief to learn the negative result, his worries regarding colon cancer had increased instead of disappearing.

Mutation-negative individuals often choose not to participate in follow-up counseling after genetic testing. However, this case suggests that it is important to evaluate the psychological outcome after genetic testing regardless of the test result, and that psychiatrists or psychologists should support the genetic counseling system.

[Case 2] Mr. B, a 59-year-old man, underwent a total colectomy for the resection of colorectal cancer. He and his 25-year old son requested predictive genetic testing 3 years later to reduce uncertainty and to help plan his son's future, since Mr. B's mother had died of colon cancer secondary to HNPCC. Mr. B and his son were provided with both an educational session explaining the genetics of hereditary diseases and counseling regarding the possible impact of positive test results. The tests revealed the presence of a mutation in the father but not in the son. Mr. B was relieved that his "bad blood" had not been passed on to his son. Later, however, he began to experience anhedonia and became depressed for several days. His primary care physician could not determine the reason for his feelings.

Mr. B was the chairperson of a hereditary cancer patient support group run by patients, their families, and health care providers. The group had been established to help families with hereditary cancer exchange information and experiences. Mr. B began to feel guilty because his son had tested negative while the family members of others in his support group had tested positive for the disease.

#### **6. Conclusion**

Cancer genetic counseling and genetic testing for HNPCC are now conducted in ordinary clinical settings. However, as mentioned above, few studies have examined the psychosocial aspects of genetic testing for HNPCC, and psychosocial assessments and long-term followup care for individuals who have undergone genetic counseling or testing and at-risk relatives with no personal history of cancer remain insufficient. To develop this field, the following problems should be examined: 1) the development of a cancer genetic counseling model, including psychosocial support; 2) the education of cancer genetic counselors; 3) the availability of appropriate information concerning cancer genetics; 4) the recruitment of subjects at risk for cancer susceptibility; and 5) the accumulation of further psycho-oncology research results. While it is by no means easy to deal with these problems, it is essential that medical oncologists, surgical oncologists, psycho-oncologists, medical geneticists, nurses, and all other health care providers involved in cancer care vigorously approach this new area in collaboration with one another.

#### **7. References**

98 Colorectal Cancer – From Prevention to Patient Care

acute stress disorder at a 1-month follow-up examination after the disclosure of a genetic test result, despite the fact that the test result had been negative. The second case is a man who felt guilty after hearing of the positive test results of family members of individuals

[Case 1] Mr. A was a 39-year-old married man without children who came for genetic counseling and testing because of a family history of colon cancer. He had no history of cancer, but his father had a history of colon cancer and his sister had died of the disease at an early age. To confirm the diagnosis of HNPCC, a blood sample was obtained and mutations in the hMSH2 and hMLH1 genes were analyzed. He then consented in writing to participate in our study, and a baseline interview was conducted. He did not meet any of

Approximately two months after the blood test, he underwent post-test counseling and was informed that no mutations had been detected in either the hMSH2 or hMLH1 gene. Four weeks after the disclosure of the test result, at a 1-month follow-up examination, he was diagnosed as having acute stress disorder according to a structured clinical interview based on the DSM-IV. The total score of the Impact of Event Scale-Revised was high. The score for Total Mood Disturbance in the Profile of Mood States was higher than that at the baseline interview. He reported that although he felt emotional relief to learn the negative result, his

Mutation-negative individuals often choose not to participate in follow-up counseling after genetic testing. However, this case suggests that it is important to evaluate the psychological outcome after genetic testing regardless of the test result, and that psychiatrists or

[Case 2] Mr. B, a 59-year-old man, underwent a total colectomy for the resection of colorectal cancer. He and his 25-year old son requested predictive genetic testing 3 years later to reduce uncertainty and to help plan his son's future, since Mr. B's mother had died of colon cancer secondary to HNPCC. Mr. B and his son were provided with both an educational session explaining the genetics of hereditary diseases and counseling regarding the possible impact of positive test results. The tests revealed the presence of a mutation in the father but not in the son. Mr. B was relieved that his "bad blood" had not been passed on to his son. Later, however, he began to experience anhedonia and became depressed for several days.

Mr. B was the chairperson of a hereditary cancer patient support group run by patients, their families, and health care providers. The group had been established to help families with hereditary cancer exchange information and experiences. Mr. B began to feel guilty because his son had tested negative while the family members of others in his support

Cancer genetic counseling and genetic testing for HNPCC are now conducted in ordinary clinical settings. However, as mentioned above, few studies have examined the psychosocial aspects of genetic testing for HNPCC, and psychosocial assessments and long-term followup care for individuals who have undergone genetic counseling or testing and at-risk relatives with no personal history of cancer remain insufficient. To develop this field, the following problems should be examined: 1) the development of a cancer genetic counseling

belonging to his support group.

the criteria for any psychiatric disorders.

group had tested positive for the disease.

**6. Conclusion** 

**5.4 Cases exhibiting adverse psychological reactions** 

worries regarding colon cancer had increased instead of disappearing.

His primary care physician could not determine the reason for his feelings.

psychologists should support the genetic counseling system.


**Part 3** 

**Nutrition** 

hereditary nonpolyposis colorectal cancer: the role of cancer history, gender, age, and psychological distress. *Genetic Testing and Molecular Biomarkers,* 15 (4), 219-225


**Part 3** 

100 Colorectal Cancer – From Prevention to Patient Care

Keller M, Jost R, Haunstetter CM, Kienle P, Knaebel HP, Gebert J, Sutter C, Knebel-Doeberitz

affected patients and family members. *Clinical Genetics,* 74 (5), 414-424 Lerman C, Narod S, Schulman K, Hughes C, Gomez-Caminero A, Bonney G, Gold K, Trock

Lerman C, Biesecker B, Benkendorf JL, Kerner J, Gomez-Caminero A, Hughes C, Reed MM.

Murakami **Y**, Okamura H, Sugano K, Yoshida T, Kazuma K, Akechi T, Uchitomi Y. (2004).

Offit K. (1998). *Clinical Cancer Genetics: Risk Counseling & Management.* ISBN: 0-471-14655-2,

Shiloh S, Koehly L, Jenkins J, Martin J, Hadley D. (2008). Monitoring coping style moderates

van Oostrom I, Meijers-Heijboer H, Duivenvoorden HJ, Bröcker-Vriends AH, van Asperen CJ,

hereditary cancer distress. *Patient Education and Counseling,* 65 (1), 58-68 Wakefield CE, Kasparian NA, Meiser B, Homewood J, Kirk J, Tucker K. (2007a). Attitudes

*Cancer Institute,* 84 (1), 24-30

*JAMA,* 275 (24), 1885-1892

Wiley-Liss, New York

colorectal cancer. *Clinical Genetics,* 66 (6), 502-511

a longitudinal study. *Psychooncology,* 17 (8), 746-755

colorectal cancer risk. *Cancer,* 113 (5), 956-965

nonpolyposis colorectal carcinoma. *Cancer,* 101 (2), 395-403

hereditary nonpolyposis colorectal cancer: the role of cancer history, gender, age, and psychological distress. *Genetic Testing and Molecular Biomarkers,* 15 (4), 219-225 Kash KM, Holland JC, Halper MS, Miller DG. Psychological distress and surveillance

behaviors of women with a family history of breast cancer. *Journal of the National* 

M, Cremer F, Mazitschek U. (2002). Comprehensive genetic counseling for families at risk for HNPCC: impact on distress and perceptions. *Genetic Testing,* 6 (4), 291-302 Keller M, Jost R, Haunstetter CM, Sattel H, Schroeter C, Bertsch U, Cremer F, Kienle P,

Tariverdian M, Kloor M, Gebert J, Brechtel A. (2008). Psychosocial outcome following genetic risk counselling for familial colorectal cancer. A comparison of

B, Main D, Lynch J, Fulmore C, Snyder C, Lemon SJ, Conway T, Tonin P, Lenoir G, Lynch H. (1996). BRCA1 testing in families with hereditary breast-ovarian cancer.

(1997). Controlled trial of pretest education approaches to enhance informed decisionmaking for BRCA1 gene testing. *Journal of the National Cancer Institute,* 89 (2), 148-157 Meiser B, Collins V, Warren R, Gaff C, St John DJ, Young MA, Harrop K, Brown J, Halliday

J. (2004). Psychological impact of genetic testing for hereditary non-polyposis

Psychologic distress after disclosure of genetic test results regarding hereditary

emotional reactions to genetic testing for hereditary nonpolyposis colorectal cancer:

Sijmons RH, Seynaeve C, Van Gool AR, Klijn JG, Tibben A. (2007). Comparison of individuals opting for BRCA1/2 or HNPCC genetic susceptibility testing with regard to coping, illness perceptions, illness experiences, family system characteristics and

toward genetic testing for cancer risk after genetic counseling and decision support: a qualitative comparison between hereditary cancer types. *Genetic Testing,* 11 (4), 401-411 Wakefield CE, Meiser B, Homewood J, Peate M, Kirk J, Warner B, Lobb E, Gaff C, Tucker K.

(2007b). Development and pilot testing of two decision aids for individuals considering genetic testing for cancer risk. *Journal of Genetic Counseling,* 16 (3), 325-339

testing Decision Aid Collaborative Group. (2008). Randomized trial of a decision aid for individuals considering genetic testing for hereditary nonpolyposis

Psychological impact and associated factors after disclosure of genetic test results concerning hereditary nonpolyposis colorectal cancer. *Stress and Health,* 24, 407-412

Wakefield CE, Meiser B, Homewood J, Ward R, O'Donnell S, Kirk J; Australian GENetic

Yamashita M, Okamura H, Murakami Y, Sugano K, Yoshida T, Uchitomi Y. (2008).

**Nutrition** 

**6** 

Martina Perše

*Slovenia* 

**Physical Activity, Dietary Fat** 

*University of Ljubljana, Faculty of Medicine, Institute of Pathology, MEC,* 

Colorectal cancer (CRC) is one of the most commonly diagnosed cancers worldwide, with over 1.2 million new cases being recorded in 2008. Global cancer statistics show that there is great (10-fold) variation in the occurrence of CRC worldwide, with the highest incidence rates in economically developed countries and regions, such as Australia, New Zealand, Europe and North America. The latest report shows that CRC incidence rates are rapidly increasing in countries within Eastern Europe and Eastern Asia, which were formerly considered low-risk areas. In some countries, e.g., the Czech Repubic and Japan, the incidence of CRC has already exceeded the peak observed in the high-risk areas. Epidemiological studies have demonstrated that the increasing incidence of CRC in these developing countries is mostly due to a higher incidence of CRC in younger age groups, which readily adopt new lifestyle habits (Jemal et al., 2011). In addition, reports have shown that persons who were born in Asia and later migrated to the United States have a higher

risk of CRC than their counterparts who have remained in Asia (Flood et al., 2000).

Changes in worldwide variations in the incidence rates, together with the results of migrant studies, provide convincing evidence that the incidence rates depend largely on environmental (i.e., non-genetic) risk factors, including lifestyle. It is estimated that most cases of CRC occur sporadically (70-80%). Approximately 15% of CRC cases develop as a result of inherited factors and 5-10% of them result from known genetic syndroms, e.g., familial adenomatous polyposis (FAP) and hereditary non-polyposis colorectal carcinoma

There are different approaches and strategies concerning how to reduce the incidence of and mortality due to CRC. Those directed toward the treatment of CRC, i.e., surgical and therapeutic measures, are mostly costly, painful and the prognosis is not promisig. Efforts have also recently been directed toward the identification and removal of precancerous lesions (visible polypoid adenomas) through screening programs, which is a promising approach and is an important step in reducing mortality due to CRC (Orlando et al., 2008). On the other hand, efforts invested into strategies directed toward public health promotion campaigns for the prevention or reduction of risk factors in populations at high risk of CRC have been few and obviously ineffective. Recent studies have shown that there is low level of awareness of the role that physical activity plays in preventing CRC among adults in the

**1. Introduction** 

(HNPCC) (Souglakos, 2007).

USA and Europe (Coups et al., 2008; Keighley et al., 2004).

**and Colorectal Cancer** 

### **Physical Activity, Dietary Fat and Colorectal Cancer**

Martina Perše

*University of Ljubljana, Faculty of Medicine, Institute of Pathology, MEC, Slovenia* 

#### **1. Introduction**

Colorectal cancer (CRC) is one of the most commonly diagnosed cancers worldwide, with over 1.2 million new cases being recorded in 2008. Global cancer statistics show that there is great (10-fold) variation in the occurrence of CRC worldwide, with the highest incidence rates in economically developed countries and regions, such as Australia, New Zealand, Europe and North America. The latest report shows that CRC incidence rates are rapidly increasing in countries within Eastern Europe and Eastern Asia, which were formerly considered low-risk areas. In some countries, e.g., the Czech Repubic and Japan, the incidence of CRC has already exceeded the peak observed in the high-risk areas. Epidemiological studies have demonstrated that the increasing incidence of CRC in these developing countries is mostly due to a higher incidence of CRC in younger age groups, which readily adopt new lifestyle habits (Jemal et al., 2011). In addition, reports have shown that persons who were born in Asia and later migrated to the United States have a higher risk of CRC than their counterparts who have remained in Asia (Flood et al., 2000).

Changes in worldwide variations in the incidence rates, together with the results of migrant studies, provide convincing evidence that the incidence rates depend largely on environmental (i.e., non-genetic) risk factors, including lifestyle. It is estimated that most cases of CRC occur sporadically (70-80%). Approximately 15% of CRC cases develop as a result of inherited factors and 5-10% of them result from known genetic syndroms, e.g., familial adenomatous polyposis (FAP) and hereditary non-polyposis colorectal carcinoma (HNPCC) (Souglakos, 2007).

There are different approaches and strategies concerning how to reduce the incidence of and mortality due to CRC. Those directed toward the treatment of CRC, i.e., surgical and therapeutic measures, are mostly costly, painful and the prognosis is not promisig. Efforts have also recently been directed toward the identification and removal of precancerous lesions (visible polypoid adenomas) through screening programs, which is a promising approach and is an important step in reducing mortality due to CRC (Orlando et al., 2008). On the other hand, efforts invested into strategies directed toward public health promotion campaigns for the prevention or reduction of risk factors in populations at high risk of CRC have been few and obviously ineffective. Recent studies have shown that there is low level of awareness of the role that physical activity plays in preventing CRC among adults in the USA and Europe (Coups et al., 2008; Keighley et al., 2004).

Physical Activity, Dietary Fat and Colorectal Cancer 105

2003) or even 40-50% (Friedenreich & Orenstein, 2002). However, estimates from metaanalyses are little lower. A meta-analysis of 19 cohort studies estimated that physical activity may reduce the risk of colon cancer on average by 20-30% (Samad et al., 2005). The World Health Organization conducted a meta-analysis using data from studies prior to 2000 and estimated that around 16% of the global colon cancer disease burden can be ascribed to physical inactivity (Bull, 2004, as cited in Wolin et al., 2009). The most reliable results are probably those from the most recent meta-analysis, which included all available case-control (24) or cohort studies (28) that had been published to June 2008, but only those that investigated the association between physical activity and colon cancer or colon and rectal cancer separately. Studies on the association between physical activity and rectal cancer or colorectal cancer combined were not included. However, this meta-analysis showed a 24% average risk reduction for colon cancer when comparing the most to the least active

The results of some studies suggest that the beneficial effects of exercise may be attenuated or less consistent in women (Kruk & Aboul-Enein, 2006; Thune & Furberg, 2001). However, most studies have found no difference in colon cancer risk according to gender (Friedenreich et al., 2006), which is in agreement with the latest meta-analysis. It was estimated that the protective effect of physical activity on colon cancer is similar for men (24%) and women (21%). The risk appeared smaller in cohort studies (men=19%; women=11%) than in case-control studies (men=28%; women=32%) (Wolin et al., 2009). It has been suggested that surveys used to measure physical activity were developed mainly for men and are thus less precise in estimating household work. Women may spend between 30 minutes to 6 hours a day on household chores and family care activities and from 4 to 16 hours a day on occupational activities, which makes it challenging to assess

Although it is clear that physical activity is associated with a decreased risk of developing colon cancer, details of the relationship are less clear. It is known that the frequency of muscle contraction (e.g., number of activities performed per day, week, month), duration (number of minutes or hours per day), intensity (how much energy is expended) and activity levels throughout the participant's entire lifetime, are important components of activity that can significantly affect the protective effects of physical activity on colon cancer risk (Friedenreich & Orenstein, 2002). Few studies have examined the type, intensity and duration of physical activity required. Since they used different criteria of physical activity in their tests for trends, meta-analyses of trends across these studies have not yet been

Physical activity is often thought of as recreational activity or exercise but it is much more than this. Physical activity is any bodily movement produced by the skeletal muscles that results in a quantifiable expenditure of energy and thereby comprises all leisure-time activities as well as occupational activities. Leisure-time physical activity refers to sports, conditioning exercises (structured and planned activity in order to improve or maintain physical fitness), household activities, self-care activities (e.g., dressing, eating, talking,

Evidence is consistent that both occupational and leisure-time physical activity can affect colon cancer risk (Wolin et al., 2009). A dose-response relationship has been noted. Higher

standing, walking, climbing stairs) etc. (Kruk & Aboul-Enein, 2006).

individuals across all studies (Wolin et al., 2009).

their physical activity accurately (Howard et al., 2008).

**2.2.1 Type, duration and intensity** 

conducted (Wolin et al., 2009).

Epidemiological studies have shown that around 30-40% of CRC may be preventable by maintaining a healthy lifestyle and suitable diet. The available evidence suggests that the population attributable risk of physical inactivity is 13-14%, which is the same as the risk due to a Western eating pattern (Coups et al., 2008). These data, and all the aforementioned facts, show that the majority of sporadic CRC cases are preventable by adjustment of appropriate environmental and lifestyle factors (dietary habits, low body index, physical activity) and that there is a need to improve strategies of public health promotion campaigns in countries with increased risk of CRC. Public health promotion campaigns, if adopted, could have a major impact in the fight against sporadic CRC and would address many health and financial challenges.

This chapter is therefore an attempt in this direction and provides a current review of the literature on the relation between physical activity or dietary fat and CRC and the mechanisms of their interaction.

#### **2. Physical activity and CRC**

The first evidence of the preventive role of physical activity against cancer appeared in 1922, when two groups of investigators, independently of each other, observed that cancer mortality rates declined with increasing physical activity required for an occupation (Lee, 2003). In spite of these results, further investigations in this area were not undertaken until the 1980s, when investigators observed that men with sedentary jobs had a higher colon cancer risk than men with jobs requiring strenuous activity (Larsson et al., 2006). Since then, the link between physical activity and cancer risk has been examined extensively, not only for cancer of the colon and rectum but also for breast and prostate cancer and, to a lesser extent also for cancers of the endometrium, lung, ovary, testis, pancreas and kidney. Nevertheless, cancer of the large bowel is the most frequently investigated cancer in relation to physical activity in humans. There have today been over 60 epidemiological studies investigating effects of physical activity on CRC in humans. Studies have been conducted in different parts of the world (North America, Europe, Australia, New Zealand and Asia-Japan) and among different populations and races. The results of these publications are convincing and clearly indicate that physical activity protects against colon cancer in all age groups, in various racial and ethnic groups and in diverse geographic areas around the world (Friedenreich & Orenstein, 2002; Kruk & Aboul-Enein, 2006; Lee, 2003; Thune & Furberg, 2001).

#### **2.1 Physical activity and rectal cancer**

Although colon and rectal cancer share some environmental risk factors, evidence of an association between rectal cancer and physical activity is inconsistent. A meta-analysis of 19 cohort studies even estimated that physical activity provides no protection against rectal cancer (Samad et al., 2005). It is therefore currently suggested that there is no association between physical activity and rectal cancer (Friedenreich & Orenstein, 2002; Kruk & Aboul-Enein, 2006; Lee, 2003; Thune & Furberg, 2001).

#### **2.2 Physical activity and colon cancer in humans**

Summarized observational epidemiological evidence on the association between physical activity and cancer suggests that the average risk reduction for colon cancer is 30-40% (Lee,

Epidemiological studies have shown that around 30-40% of CRC may be preventable by maintaining a healthy lifestyle and suitable diet. The available evidence suggests that the population attributable risk of physical inactivity is 13-14%, which is the same as the risk due to a Western eating pattern (Coups et al., 2008). These data, and all the aforementioned facts, show that the majority of sporadic CRC cases are preventable by adjustment of appropriate environmental and lifestyle factors (dietary habits, low body index, physical activity) and that there is a need to improve strategies of public health promotion campaigns in countries with increased risk of CRC. Public health promotion campaigns, if adopted, could have a major impact in the fight against sporadic CRC and would address many

This chapter is therefore an attempt in this direction and provides a current review of the literature on the relation between physical activity or dietary fat and CRC and the

The first evidence of the preventive role of physical activity against cancer appeared in 1922, when two groups of investigators, independently of each other, observed that cancer mortality rates declined with increasing physical activity required for an occupation (Lee, 2003). In spite of these results, further investigations in this area were not undertaken until the 1980s, when investigators observed that men with sedentary jobs had a higher colon cancer risk than men with jobs requiring strenuous activity (Larsson et al., 2006). Since then, the link between physical activity and cancer risk has been examined extensively, not only for cancer of the colon and rectum but also for breast and prostate cancer and, to a lesser extent also for cancers of the endometrium, lung, ovary, testis, pancreas and kidney. Nevertheless, cancer of the large bowel is the most frequently investigated cancer in relation to physical activity in humans. There have today been over 60 epidemiological studies investigating effects of physical activity on CRC in humans. Studies have been conducted in different parts of the world (North America, Europe, Australia, New Zealand and Asia-Japan) and among different populations and races. The results of these publications are convincing and clearly indicate that physical activity protects against colon cancer in all age groups, in various racial and ethnic groups and in diverse geographic areas around the world (Friedenreich & Orenstein, 2002; Kruk & Aboul-Enein, 2006; Lee, 2003; Thune &

Although colon and rectal cancer share some environmental risk factors, evidence of an association between rectal cancer and physical activity is inconsistent. A meta-analysis of 19 cohort studies even estimated that physical activity provides no protection against rectal cancer (Samad et al., 2005). It is therefore currently suggested that there is no association between physical activity and rectal cancer (Friedenreich & Orenstein, 2002; Kruk & Aboul-

Summarized observational epidemiological evidence on the association between physical activity and cancer suggests that the average risk reduction for colon cancer is 30-40% (Lee,

health and financial challenges.

mechanisms of their interaction.

**2. Physical activity and CRC** 

Furberg, 2001).

**2.1 Physical activity and rectal cancer** 

Enein, 2006; Lee, 2003; Thune & Furberg, 2001).

**2.2 Physical activity and colon cancer in humans** 

2003) or even 40-50% (Friedenreich & Orenstein, 2002). However, estimates from metaanalyses are little lower. A meta-analysis of 19 cohort studies estimated that physical activity may reduce the risk of colon cancer on average by 20-30% (Samad et al., 2005). The World Health Organization conducted a meta-analysis using data from studies prior to 2000 and estimated that around 16% of the global colon cancer disease burden can be ascribed to physical inactivity (Bull, 2004, as cited in Wolin et al., 2009). The most reliable results are probably those from the most recent meta-analysis, which included all available case-control (24) or cohort studies (28) that had been published to June 2008, but only those that investigated the association between physical activity and colon cancer or colon and rectal cancer separately. Studies on the association between physical activity and rectal cancer or colorectal cancer combined were not included. However, this meta-analysis showed a 24% average risk reduction for colon cancer when comparing the most to the least active individuals across all studies (Wolin et al., 2009).

The results of some studies suggest that the beneficial effects of exercise may be attenuated or less consistent in women (Kruk & Aboul-Enein, 2006; Thune & Furberg, 2001). However, most studies have found no difference in colon cancer risk according to gender (Friedenreich et al., 2006), which is in agreement with the latest meta-analysis. It was estimated that the protective effect of physical activity on colon cancer is similar for men (24%) and women (21%). The risk appeared smaller in cohort studies (men=19%; women=11%) than in case-control studies (men=28%; women=32%) (Wolin et al., 2009). It has been suggested that surveys used to measure physical activity were developed mainly for men and are thus less precise in estimating household work. Women may spend between 30 minutes to 6 hours a day on household chores and family care activities and from 4 to 16 hours a day on occupational activities, which makes it challenging to assess their physical activity accurately (Howard et al., 2008).

#### **2.2.1 Type, duration and intensity**

Although it is clear that physical activity is associated with a decreased risk of developing colon cancer, details of the relationship are less clear. It is known that the frequency of muscle contraction (e.g., number of activities performed per day, week, month), duration (number of minutes or hours per day), intensity (how much energy is expended) and activity levels throughout the participant's entire lifetime, are important components of activity that can significantly affect the protective effects of physical activity on colon cancer risk (Friedenreich & Orenstein, 2002). Few studies have examined the type, intensity and duration of physical activity required. Since they used different criteria of physical activity in their tests for trends, meta-analyses of trends across these studies have not yet been conducted (Wolin et al., 2009).

Physical activity is often thought of as recreational activity or exercise but it is much more than this. Physical activity is any bodily movement produced by the skeletal muscles that results in a quantifiable expenditure of energy and thereby comprises all leisure-time activities as well as occupational activities. Leisure-time physical activity refers to sports, conditioning exercises (structured and planned activity in order to improve or maintain physical fitness), household activities, self-care activities (e.g., dressing, eating, talking, standing, walking, climbing stairs) etc. (Kruk & Aboul-Enein, 2006).

Evidence is consistent that both occupational and leisure-time physical activity can affect colon cancer risk (Wolin et al., 2009). A dose-response relationship has been noted. Higher

Physical Activity, Dietary Fat and Colorectal Cancer 107

As shown above, physical activity has an important role in the prevention of colon cancer. Does physical activity also have a beneficial effect in CRC patients and survivors, though? A literature search shows that most attention on the efficacy of physical activity in colon cancer has been paid to cancer prevention. There have been few studies investigating the efficacy of activity in CRC patients. Nevertheless, available evidence suggests that physical activity may affect cancer treatment and outcome. It has been proposed that exercise during the pretreatment period may increase (boost) physical and psychological functioning, resulting in better physical preparation for treatment (shown in detail in Friedenreich & Orenstein, 2002). A study evaluating the benefits of physical activity in cancer patients and survivors shows improved functional capacity and quality of life (Johnson et al., 2009).

The effect of physical activity on CRC has mainly been evaluated using two rodent models of CRC, i.e., DMH/AOM animal model and ApcMin mice. The first model is a chemically induced animal model. Animals develop colon lesions after the application of a carcinogen (dimethylhydrazine (DMH) or azoxymethane (AOM)). Colorectal carcinogenesis in this model is a multistep process with molecular, morphological and histological features similar to those seen in human sporadic colon carcinogenesis (Perse & Cerar, 2011). The second model is a genetically predisposed model. ApcMin mice carry a dominant heterozygous nonsense mutation at codon 850 of the mouse homologue of the human tumor suppressor gene, APC, which results in the development of multiple adenomas throughout the intestinal tract. This mutation is implicated in both sporadic and inherited human colorectal

The first studies on carcinogen treated rats were directed at evaluating the effects of different voluntary or forced exercise (swimming, treadmill running, voluntary wheelrunning) on colorectal lesions in the later stages of development (tumors, adenomas, carcinomas). These studies found that exercise significantly reduced the incidence and multiplicity of tumors, as well as the incidence and multiplicity of adenocarcinomas but had little or no effect on the incidence and multiplicity of adenomas (Basterfield et al., 2005). Various studies have recently evaluated the effects of exercise on aberrant crypt foci (ACF), which are the first microscopically visible precursor lesions of CRC. They found that moderate-intensity exercise reduced the number of ACF in colons of DMH–treated rats (Fuku et al., 2007), low intensity exercise had no significant effect on the incidence of ACF (Lunz et al., 2008), while excessive and exhausting exercise significantly increased the number of ACF and, consequently, also the risk of development of colon cancer (Demarzo &

In contrast to the results obtained from the chemically induced model, the results of various studies on ApcMin mice are inconsistent. Two studies reported that exercise (treadmill running) had no effect on the incidence of intestinal adenomas in females, while a tendency toward a reduced incidence in males was observed (Colbert et al., 2000; Colbert et al., 2003). It was then found that the beneficial effects of exercise may be related to the exercise mode (treadmill/wheel running), gender (Mehl et al., 2005) and energy balance (Colbert et al., 2006). A recent study found that voluntary wheel running exercise also inhibited tumor

The reasons for the inconsistent results are not clear. In has been suggested that different types of exercise may elicit different physiological changes related to stress hormone release

**2.2.3 Physical activity may affect cancer treatment and outcome** 

**2.3 Physical activity and CRC in animal models** 

formation in female ApcMin mice (Ju et al., 2008).

carcinogenesis.

Garcia, 2004).

activity has been related to a reduced risk of colon cancer for both leisure-time and occupational physical activities (Friedenreich & Orenstein, 2002; Kruk & Aboul-Enein, 2006; Thune & Furberg, 2001; Wolin et al., 2009). Since little information is available, conclusions cannot be made about the type, intensity, frequency and duration of physical activity that is most beneficial.

Based on the current level of knowledge, it is believed that 30-60 min/day of regular physical activity of moderate to vigorous intensity is sufficient to decrease the risk of colon cancer in the general population (Friedenreich & Orenstein, 2002; Lee, 2003; Thune & Furberg, 2001).

#### **2.2.2 Other factors**

More detailed investigations on the effects of physical activity in relation to colon cancer site have recently been conducted. A few studies have evaluated the association between physical activity and colon cancer risk by anatomic site (proximal versus distal) and produced contradictory results. One study found no significant difference in risk estimates among different parts of the colon (Mai et al., 2007), while other studies found a reduction of risk only in transverse or sigmoid segments of the colon (Nilsen et al., 2008) or predominantly in the proximal colon (Lee et al., 2007).

It is believed that physical activity is associated with a reduced risk of colon cancer independently of diet or other environmental risk factors. This is supported by studies that have found that adjustment for potentially confounding factors, such as age, diet and obesity or body mass index, does not diminish the observed protective association (Friedenreich & Orenstein, 2002; Thune & Furberg, 2001).

However, it has been suggested that determination of potentially confounding variables is not always easy. When there are multiple hypothesized mechanisms, some of which may be in the causal path, the determination of confounding variables may be especially difficult. "For instance, if physical activity is associated with colon cancer through its ability to help maintain body weight, adjustment for body mass index would be inappropriate. If physical activity acts through other mechanisms, body mass index may be an important confounding variable because it is associated both with physical activity and colon cancer." An understanding of the biological mechanisms involved in the association between physical activity and colon cancer is therefore fundamental to evaluating confounding factors. In order to identify and understand the modifying effects of physical activity on other risk factors, the use of effect modification is advised (Slattery & Potter, 2002).

Slattery et al. examined confounding effect modification and observed the relative importance of high-risk diet, body mass index, energy intake and glycemic index in colon cancer prevention, which were found to be dependent on the level of physical activity (Slattery & Potter, 2002). Some studies have suggested a greater protective effect of physical activity in lean than in obese persons (Friedenreich et al., 2006; Larsson et al., 2006; Thune & Furberg, 2001). The findings of one cohort study even indicated that sedentary behavior, in particular television or video watching, is associated with an increased risk of colon cancer, independent of the time spent participating in physical activity and body mass index (Howard et al., 2008). Little is actually known about how physical activity may modify or be modified by other dietary and lifestyle factors, so conclusions based on currently available evidence can be misleading. Additional research in this direction is needed to provide public health recommendations regarding physical activity as a means of primary prevention of CRC.

activity has been related to a reduced risk of colon cancer for both leisure-time and occupational physical activities (Friedenreich & Orenstein, 2002; Kruk & Aboul-Enein, 2006; Thune & Furberg, 2001; Wolin et al., 2009). Since little information is available, conclusions cannot be made about the type, intensity, frequency and duration of physical activity that is

Based on the current level of knowledge, it is believed that 30-60 min/day of regular physical activity of moderate to vigorous intensity is sufficient to decrease the risk of colon cancer in the general population (Friedenreich & Orenstein, 2002; Lee, 2003; Thune &

More detailed investigations on the effects of physical activity in relation to colon cancer site have recently been conducted. A few studies have evaluated the association between physical activity and colon cancer risk by anatomic site (proximal versus distal) and produced contradictory results. One study found no significant difference in risk estimates among different parts of the colon (Mai et al., 2007), while other studies found a reduction of risk only in transverse or sigmoid segments of the colon (Nilsen et al., 2008) or

It is believed that physical activity is associated with a reduced risk of colon cancer independently of diet or other environmental risk factors. This is supported by studies that have found that adjustment for potentially confounding factors, such as age, diet and obesity or body mass index, does not diminish the observed protective association

However, it has been suggested that determination of potentially confounding variables is not always easy. When there are multiple hypothesized mechanisms, some of which may be in the causal path, the determination of confounding variables may be especially difficult. "For instance, if physical activity is associated with colon cancer through its ability to help maintain body weight, adjustment for body mass index would be inappropriate. If physical activity acts through other mechanisms, body mass index may be an important confounding variable because it is associated both with physical activity and colon cancer." An understanding of the biological mechanisms involved in the association between physical activity and colon cancer is therefore fundamental to evaluating confounding factors. In order to identify and understand the modifying effects of physical activity on other risk

Slattery et al. examined confounding effect modification and observed the relative importance of high-risk diet, body mass index, energy intake and glycemic index in colon cancer prevention, which were found to be dependent on the level of physical activity (Slattery & Potter, 2002). Some studies have suggested a greater protective effect of physical activity in lean than in obese persons (Friedenreich et al., 2006; Larsson et al., 2006; Thune & Furberg, 2001). The findings of one cohort study even indicated that sedentary behavior, in particular television or video watching, is associated with an increased risk of colon cancer, independent of the time spent participating in physical activity and body mass index (Howard et al., 2008). Little is actually known about how physical activity may modify or be modified by other dietary and lifestyle factors, so conclusions based on currently available evidence can be misleading. Additional research in this direction is needed to provide public health recommendations regarding physical activity as a means of primary

most beneficial.

Furberg, 2001).

**2.2.2 Other factors** 

prevention of CRC.

predominantly in the proximal colon (Lee et al., 2007).

(Friedenreich & Orenstein, 2002; Thune & Furberg, 2001).

factors, the use of effect modification is advised (Slattery & Potter, 2002).

#### **2.2.3 Physical activity may affect cancer treatment and outcome**

As shown above, physical activity has an important role in the prevention of colon cancer. Does physical activity also have a beneficial effect in CRC patients and survivors, though? A literature search shows that most attention on the efficacy of physical activity in colon cancer has been paid to cancer prevention. There have been few studies investigating the efficacy of activity in CRC patients. Nevertheless, available evidence suggests that physical activity may affect cancer treatment and outcome. It has been proposed that exercise during the pretreatment period may increase (boost) physical and psychological functioning, resulting in better physical preparation for treatment (shown in detail in Friedenreich & Orenstein, 2002). A study evaluating the benefits of physical activity in cancer patients and survivors shows improved functional capacity and quality of life (Johnson et al., 2009).

#### **2.3 Physical activity and CRC in animal models**

The effect of physical activity on CRC has mainly been evaluated using two rodent models of CRC, i.e., DMH/AOM animal model and ApcMin mice. The first model is a chemically induced animal model. Animals develop colon lesions after the application of a carcinogen (dimethylhydrazine (DMH) or azoxymethane (AOM)). Colorectal carcinogenesis in this model is a multistep process with molecular, morphological and histological features similar to those seen in human sporadic colon carcinogenesis (Perse & Cerar, 2011). The second model is a genetically predisposed model. ApcMin mice carry a dominant heterozygous nonsense mutation at codon 850 of the mouse homologue of the human tumor suppressor gene, APC, which results in the development of multiple adenomas throughout the intestinal tract. This mutation is implicated in both sporadic and inherited human colorectal carcinogenesis.

The first studies on carcinogen treated rats were directed at evaluating the effects of different voluntary or forced exercise (swimming, treadmill running, voluntary wheelrunning) on colorectal lesions in the later stages of development (tumors, adenomas, carcinomas). These studies found that exercise significantly reduced the incidence and multiplicity of tumors, as well as the incidence and multiplicity of adenocarcinomas but had little or no effect on the incidence and multiplicity of adenomas (Basterfield et al., 2005).

Various studies have recently evaluated the effects of exercise on aberrant crypt foci (ACF), which are the first microscopically visible precursor lesions of CRC. They found that moderate-intensity exercise reduced the number of ACF in colons of DMH–treated rats (Fuku et al., 2007), low intensity exercise had no significant effect on the incidence of ACF (Lunz et al., 2008), while excessive and exhausting exercise significantly increased the number of ACF and, consequently, also the risk of development of colon cancer (Demarzo & Garcia, 2004).

In contrast to the results obtained from the chemically induced model, the results of various studies on ApcMin mice are inconsistent. Two studies reported that exercise (treadmill running) had no effect on the incidence of intestinal adenomas in females, while a tendency toward a reduced incidence in males was observed (Colbert et al., 2000; Colbert et al., 2003). It was then found that the beneficial effects of exercise may be related to the exercise mode (treadmill/wheel running), gender (Mehl et al., 2005) and energy balance (Colbert et al., 2006). A recent study found that voluntary wheel running exercise also inhibited tumor formation in female ApcMin mice (Ju et al., 2008).

The reasons for the inconsistent results are not clear. In has been suggested that different types of exercise may elicit different physiological changes related to stress hormone release

Physical Activity, Dietary Fat and Colorectal Cancer 109

(Giovannucci, 2001). Based on meta-analysis of case-control and cohort studies, individuals with diabetes have an approximately 30% increased relative risk of developing CRC compared to non-diabetic individuals, regardless of gender or the anatomical site of CRC (Larsson et al., 2005). Preliminary results have shown that CRC is more common in people with increased circulating levels of insulin and glucose. A long-term increase in circulating levels of insulin may influence every step of colon carcinogenesis by stimulating cell proliferation or inhibiting apoptosis (Pisani, 2008). In addition to type 2 diabetes, obesity may cause problems with insulin metabolism and an alteration in blood glucose (explained

Physical activity can contribute to increased insulin sensitivity in skeletal muscles, both directly and indirectly through its influence on body weight (Giovannucci, 2001). Regular physical activity significantly lowers insulin levels by stimulation of the signaling pathways that contribute to increased expression and translocation of GLUT 4, which is responsible for basal and insulin-stimulated glucose uptake into the cells (explained in detail in Kramer

An increasing body of evidence suggests that variations not only in the levels of insulin but also in the levels of insulin-like growth factors (IGF) may account for colon cancer and for its high incidence in Western countries. The IGF family of proteins (peptide ligands, receptors, binding proteins and proteases) are involved in the regulation of somatic growth, cell proliferation, transformation and apoptosis. Among them, IGF-1 and IGFBP-3 have been most frequently investigated in relation to CRC. It has been hypothesized that IGF-1 is implicated in the etiology of CRC as a potent mediator of cell survival and growth (for more detail see Sandhu et al., 2002). In spite of this, current evidence does not support an association between the blood level of IGF-1 and CRC. Among exercise studies in humans, 50% have found no change in IGF-1, 37% an increase in IGF-1 and 13% a decrease in IGF-1 (Friedenreich & Orenstein, 2004). Likewise, no significant association between circulating levels of IGF-1 and exercise in animal models of CRC has been found (Colbert et al., 2000;

A number of studies have demonstrated that regular exercise has anti-inflammatory effects, which may play a significant role in the prevention of colon carcinogenesis, as well as many other diseases, such as atherosclerosis, type 2 diabetes and breast cancer. A marked increase in circulating levels of interleukin (IL)-6 after exercise, without any muscle damage, has been observed. It was found that the level of circulating IL-6 increases in an exponential fashion (up to 100-fold) in response to exercise and declines after exercise. It has been demonstrated that plasma IL-6 increase is related to exercise intensity, duration, the mass of muscle recruited and one's endurance capacity. Recent data demonstrate that IL-6 released from contracting human skeletal muscle has anti-inflammatory, immunosuppressive, metabolic and hypertrophic effects in humans (Petersen & Pedersen, 2005; Petersen &

Until recently, IL-6 was generally considered to be a pro-inflammatory cytokine released primarily from immune cells. However, dramatic increases in circulating IL-6 during exercise have led to the finding that skeletal muscles are a primary source of IL-6. Skeletal muscle has thus been found to be an immunogenic and an endocrine organ, which by contraction stimulates the production and release of cytokines, which can influence

Colbert et al., 2003; Colbert et al., 2006; Ju et al., 2008; Mehl et al., 2005).

**2.4.3 Effects on inflammatory modulators** 

in Murthy et al., 2009).

& Goodyear, 2007).

Pedersen, 2006).

and may alter the inflammatory effects (Mehl et al., 2005). Another possible reason is the large variation in tumor yield among individual ApcMin mice, which may have resulted in false-negative or non-significant results when a small number of animals were used (Ju et al., 2008). Finally, it is likely that this model may be suitable for investigating and assessing the effect of physical activity on CRC development in organisms with an inherited mutation or genetic predisposition.

However, experimental studies investigating the modifying effect of other dietary and lifestyle factors in relation to the beneficial effect of exercise are scarce. One study investigated the effect of exercise in animal models maintained on different types of high-fat diet. It was found that a different type of high-fat diet (21% coconut + 2% corn oil versus 23 % corn oil) may be associated with a different outcome of colon carcinogenesis in carcinogen treated rats exposed to exercise (Thorling et al., 1994). A second study reported that 6 weeks and 9 weeks of voluntary exercise (wheel running) successfully decreased the number of intestinal polyps in ApcMin mice on low and high fat diets, respectively (Ju et al., 2008). We recently found that exercise has a protective role in colon carcinogenesis in carcinogen treated rats, in the case of both low and high fat consumption diets. However, in terms of the combined effect of dietary fat and exercise, our results suggest that the protective role of exercise on colon carcinogenesis may be reduced in relation to the amount and type of fat in the diet (Perse, 2010). The lack of understanding of the biological mechanisms operating between physical activity and other risk factors warrants further research.

#### **2.4 Mechanisms of physical activity modulation**

A number of plausible biological mechanisms for the protective effect of physical activity against colon cancer have been suggested. They are mostly based on various experimental results. There are currently few empirical clinical data to support any of the hypothesized biological mechanisms for the protective effect of exercise on colon cancer.

#### **2.4.1 Effects on gastrointestinal transit time and gut microbiota**

The most frequently quoted explanation for reduced colon cancer among physically active people is that physical activity accelerates the movement of stool through the colon and shortens the gastrointestinal transit time, thereby reducing the contact of potential carcinogens and cancer promoters with colon mucosa (Kruk & Aboul-Enein, 2006). Although plausible, the epidemiological evidence of an association between gastrointestinal transit time and colon cancer risk has so far been inconsistent and this explanation has not yet been directly confirmed (Friedenreich et al., 2006).

The colon contains a vast population of many types of bacteria, which have potentially important functions and may contribute to cancer development (Tammariello & Milner, 2010). It was recently found that voluntary wheel running influnced the composition of cecal microbiota, which in turn produced higher concentrations of n-butyrate. Butyrate is a short-chain fatty acid end product of bacterial fermentation in the intestines, which has been related to intestinal motility and an inhibitory effect on tumor development (Matsumoto et al., 2008).

#### **2.4.2 Effects on blood insulin, IGF-1 and body weight**

Similarities in geographic patterns and dietary and lifestyle risk factors for CRC and type 2 diabetes have led to the suggestion that there is an association between the two diseases

and may alter the inflammatory effects (Mehl et al., 2005). Another possible reason is the large variation in tumor yield among individual ApcMin mice, which may have resulted in false-negative or non-significant results when a small number of animals were used (Ju et al., 2008). Finally, it is likely that this model may be suitable for investigating and assessing the effect of physical activity on CRC development in organisms with an inherited mutation

However, experimental studies investigating the modifying effect of other dietary and lifestyle factors in relation to the beneficial effect of exercise are scarce. One study investigated the effect of exercise in animal models maintained on different types of high-fat diet. It was found that a different type of high-fat diet (21% coconut + 2% corn oil versus 23 % corn oil) may be associated with a different outcome of colon carcinogenesis in carcinogen treated rats exposed to exercise (Thorling et al., 1994). A second study reported that 6 weeks and 9 weeks of voluntary exercise (wheel running) successfully decreased the number of intestinal polyps in ApcMin mice on low and high fat diets, respectively (Ju et al., 2008). We recently found that exercise has a protective role in colon carcinogenesis in carcinogen treated rats, in the case of both low and high fat consumption diets. However, in terms of the combined effect of dietary fat and exercise, our results suggest that the protective role of exercise on colon carcinogenesis may be reduced in relation to the amount and type of fat in the diet (Perse, 2010). The lack of understanding of the biological mechanisms operating

A number of plausible biological mechanisms for the protective effect of physical activity against colon cancer have been suggested. They are mostly based on various experimental results. There are currently few empirical clinical data to support any of the hypothesized

The most frequently quoted explanation for reduced colon cancer among physically active people is that physical activity accelerates the movement of stool through the colon and shortens the gastrointestinal transit time, thereby reducing the contact of potential carcinogens and cancer promoters with colon mucosa (Kruk & Aboul-Enein, 2006). Although plausible, the epidemiological evidence of an association between gastrointestinal transit time and colon cancer risk has so far been inconsistent and this explanation has not

The colon contains a vast population of many types of bacteria, which have potentially important functions and may contribute to cancer development (Tammariello & Milner, 2010). It was recently found that voluntary wheel running influnced the composition of cecal microbiota, which in turn produced higher concentrations of n-butyrate. Butyrate is a short-chain fatty acid end product of bacterial fermentation in the intestines, which has been related to intestinal motility and an inhibitory effect on tumor development (Matsumoto et

Similarities in geographic patterns and dietary and lifestyle risk factors for CRC and type 2 diabetes have led to the suggestion that there is an association between the two diseases

between physical activity and other risk factors warrants further research.

biological mechanisms for the protective effect of exercise on colon cancer.

**2.4.1 Effects on gastrointestinal transit time and gut microbiota** 

**2.4 Mechanisms of physical activity modulation** 

yet been directly confirmed (Friedenreich et al., 2006).

**2.4.2 Effects on blood insulin, IGF-1 and body weight** 

al., 2008).

or genetic predisposition.

(Giovannucci, 2001). Based on meta-analysis of case-control and cohort studies, individuals with diabetes have an approximately 30% increased relative risk of developing CRC compared to non-diabetic individuals, regardless of gender or the anatomical site of CRC (Larsson et al., 2005). Preliminary results have shown that CRC is more common in people with increased circulating levels of insulin and glucose. A long-term increase in circulating levels of insulin may influence every step of colon carcinogenesis by stimulating cell proliferation or inhibiting apoptosis (Pisani, 2008). In addition to type 2 diabetes, obesity may cause problems with insulin metabolism and an alteration in blood glucose (explained in Murthy et al., 2009).

Physical activity can contribute to increased insulin sensitivity in skeletal muscles, both directly and indirectly through its influence on body weight (Giovannucci, 2001). Regular physical activity significantly lowers insulin levels by stimulation of the signaling pathways that contribute to increased expression and translocation of GLUT 4, which is responsible for basal and insulin-stimulated glucose uptake into the cells (explained in detail in Kramer & Goodyear, 2007).

An increasing body of evidence suggests that variations not only in the levels of insulin but also in the levels of insulin-like growth factors (IGF) may account for colon cancer and for its high incidence in Western countries. The IGF family of proteins (peptide ligands, receptors, binding proteins and proteases) are involved in the regulation of somatic growth, cell proliferation, transformation and apoptosis. Among them, IGF-1 and IGFBP-3 have been most frequently investigated in relation to CRC. It has been hypothesized that IGF-1 is implicated in the etiology of CRC as a potent mediator of cell survival and growth (for more detail see Sandhu et al., 2002). In spite of this, current evidence does not support an association between the blood level of IGF-1 and CRC. Among exercise studies in humans, 50% have found no change in IGF-1, 37% an increase in IGF-1 and 13% a decrease in IGF-1 (Friedenreich & Orenstein, 2004). Likewise, no significant association between circulating levels of IGF-1 and exercise in animal models of CRC has been found (Colbert et al., 2000; Colbert et al., 2003; Colbert et al., 2006; Ju et al., 2008; Mehl et al., 2005).

#### **2.4.3 Effects on inflammatory modulators**

A number of studies have demonstrated that regular exercise has anti-inflammatory effects, which may play a significant role in the prevention of colon carcinogenesis, as well as many other diseases, such as atherosclerosis, type 2 diabetes and breast cancer. A marked increase in circulating levels of interleukin (IL)-6 after exercise, without any muscle damage, has been observed. It was found that the level of circulating IL-6 increases in an exponential fashion (up to 100-fold) in response to exercise and declines after exercise. It has been demonstrated that plasma IL-6 increase is related to exercise intensity, duration, the mass of muscle recruited and one's endurance capacity. Recent data demonstrate that IL-6 released from contracting human skeletal muscle has anti-inflammatory, immunosuppressive, metabolic and hypertrophic effects in humans (Petersen & Pedersen, 2005; Petersen & Pedersen, 2006).

Until recently, IL-6 was generally considered to be a pro-inflammatory cytokine released primarily from immune cells. However, dramatic increases in circulating IL-6 during exercise have led to the finding that skeletal muscles are a primary source of IL-6. Skeletal muscle has thus been found to be an immunogenic and an endocrine organ, which by contraction stimulates the production and release of cytokines, which can influence

Physical Activity, Dietary Fat and Colorectal Cancer 111

It has been reported that physical exercise decreased COX-2 expression in the colon mucosa of healthy untreated rats (Buehlmeyer et al., 2007) and DMH-treated rats (Demarzo et al., 2008). Exercise was found to inhibit one of the products of COX activities, PGE2, in intestinal tumors and serum of ApcMin mice (Ju et al., 2008). In rat colon mucosa, exercise was found to reduce the expression of iPLA2, which is one of the PLA2 implicated in arachidonic acid release (Buehlmeyer et al., 2008). Exercise has also been found to reduce PGE2 levels in rectal tissue among individuals with higher levels of self-reported exercise. On the other hand, in another study, exercise had no significant effect on PGE2 levels in colon mucosa

An alteration in the control of cellular proliferation and survival may be an important step in the development of colonic neoplasms. New cells are produced rapidly and continuously from stem cells at the base of the colonic crypt. Older cells undergo apoptosis (programmed cell death) and are sloughed into the colonic lumen. To maintain crypt organization and structure, cellular proliferation and apoptosis must be tightly controlled. Failure of these controls may lead to the formation of colonic neoplasms. It has been hypothesized that exercise-induced colon cancer risk reduction might be through alterations to colon crypt cell architecture and proliferation. It has been reported that a 12-month moderate-to-vigorous intensity exercise program (60 minutes per day, 6 days per week) increased colon crypt height and decreased proliferation in men (McTiernan et al., 2006) and changed the

There is growing support for the concept that reactive oxygen species (ROS), which are already implicated in a range of diseases, may be important progenitors in the pathogenesis of colon cancer. Namely, an excess of intracellular ROS results in an environment that modulates gene expression, damages cellular molecules, including DNA, which ultimately leads to mutations. In order to counteract these deleterious actions of increased levels of ROS, cells possess an antioxidant defence system, which plays a central role in protecting cells from oxidative injury. It is belived that exercise may help to prevent colon cancer due to an improvement in the cell's antioxidant defence system. It has already been demonstrated that exercise improves the antioxidant defence system in various tissues. Exercise stimulates various signaling pathways in cells, such as MAPK and NFκB, which results in increased expression of important enzymes associated with cell defence (MnSOD and GPx) and adaptation to exercise (eNOS and iNOS). Many of the biological effects of antioxidants appear to be related to their ability not only to scavenge deleterious free radicals but also to modulate cell-signalling pathways. The modulation of signalling pathways by antioxidants could thus help to prevent cancer by preserving normal cell cycle regulation, inhibiting proliferation and inducing apoptosis, inhibiting tumor invasion and angiogenesis, suppressing inflammation and stimulating detoxification enzyme activity (Kramer & Goodyear, 2007; Scheele et al., 2009; Valko et al., 2007). Exercise has been found to decrease the expression of inducible nitric oxide synthase (iNOS), as well as TNF-α, in the

The body of evidence is currently too limited to reach any final conclusions.

expression of Bcl-2 and Bax protein in colonic crypts (Campbell et al., 2007).

**2.4.6 Effects on apoptosis, proliferation, gene expression** 

(Abrahamson et al., 2007).

**2.4.7 Effects on oxidative status** 

colon of AOM-treated mice (Aoi et al., 2010).

metabolism and modify cytokine production in tissues and organs (Mathur & Pedersen, 2008).

It has been found that IL-6 induces the production of cytokine inhibitors, such as IL-1 receptor antagonist (IL-1ra) and IL-10, which are anti-inflammatory molecules. IL-1ra inhibits signaling transduction through the IL-1 receptor complex, while IL-10 inhibits the production of cytokines (IL-1α, IL-1β, TNF-α) and chemokines (IL-8, protein α), which play a critical role in the activation of granulocytes, monocytes, natural killer cells and T and B cells and in their recruitment to sites of inflammation (Petersen & Pedersen, 2005; Petersen & Pedersen, 2006).

IL-6 may increase basal and insulin-stimulated glucose uptake via increased GLUT 4 translocation. IL-6 has been shown to enhance AMP-activated protein kinase (AMPK) in both skeletal muscle and adipose tissue, which stimulates fatty acid oxidation and increases glucose uptake (Nielsen & Pedersen, 2008). TNF-α has been implicated in the pathogenesis of insulin resistance related to obesity (Steinberg, 2007). Evidence exists that TNF-α blocks AMPK signaling. However, exercise may also suppress TNF-α production via IL-6 independent pathways (Petersen & Pedersen, 2006).

#### **2.4.4 Effects on immune function**

It has been suggested that the immune system plays a role in reducing cancer risk by recognition and elimination of abnormal cells through immune components. Increased inflammation and/or depressed immune function are important risk factors that may lead to several cancers, including CRC. This is in accordance with the finding of an increased incidence of cancers among patients with inflammatory bowel disease (IBD) or AIDS. AIDS patients show an increased risk not only of AIDS-related malignancies (e.g., Kaposi's sarcoma) but also other cancers, such as lung and colon. An intact immune system is usually able to destroy cancer cells as soon as they appear.

It has been demonstrated that lifestyle factors can significantly affect immune function. Regular physical activity can enhance both the functionality and number of innate immune cell components, such as cytotoxic T lymphocytes, natural killer cells, lymphokine-activated killer cells and macrophages. Moderate physical activity results in enhanced immune function, whereas exhausting exercise, overtraining or high-intensity exercise may lead to suppression of the immune function (Pedersen & Hoffman-Goetz, 2000).

#### **2.4.5 Effects on arachidonic acid metabolism**

There have been studies suggesting that exercise affects enzymes that are implicated in arachidonic acid metabolism. Arachidonic acid is part of the phospholipids in the membranes of cells and in its free form serves as a precursor in the production of eicosanoids. After liberation (by the enzyme phospholipase A2 (PLA2)), arachidonic acid is available as a substrate for cyclooxygenases (COX) and lipoxygenases (LOX) to form prostaglandins (PG) and leukotriens (LT). Increased levels of COX-2 and PGE2 have been found to promote the development of CRC by increasing proliferation and decreasing colonic motility and apoptosis and have been associated with aggressive tumor progression. A relationship between PG and CRC is also supported by studies showing a reduced risk of CRC with aspirin and other non-steroidal anti-inflammatory drugs (NSAID), which inhibit COX-2, thereby inhibiting PG production (Jones et al., 2003).

metabolism and modify cytokine production in tissues and organs (Mathur & Pedersen,

It has been found that IL-6 induces the production of cytokine inhibitors, such as IL-1 receptor antagonist (IL-1ra) and IL-10, which are anti-inflammatory molecules. IL-1ra inhibits signaling transduction through the IL-1 receptor complex, while IL-10 inhibits the production of cytokines (IL-1α, IL-1β, TNF-α) and chemokines (IL-8, protein α), which play a critical role in the activation of granulocytes, monocytes, natural killer cells and T and B cells and in their recruitment to sites of inflammation (Petersen & Pedersen, 2005; Petersen

IL-6 may increase basal and insulin-stimulated glucose uptake via increased GLUT 4 translocation. IL-6 has been shown to enhance AMP-activated protein kinase (AMPK) in both skeletal muscle and adipose tissue, which stimulates fatty acid oxidation and increases glucose uptake (Nielsen & Pedersen, 2008). TNF-α has been implicated in the pathogenesis of insulin resistance related to obesity (Steinberg, 2007). Evidence exists that TNF-α blocks AMPK signaling. However, exercise may also suppress TNF-α production via IL-6

It has been suggested that the immune system plays a role in reducing cancer risk by recognition and elimination of abnormal cells through immune components. Increased inflammation and/or depressed immune function are important risk factors that may lead to several cancers, including CRC. This is in accordance with the finding of an increased incidence of cancers among patients with inflammatory bowel disease (IBD) or AIDS. AIDS patients show an increased risk not only of AIDS-related malignancies (e.g., Kaposi's sarcoma) but also other cancers, such as lung and colon. An intact immune system is usually

It has been demonstrated that lifestyle factors can significantly affect immune function. Regular physical activity can enhance both the functionality and number of innate immune cell components, such as cytotoxic T lymphocytes, natural killer cells, lymphokine-activated killer cells and macrophages. Moderate physical activity results in enhanced immune function, whereas exhausting exercise, overtraining or high-intensity exercise may lead to

There have been studies suggesting that exercise affects enzymes that are implicated in arachidonic acid metabolism. Arachidonic acid is part of the phospholipids in the membranes of cells and in its free form serves as a precursor in the production of eicosanoids. After liberation (by the enzyme phospholipase A2 (PLA2)), arachidonic acid is available as a substrate for cyclooxygenases (COX) and lipoxygenases (LOX) to form prostaglandins (PG) and leukotriens (LT). Increased levels of COX-2 and PGE2 have been found to promote the development of CRC by increasing proliferation and decreasing colonic motility and apoptosis and have been associated with aggressive tumor progression. A relationship between PG and CRC is also supported by studies showing a reduced risk of CRC with aspirin and other non-steroidal anti-inflammatory drugs (NSAID), which inhibit

suppression of the immune function (Pedersen & Hoffman-Goetz, 2000).

2008).

& Pedersen, 2006).

independent pathways (Petersen & Pedersen, 2006).

able to destroy cancer cells as soon as they appear.

**2.4.5 Effects on arachidonic acid metabolism** 

COX-2, thereby inhibiting PG production (Jones et al., 2003).

**2.4.4 Effects on immune function** 

It has been reported that physical exercise decreased COX-2 expression in the colon mucosa of healthy untreated rats (Buehlmeyer et al., 2007) and DMH-treated rats (Demarzo et al., 2008). Exercise was found to inhibit one of the products of COX activities, PGE2, in intestinal tumors and serum of ApcMin mice (Ju et al., 2008). In rat colon mucosa, exercise was found to reduce the expression of iPLA2, which is one of the PLA2 implicated in arachidonic acid release (Buehlmeyer et al., 2008). Exercise has also been found to reduce PGE2 levels in rectal tissue among individuals with higher levels of self-reported exercise. On the other hand, in another study, exercise had no significant effect on PGE2 levels in colon mucosa (Abrahamson et al., 2007).

The body of evidence is currently too limited to reach any final conclusions.

#### **2.4.6 Effects on apoptosis, proliferation, gene expression**

An alteration in the control of cellular proliferation and survival may be an important step in the development of colonic neoplasms. New cells are produced rapidly and continuously from stem cells at the base of the colonic crypt. Older cells undergo apoptosis (programmed cell death) and are sloughed into the colonic lumen. To maintain crypt organization and structure, cellular proliferation and apoptosis must be tightly controlled. Failure of these controls may lead to the formation of colonic neoplasms. It has been hypothesized that exercise-induced colon cancer risk reduction might be through alterations to colon crypt cell architecture and proliferation. It has been reported that a 12-month moderate-to-vigorous intensity exercise program (60 minutes per day, 6 days per week) increased colon crypt height and decreased proliferation in men (McTiernan et al., 2006) and changed the expression of Bcl-2 and Bax protein in colonic crypts (Campbell et al., 2007).

#### **2.4.7 Effects on oxidative status**

There is growing support for the concept that reactive oxygen species (ROS), which are already implicated in a range of diseases, may be important progenitors in the pathogenesis of colon cancer. Namely, an excess of intracellular ROS results in an environment that modulates gene expression, damages cellular molecules, including DNA, which ultimately leads to mutations. In order to counteract these deleterious actions of increased levels of ROS, cells possess an antioxidant defence system, which plays a central role in protecting cells from oxidative injury. It is belived that exercise may help to prevent colon cancer due to an improvement in the cell's antioxidant defence system. It has already been demonstrated that exercise improves the antioxidant defence system in various tissues. Exercise stimulates various signaling pathways in cells, such as MAPK and NFκB, which results in increased expression of important enzymes associated with cell defence (MnSOD and GPx) and adaptation to exercise (eNOS and iNOS). Many of the biological effects of antioxidants appear to be related to their ability not only to scavenge deleterious free radicals but also to modulate cell-signalling pathways. The modulation of signalling pathways by antioxidants could thus help to prevent cancer by preserving normal cell cycle regulation, inhibiting proliferation and inducing apoptosis, inhibiting tumor invasion and angiogenesis, suppressing inflammation and stimulating detoxification enzyme activity (Kramer & Goodyear, 2007; Scheele et al., 2009; Valko et al., 2007). Exercise has been found to decrease the expression of inducible nitric oxide synthase (iNOS), as well as TNF-α, in the colon of AOM-treated mice (Aoi et al., 2010).

Physical Activity, Dietary Fat and Colorectal Cancer 113

Lipids and fatty acids obtained from dietary fats are metabolized and incorporated into the phospholipids of the cell membranes of many cell types and serve as precursors for many biologically active molecules, as well as being important for cell signaling (Jones et al., 2003). It is generally accepted that the balance of n-6 to n-3 PUFA in the diet is of importance to human health and disease, including CRC. An alteration in fatty acid composition in the cell as a result of altered dietary fat consumption may lead to changes in all these functions,

Experimental and epidemiological studies have shown that diets high in beef tallow, lard or corn oil increase the concentration of colonic luminal (fecal) secondary bile acids, i.e. deoxycholic acid (DOC) and lithocholic acid, whereas high dietary fish oil has no such enhancing effect. It has been found and confirmed that these secondary bile acids induce cell proliferation and act as promoters in colon carcinogenesis. Recent experiments have provided new insight into their effects on colonic epithelial cells. The results indicate that secondary bile acid DOC may act as a carcinogen, not merely a promoter (explained in

Energy balance has become an important concept in exploring the etiology of a number of chronic diseases, including cancer, because of its close association with weight gain and

The amount of energy that is required depends in part on the composition of the food. In this regard, it is worth noting that dietary fats are more readily converted to body fat and require less energy for transformation than carbohydrates. A high fat diet therefore

One of the most thoroughly evaluated associations between nutrition and the immune system is that related to dietary fat. Although total fat intake has been found to increase the risk of various types of cancer, it is the type of fat that has a more important effect on the immune response and, consequently, on cancer development. PUFA have been shown to modulate cytokine production, lymphocyte proliferation, expression of surface molecules, phagocytosis, apoptosis and natural killer cell activity (the last two are closely related to cancer development). An increase in n-3 PUFA helps control the production of proinflammatory eicosanoids, as well as cytokine production (Valdes-Ramos & Benitez-

It has been suggested that dietary n-3 PUFA has an anti-carcinogenic role in reduction of n-6

Dietary n-6 PUFA incorporates into the membrane phospholipids as arachidonic acid (AA), while dietary n-3 PUFA does so as eicosapentaenoic acid (EPA). AA and EPA compete for prostaglandin and leukotriene synthesis. Pro-inflammatory eicosanoids of AA metabolism are released from membrane phospholipids in response to inflammatory activation. EPA is released to compete with AA for enzymatic metabolism, inducing the production of less

PUFA-derived eicosanoid biosynthesis and direct inhibition of COX-2.

overweight - conditions known to increase the risk of many chronic diseases.

contributes indirectly to CRC due to increased body mass index.

which are briefly outlined below.

detail in Bernstein et al., 2011).

**3.1.2 Effects on energy balance** 

**3.1.3 Effects on immune function** 

**3.1.4 Effects on arachidonic acid metabolism** 

Arciniega, 2007).

**3.1.1 Effects on the concentration of secondary bile acids** 

### **3. Dietary fat and CRC**

In contrast to physical activity, the association between fat intake and CRC is less conclusive. In the past, dietary fat has received considerable attention as a possible risk factor in the etiology of CRC but subsequent analysis of case control studies has indicated that the positive association was at least in part due to increased energy intake (Johnson & Lund, 2007).

While epidemiological studies have produced contradictory results (Johnson & Lund, 2007), experimental studies under isocaloric conditions have provided unequivocal evidence that a diet high in saturated fatty acids (SFA), such as lard or beef tallow, and n-6 polyunsaturated fatty acid (PUFA), such as corn or sunflower oil, increases the risk of developing CRC (Dai et al., 2002; Reddy, 2000). It was recently shown that long-term consumption of a high-fat, low-calcium and vitamin D diet induces colon neoplasia in mice, without any other treatment (Erdelyi et al., 2009). Interestingly, a recent expert review on nutrition and cancer published by the World Cancer Research Fund (American Institute for Cancer Research, 2007) found suggestive evidence that food rich in animal fat (rich in SFA) is associated with an increased risk of CRC. This means that epidemiological studies are mainly supportive but are limited in quantity, quality or consistency.

In contrast, diets high in olive oil or n-9 monounsaturated fatty acid (MUFA) have shown a protective or no effect on colon carcinogenesis in animal models, while diets with fish oil or n-3 PUFA have been shown to reduce colon tumorigenesis in both initiation and postinitiation phases (reviewed in Perše, 2010). Epidemiological reports investigating the effect of n-3 PUFA on CRC are scarce. However, some studies have shown that an n-3 PUFA-rich diet suppressed the risk of colon cancer in humans. The preventive or inhibitory effect of n-3 PUFA on experimental colon carcinogenesis has been widely evaluated (Biondo et al., 2008). All these results suggest that the composition of ingested dietary fatty acids is a more critical risk factor than the total amount of fat. This is further supported by their different mode of action, which is described in the following section.

However, at the same time, it is worth emphasizing that studies on animal models have shown that the promoting effects of SFA and n-6 PUFA on CRC can be modified by various dietary factors. A relatively small fraction of n-3 PUFA (25%) in total dietary fat or supplemental calcium or antioxidants, such as vitamin D (Pence & Buddingh, 1988), vitamin A (Delage et al., 2004), as well as green tea, vitamin B6 (Ju et al., 2003) and poliphenolic extract of red wine (Femia et al., 2005), have shown an appreciable beneficial effect in lowering the risk of CRC in animal models on a high fat diet. The influence of different amounts of calcium, antioxidants and other beneficial compounds in combination with dietary lipids may therefore be complex and difficult to elucidate, particularly in epidemiological investigations. Because many dietary, as well as environmental or lifestyle factors such as physical activity, can modify the promoting effects of dietary fat on CRC, results obtained from animal models under standardized conditions may represent an important contribution to understanding the mechanisms of dietary fat involvement in colorectal carcinogenesis (Hoffman-Goetz, 2003).

#### **3.1 Mechanisms of dietary fat modulation**

Dietary fats are an important energy reserve in an organism. However, this is not their only function. Linoleic acid (n-6 PUFA) and linolenic acid (n-3 PUFA) are considered essential, since they can not be synthesized by mammals and must therefore be obtained from diet.

In contrast to physical activity, the association between fat intake and CRC is less conclusive. In the past, dietary fat has received considerable attention as a possible risk factor in the etiology of CRC but subsequent analysis of case control studies has indicated that the positive association was at least in part due to increased energy intake (Johnson &

While epidemiological studies have produced contradictory results (Johnson & Lund, 2007), experimental studies under isocaloric conditions have provided unequivocal evidence that a diet high in saturated fatty acids (SFA), such as lard or beef tallow, and n-6 polyunsaturated fatty acid (PUFA), such as corn or sunflower oil, increases the risk of developing CRC (Dai et al., 2002; Reddy, 2000). It was recently shown that long-term consumption of a high-fat, low-calcium and vitamin D diet induces colon neoplasia in mice, without any other treatment (Erdelyi et al., 2009). Interestingly, a recent expert review on nutrition and cancer published by the World Cancer Research Fund (American Institute for Cancer Research, 2007) found suggestive evidence that food rich in animal fat (rich in SFA) is associated with an increased risk of CRC. This means that epidemiological studies are mainly supportive

In contrast, diets high in olive oil or n-9 monounsaturated fatty acid (MUFA) have shown a protective or no effect on colon carcinogenesis in animal models, while diets with fish oil or n-3 PUFA have been shown to reduce colon tumorigenesis in both initiation and postinitiation phases (reviewed in Perše, 2010). Epidemiological reports investigating the effect of n-3 PUFA on CRC are scarce. However, some studies have shown that an n-3 PUFA-rich diet suppressed the risk of colon cancer in humans. The preventive or inhibitory effect of n-3 PUFA on experimental colon carcinogenesis has been widely evaluated (Biondo et al., 2008). All these results suggest that the composition of ingested dietary fatty acids is a more critical risk factor than the total amount of fat. This is further supported by their different mode of

However, at the same time, it is worth emphasizing that studies on animal models have shown that the promoting effects of SFA and n-6 PUFA on CRC can be modified by various dietary factors. A relatively small fraction of n-3 PUFA (25%) in total dietary fat or supplemental calcium or antioxidants, such as vitamin D (Pence & Buddingh, 1988), vitamin A (Delage et al., 2004), as well as green tea, vitamin B6 (Ju et al., 2003) and poliphenolic extract of red wine (Femia et al., 2005), have shown an appreciable beneficial effect in lowering the risk of CRC in animal models on a high fat diet. The influence of different amounts of calcium, antioxidants and other beneficial compounds in combination with dietary lipids may therefore be complex and difficult to elucidate, particularly in epidemiological investigations. Because many dietary, as well as environmental or lifestyle factors such as physical activity, can modify the promoting effects of dietary fat on CRC, results obtained from animal models under standardized conditions may represent an important contribution to understanding the mechanisms of dietary fat involvement in

Dietary fats are an important energy reserve in an organism. However, this is not their only function. Linoleic acid (n-6 PUFA) and linolenic acid (n-3 PUFA) are considered essential, since they can not be synthesized by mammals and must therefore be obtained from diet.

**3. Dietary fat and CRC** 

but are limited in quantity, quality or consistency.

action, which is described in the following section.

colorectal carcinogenesis (Hoffman-Goetz, 2003).

**3.1 Mechanisms of dietary fat modulation** 

Lund, 2007).

Lipids and fatty acids obtained from dietary fats are metabolized and incorporated into the phospholipids of the cell membranes of many cell types and serve as precursors for many biologically active molecules, as well as being important for cell signaling (Jones et al., 2003). It is generally accepted that the balance of n-6 to n-3 PUFA in the diet is of importance to human health and disease, including CRC. An alteration in fatty acid composition in the cell as a result of altered dietary fat consumption may lead to changes in all these functions, which are briefly outlined below.

#### **3.1.1 Effects on the concentration of secondary bile acids**

Experimental and epidemiological studies have shown that diets high in beef tallow, lard or corn oil increase the concentration of colonic luminal (fecal) secondary bile acids, i.e. deoxycholic acid (DOC) and lithocholic acid, whereas high dietary fish oil has no such enhancing effect. It has been found and confirmed that these secondary bile acids induce cell proliferation and act as promoters in colon carcinogenesis. Recent experiments have provided new insight into their effects on colonic epithelial cells. The results indicate that secondary bile acid DOC may act as a carcinogen, not merely a promoter (explained in detail in Bernstein et al., 2011).

#### **3.1.2 Effects on energy balance**

Energy balance has become an important concept in exploring the etiology of a number of chronic diseases, including cancer, because of its close association with weight gain and overweight - conditions known to increase the risk of many chronic diseases.

The amount of energy that is required depends in part on the composition of the food. In this regard, it is worth noting that dietary fats are more readily converted to body fat and require less energy for transformation than carbohydrates. A high fat diet therefore contributes indirectly to CRC due to increased body mass index.

#### **3.1.3 Effects on immune function**

One of the most thoroughly evaluated associations between nutrition and the immune system is that related to dietary fat. Although total fat intake has been found to increase the risk of various types of cancer, it is the type of fat that has a more important effect on the immune response and, consequently, on cancer development. PUFA have been shown to modulate cytokine production, lymphocyte proliferation, expression of surface molecules, phagocytosis, apoptosis and natural killer cell activity (the last two are closely related to cancer development). An increase in n-3 PUFA helps control the production of proinflammatory eicosanoids, as well as cytokine production (Valdes-Ramos & Benitez-Arciniega, 2007).

#### **3.1.4 Effects on arachidonic acid metabolism**

It has been suggested that dietary n-3 PUFA has an anti-carcinogenic role in reduction of n-6 PUFA-derived eicosanoid biosynthesis and direct inhibition of COX-2.

Dietary n-6 PUFA incorporates into the membrane phospholipids as arachidonic acid (AA), while dietary n-3 PUFA does so as eicosapentaenoic acid (EPA). AA and EPA compete for prostaglandin and leukotriene synthesis. Pro-inflammatory eicosanoids of AA metabolism are released from membrane phospholipids in response to inflammatory activation. EPA is released to compete with AA for enzymatic metabolism, inducing the production of less

Physical Activity, Dietary Fat and Colorectal Cancer 115

and RAR, RXR) are key transcription factors regulating gene expression in response to nutrient-activated signals. A high-fat diet containing various sources of fat, such as commonly consumed in Western countries (the majority SFA), induced PPARγ and RARβ expression, concomitant with an increase in levels of COX-2 and β-catenin in colon mucosa of DMH treated rats (Delage et al., 2004). Various fatty acids have different effects on the Wnt signaling pathway (Kim & Milner, 2007). Long-term (44 wks) high intake of corn oil and beef tallow enhanced Wnt signaling. Dietary corn oil and beef tallow increased the expression of cytosolic β-catenin and cyclin D1. Expressions of Wnt2 and Wnt3 in rats fed with beef tallow and Wnt5a in rats fed with corn oil increased, with or without AOM-

It has been demonstrated that dietary fatty acids affect the lipid content of tissue and the lipid peroxidation process, due to the ratio of polyunsaturated versus saturated fatty acid. A substantial increase in the PUFA content may overcome the protective action of the antioxidant system and increase susceptibility to lipid peroxidation (Avula & Fernandes, 1999). We have recently demonstrated that long-term consumption of an high-fat mixedlipid (HFML) diet together with physical inactivity significantly increased the production of lipid peroxides in the skeletal muscle (Perse et al., 2009), suggesting that an HFML diet is an important contributor to the development of chronic diseases, including CRC. There is growing support for the concept that an excess of intracellular ROS, results in an environment that modulates gene expression, damages cellular molecules, including DNA, which ultimately leads to mutations (Valko et al., 2006). On the other hand, fish oil has been

The large intestine is constantly exposed to ROS originating from endogenous and exogenous sources, due to oxidized food debris, toxins and high levels of iron. It has been demonstrated that dietary fatty acids affect the lipid content of tissue and result in differential susceptibility to peroxidation (Kuratko & Pence, 1991; Kuratko & Becker, 1998; Kuratko & Constante, 1998; Wu et al., 2004). It was recently shown that a high-fat, lowcalcium and vitamin D diet induces oxidative stress in the colon (Erdelyi et al., 2009).

Based on considerable evidence showing different beneficial effects of n-3 PUFA, it has been suggested that n-3 PUFA may improve the outcome of patients undergoing abdominal cancer surgery (Valdes-Ramos & Benitez-Arciniega, 2007). Since n-3 PUFA enrichment can affect the physical properties of cell membranes, altering membrane fluidity and increasing the permeability of tumor cells, it has been proposed that n-3 PUFA consumption may modify the influx and efflux of drugs into or out of tumor cells. However, elucidation of mechanisms is essential for ensuring both the optimal efficacy of a drug and for identifying the target level at which to modify the diet or supplement with n-3 PUFA, in order to

Evidence that physical activity affects the risk of colon cancer has been provided by numerous epidemiological and experimental studies and reviewed extensively. Although strong evidence exists that regular physical activity is associated with decreased risk of

found to reduce oxidative DNA damage (Bancroft et al., 2003; Wu et al., 2004).

**3.1.8 Beneficial role of n-3 PUFA before or during chemotherapy** 

optimize the benefits to the patient (Biondo et.al, 2008).

**4. Conclusion** 

treatment (Fujise et al., 2007).

**3.1.7 Effects on oxidative status** 

inflammatory and chemotactic derivatives. Eicosanoids produced from EPA are much less potent (up to 100-fold) than the analogues produced from AA and even have antithrombotic and anti-inflammatory properties. The relative amounts of n-6 and n-3 PUFA provided by the diet, and so present in blood and tissues, may thus be of importance in the development of inflammatory diseases and cancers. The production of inflammatory eicosanoids is increased in response to many inflammatory stimuli (Simopoulos, 2002a). When the production of these substances is excessive, it may lead to chronic inflammation and an increased risk of cancer, since inflammation has been linked to the promotion phase of carcinogenesis (Federico et al., 2007). The increased n-6/n-3 ratio in Western diets probably contributes to reduced levels of EPA in phospholipids and, consequently, to an increased incidence of cardiovascular disease and inflammatory disorders (Simopoulos, 2002b).

Another indication of the importance of diet and n-6 PUFA in the induction and progression of CRC may be the upregulation in fatty acid binding protein (FABP)-5 during tumorigenesis, with concomitant inhibition of Δ6 desaturase activity, which are important steps in the production of AA (explained in Jones et al., 2003). Most AA in the human body derives from dietary linoleic acid (essential n-6 PUFA), which comes from vegetable oils and animal fats.

Animal studies have demonstrated that a high fat diet significantly increases the expression of PLA2, COX-2 and PGE2 in colon mucosa and tumors of carcinogen treated rats (Rao et al., 1996; Rao et al., 2001).

#### **3.1.5 Effects on cell proliferation and apoptosis**

The expression of Polo-like kinase-3 (PLK-3) results in cell cycle arrest or induces apoptosis. It is significantly suppressed in tumor tissue of the colon but has been found to be unchanged in colon mucosa isolated from rats on different diets. Suppression of PLK-3 was lower in tumors from rats fed n-3 PUFA than those fed n-6 PUFA (Dai et al., 2002).

Dietary corn oil and beef tallow increased BrdU incorporation and decreased apoptosis of the colon mucosa. Long-term (44 wks) high intake of corn oil and beef tallow enhanced cell proliferation through Wnt signaling and modulated the distribution of proliferating cells (Fujise et al., 2007).

High corn oil consumption decreased apoptosis and increased cell proliferation in colon of AOM treated rats (Wu et al., 2004). On the other hand, studies have indicated that n-3 PUFA has an inhibitory effect at least in part due to increased apoptosis in colonic mucosa (Hong et al., 2000; Wu et al., 2004).

#### **3.1.6 Effects on cell signaling pathways**

Studies have suggested that different types of fat may be implicated in different cell signaling pathways, rather than at the level of mutations. n-3 PUFA may interfere with Ras activation by decreasing its membrane localization and may thereby potentiate the effects of anti-Ras therapies. EPA and/or docosahexaenoic acid (DHA; another n-3 PUFA) have also been reported to prevent Akt phosphorylation or activation. n-3 PUFA incorporation into rafts or caveolae may alter the distribution or function of raft-associated signaling proteins – reduced epithelial growth factor receptor (EGFR) levels in the rafts, decreased levels of Hras and eNOS in colonic caveolae. Alterations in raft lipid composition by PUFA have also been shown to displace signaling proteins from rafts in immune cells. n-3 PUFA decreases NFκB activity or expression in cancer cells, as well as monocytes, macrophages and T cells (Biondo et al., 2008). Peroxisome proliferators and retinoic acid–activated receptors (PPAR

inflammatory and chemotactic derivatives. Eicosanoids produced from EPA are much less potent (up to 100-fold) than the analogues produced from AA and even have antithrombotic and anti-inflammatory properties. The relative amounts of n-6 and n-3 PUFA provided by the diet, and so present in blood and tissues, may thus be of importance in the development of inflammatory diseases and cancers. The production of inflammatory eicosanoids is increased in response to many inflammatory stimuli (Simopoulos, 2002a). When the production of these substances is excessive, it may lead to chronic inflammation and an increased risk of cancer, since inflammation has been linked to the promotion phase of carcinogenesis (Federico et al., 2007). The increased n-6/n-3 ratio in Western diets probably contributes to reduced levels of EPA in phospholipids and, consequently, to an increased

incidence of cardiovascular disease and inflammatory disorders (Simopoulos, 2002b).

animal fats.

1996; Rao et al., 2001).

(Fujise et al., 2007).

et al., 2000; Wu et al., 2004).

**3.1.6 Effects on cell signaling pathways** 

**3.1.5 Effects on cell proliferation and apoptosis** 

Another indication of the importance of diet and n-6 PUFA in the induction and progression of CRC may be the upregulation in fatty acid binding protein (FABP)-5 during tumorigenesis, with concomitant inhibition of Δ6 desaturase activity, which are important steps in the production of AA (explained in Jones et al., 2003). Most AA in the human body derives from dietary linoleic acid (essential n-6 PUFA), which comes from vegetable oils and

Animal studies have demonstrated that a high fat diet significantly increases the expression of PLA2, COX-2 and PGE2 in colon mucosa and tumors of carcinogen treated rats (Rao et al.,

The expression of Polo-like kinase-3 (PLK-3) results in cell cycle arrest or induces apoptosis. It is significantly suppressed in tumor tissue of the colon but has been found to be unchanged in colon mucosa isolated from rats on different diets. Suppression of PLK-3 was

Dietary corn oil and beef tallow increased BrdU incorporation and decreased apoptosis of the colon mucosa. Long-term (44 wks) high intake of corn oil and beef tallow enhanced cell proliferation through Wnt signaling and modulated the distribution of proliferating cells

High corn oil consumption decreased apoptosis and increased cell proliferation in colon of AOM treated rats (Wu et al., 2004). On the other hand, studies have indicated that n-3 PUFA has an inhibitory effect at least in part due to increased apoptosis in colonic mucosa (Hong

Studies have suggested that different types of fat may be implicated in different cell signaling pathways, rather than at the level of mutations. n-3 PUFA may interfere with Ras activation by decreasing its membrane localization and may thereby potentiate the effects of anti-Ras therapies. EPA and/or docosahexaenoic acid (DHA; another n-3 PUFA) have also been reported to prevent Akt phosphorylation or activation. n-3 PUFA incorporation into rafts or caveolae may alter the distribution or function of raft-associated signaling proteins – reduced epithelial growth factor receptor (EGFR) levels in the rafts, decreased levels of Hras and eNOS in colonic caveolae. Alterations in raft lipid composition by PUFA have also been shown to displace signaling proteins from rafts in immune cells. n-3 PUFA decreases NFκB activity or expression in cancer cells, as well as monocytes, macrophages and T cells (Biondo et al., 2008). Peroxisome proliferators and retinoic acid–activated receptors (PPAR

lower in tumors from rats fed n-3 PUFA than those fed n-6 PUFA (Dai et al., 2002).

and RAR, RXR) are key transcription factors regulating gene expression in response to nutrient-activated signals. A high-fat diet containing various sources of fat, such as commonly consumed in Western countries (the majority SFA), induced PPARγ and RARβ expression, concomitant with an increase in levels of COX-2 and β-catenin in colon mucosa of DMH treated rats (Delage et al., 2004). Various fatty acids have different effects on the Wnt signaling pathway (Kim & Milner, 2007). Long-term (44 wks) high intake of corn oil and beef tallow enhanced Wnt signaling. Dietary corn oil and beef tallow increased the expression of cytosolic β-catenin and cyclin D1. Expressions of Wnt2 and Wnt3 in rats fed with beef tallow and Wnt5a in rats fed with corn oil increased, with or without AOMtreatment (Fujise et al., 2007).

#### **3.1.7 Effects on oxidative status**

It has been demonstrated that dietary fatty acids affect the lipid content of tissue and the lipid peroxidation process, due to the ratio of polyunsaturated versus saturated fatty acid. A substantial increase in the PUFA content may overcome the protective action of the antioxidant system and increase susceptibility to lipid peroxidation (Avula & Fernandes, 1999). We have recently demonstrated that long-term consumption of an high-fat mixedlipid (HFML) diet together with physical inactivity significantly increased the production of lipid peroxides in the skeletal muscle (Perse et al., 2009), suggesting that an HFML diet is an important contributor to the development of chronic diseases, including CRC. There is growing support for the concept that an excess of intracellular ROS, results in an environment that modulates gene expression, damages cellular molecules, including DNA, which ultimately leads to mutations (Valko et al., 2006). On the other hand, fish oil has been found to reduce oxidative DNA damage (Bancroft et al., 2003; Wu et al., 2004).

The large intestine is constantly exposed to ROS originating from endogenous and exogenous sources, due to oxidized food debris, toxins and high levels of iron. It has been demonstrated that dietary fatty acids affect the lipid content of tissue and result in differential susceptibility to peroxidation (Kuratko & Pence, 1991; Kuratko & Becker, 1998; Kuratko & Constante, 1998; Wu et al., 2004). It was recently shown that a high-fat, lowcalcium and vitamin D diet induces oxidative stress in the colon (Erdelyi et al., 2009).

#### **3.1.8 Beneficial role of n-3 PUFA before or during chemotherapy**

Based on considerable evidence showing different beneficial effects of n-3 PUFA, it has been suggested that n-3 PUFA may improve the outcome of patients undergoing abdominal cancer surgery (Valdes-Ramos & Benitez-Arciniega, 2007). Since n-3 PUFA enrichment can affect the physical properties of cell membranes, altering membrane fluidity and increasing the permeability of tumor cells, it has been proposed that n-3 PUFA consumption may modify the influx and efflux of drugs into or out of tumor cells. However, elucidation of mechanisms is essential for ensuring both the optimal efficacy of a drug and for identifying the target level at which to modify the diet or supplement with n-3 PUFA, in order to optimize the benefits to the patient (Biondo et.al, 2008).

#### **4. Conclusion**

Evidence that physical activity affects the risk of colon cancer has been provided by numerous epidemiological and experimental studies and reviewed extensively. Although strong evidence exists that regular physical activity is associated with decreased risk of

Physical Activity, Dietary Fat and Colorectal Cancer 117

Buehlmeyer, K., Doering, F., Daniel, H., Schulz, T., & Michna, H. (2007) Exercise Associated

Campbell, K.L., McTiernan, A., Li, S.S., Sorensen, B.E., Yasui, Y., Lampe, J.W., King, I.B.,

Colbert, L.H., Davis, J.M., Essig, D.A., Ghaffar, A., & Mayer, E.P. (2000) Exercise and Tumor

Colbert, L.H., Mai, V., Perkins, S.N., Berrigan, D., Lavigne, J.A., Wimbrow, H.H., Alvord,

Colbert, L.H., Mai, V., Tooze, J.A., Perkins, S.N., Berrigan, D., & Hursting, S.D. (2006)

Development in APCMin Mice. *Carcinogenesis*, Vol.27, No.10, pp. 2103-2107, Coups, E.J., Hay, J., & Ford, J.S. (2008) Awareness of the Role of Physical Activity in Colon

Dai, W., Liu, T., Wang, Q., Rao, C.V., & Reddy, B.S. (2002) Down-Regulation of PLK3 Gene

Delage, B., Groubet, R., Pallet, V., Bairras, C., Higueret, P., & Cassand, P. (2004) Vitamin A

Demarzo, M.M. & Garcia, S.B. (2004) Exhaustive Physical Exercise Increases the Number of

Demarzo, M.M., Martins, L.V., Fernandes, C.R., Herrero, F.A., Perez, S.E., Turatti, A., &

Federico, A., Morgillo, F., Tuccillo, C., Ciardiello, F., & Loguercio, C. (2007) Chronic

Femia, A.P., Caderni, G., Vignali, F., Salvadori, M., Giannini, A., Biggeri, A., Gee, J.,

Colon Carcinogenesis in Rats. *Eur J Nutr*, Vol.44, No.2, pp. 79-84,

Colon Carcinogenesis. *Med Sci Sports Exerc*, Vol.40, No.4, pp. 618-621, Erdelyi, I., Levenkova, N., Lin, E.Y., Pinto, J.T., Lipkin, M., Quimby, F.W., & Holt, P.R. (2009)

Cancer Prevention. *Patient Educ Couns*, Vol.72, No.2, pp. 246-251,

*Cancer Epidemiol Biomarkers Prev*, Vol.16, No.9, pp. 1767-1774,

*Med*, Vol.28, No.5, pp. 361-367,

Vol.20, No.1, pp. 121-126,

pp. 2072-2078,

No.11, pp. 2381-2386,

*Cancer*, Vol.48, No.1, pp. 28-36,

Carcinogen. *Cancer Lett*, Vol.216, No.1, pp. 31-34,

1669,

*Sports Exerc*, Vol.32, No.10, pp. 1704-1708,

Genes in Rat Colon Mucosa: Upregulation of Ornithin Decarboxylase-1. *Int J Sports* 

Ulrich, C.M., Rudolph, R.E., Irwin, M.L., Surawicz, C., Ayub, K., Potter, J.D., & Lampe, P.D. (2007) Effect of a 12-Month Exercise Intervention on the Apoptotic Regulating Proteins Bax and Bcl-2 in Colon Crypts: a Randomized Controlled Trial.

Development in a Mouse Predisposed to Multiple Intestinal Adenomas. *Med Sci* 

W.G., Haines, D.C., Srinivas, P., & Hursting, S.D. (2003) Exercise and Intestinal Polyp Development in APCMin Mice. *Med Sci Sports Exerc*, Vol.35, No.10, pp. 1662-

Negative Energy Balance Induced by Voluntary Wheel Running Inhibits Polyp

Expression by Types and Amount of Dietary Fat in Rat Colon Tumors. *Int J Oncol*,

Prevents High Fat Diet-Induced ACF Development and Modifies the Pattern of Expression of Peroxisome Proliferator and Retinoic Acid Receptor M-RNA. *Nutr* 

Colonic Preneoplastic Lesions in Untrained Rats Treated With a Chemical

Garcia, S.B. (2008) Exercise Reduces Inflammation and Cell Proliferation in Rat

Western-Style Diets Induce Oxidative Stress and Dysregulate Immune Responses in the Colon in a Mouse Model of Sporadic Colon Cancer. *J Nutr*, Vol.139, No.11,

Inflammation and Oxidative Stress in Human Carcinogenesis. *Int J Cancer*, Vol.121,

Przybylska, K., Cheynier, V., & Dolara, P. (2005) Effect of Polyphenolic Extracts From Red Wine and 4-OH-Coumaric Acid on 1,2-Dimethylhydrazine-Induced

colon cancer, little is known about the type, intensity, frequency and duration of physical activity that is most benefical. Evidence is consistent that both occupational and leisure-time physical activity can affect the risk of colon cancer. Based on the current level of knowledge, it is believed that 30-60 min/day of regular moderate to vigorous intensity physical activity is sufficient to decrease the risk of colon cancer in the general population.

The relation between dietary fat and CRC is less conclusive. Experimental studies suggest that the composition of ingested dietary fatty acids is a more critical risk factor than the total amount of fat. This is further supported by their different modes of action. It has been proposed that the increased n-6/n-3 ratio in Western diets probably contributes to an increased incidence of cardiovascular disease and inflammatory disorders, as well as at least in part CRC. There is increasing body of evidence that the consumption of n-3 PUFA can impact on immune functions, as well as alter gene expression and transcription factor activity in normal and cancer cells. It has also been found to reduce CRC risk and is suggested to have a beneficial role before or during chemotherapy, and even improve drug uptake.

#### **5. Acknowledgement**

This work was in part supported by ARRS (Slovenian Research Agency, Program P3-054).

#### **6. References**


colon cancer, little is known about the type, intensity, frequency and duration of physical activity that is most benefical. Evidence is consistent that both occupational and leisure-time physical activity can affect the risk of colon cancer. Based on the current level of knowledge, it is believed that 30-60 min/day of regular moderate to vigorous intensity physical activity

The relation between dietary fat and CRC is less conclusive. Experimental studies suggest that the composition of ingested dietary fatty acids is a more critical risk factor than the total amount of fat. This is further supported by their different modes of action. It has been proposed that the increased n-6/n-3 ratio in Western diets probably contributes to an increased incidence of cardiovascular disease and inflammatory disorders, as well as at least in part CRC. There is increasing body of evidence that the consumption of n-3 PUFA can impact on immune functions, as well as alter gene expression and transcription factor activity in normal and cancer cells. It has also been found to reduce CRC risk and is suggested to have a beneficial role before or during chemotherapy, and even improve drug

This work was in part supported by ARRS (Slovenian Research Agency, Program P3-054).

Abrahamson, P.E., King, I.B., Ulrich, C.M., Rudolph, R.E., Irwin, M.L., Yasui, Y., Surawicz,

Aoi, W., Naito, Y., Takagi, T., Kokura, S., Mizushima, K., Takanami, Y., Kawai, Y.,

Avula, C.P. & Fernandes, G. (1999) Modulation of Antioxidant Enzymes and Apoptosis in

Bancroft, L.K., Lupton, J.R., Davidson, L.A., Taddeo, S.S., Murphy, M.E., Carroll, R.J., &

Basterfield, L., Reul, J.M., & Mathers, J.C. (2005) Impact of Physical Activity on Intestinal Cancer Development in Mice. *J Nutr*, Vol.135, No.12 Suppl, pp. 3002S-3008S, Biondo, P.D., Brindley, D.N., Sawyer, M.B., & Field, C.J. (2008) The Potential for Treatment

Buehlmeyer, K., Doering, F., Daniel, H., Kindermann, B., Schulz, T., & Michna, H. (2008)

INOS. *Biochem Biophys Res Commun*, Vol.399, No.1, pp. 14-19,

Colonocytes. *Free Radic Biol Med*, Vol.35, No.2, pp. 149-159,

*Biomarkers Prev*, Vol.16, No.11, pp. 2351-2356,

*J Nutr Biochem*, Vol.19, No.12, pp. 787-796,

Vol.190, No.1, pp. 71-80,

C., Lampe, J.W., Lampe, P.D., Morgan, A., Sorensen, B.E., Ayub, K., Potter, J.D., & McTiernan, A. (2007) No Effect of Exercise on Colon Mucosal Prostaglandin Concentrations: a 12-Month Randomized Controlled Trial. *Cancer Epidemiol* 

Tanimura, Y., Hung, L.P., Koyama, R., Ichikawa, H., & Yoshikawa, T. (2010) Regular Exercise Reduces Colon Tumorigenesis Associated With Suppression of

Mice by Dietary Lipids and Treadmill Exercise. *J Clin Immunol*, Vol.19, No.1, pp. 35-

Chapkin, R.S. (2003) Dietary Fish Oil Reduces Oxidative DNA Damage in Rat

With Dietary Long-Chain Polyunsaturated N-3 Fatty Acids During Chemotherapy.

Alteration of Gene Expression in Rat Colon Mucosa After Exercise. *Ann Anat*,

is sufficient to decrease the risk of colon cancer in the general population.

uptake.

**5. Acknowledgement** 

**6. References** 

44,


Physical Activity, Dietary Fat and Colorectal Cancer 119

Jones, R., Adel-Alvarez, L.A., Alvarez, O.R., Broaddus, R., & Das, S. (2003) Arachidonic Acid and Colorectal Carcinogenesis. *Mol Cell Biochem*, Vol.253, No.1-2, pp. 141-149, Ju, J., Liu, Y., Hong, J., Huang, M.T., Conney, A.H., & Yang, C.S. (2003) Effects of Green Tea

Ju, J., Nolan, B., Cheh, M., Bose, M., Lin, Y., Wagner, G.C., & Yang, C.S. (2008) Voluntary

Keighley, M.R., O'Morain, C., Giacosa, A., Ashorn, M., Burroughs, A., Crespi, M., Delvaux,

Kim, Y.S. & Milner, J.A. (2007) Dietary Modulation of Colon Cancer Risk. *J Nutr*, Vol.137,

Kramer, H.F. & Goodyear, L.J. (2007) Exercise, MAPK, and NF-KappaB Signaling in Skeletal

Kruk, J. & Aboul-Enein, H.Y. (2006) Physical Activity in the Prevention of Cancer. *Asian Pac* 

Kuratko, C. & Pence, B.C. (1991) Changes in Colonic Antioxidant Status in Rats During

Kuratko, C.N. & Becker, S.A. (1998) Dietary Lipids Alter Fatty Acid Composition and PGE2 Production in Colonic Lymphocytes. *Nutr Cancer*, Vol.31, No.1, pp. 56-61, Kuratko, C.N. & Constante, B.J. (1998) Linoleic Acid and Tumor Necrosis Factor-Alpha

Larsson, S.C., Orsini, N., & Wolk, A. (2005) Diabetes Mellitus and Risk of Colorectal Cancer:

Larsson, S.C., Rutegard, J., Bergkvist, L., & Wolk, A. (2006) Physical Activity, Obesity, and

Lee, I.M. (2003) Physical Activity and Cancer Prevention-Data From Epidemiologic Studies.

Lee, K.J., Inoue, M., Otani, T., Iwasaki, M., Sasazuki, S., & Tsugane, S. (2007) Physical

Lunz, W., Peluzio, M.C., Dias, C.M., Moreira, A.P., & Natali, A.J. (2008) Long-Term Aerobic

Mai, P.L., Sullivan-Halley, J., Ursin, G., Stram, D.O., Deapen, D., Villaluna, D., Horn-Ross,

a Meta-Analysis. *J Natl Cancer Inst*, Vol.97, No.22, pp. 1679-1687,

*Med Sci Sports Exerc*, Vol.35, No.11, pp. 1823-1827,

Long-Term Feeding of Different High Fat Diets. *J Nutr*, Vol.121, No.10, pp. 1562-

Increase Manganese Superoxide Dismutase Activity in Intestinal Cells. *Cancer Lett*,

Risk of Colon and Rectal Cancer in a Cohort of Swedish Men. *Eur J Cancer*, Vol.42,

Activity and Risk of Colorectal Cancer in Japanese Men and Women: the Japan Public Health Center-Based Prospective Study. *Cancer Causes Control*, Vol.18, No.2,

Swimming Training by Rats Reduces the Number of Aberrant Crypt Foci in 1,2- Dimethylhydrazine-Induced Colon Cancer. *Braz J Med Biol Res*, Vol.41, No.11, pp.

P.L., Clarke, C.A., Reynolds, P., Ross, R.K., West, D.W., Anton-Culver, H., Ziogas, A., & Bernstein, L. (2007) Physical Activity and Colon Cancer Risk Among Women

*Cancer*, Vol.46, No.2, pp. 172-178,

No.11 Suppl, pp. 2576S-2579S,

Vol.130, No.1-2, pp. 191-196,

No.15, pp. 2590-2597,

pp. 199-209,

1000-1004,

*J Cancer Prev*, Vol.7, No.1, pp. 11-21,

Muscle. *J Appl Physiol*, Vol.103, No.1, pp. 388-395,

No.4, pp. 257-262,

1569,

and High-Fat Diet on Arachidonic Acid Metabolism and Aberrant Crypt Foci Formation in an Azoxymethane-Induced Colon Carcinogenesis Mouse Model. *Nutr* 

Exercise Inhibits Intestinal Tumorigenesis in Apc(Min/+) Mice and Azoxymethane/Dextran Sulfate Sodium-Treated Mice. *BMC Cancer*, Vol.8, pp. 316-

M., Faivre, J., Hagenmuller, F., Lamy, V., Manger, F., Mills, H.T., Neumann, C., Nowak, A., Pehrsson, A., Smits, S., & Spencer, K. (2004) Public Awareness of Risk Factors and Screening for Colorectal Cancer in Europe. *Eur J Cancer Prev*, Vol.13,


Flood, D.M., Weiss, N.S., Cook, L.S., Emerson, J.C., Schwartz, S.M., & Potter, J.D. (2000)

Friedenreich, C., Norat, T., Steindorf, K., Boutron-Ruault, M.C., Pischon, T., Mazuir, M.,

Friedenreich, C. & Orenstein, M.R. (2004) Review of Physical Activity and the IGF Family.

Friedenreich, C.M. & Orenstein, M.R. (2002) Physical Activity and Cancer Prevention:

Fujise, T., Iwakiri, R., Kakimoto, T., Shiraishi, R., Sakata, Y., Wu, B., Tsunada, S., Ootani, A.,

Fuku, N., Ochiai, M., Terada, S., Fujimoto, E., Nakagama, H., & Tabata, I. (2007) Effect of

Giovannucci, E. (2001) Insulin, Insulin-Like Growth Factors and Colon Cancer: a Review of

Hoffman-Goetz, L. (2003) Physical Activity and Cancer Prevention: Animal-Tumor Models.

Hong, M.Y., Lupton, J.R., Morris, J.S., Wang, N., Carroll, R.J., Davidson, L.A., Elder, R.H., &

Howard, R.A., Freedman, D.M., Park, Y., Hollenbeck, A., Schatzkin, A., & Leitzmann, M.F.

Jemal, A., Bray, F., Center, M.M., Ferlay, J., Ward, E., & Forman, D. (2011) Global Cancer

Johnson, B.L., Trentham-Dietz, A., Koltyn, K.F., & Colbert, L.H. (2009) Physical Activity and

Johnson, I.T. & Lund, E.K. (2007) Review Article: Nutrition, Obesity and Colorectal Cancer.

the Evidence. *J Nutr*, Vol.131, No.11 Suppl, pp. 3109S-3120S,

*Cancer Epidemiol Biomarkers Prev*, Vol.9, No.8, pp. 819-826,

*Med Sci Sports Exerc*, Vol.35, No.11, pp. 1828-1833,

Statistics. *CA Cancer J Clin*, Vol.61, No.2, pp. 69-90,

*Aliment Pharmacol Ther*, Vol.26, No.2, pp. 161-181,

Etiologic Evidence and Biological Mechanisms. *J Nutr*, Vol.132, No.11 Suppl, pp.

& Fujimoto, K. (2007) Long-Term Feeding of Various Fat Diets Modulates Azoxymethane-Induced Colon Carcinogenesis Through Wnt/Beta-Catenin Signaling in Rats. *Am J Physiol Gastrointest Liver Physiol*, Vol.292, No.4, pp. G1150-

Running Training on DMH-Induced Aberrant Crypt Foci in Rat Colon. *Med Sci* 

Chapkin, R.S. (2000) Dietary Fish Oil Reduces O6-Methylguanine DNA Adduct Levels in Rat Colon in Part by Increasing Apoptosis During Tumor Initiation.

(2008) Physical Activity, Sedentary Behavior, and the Risk of Colon and Rectal Cancer in the NIH-AARP Diet and Health Study. *Cancer Causes Control*, Vol.19,

Function in Older, Long-Term Colorectal Cancer Survivors. *Cancer Causes Control*,

Descendants. *Cancer Causes Control*, Vol.11, No.5, pp. 403-411,

*Journal of Physical Activity and Health*, Vol.1, No.4, pp. 291-320,

No.12, pp. 2398-2407,

*Sports Exerc*, Vol.39, No.1, pp. 70-74,

3456S-3464S,

No.9, pp. 939-953,

Vol.20, No.5, pp. 775-784,

G1156,

Colorectal Cancer Incidence in Asian Migrants to the United States and Their

Clavel-Chapelon, F., Linseisen, J., Boeing, H., Bergman, M., Johnsen, N.F., Tjonneland, A., Overvad, K., Mendez, M., Quiros, J.R., Martinez, C., Dorronsoro, M., Navarro, C., Gurrea, A.B., Bingham, S., Khaw, K.T., Allen, N., Key, T., Trichopoulou, A., Trichopoulos, D., Orfanou, N., Krogh, V., Palli, D., Tumino, R., Panico, S., Vineis, P., Bueno-de-Mesquita, H.B., Peeters, P.H., Monninkhof, E., Berglund, G., Manjer, J., Ferrari, P., Slimani, N., Kaaks, R., & Riboli, E. (2006) Physical Activity and Risk of Colon and Rectal Cancers: the European Prospective Investigation into Cancer and Nutrition. *Cancer Epidemiol Biomarkers Prev*, Vol.15,


Physical Activity, Dietary Fat and Colorectal Cancer 121

Rao, C.V., Hirose, Y., Indranie, C., & Reddy, B.S. (2001) Modulation of Experimental Colon

Rao, C.V., Simi, B., Wynn, T.T., Garr, K., & Reddy, B.S. (1996) Modulating Effect of Amount

Reddy, B.S. (2000) The Fourth DeWitt S. Goodman Lecture. Novel Approaches to the

Samad, A.K., Taylor, R.S., Marshall, T., & Chapman, M.A. (2005) A Meta-Analysis of the

Sandhu, M.S., Dunger, D.B., & Giovannucci, E.L. (2002) Insulin, Insulin-Like Growth Factor-

Scheele, C., Nielsen, S., & Pedersen, B.K. (2009) ROS and Myokines Promote Muscle Adaptation to Exercise. *Trends Endocrinol Metab*, Vol.20, No.3, pp. 95-99, Simopoulos, A.P. (2002a) Omega-3 Fatty Acids in Inflammation and Autoimmune Diseases.

Simopoulos, A.P. (2002b) The Importance of the Ratio of Omega-6/Omega-3 Essential Fatty

Slattery, M.L. & Potter, J.D. (2002) Physical Activity and Colon Cancer: Confounding or

Souglakos, J. (2007) Genetic Alterations in Sporadic and Hereditary Colorectal Cancer: Implementations for Screening and Follow-Up. *Dig Dis*, Vol.25, No.1, pp. 9-19, Steinberg, G.R. (2007) Inflammation in Obesity Is the Common Link Between Defects in Fatty Acid Metabolism and Insulin Resistance. *Cell Cycle*, Vol.6, No.8, pp. 888-894, Tammariello, A.E. & Milner, J.A. (2010) Mouse Models for Unraveling the Importance of Diet in Colon Cancer Prevention. *J Nutr Biochem*, Vol.21, No.2, pp. 77-88, Thorling, E.B., Jacobsen, N.O., & Overvad, K. (1994) The Effect of Treadmill Exercise on

Thune, I. & Furberg, A.S. (2001) Physical Activity and Cancer Risk: Dose-Response and

Valdes-Ramos, R. & Benitez-Arciniega, A.D. (2007) Nutrition and Immunity in Cancer. *Br J* 

Valko, M., Leibfritz, D., Moncol, J., Cronin, M.T., Mazur, M., & Telser, J. (2007) Free Radicals

Valko, M., Rhodes, C.J., Moncol, J., Izakovic, M., & Mazur, M. (2006) Free Radicals, Metals

Azoxymethane-Induced Intestinal Neoplasia in the Male Fischer Rat on Two

Cancer, All Sites and Site-Specific. *Med Sci Sports Exerc*, Vol.33, No.6 Suppl, pp.

and Antioxidants in Normal Physiological Functions and Human Disease. *Int J* 

and Antioxidants in Oxidative Stress-Induced Cancer. *Chem Biol Interact*, Vol.160,

F344 Rats. *Cancer Res*, Vol.56, No.3, pp. 532-537,

*Natl Cancer Inst*, Vol.94, No.13, pp. 972-980,

*J Am Coll Nutr*, Vol.21, No.6, pp. 495-505,

*Nutr*, Vol.98 Suppl 1, pp. S127-S132,

*Biochem Cell Biol*, Vol.39, No.1, pp. 44-84,

Acids. *Biomed Pharmacother*, Vol.56, No.8, pp. 365-379,

Interaction? *Med Sci Sports Exerc*, Vol.34, No.6, pp. 913-919,

Different High-Fat Diets. *Nutr Cancer*, Vol.22, No.1, pp. 31-41,

*Cancer Epidemiol Biomarkers Prev*, Vol.9, No.3, pp. 239-247,

No.5, pp. 1927-1933,

*Dis*, Vol.7, No.3, pp. 204-213,

S530-S550,

No.1, pp. 1-40,

Tumorigenesis by Types and Amounts of Dietary Fatty Acids. *Cancer Res*, Vol.61,

and Types of Dietary Fat on Colonic Mucosal Phospholipase A2, Phosphatidylinositol-Specific Phospholipase C Activities, and Cyclooxygenase Metabolite Formation During Different Stages of Colon Tumor Promotion in Male

Prevention of Colon Cancer by Nutritional Manipulation and Chemoprevention.

Association of Physical Activity With Reduced Risk of Colorectal Cancer. *Colorectal* 

I (IGF-I), IGF Binding Proteins, Their Biologic Interactions, and Colorectal Cancer. *J* 

in the California Teachers Study. *Cancer Epidemiol Biomarkers Prev*, Vol.16, No.3, pp. 517-525,


Mathur, N. & Pedersen, B.K. (2008) Exercise As a Mean to Control Low-Grade Systemic

Matsumoto, M., Inoue, R., Tsukahara, T., Ushida, K., Chiji, H., Matsubara, N., & Hara, H.

McTiernan, A., Yasui, Y., Sorensen, B., Irwin, M.L., Morgan, A., Rudolph, R.E., Surawicz, C.,

Mehl, K.A., Davis, J.M., Clements, J.M., Berger, F.G., Pena, M.M., & Carson, J.A. (2005)

Murthy, N.S., Mukherjee, S., Ray, G., & Ray, A. (2009) Dietary Factors and Cancer

Nielsen, S. & Pedersen, B.K. (2008) Skeletal Muscle As an Immunogenic Organ. *Curr Opin* 

Nilsen, T.I., Romundstad, P.R., Petersen, H., Gunnell, D., & Vatten, L.J. (2008) Recreational

Health Study). *Cancer Epidemiol Biomarkers Prev*, Vol.17, No.1, pp. 183-188, Orlando, F.A., Tan, D., Baltodano, J.D., Khoury, T., Gibbs, J.F., Hassid, V.J., Ahmed, B.H., &

Pedersen, B.K. & Hoffman-Goetz, L. (2000) Exercise and the Immune System: Regulation, Integration, and Adaptation. *Physiol Rev*, Vol.80, No.3, pp. 1055-1081, Pence, B.C. & Buddingh, F. (1988) Inhibition of Dietary Fat-Promoted Colon Carcinogenesis

Perse, M. & Cerar, A. (2011) Morphological and Molecular Alterations in 1,2

Perse, M., Injac, R., Strukelj, B., & Cerar, A. (2009) Effects of High-Fat Mixed-Lipid Diet and

Perse, M. (2010) *Effects of Physical Activity and High-fat-mixed-Lipid Diet on the Development of* 

Petersen, A.M. & Pedersen, B.K. (2005) The Anti-Inflammatory Effect of Exercise. *J Appl* 

Petersen, A.M. & Pedersen, B.K. (2006) The Role of IL-6 in Mediating the Anti-Inflammatory

Pisani, P. (2008) Hyper-Insulinaemia and Cancer, Meta-Analyses of Epidemiological

Effects of Exercise. *J Physiol Pharmacol*, Vol.57 Suppl 10, pp. 43-51,

Studies. *Arch Physiol Biochem*, Vol.114, No.1, pp. 63-70,

Colon Carcinoma. *Pharmacol Rep*, Vol.61, No.5, pp. 909-916,

ApcMin/+ Mice. *J Appl Physiol*, Vol.98, No.6, pp. 2219-2225,

Progression. *J Surg Oncol*, Vol.98, No.3, pp. 207-213,

Inflammation. *Mediators Inflamm*, Vol.2008, pp. 109502-

517-525,

pp. 572-576,

1588-1597,

187-190,

Vol.55, No.1, pp. 45-54,

*Pharmacol*, Vol.8, No.3, pp. 346-351,

*Biomed Biotechnol*, Vol.2011, pp. 473964-

*Physiol*, Vol.98, No.4, pp. 1154-1162,

in the California Teachers Study. *Cancer Epidemiol Biomarkers Prev*, Vol.16, No.3, pp.

(2008) Voluntary Running Exercise Alters Microbiota Composition and Increases N-Butyrate Concentration in the Rat Cecum. *Biosci Biotechnol Biochem*, Vol.72, No.2,

Lampe, J.W., Ayub, K., Potter, J.D., & Lampe, P.D. (2006) Effect of a 12-Month Exercise Intervention on Patterns of Cellular Proliferation in Colonic Crypts: a Randomized Controlled Trial. *Cancer Epidemiol Biomarkers Prev*, Vol.15, No.9, pp.

Decreased Intestinal Polyp Multiplicity Is Related to Exercise Mode and Gender in

Chemoprevention: an Overview of Obesity-Related Malignancies. *J Postgrad Med*,

Physical Activity and Cancer Risk in Subsites of the Colon (the Nord-Trondelag

Alrawi, S.J. (2008) Aberrant Crypt Foci As Precursors in Colorectal Cancer

in Rats by Supplemental Calcium or Vitamin D3. *Carcinogenesis*, Vol.9, No.1, pp.

Dimethylhydrazine and Azoxymethane Induced Colon Carcinogenesis in Rats. *J* 

Exercise on the Antioxidant System in Skeletal and Cardiac Muscles of Rats With

*Chemically Induced Colorectal Tumors in Wistar Rat*, University of Ljubljana: Ljubljana


**7** 

**Dietary Anthocyanins: Impact on Colorectal** 

Colorectal cancer is the third most common malignancy in males and the second most common in females, with significant variations in the worldwide distribution, and remains among four leading causes of cancer deaths overall, shows global cancer statistics. The highest incident rates are found in economically developed countries, whereas the lowest rates are noted in Africa and South-Central (Jemal et al., 2011). However, striking increase in colorectal cancer incident trends is observed in areas historically at low risk, such as Spain and some Eastern European (the Czech Republic and Slovakia) and Eastern Asian countries (Japan). On the other hand, generally high incident rates over the past several decades are going down in the Unites States (Center et al., 2009). These recent "perturbations" in colorectal cancer trends probably result from a combination of risk factors, including obesity, sedentary lifestyle, increased prevalence of smoking, excessive alcohol consumption and "westernization" in dietary habits - a diet rich in red and processed meat and low intake of fruits and vegetables (Center et al., 2009; Chao et al., 2005; Jemal et al., 2011). Decreasing incident and mortality rates are mainly associated with colorectal cancer screening and

Prognosis of these patients depends on the stage of the cancer at diagnosis. As the AJCC (American Joint Committee on Cancer) stage increases from stage I to stage IV, the 5-year overall survival rates decrease dramatically, reaching 90% if the disease is detected early when still localized, though just 39% of colorectal cancers are found at this stage. Almost 25% of patients have a metastatic disease at diagnosis, with a 5-year survival of less than 10% (Goldberg et al., 2007). The primary treatment for colorectal cancer is surgical resection. More than two-thirds of patients undergo radical surgery, but 30-50% of patients who present with stage II or III tumors ultimately experience disease recurrence and distant metastases (Rodriguez-Moranta et al., 2006). Although a broader base of treatment options for metastatic colorectal cancer (mCRC) has evolved in recent years, 50 - 70% of mCRC

**1. Introduction** 

improved treatment.

**Cancer and Mechanisms of Action** 

Sabina Passamonti1 and Jovana Cvorovic4

*3Federal University of Technology, Ondo State,* 

*1University of Trieste, <sup>2</sup>University of Ljubljana,*

*4University of Trieste,* 

*1,4Italy 2Slovenia 3Nigeria* 

Federica Tramer1, Spela Moze2, Ayokunle O. Ademosun3,


### **Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action**

Federica Tramer1, Spela Moze2, Ayokunle O. Ademosun3,

Sabina Passamonti1 and Jovana Cvorovic4 *1University of Trieste, <sup>2</sup>University of Ljubljana, 3Federal University of Technology, Ondo State, 4University of Trieste, 1,4Italy 2Slovenia 3Nigeria* 

#### **1. Introduction**

122 Colorectal Cancer – From Prevention to Patient Care

Wolin, K.Y., Yan, Y., Colditz, G.A., & Lee, I.M. (2009) Physical Activity and Colon Cancer Prevention: a Meta-Analysis. *Br J Cancer*, Vol.100, No.4, pp. 611-616, World Cancer Research Fund/American Institute for Cancer Research. (2007) *Food, nutrition,* 

Wu, B., Iwakiri, R., Ootani, A., Tsunada, S., Fujise, T., Sakata, Y., Sakata, H., Toda, S., &

American Institute of Cancer Research,

*(Maywood )*, Vol.229, No.10, pp. 1017-1025,

*physical activity, and prevention of cancer: a global perspective*. Washington, DC:

Fujimoto, K. (2004) Dietary Corn Oil Promotes Colon Cancer by Inhibiting Mitochondria-Dependent Apoptosis in Azoxymethane-Treated Rats. *Exp Biol Med* 

> Colorectal cancer is the third most common malignancy in males and the second most common in females, with significant variations in the worldwide distribution, and remains among four leading causes of cancer deaths overall, shows global cancer statistics. The highest incident rates are found in economically developed countries, whereas the lowest rates are noted in Africa and South-Central (Jemal et al., 2011). However, striking increase in colorectal cancer incident trends is observed in areas historically at low risk, such as Spain and some Eastern European (the Czech Republic and Slovakia) and Eastern Asian countries (Japan). On the other hand, generally high incident rates over the past several decades are going down in the Unites States (Center et al., 2009). These recent "perturbations" in colorectal cancer trends probably result from a combination of risk factors, including obesity, sedentary lifestyle, increased prevalence of smoking, excessive alcohol consumption and "westernization" in dietary habits - a diet rich in red and processed meat and low intake of fruits and vegetables (Center et al., 2009; Chao et al., 2005; Jemal et al., 2011). Decreasing incident and mortality rates are mainly associated with colorectal cancer screening and improved treatment.

> Prognosis of these patients depends on the stage of the cancer at diagnosis. As the AJCC (American Joint Committee on Cancer) stage increases from stage I to stage IV, the 5-year overall survival rates decrease dramatically, reaching 90% if the disease is detected early when still localized, though just 39% of colorectal cancers are found at this stage. Almost 25% of patients have a metastatic disease at diagnosis, with a 5-year survival of less than 10% (Goldberg et al., 2007). The primary treatment for colorectal cancer is surgical resection. More than two-thirds of patients undergo radical surgery, but 30-50% of patients who present with stage II or III tumors ultimately experience disease recurrence and distant metastases (Rodriguez-Moranta et al., 2006). Although a broader base of treatment options for metastatic colorectal cancer (mCRC) has evolved in recent years, 50 - 70% of mCRC

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 125

never and former smokers and a statistically non signicant positive association for current

Key, on the other hand, by summarizing data recorded from large prospective studies or pooled analyses, recommended a diet which contains moderate amounts of fruit and vegetables in order to prevent deficiencies of any nutrients. Nevertheless, the available data suggest that, at least in relatively well-nourished populations, general increases in fruit and

Due, at least in part, to their anti-oxidant and anti-inflammatory activities, epidemiologic studies suggest that the consumption of anthocyanins lowers the risk of cardiovascular disease, diabetes, arthritis and cancer (Prior and Wu, 2006). Their activities are associated to their action at different molecular level: direct ability to scavenge reactive oxygen species (Wang and Jiao, 2000) or to induce phase II antioxidant and detoxifying enzymes (Shih et al.,

Cancer cells differ from normal cells due to the following properties: unlimited replication potential, the absence of apoptosis, the absence of telomere shortening, angiogenesis and metastasis. Dietary compounds have been shown to affect molecular events involved in the initiation, promotion and progression of cancer, thereby inhibiting carcinogenesis. Furthermore, their inhibitory activity may ultimately suppress the final steps of carcinogenesis as well, namely angiogenesis and metastasis. The relationship between the frequency of consumption of vegetables and fruit and cancer risk is linked to a class of

The unlimited replication potential of cancer cells is a result of the inactivation of tumour suppressor genes. For instance, mutated p21 gene products are no longer able to bind to cyclin, thus cyclin-dependent kinase remains active and cell division becomes uncontrolled. Targeting these protein kinases using natural products has been seen as a promising approach in solving the cancer menace (Omura et al., 1995; Yasuzawa et al., 1986). Although research on protein kinases is still at an early stage, there is enough evidence that dietary compounds have useful potency and specificity against protein kinases of medicinal

Resveratol has been shown by numerous reports to inhibit cell proliferation through the inhibition of cell-cycle progression at different stages (Aggarwal and Shishodia, 2006; Liang et al., 2003; Takagaki et al., 2005). Down-regulation of the cyclin D1/Cdk4 complex by resveratrol has been reported in colon cancer cell lines (Wolter et al., 2001) as well as resveratrol-induced G2 arrest through the inhibition of Cdk7 and Cdc2 kinases in colon carcinoma HT-29cells (Liang et al., 2003). Furthermore, an anthocyanin-rich extract caused cell cycle arrest and increased expression of the p27kip1 and p21WAF1/Cip1 genes and a

Abnormalities in the ubiquitin-proteasome system have been implicated in many protein degradation disorders, including several types of cancer. This has made the proteasome an important target for anti-cancer drug discovery. Proteasome inhibitors can be categorized as synthetic and natural, where natural molecules are often more specific and potent than synthetic ones (D'Alessandro et al., 2009). Chen and colleagues showed that dietary flavonoids apigenin and quercetin inhibit proteasome, and this inhibition may contribute to their cancer-preventative effects (Chen et al., 2005). Furthermore, Kazi and colleagues also

vegetable intake would not have much effect on cancer rates (Key, 2011).

smokers was observed (van Duijnhoven et al., 2009).

**2.2 Dietary compounds and tumor progression** 

phytochemicals which flavonoids belong to.

60% cancer cell growth inhibition (Malik et al., 2003).

2005; Shih et al., 2007).

importance.

patients still cannot be subjected to radical resection of metastases and are candidates for palliative therapy only (Fornaro et al., 2010).

The drugs commonly used to treat mCRC are fluoropyrimidines (fluorouracil and capecitabine), irinotecan – a semisynthetic derivative of the natural alkaloid camptothecin, and oxaliplatin – a diaminocyclohexane platinum compound. More recently, two monoclonal antibodies have been approved for the treatment of advanced stages of colorectal cancer. Bevacizumab, a humanized monoclonal antibody against vascular endothelial growth factor (VEGF), is broadly used in combination with fluoropyrimidinebased chemotherapy. Cetuximab, a chimeric monoclonal antibody against the epidermal growth factor receptor (EGFR), is used as monotherapy or together with irinotecan in irinotecan-resistant patients (Hess et al., 2010; Tol and Punt, 2010; Van Cutsem et al., 2009). These chemotherapy agents have significantly improved the prognoses and median overall survival. However, chemotherapy drug resistance occurs in nearly all patients and remains the most frequent cause of treatment failure (Candeil et al., 2004; Dallas et al., 2009), calling for finding novel agents capable of killing drug-resistant colorectal cancer cells.

#### **2. Dietary compounds and cancer**

#### **2.1 Cancer prevention by diet**

The possibility that fruit and vegetables might help to reduce the risk for various types of cancer raised great interest already in the 1970s. The first studies conducted to assess differences in cancer rates and diet between countries suggested that various dietary factors might have important effects on cancer risk (Armstrong and Doll, 1975; Bjelke, 1975).

In 1992, an epidemiological research with 156 studies on connection between the consumption of fruit and vegetables and cancer concluded that persons with a low fruit and vegetable intake face up to twice the risk of developing cancer compared to those with a high intake (Block et al., 1992). Several years later, a joint report by the World Cancer Research Fund together with the American Institute of Cancer Research found 'convincing' evidence that a high fruit and vegetable intake would reduce cancer of the colon and rectum (AIRC, 1997).

Unfortunately, 10 years later, an updated report released by the same organization and based on large prospective studies instead on case-control studies, downgraded these previous conclusions. The evidence that high intakes of fruit and/or vegetables decrease the risk for cancers of the mouth and pharynx, esophagus, stomach, colorectum and lung were judged 'probable' or 'limited- suggestive', so researchers did not confirm the earlier results (AIRC, 2007).

In a randomized dietary intervention trial, called The Polyp Prevention Trial, it was examined the effectiveness of a low-fat, high-fiber, high-fruit, and high-vegetable diet on adenoma recurrence. This study was the first to examine the association between flavonoid intake and colorectal adenoma recurrence. It was found that total flavonoid intake was not associated with colorectal adenoma recurrence, but they also detected during the trial a reduced risk of advanced adenoma recurrence with greater flavonol consumption (Bobe et al., 2008).

The European Prospective Investigation into Cancer and Nutrition in 2009 suggested that a high consumption of fruit and vegetables is associated with a reduced risk of CRC, especially of colon cancer but differs according to smoking status. An inverse association for never and former smokers and a statistically non signicant positive association for current smokers was observed (van Duijnhoven et al., 2009).

Key, on the other hand, by summarizing data recorded from large prospective studies or pooled analyses, recommended a diet which contains moderate amounts of fruit and vegetables in order to prevent deficiencies of any nutrients. Nevertheless, the available data suggest that, at least in relatively well-nourished populations, general increases in fruit and vegetable intake would not have much effect on cancer rates (Key, 2011).

Due, at least in part, to their anti-oxidant and anti-inflammatory activities, epidemiologic studies suggest that the consumption of anthocyanins lowers the risk of cardiovascular disease, diabetes, arthritis and cancer (Prior and Wu, 2006). Their activities are associated to their action at different molecular level: direct ability to scavenge reactive oxygen species (Wang and Jiao, 2000) or to induce phase II antioxidant and detoxifying enzymes (Shih et al., 2005; Shih et al., 2007).

#### **2.2 Dietary compounds and tumor progression**

124 Colorectal Cancer – From Prevention to Patient Care

patients still cannot be subjected to radical resection of metastases and are candidates for

The drugs commonly used to treat mCRC are fluoropyrimidines (fluorouracil and capecitabine), irinotecan – a semisynthetic derivative of the natural alkaloid camptothecin, and oxaliplatin – a diaminocyclohexane platinum compound. More recently, two monoclonal antibodies have been approved for the treatment of advanced stages of colorectal cancer. Bevacizumab, a humanized monoclonal antibody against vascular endothelial growth factor (VEGF), is broadly used in combination with fluoropyrimidinebased chemotherapy. Cetuximab, a chimeric monoclonal antibody against the epidermal growth factor receptor (EGFR), is used as monotherapy or together with irinotecan in irinotecan-resistant patients (Hess et al., 2010; Tol and Punt, 2010; Van Cutsem et al., 2009). These chemotherapy agents have significantly improved the prognoses and median overall survival. However, chemotherapy drug resistance occurs in nearly all patients and remains the most frequent cause of treatment failure (Candeil et al., 2004; Dallas et al., 2009), calling

for finding novel agents capable of killing drug-resistant colorectal cancer cells.

The possibility that fruit and vegetables might help to reduce the risk for various types of cancer raised great interest already in the 1970s. The first studies conducted to assess differences in cancer rates and diet between countries suggested that various dietary factors

In 1992, an epidemiological research with 156 studies on connection between the consumption of fruit and vegetables and cancer concluded that persons with a low fruit and vegetable intake face up to twice the risk of developing cancer compared to those with a high intake (Block et al., 1992). Several years later, a joint report by the World Cancer Research Fund together with the American Institute of Cancer Research found 'convincing' evidence that a high fruit and vegetable intake would reduce cancer of the colon and rectum

Unfortunately, 10 years later, an updated report released by the same organization and based on large prospective studies instead on case-control studies, downgraded these previous conclusions. The evidence that high intakes of fruit and/or vegetables decrease the risk for cancers of the mouth and pharynx, esophagus, stomach, colorectum and lung were judged 'probable' or 'limited- suggestive', so researchers did not confirm the earlier results

In a randomized dietary intervention trial, called The Polyp Prevention Trial, it was examined the effectiveness of a low-fat, high-fiber, high-fruit, and high-vegetable diet on adenoma recurrence. This study was the first to examine the association between flavonoid intake and colorectal adenoma recurrence. It was found that total flavonoid intake was not associated with colorectal adenoma recurrence, but they also detected during the trial a reduced risk of advanced adenoma recurrence with greater flavonol consumption (Bobe et

The European Prospective Investigation into Cancer and Nutrition in 2009 suggested that a high consumption of fruit and vegetables is associated with a reduced risk of CRC, especially of colon cancer but differs according to smoking status. An inverse association for

might have important effects on cancer risk (Armstrong and Doll, 1975; Bjelke, 1975).

palliative therapy only (Fornaro et al., 2010).

**2. Dietary compounds and cancer** 

**2.1 Cancer prevention by diet** 

(AIRC, 1997).

(AIRC, 2007).

al., 2008).

Cancer cells differ from normal cells due to the following properties: unlimited replication potential, the absence of apoptosis, the absence of telomere shortening, angiogenesis and metastasis. Dietary compounds have been shown to affect molecular events involved in the initiation, promotion and progression of cancer, thereby inhibiting carcinogenesis. Furthermore, their inhibitory activity may ultimately suppress the final steps of carcinogenesis as well, namely angiogenesis and metastasis. The relationship between the frequency of consumption of vegetables and fruit and cancer risk is linked to a class of phytochemicals which flavonoids belong to.

The unlimited replication potential of cancer cells is a result of the inactivation of tumour suppressor genes. For instance, mutated p21 gene products are no longer able to bind to cyclin, thus cyclin-dependent kinase remains active and cell division becomes uncontrolled. Targeting these protein kinases using natural products has been seen as a promising approach in solving the cancer menace (Omura et al., 1995; Yasuzawa et al., 1986). Although research on protein kinases is still at an early stage, there is enough evidence that dietary compounds have useful potency and specificity against protein kinases of medicinal importance.

Resveratol has been shown by numerous reports to inhibit cell proliferation through the inhibition of cell-cycle progression at different stages (Aggarwal and Shishodia, 2006; Liang et al., 2003; Takagaki et al., 2005). Down-regulation of the cyclin D1/Cdk4 complex by resveratrol has been reported in colon cancer cell lines (Wolter et al., 2001) as well as resveratrol-induced G2 arrest through the inhibition of Cdk7 and Cdc2 kinases in colon carcinoma HT-29cells (Liang et al., 2003). Furthermore, an anthocyanin-rich extract caused cell cycle arrest and increased expression of the p27kip1 and p21WAF1/Cip1 genes and a 60% cancer cell growth inhibition (Malik et al., 2003).

Abnormalities in the ubiquitin-proteasome system have been implicated in many protein degradation disorders, including several types of cancer. This has made the proteasome an important target for anti-cancer drug discovery. Proteasome inhibitors can be categorized as synthetic and natural, where natural molecules are often more specific and potent than synthetic ones (D'Alessandro et al., 2009). Chen and colleagues showed that dietary flavonoids apigenin and quercetin inhibit proteasome, and this inhibition may contribute to their cancer-preventative effects (Chen et al., 2005). Furthermore, Kazi and colleagues also

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 127

acid and hydroxycinnamate esters from cranberry fruit strongly inhibited expression of MMP-2 and MMP-9 activities at micromolar concentrations in fibrosarcoma cells (Cha et al., 1996). Anthocyanins from mulberry fruits and highbush blueberry (*V. angustifolium*) inhibited MMP-2 and MMP-9 activities (Huang et al., 2008; Matchett et al., 2005; Matchett et al., 2006). Delphinidin can inhibit invasion of human fibrosarcoma cells through downregulation of MMP-2 and MMP-9, expression (Nagase et al., 1998). More recently, it has been demonstrated that black rice anthocyanins, cyanidin 3-glucoside and peonidin 3 glucoside, significantly reduce the expression of MMP-9 in diverse types of cancer cells (Chen et al., 2006). Furthermore, it was also demonstrated that the activities of MMP-2 and - 9 were dose-dependently suppressed by anthocyanin treatment on HT-29 human colon cancer cells (Yun et al., 2010) and in HCT-116 human colon cancer cells through the

activation of 38-MAPK and suppression of the PI3K/Akt pathway (Shin et al., 2011).

E, flavonoids, and dietary fiber (Aggarwal and Shishodia, 2006).

constituents of fruits and vegetables (Wang and Stoner, 2008).

apoptosis, and angiogenesis (Yao et al., 2011).

**2.3.1 Anthocyanins: Chemistry** 

Bioactive compounds that impart protective properties to plants against various pathological conditions are grouped under the name of phytochemicals. Active components of dietary phytochemicals which have been identified to protect against cancer include curcumin, resveratrol, diallyl sulfide, S-allyl cysteine, allicin, lycopene, capsaicin, diosgenin, 6-gingerol, ellagic acid, ursolic acid, silymarin, anethol, eugenol, isoeugenol, dithiolthiones, isothiocyanates, indole-3-carbinol, protease inhibitors, saponins, phytosterols, inositol hexaphosphate, Vitamin C, D-limonene, lutein, folic acid, beta carotene, selenium, Vitamin

Flavonoids represent one of the largest groups of secondary metabolites whose name refers to a class of more than 6500 molecules based upon a 15-carbon skeleton (Harborne and Williams, 2000; Ververidis et al., 2007). They are divided into six major classes: flavanols, flavonones, flavones, isoflavones, flavonols and anthocyanins. Flavonoids are not synthesized in animal cells, thus their detection in animal tissues is indicative of plant ingestion (Mennen et al., 2008). Dietary flavonoids play an important role in cancer prevention and inhibition influencing various cellular processes, such as reactive oxygen species production and cell signal transduction pathways related to cellular proliferation,

Flavonoids compounds are the most studied anticarcinogens among phytochemicals. Anthocyanins, a particular class of this group of molecules, are the most abundant flavonoid

Anthocyanins (Greek anthos = flower and kyanos = blue) are water-soluble pigments in fruits and vegetables, responsible for red, blue and purple colors. In plant cells, they are present in vacuoles in the form of various sized granules. Their basic anthocyanidin aglycone structures consist of an aromatic ring A bonded to a heterocyclic ring C that contains oxygen, which is also bound by carbon-carbon bond to a third aromatic ring B (Figure 1). Anthocyanins normally occur in nature in glycoside forms. The sugar moiety is mainly attached to the C ring (in the 3-position) or to the A ring (in the 5, 7-position). Glucose, galactose, arabinose, rhamnose and xylose are the most common sugars bonded to the anthocyanidins. These glycosylated forms are known as anthocyanins. More than 500 different anthocyanins have been found, among which the most common is cyanidin

**2.3 Flavonoids** 

showed that the tumor cell apoptosis-inducing ability of genistein (a soy flavonoid) is associated with its inhibition of proteosome activity (Kazi et al., 2003).

Apoptosis is triggered when normal cells are worn out. In cancer cells, the telomerase activity allows them to evade apoptosis by stabilizing and elongating telomeres through synthesis of de novo telomeric DNA (Naasani et al., 2003). Telomerase activity has been identified in most human tumors (Kim et al., 1994). A high telomerase activity has been linked to the degree of malignancy and likelihood of tumor progression (Fujiwara et al., 2000; Hiyama et al., 1995). Tea catechins, especially the degradation products of epigallocatechin gallate, epicatechin, quercetin, naringin and naringinin, have been found to inhibit telomerase activity (Naasani et al., 1998).

Angiogenesis, one of the hallmarks of cancer, vital to tumor growth and metastasis, is characterized by growth of new capillaries from preexisting vessels (Folkman, 1995). Cancer cells release vascular epithelial growth factor (VEGF), an angiogenic cytokine which stimulates blood vessel growth. Inhibition of VEGF has therefore become a primary target for anti-angiogenic strategies, and inhibitors directed against either VEGF or its receptor VEGFR-2, have been demonstrated to prevent vascularization and growth of a large number of experimental tumor types (Labrecque et al., 2005; Underiner et al., 2004). Ellagic acid (naturally occurring phenolic constituent in fruits and nuts) has been shown to inhibit VEGF-induced migration of endothelial cells (Labrecque et al., 2005). Green tea catechins inhibit vascular endothelial growth factor receptor phosphorylation (Lamy et al., 2002), and resveratol also inhibits vascular endothelial growth factor (VEGF)-induced angiogenic effects in the human umbilical vein endothelial cells through the abrogation of VEGFmediated tyrosine phosphorylation of vascular endothelial (VE)-cadherin and its complex partner, b-catenin (Aggarwal and Shishodia, 2006; Lin et al., 2003). In addition, the flavonoid luteolin also inhibited both VEGF-induced survival and proliferation of the human umbilical vein endothelial cells (Bagchi et al., 2004). *In vitro* studies have shown that anthocyanin-rich berry extract formula exhibited a potent inhibitory effect on H2O2-induced VEGF expression. Anthocyanins suppress angiogenesis through the inhibition of H2O2- and tumor necrosis factor alpha (TNF-a)-induced VEGF expression, as well as through the inhibition of VEGF and VEGF receptor expression (Bagchi et al., 2004).

Metastasis occurs when cancer cells invade blood and lymphatic vessels and are transported to other cells and tissues in the body. Cancer cells produce proteinase enzymes that allow them to invade blood and lymphatic vessels. The matrix metalloproteinases (MMP) are a group of proteolytic enzymes that degrade the extracellular matrix (ECM) components (Nabeshima et al., 2002). MMP-2 and MMP-9 are two important MMPs in cell invasion as cancerous tissues and tumor cells have shown increased levels and activities of both MMP-2 and MMP-9 (Bernardo and Fridman, 2003).

Proanthocyanidins and flavonoids from cranberry and other *Vaccinium* berries have been shown to inhibit the expression of MMPs involved in remodeling the extracellular matrix (Pupa et al., 2002). Curcumin inhibits MMP-2, which is implicated in the formation of loose and primitive looking meshwork formed by aggressive cancers such as melanoma and prostate cancers (Aggarwal and Shishodia, 2006). Resveratrol has been found to cause a dose-dependent inhibition of PMA (Phorbol Myristate Aacetate)-induced increases in MMP-9 expression and activity and also the suppression of MMP-9 mRNA expression. Furthermore, Rose and colleagues found that phytochemicals from broccoli and rorripa have anti-invasive and anti-metalloproteinase activities (Rose et al., 2005). Purified ursolic acid and hydroxycinnamate esters from cranberry fruit strongly inhibited expression of MMP-2 and MMP-9 activities at micromolar concentrations in fibrosarcoma cells (Cha et al., 1996). Anthocyanins from mulberry fruits and highbush blueberry (*V. angustifolium*) inhibited MMP-2 and MMP-9 activities (Huang et al., 2008; Matchett et al., 2005; Matchett et al., 2006). Delphinidin can inhibit invasion of human fibrosarcoma cells through downregulation of MMP-2 and MMP-9, expression (Nagase et al., 1998). More recently, it has been demonstrated that black rice anthocyanins, cyanidin 3-glucoside and peonidin 3 glucoside, significantly reduce the expression of MMP-9 in diverse types of cancer cells (Chen et al., 2006). Furthermore, it was also demonstrated that the activities of MMP-2 and - 9 were dose-dependently suppressed by anthocyanin treatment on HT-29 human colon cancer cells (Yun et al., 2010) and in HCT-116 human colon cancer cells through the activation of 38-MAPK and suppression of the PI3K/Akt pathway (Shin et al., 2011).

#### **2.3 Flavonoids**

126 Colorectal Cancer – From Prevention to Patient Care

showed that the tumor cell apoptosis-inducing ability of genistein (a soy flavonoid) is

Apoptosis is triggered when normal cells are worn out. In cancer cells, the telomerase activity allows them to evade apoptosis by stabilizing and elongating telomeres through synthesis of de novo telomeric DNA (Naasani et al., 2003). Telomerase activity has been identified in most human tumors (Kim et al., 1994). A high telomerase activity has been linked to the degree of malignancy and likelihood of tumor progression (Fujiwara et al., 2000; Hiyama et al., 1995). Tea catechins, especially the degradation products of epigallocatechin gallate, epicatechin, quercetin, naringin and naringinin, have been found to

Angiogenesis, one of the hallmarks of cancer, vital to tumor growth and metastasis, is characterized by growth of new capillaries from preexisting vessels (Folkman, 1995). Cancer cells release vascular epithelial growth factor (VEGF), an angiogenic cytokine which stimulates blood vessel growth. Inhibition of VEGF has therefore become a primary target for anti-angiogenic strategies, and inhibitors directed against either VEGF or its receptor VEGFR-2, have been demonstrated to prevent vascularization and growth of a large number of experimental tumor types (Labrecque et al., 2005; Underiner et al., 2004). Ellagic acid (naturally occurring phenolic constituent in fruits and nuts) has been shown to inhibit VEGF-induced migration of endothelial cells (Labrecque et al., 2005). Green tea catechins inhibit vascular endothelial growth factor receptor phosphorylation (Lamy et al., 2002), and resveratol also inhibits vascular endothelial growth factor (VEGF)-induced angiogenic effects in the human umbilical vein endothelial cells through the abrogation of VEGFmediated tyrosine phosphorylation of vascular endothelial (VE)-cadherin and its complex partner, b-catenin (Aggarwal and Shishodia, 2006; Lin et al., 2003). In addition, the flavonoid luteolin also inhibited both VEGF-induced survival and proliferation of the human umbilical vein endothelial cells (Bagchi et al., 2004). *In vitro* studies have shown that anthocyanin-rich berry extract formula exhibited a potent inhibitory effect on H2O2-induced VEGF expression. Anthocyanins suppress angiogenesis through the inhibition of H2O2- and tumor necrosis factor alpha (TNF-a)-induced VEGF expression, as well as through the

associated with its inhibition of proteosome activity (Kazi et al., 2003).

inhibition of VEGF and VEGF receptor expression (Bagchi et al., 2004).

and MMP-9 (Bernardo and Fridman, 2003).

Metastasis occurs when cancer cells invade blood and lymphatic vessels and are transported to other cells and tissues in the body. Cancer cells produce proteinase enzymes that allow them to invade blood and lymphatic vessels. The matrix metalloproteinases (MMP) are a group of proteolytic enzymes that degrade the extracellular matrix (ECM) components (Nabeshima et al., 2002). MMP-2 and MMP-9 are two important MMPs in cell invasion as cancerous tissues and tumor cells have shown increased levels and activities of both MMP-2

Proanthocyanidins and flavonoids from cranberry and other *Vaccinium* berries have been shown to inhibit the expression of MMPs involved in remodeling the extracellular matrix (Pupa et al., 2002). Curcumin inhibits MMP-2, which is implicated in the formation of loose and primitive looking meshwork formed by aggressive cancers such as melanoma and prostate cancers (Aggarwal and Shishodia, 2006). Resveratrol has been found to cause a dose-dependent inhibition of PMA (Phorbol Myristate Aacetate)-induced increases in MMP-9 expression and activity and also the suppression of MMP-9 mRNA expression. Furthermore, Rose and colleagues found that phytochemicals from broccoli and rorripa have anti-invasive and anti-metalloproteinase activities (Rose et al., 2005). Purified ursolic

inhibit telomerase activity (Naasani et al., 1998).

Bioactive compounds that impart protective properties to plants against various pathological conditions are grouped under the name of phytochemicals. Active components of dietary phytochemicals which have been identified to protect against cancer include curcumin, resveratrol, diallyl sulfide, S-allyl cysteine, allicin, lycopene, capsaicin, diosgenin, 6-gingerol, ellagic acid, ursolic acid, silymarin, anethol, eugenol, isoeugenol, dithiolthiones, isothiocyanates, indole-3-carbinol, protease inhibitors, saponins, phytosterols, inositol hexaphosphate, Vitamin C, D-limonene, lutein, folic acid, beta carotene, selenium, Vitamin E, flavonoids, and dietary fiber (Aggarwal and Shishodia, 2006).

Flavonoids represent one of the largest groups of secondary metabolites whose name refers to a class of more than 6500 molecules based upon a 15-carbon skeleton (Harborne and Williams, 2000; Ververidis et al., 2007). They are divided into six major classes: flavanols, flavonones, flavones, isoflavones, flavonols and anthocyanins. Flavonoids are not synthesized in animal cells, thus their detection in animal tissues is indicative of plant ingestion (Mennen et al., 2008). Dietary flavonoids play an important role in cancer prevention and inhibition influencing various cellular processes, such as reactive oxygen species production and cell signal transduction pathways related to cellular proliferation, apoptosis, and angiogenesis (Yao et al., 2011).

Flavonoids compounds are the most studied anticarcinogens among phytochemicals. Anthocyanins, a particular class of this group of molecules, are the most abundant flavonoid constituents of fruits and vegetables (Wang and Stoner, 2008).

#### **2.3.1 Anthocyanins: Chemistry**

Anthocyanins (Greek anthos = flower and kyanos = blue) are water-soluble pigments in fruits and vegetables, responsible for red, blue and purple colors. In plant cells, they are present in vacuoles in the form of various sized granules. Their basic anthocyanidin aglycone structures consist of an aromatic ring A bonded to a heterocyclic ring C that contains oxygen, which is also bound by carbon-carbon bond to a third aromatic ring B (Figure 1). Anthocyanins normally occur in nature in glycoside forms. The sugar moiety is mainly attached to the C ring (in the 3-position) or to the A ring (in the 5, 7-position). Glucose, galactose, arabinose, rhamnose and xylose are the most common sugars bonded to the anthocyanidins. These glycosylated forms are known as anthocyanins. More than 500 different anthocyanins have been found, among which the most common is cyanidin

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 129

In the stomach, anthocyanins remain intact due to the low pH that shifts the molecules toward the most stabile flavylium cation. Anthocyanins absorption takes place in the stomach through active transport (including transport carriers such as bilitranslocase (Passamonti et al., 2003b) and sodium dependent glucose transporter (Felgines et al., 2008))

At the intestine neutral pH anthocyanins exist in equilibrium of four molecular forms (flavylium cation, quinoidal base, carbinol pseudobase and chalcone pseudobase) thus they can be easily exposed to degradation (McDougall et al., 2005). First studies showed degradation of anthocyanins from tart cherries to phenolic acids (Seeram et al., 2001). Later, their degradation was demonstrated by two steps. The first step is deglycosylation of anthocyanins to anthocyanidin aglycon while the second step is degradation of the formed

Deglycosylation is the cleavage of the glycosyl moiety from anthocyanins structure to form

These reactions could take place due to intestinal microflora (Aura et al., 2005b; Ávila et al., 2009; Fleschhut et al., 2006), under intestinal conditions at pH 7 (Fleschhut et al., 2006), or spontaneously in the presence of intestinal epithelial cells (Hassimotto et al., 2008; Kay et al.,

Degradation of anthocyanidin aglycon, achieved spontaneously or by microflora (Ávila et al., 2009; Fleschhut et al., 2006; Forester and Waterhouse, 2008), represents the breakdown of its heterocycle and cleavage of the C-ring to form phenolic acid and aldehyde (Keppler and Humpf, 2005). Spontaneous degradation is a consequence of the neutral pH because anthocyanidin aglycones are observed in chalcone form which is rather unstable and can be easily degraded (Fleschhut et al., 2006; Keppler and Humpf, 2005). Data showed that major degradation products of anthocyanidin aglycons degraded to corresponding phenolic acids (Table 2), as well to some other less present products still unidentified (Ávila et al., 2009; Fleschhut et al., 2006; Forester and Waterhouse, 2008). Further phenolic acids can be transformed to the benzoic acids in the presence of intestinal bacteria by cleavage of the

The fastest degraded were anthocyanidin aglycons, much faster than anthocyanin monoglycosides (Keppler and Humpf, 2005). As well anthocyanin degradation by intestinal

Anthocyanins that are not absorbed or degraded in the gastrointestinal tract can be excreted as intact forms. Unchanged anthocyanins were detected in human fecal samples 24 hours after blood orange juice consumption (Vitaglione et al., 2007), as well as in fecal samples collected

microflora was much faster than spontaneous one (Forester and Waterhouse, 2008).

from rats previously fattened by chokeberries, bilberries and grapes (He et al., 2005).

aglycon to phenolic acid and aldehyde (Ávila et al., 2009; Fleschhut et al., 2006).

hydroxyl group in the 4-position (Aura et al., 2005a; Selma et al., 2009).

Cyanidin Protocatechuic acid

Pelargonidin 4-hydroxybenzoic acid

Petunidin 3-*O*-methylgallic acid

Delphinidin Gallic acid

Table 2. Degradation products of anthocyanidin aglycons

Malvidin Syringic acid Peonidin Vanilic acid

Anthocyanidin aglycon Corresponding phenolic acid

and continues in the small intestine (Talavera et al., 2004).

anthocyanidin aglycons.

2009).

3-glucoside. The most common anthocyanidins (anthocyanins aglycones) found in nature are pelargonidin, peonidin, cyanidin, malvidin, petunidin and delphinidin (Figure 1) (Castañeda-Ovando et al., 2009; Manach et al., 2004; Szajdek and Borowska, 2008).


Fig. 1. Chemical structures of anthocyanidins (Prior and Wu, 2006)**.**

#### **2.3.2 Fate of anthocyanins in the gastro-intestinal tract**

The lack of the knowledge of anthocyanin metabolism in the gastrointestinal tract has been studied by many authors (Aura, 2005; Hassimotto et al., 2008; He et al., 2009; McGhie and Walton, 2007; Vitaglione et al., 2007). The fate of anthocyanins in the gastrointestinal tract is summarized in Table 1.


Table 1. Fate of anthocyanins through the gastrointestinal pathway

There are no data of the effect of saliva on anthocyanins but some publications suggest that flavonoid glycosides are hydrolyzed to corresponding aglycons (McGhie and Walton, 2007; Selma et al., 2009).

3-glucoside. The most common anthocyanidins (anthocyanins aglycones) found in nature are pelargonidin, peonidin, cyanidin, malvidin, petunidin and delphinidin (Figure 1)

+

OH

R1

B

R3

R2

(Castañeda-Ovando et al., 2009; Manach et al., 2004; Szajdek and Borowska, 2008).

HO O

OH

Fig. 1. Chemical structures of anthocyanidins (Prior and Wu, 2006)**.**

Small intestine Deglycosylation, degradation, absorption

Table 1. Fate of anthocyanins through the gastrointestinal pathway

Large intestine Deglycosylation, degradation (Talavera et al., 2004)

There are no data of the effect of saliva on anthocyanins but some publications suggest that flavonoid glycosides are hydrolyzed to corresponding aglycons (McGhie and Walton, 2007;

**2.3.2 Fate of anthocyanins in the gastro-intestinal tract** 

Part of gastrointestinal tact Anthocyanin fate Mouth Deglycosylation?

Stomach Chemical stability

summarized in Table 1.

Selma et al., 2009).

A C

Anthocyanidin R1 R2 R3 Pelargonidin H OH H Cyanidin OH OH H Delphinidin OH OH OH Peonidin OCH3 OH H Petunidin OCH3 OH OH Malvidin OCH3 OH OCH3

The lack of the knowledge of anthocyanin metabolism in the gastrointestinal tract has been studied by many authors (Aura, 2005; Hassimotto et al., 2008; He et al., 2009; McGhie and Walton, 2007; Vitaglione et al., 2007). The fate of anthocyanins in the gastrointestinal tract is

(McGhie and Walton, 2007; Selma et al., 2009)

(Felgines et al., 2008; Passamonti et al., 2003b)

(Hassimotto et al., 2008; McDougall et al., 2005) Absorption

In the stomach, anthocyanins remain intact due to the low pH that shifts the molecules toward the most stabile flavylium cation. Anthocyanins absorption takes place in the stomach through active transport (including transport carriers such as bilitranslocase (Passamonti et al., 2003b) and sodium dependent glucose transporter (Felgines et al., 2008)) and continues in the small intestine (Talavera et al., 2004).

At the intestine neutral pH anthocyanins exist in equilibrium of four molecular forms (flavylium cation, quinoidal base, carbinol pseudobase and chalcone pseudobase) thus they can be easily exposed to degradation (McDougall et al., 2005). First studies showed degradation of anthocyanins from tart cherries to phenolic acids (Seeram et al., 2001). Later, their degradation was demonstrated by two steps. The first step is deglycosylation of anthocyanins to anthocyanidin aglycon while the second step is degradation of the formed aglycon to phenolic acid and aldehyde (Ávila et al., 2009; Fleschhut et al., 2006).

Deglycosylation is the cleavage of the glycosyl moiety from anthocyanins structure to form anthocyanidin aglycons.

These reactions could take place due to intestinal microflora (Aura et al., 2005b; Ávila et al., 2009; Fleschhut et al., 2006), under intestinal conditions at pH 7 (Fleschhut et al., 2006), or spontaneously in the presence of intestinal epithelial cells (Hassimotto et al., 2008; Kay et al., 2009).

Degradation of anthocyanidin aglycon, achieved spontaneously or by microflora (Ávila et al., 2009; Fleschhut et al., 2006; Forester and Waterhouse, 2008), represents the breakdown of its heterocycle and cleavage of the C-ring to form phenolic acid and aldehyde (Keppler and Humpf, 2005). Spontaneous degradation is a consequence of the neutral pH because anthocyanidin aglycones are observed in chalcone form which is rather unstable and can be easily degraded (Fleschhut et al., 2006; Keppler and Humpf, 2005). Data showed that major degradation products of anthocyanidin aglycons degraded to corresponding phenolic acids (Table 2), as well to some other less present products still unidentified (Ávila et al., 2009; Fleschhut et al., 2006; Forester and Waterhouse, 2008). Further phenolic acids can be transformed to the benzoic acids in the presence of intestinal bacteria by cleavage of the hydroxyl group in the 4-position (Aura et al., 2005a; Selma et al., 2009).


Table 2. Degradation products of anthocyanidin aglycons

The fastest degraded were anthocyanidin aglycons, much faster than anthocyanin monoglycosides (Keppler and Humpf, 2005). As well anthocyanin degradation by intestinal microflora was much faster than spontaneous one (Forester and Waterhouse, 2008).

Anthocyanins that are not absorbed or degraded in the gastrointestinal tract can be excreted as intact forms. Unchanged anthocyanins were detected in human fecal samples 24 hours after blood orange juice consumption (Vitaglione et al., 2007), as well as in fecal samples collected from rats previously fattened by chokeberries, bilberries and grapes (He et al., 2005).

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 131

demonstrated in the same animal model that anthocyanin-rich tart cherry extract added to the drinking water was associated with fewer and smaller tumors in the cecum, but none of the tested treatments inuenced the number of tumors in the small intestine or the number or burden of tumors in the colon. It was supposed, therefore, that lack of effect of anthocyanins on colonic tumor development may be a consequence of their metabolism by intestinal bacteria or their spontaneous degradation in the cecal and colonic environment (Bobe et al., 2006; Kang et al., 2003). Moreover, it was shown in ApcMin mice that dietary consumption of anthocyanins in the form of either a mixture (Mirtoselect) or as a pure compound (cyanidin-3-glucoside) interferes with small intestinal adenoma development in a dose-dependent fashion. Authors remarked the presence of measurable levels of anthocyanins in the target organ and in the urine, and in concentrations near or below the detection limit in the systemic circulation. Unfortunately, the dietary dose, at which either agent was signicantly efficacious when extrapolated by dose/ surface area comparison, suggested that equivalent for humans can be found in 740 g bilberries, that is a hefty dose. In terms of absolute dose of agent, cyanidin-3-glucoside was less efficacious than the Mirtoselect mixture. Furthermore, authors suggested that different results obtained, in comparison with other studies (Kang et al., 2003) were possibly due in part to the higher dose of anthocyanins employed but also due to the different way of administration since anthocyanins tend to be unstable in aqueous solution at neutral pH (Cooke et al., 2006). Recently, bilberry [(BB),Vaccinium myrtillus], lingonberry (LB, Vaccinium vitis-idaea), and cloudberry (CB, Rubus chamaemorus), rich in anthocyanins, proanthocyanidins and ellagic acid respectively, proved to be chemopreventive as demonstrated by a signicant reduction in the number of intestinal tumors in Min/1 mice. Concerning their different chemical composition, authors suggested that the effects seen, may rather be a result of a mixture of compounds acting in synergy than an effect of a single active substance. Moreover, since the cellular levels of b-catenin are increased at all stages of colon carcinogenesis, it was demonstrated that two of these berries, LB and CB, markedly inhibited the growth of the adenomas and accumulation of nuclear b-catenin and cyclin D1. Unfortunately, also in this study, the amount of berries in the diets was high and could not be easily reached in a

Concerning other tumor models, the incidence, multiplicity and final mass of mammary tumors were significantly reduced in rats that would receive grape juice containing 15 different anthocyanins (Singletary et al., 2003). Cyanidin-3-glucoside reduced the size of lung cancer xenografts and significantly inhibited metastasis in nude mice (Ding et al., 2006). Lyophilized black raspberries prevented the development of NMBA (Nnitrosomethylbenzylamine)-induced esophageal tumors (Stoner et al., 2007), just like anthocyanin-containing pomegranate extract delayed the onset and reduced the incidence of

Epidemiological studies in humans are, however, still scarce and contradictory. Biopsies of tumor and normal-appearing tissues in colon cancer patients consuming black raspberry powder daily during several weeks, showed reduced proliferation and increased apoptosis in cancerous but not in normal tissue. Antiangiogenic effect was also observed in these patients (Wang et al., 2007). A phase I pilot study in colorectal cancer patients demonstrated that treatment with black raspberries caused positive modulation of biomarkers of tumor development, including cell proliferation, apoptosis, angiogenesis and Wnt pathway in both colorectal adenocarcinomas and adjacent normal tissues (Wang et al., 2007). In a clinical pilot study twenty-five colorectal cancer patients, scheduled to undergo resection of primary

DMBA (7,12-dimethylbenzanthracene)-induced skin tumors in CD-1 mice.

human diet (Misikangas et al., 2007).

#### **3. Citotoxicity/apoptosis of anthocyanins on colon cancer cells**

As mentioned above, naturally occurring dietary substances, in particular, flavonoids, have gained increased attention as agents interfering with processes involved in cancer development and progression. Among them, anthocyanins might be of particular interest since their daily intake is remarkably high compared to other flavonoids - it is estimated to vary between 180 and 215 mg (Hou, 2003) whereas the intake of other flavonoids reaches only 20-25 mg/day. Numerous recent studies indicate that anthocyanins are able to inhibit the growth of embryonic fibroblasts and of different cancer cells derived from malignant human tissues, suggesting their possible role as chemopreventive agents. This brings in focus their possible importance for public health as dietary components with preventive impact on cancer as well as effective, cheap and safe anticancer supplements.

#### **3.1 Cytotoxicity in colon and other cancer cell lines**

There are few reports on the inhibitory effects of anthocyanins on colon cancer cell growth. Extracts of grapes, bilberries and chokeberries rich in anthocyanins have been shown to inhibit the growth of human malignant HT-29 colon cancer cells but did not affect the growth of non-malignant colon-derived cells (Zhao et al., 2004). Similar effect was observed in highly and low tumorigenic colon cancer cell lines, LoVo/Adr and LoVo. While delphinidin and cyanidin were cytotoxic and induced apoptosis in the former, they failed to demonstrate a similar effect in the latter (Cvorovic et al., 2010). Anthocyanins from tart cherries significantly reduced proliferation of human colon cancer cells HT29 and HCT-116 as well (Kang et al., 2003; Marko et al., 2004). An anthocyanins extract from Vaccinium uliginosum suppressed the growth of human colorectal cancer cells DLD-1 and COLO205 in a dose-dependent manner through the induction of apoptosis. It was hypothesized that the anticancer efficacy might be attributed to its high percentage of malvidin (Zu et al., 2010). On the other hand, the antiproliferative and the anti-cancer potential of several berry extracts containing different profiles of phenolic compounds (anthocyanins, flavonols, and ellagitannins) was studied in human colon cancer HT-29 cells. All the berry extracts studied decreased the proliferation and the number of HT-29 cells in the G0/G1 phase of the cell cycle. This correlated with their anthocyanin concentration supporting the fact that the inhibitory effect of berry extracts is based on the concentration rather than the composition of anthocyanins (Coates et al., 2007; Johnson et al., 2011; Wu et al., 2007).

Numerous studies reported antiproliferative activity of anthocyanins in human cancer cells derived from malignant tissues of various origins such as breast, lung, uterus, stomach, central nervous system, vulva, prostate (Lazze et al., 2004; Meiers et al., 2001; Olsson et al., 2004; Seeram et al., 2004; Zhang et al., 2005). Anthocyanins were potent and selective in inhibiting human promyelocytic leukemia cell proliferation as well (Feng et al., 2007; Hou et al., 2003; Katsube et al., 2003).

Animal studies have also reported anticarcinogenic properties of anthocyanins. In induced rat colon cancer cell models they significantly decreased total tumors as well as aberrant crypts (Hagiwara et al., 2001; Hagiwara et al., 2002; Harris et al., 2001; Lala et al., 2006; Magnusson et al., 2003). Cai and colleagues demonstrated that Red grape pomace extract (oenocyanin) interferes with adenoma development in the ApcMin mouse by affecting tumor burden more prominently than tumor number. Oenocyanin efcacy was accompanied by the decreased adenoma cell proliferation and down-regulation of expression of the PI3 pathway component Akt, which supports cell proliferation (Cai et al., 2010). It was also

As mentioned above, naturally occurring dietary substances, in particular, flavonoids, have gained increased attention as agents interfering with processes involved in cancer development and progression. Among them, anthocyanins might be of particular interest since their daily intake is remarkably high compared to other flavonoids - it is estimated to vary between 180 and 215 mg (Hou, 2003) whereas the intake of other flavonoids reaches only 20-25 mg/day. Numerous recent studies indicate that anthocyanins are able to inhibit the growth of embryonic fibroblasts and of different cancer cells derived from malignant human tissues, suggesting their possible role as chemopreventive agents. This brings in focus their possible importance for public health as dietary components with preventive

There are few reports on the inhibitory effects of anthocyanins on colon cancer cell growth. Extracts of grapes, bilberries and chokeberries rich in anthocyanins have been shown to inhibit the growth of human malignant HT-29 colon cancer cells but did not affect the growth of non-malignant colon-derived cells (Zhao et al., 2004). Similar effect was observed in highly and low tumorigenic colon cancer cell lines, LoVo/Adr and LoVo. While delphinidin and cyanidin were cytotoxic and induced apoptosis in the former, they failed to demonstrate a similar effect in the latter (Cvorovic et al., 2010). Anthocyanins from tart cherries significantly reduced proliferation of human colon cancer cells HT29 and HCT-116 as well (Kang et al., 2003; Marko et al., 2004). An anthocyanins extract from Vaccinium uliginosum suppressed the growth of human colorectal cancer cells DLD-1 and COLO205 in a dose-dependent manner through the induction of apoptosis. It was hypothesized that the anticancer efficacy might be attributed to its high percentage of malvidin (Zu et al., 2010). On the other hand, the antiproliferative and the anti-cancer potential of several berry extracts containing different profiles of phenolic compounds (anthocyanins, flavonols, and ellagitannins) was studied in human colon cancer HT-29 cells. All the berry extracts studied decreased the proliferation and the number of HT-29 cells in the G0/G1 phase of the cell cycle. This correlated with their anthocyanin concentration supporting the fact that the inhibitory effect of berry extracts is based on the concentration rather than the composition

**3. Citotoxicity/apoptosis of anthocyanins on colon cancer cells** 

impact on cancer as well as effective, cheap and safe anticancer supplements.

of anthocyanins (Coates et al., 2007; Johnson et al., 2011; Wu et al., 2007).

al., 2003; Katsube et al., 2003).

Numerous studies reported antiproliferative activity of anthocyanins in human cancer cells derived from malignant tissues of various origins such as breast, lung, uterus, stomach, central nervous system, vulva, prostate (Lazze et al., 2004; Meiers et al., 2001; Olsson et al., 2004; Seeram et al., 2004; Zhang et al., 2005). Anthocyanins were potent and selective in inhibiting human promyelocytic leukemia cell proliferation as well (Feng et al., 2007; Hou et

Animal studies have also reported anticarcinogenic properties of anthocyanins. In induced rat colon cancer cell models they significantly decreased total tumors as well as aberrant crypts (Hagiwara et al., 2001; Hagiwara et al., 2002; Harris et al., 2001; Lala et al., 2006; Magnusson et al., 2003). Cai and colleagues demonstrated that Red grape pomace extract (oenocyanin) interferes with adenoma development in the ApcMin mouse by affecting tumor burden more prominently than tumor number. Oenocyanin efcacy was accompanied by the decreased adenoma cell proliferation and down-regulation of expression of the PI3 pathway component Akt, which supports cell proliferation (Cai et al., 2010). It was also

**3.1 Cytotoxicity in colon and other cancer cell lines** 

demonstrated in the same animal model that anthocyanin-rich tart cherry extract added to the drinking water was associated with fewer and smaller tumors in the cecum, but none of the tested treatments inuenced the number of tumors in the small intestine or the number or burden of tumors in the colon. It was supposed, therefore, that lack of effect of anthocyanins on colonic tumor development may be a consequence of their metabolism by intestinal bacteria or their spontaneous degradation in the cecal and colonic environment (Bobe et al., 2006; Kang et al., 2003). Moreover, it was shown in ApcMin mice that dietary consumption of anthocyanins in the form of either a mixture (Mirtoselect) or as a pure compound (cyanidin-3-glucoside) interferes with small intestinal adenoma development in a dose-dependent fashion. Authors remarked the presence of measurable levels of anthocyanins in the target organ and in the urine, and in concentrations near or below the detection limit in the systemic circulation. Unfortunately, the dietary dose, at which either agent was signicantly efficacious when extrapolated by dose/ surface area comparison, suggested that equivalent for humans can be found in 740 g bilberries, that is a hefty dose. In terms of absolute dose of agent, cyanidin-3-glucoside was less efficacious than the Mirtoselect mixture. Furthermore, authors suggested that different results obtained, in comparison with other studies (Kang et al., 2003) were possibly due in part to the higher dose of anthocyanins employed but also due to the different way of administration since anthocyanins tend to be unstable in aqueous solution at neutral pH (Cooke et al., 2006). Recently, bilberry [(BB),Vaccinium myrtillus], lingonberry (LB, Vaccinium vitis-idaea), and cloudberry (CB, Rubus chamaemorus), rich in anthocyanins, proanthocyanidins and ellagic acid respectively, proved to be chemopreventive as demonstrated by a signicant reduction in the number of intestinal tumors in Min/1 mice. Concerning their different chemical composition, authors suggested that the effects seen, may rather be a result of a mixture of compounds acting in synergy than an effect of a single active substance. Moreover, since the cellular levels of b-catenin are increased at all stages of colon carcinogenesis, it was demonstrated that two of these berries, LB and CB, markedly inhibited the growth of the adenomas and accumulation of nuclear b-catenin and cyclin D1. Unfortunately, also in this study, the amount of berries in the diets was high and could not be easily reached in a human diet (Misikangas et al., 2007).

Concerning other tumor models, the incidence, multiplicity and final mass of mammary tumors were significantly reduced in rats that would receive grape juice containing 15 different anthocyanins (Singletary et al., 2003). Cyanidin-3-glucoside reduced the size of lung cancer xenografts and significantly inhibited metastasis in nude mice (Ding et al., 2006). Lyophilized black raspberries prevented the development of NMBA (Nnitrosomethylbenzylamine)-induced esophageal tumors (Stoner et al., 2007), just like anthocyanin-containing pomegranate extract delayed the onset and reduced the incidence of DMBA (7,12-dimethylbenzanthracene)-induced skin tumors in CD-1 mice.

Epidemiological studies in humans are, however, still scarce and contradictory. Biopsies of tumor and normal-appearing tissues in colon cancer patients consuming black raspberry powder daily during several weeks, showed reduced proliferation and increased apoptosis in cancerous but not in normal tissue. Antiangiogenic effect was also observed in these patients (Wang et al., 2007). A phase I pilot study in colorectal cancer patients demonstrated that treatment with black raspberries caused positive modulation of biomarkers of tumor development, including cell proliferation, apoptosis, angiogenesis and Wnt pathway in both colorectal adenocarcinomas and adjacent normal tissues (Wang et al., 2007). In a clinical pilot study twenty-five colorectal cancer patients, scheduled to undergo resection of primary

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 133

blood mononuclear cells. Parallel with the accumulation of ROS, Feng and colleagues demonstrated the increase of peroxides, but not superoxides in these cells, suggesting the reaction with the glutathione antioxidant system as one of the possible mechanisms for this prooxidant activity, together with ROS-dependent activation of p38 and JNK (Feng et al., 2007). Similarly, both delphinidin and cyanidin, showed prooxidant activity and induced apoptotic changes and cytotoxic effect in metastatic colorectal drug-resistant cells (LoVo/ADR), but not in cells originating from primary tumor site, Caco-2 (Cvorovic et al., 2010). This "inconsistent" behavior of the anthocyanidins might be influenced by cellular energy metabolism changes associated with neoplastic transformation (Warburg, 1956b). Indeed, the rate of lactate production is significantly higher in highly tumorigenic LoVo/ADR than in low tumorigenic LoVo cells (Fanciulli et al., 2000), and, presumably, in CaCo-2 as well. And even a slight decrease of pH might favor protonation of anthocyanidins, a mechanism causing loss of their free-radical scavenging activity (Borkowski et al., 2005). However, it is not clear if anthocyanidins directly promote oxidative stress in LoVo/ADR cells. One of the possible mechanisms proposed in this study is the interference with the glutathione antioxidant system. Delphinidin and cyanidin inhibited glutathione reductase (GR) activity in LoVo/ADR cells and significantly depleted their intracellular glutathione levels, while failing to induce any similar effect in CaCo-2 cells. These studies give evidence that anthocyanins preferentially kill cancer cells with high malignant characteristics and resistant to conventional treatment regimens, which could set the basis for the development of new sensitizing agents in the treatment of metastatic

**4. Biochemical features accompanying cytotoxicity/apoptosis** 

understood (Milane et al., 2007; Talavera et al., 2004; Wolffram et al., 2002).

that this behavior could increase their own bioavailability (Faria et al., 2009).

All the metabolic actions exerted by anthocyanins imply their cellular bioavailability. Previous *in vivo* studies have reported that anthocyanins are absorbed in the stomach and small intestine (Passamonti et al., 2003a; Talavera et al., 2003; Talavera et al., 2004). Felgines and colleagues administered an oral dose of a radiolabelled cyanidin 3-O-glucoside, demonstrating that the major site of absorption in mice is the intestine with a minimal accumulation of the radioactivity in tissues out of the gastrointestinal tract (Felgines et al.,

Intestinal barrier is impermeable to most flavonoid glucosides because, based on their molecular structure, anthocyanins and their aglycones cannot cross the cell membrane passively (Dreiseitel et al., 2009). Among the influx carriers, the hexose transporters SGLT1 and GLUT 5 are expressed on apical membrane of the intestinal epithelium. Different groups suggested that anthocyanins, based on their glycosides moiety, could be transported by glucose carrier SGLT1. However, the involvement of this protein it is not completely

Recently, it was also demonstrated that GLUT2, another glucose transporter, is expressed not only at the basolateral but also at apical membranes of intestinal cells. Faria and colleagues showed that kinetic parameters of 3H-2-deoxy-D-glucose-uptake of GLUT2 are changed after acute treatment with anthocyanins, supporting a favorable use of anthocyanins in diabetic population. Interestingly they also observed an increased GLUT2 expression (not for SGLT1 or GLUT5) after a chronic exposure to anthocyanins speculating

**4.1 Membrane transport of anthocyanins** 

disease.

2010).

tumor or liver metastases, received different amount of mirtocyan. This is a standardized anthocyanin mixture extracted from bilberries administered daily for 7 days before surgery. In the immunohistochemical observations of colorectal tumors from all patients who had received mirtocyan, in comparison with the preintervention biopsy, the proliferation index, reflected by Ki-67 staining, was significantly decreased by 7%. The apoptotic index in colorectal cancer samples from all patients increased from 3.6% to 5.3% of epithelial cells. However, in the absence of a zero dose control group, authors couldn't determine if this increase could, at least, to some extent, be the consequence of inherent procedural differences in measurements. Nevertheless, the pharmacodynamic changes observed seemed to be more prominent in patients at a dose of anthocyanins, which elicited target tissue levels below the detection limit, than at higher one, which furnished detectable anthocyanin levels in colorectal tissue (Thomasset et al., 2009).

However, an Italian study aimed at investigating the relationship between anthocyanidins intake and risk for oral or pharyngeal cancer did not show any significant association (Rossi et al., 2007). There was no protective effect demonstrated on the development of prostate cancer either (Bosetti et al., 2006). Optimal tumor inhibition occurs when the berry anthocyanins are added to the diet before, during and after treatment with carcinogens, suggesting that consumption of berries throughout life may maximize their chemopreventive effectiveness in humans. The fact that berry diets show a variable effect on tumorigenesis suggests that the inhibitory components of berry extract are not completely absorbed and/or that molecules housed in berry extracts do not affect certain critical signaling pathways of carcinogenesis (Stoner, 2009). Although further proves are needed, these studies open a possibility for anthocyanins to be considered for use in cancer treatment in combination with other therapeutic methods.

#### **3.2 How anthocyanins work – The mechanisms**

Antimutagenic and anticarcinogenic activity of anthocyanins is generally ascribed to their antioxidant properties conveyed by their phenolic structure. The double bonds in the ring and the hydroxyl side chains confers them potent free-radical scavenging activities (the positively charged oxygen atom in their molecule makes them more generous hydrogendonating antioxidants compared to other flavonoids), but also enables their metal chelation and protein binding properties (Kong et al., 2003). Apart from acting as direct free-radical scavengers, anthocyanins have been demonstrated to affect the activity of phase II enzymes well-known for their detoxifying and antioxidant properties and therefore important in cancer prevention. *In vivo* studies showed that the diet supplemented with freeze-dried blueberries or black raspberries, both rich in anthocyanins, led to increased glutathione Stransferase (GST) activity in rats (Boateng et al., 2007; Reen et al., 2006). On the other hand, intake of an anthocyanin-rich mixed berry juice reduced oxidative DNA damage in peripheral-blood mononuclear cells and significantly increased total glutathione (GSH) level and GSH status in whole blood in male healthy non-smoking probands (Weisel et al., 2006). All this speaks in favor of a multi-level antioxidant activity of anthocyanins.

However, numerous recent studies, on the anthocyanins role in tumor growth inhibition, point at their prooxidant properties. It has been shown that the apoptotic effect of anthocyanins in malignant cells could be result of their ability to induce ROS accumulation in these cells. Moreover, the apoptotic activity was directly correlated to the number of hydroxyl groups at the B-ring (Hou et al., 2003). Interestingly, ROS generation was observed in leukemia cells treated with cyanidin-3-rutinoside, but not in normal human peripheral-

tumor or liver metastases, received different amount of mirtocyan. This is a standardized anthocyanin mixture extracted from bilberries administered daily for 7 days before surgery. In the immunohistochemical observations of colorectal tumors from all patients who had received mirtocyan, in comparison with the preintervention biopsy, the proliferation index, reflected by Ki-67 staining, was significantly decreased by 7%. The apoptotic index in colorectal cancer samples from all patients increased from 3.6% to 5.3% of epithelial cells. However, in the absence of a zero dose control group, authors couldn't determine if this increase could, at least, to some extent, be the consequence of inherent procedural differences in measurements. Nevertheless, the pharmacodynamic changes observed seemed to be more prominent in patients at a dose of anthocyanins, which elicited target tissue levels below the detection limit, than at higher one, which furnished detectable

However, an Italian study aimed at investigating the relationship between anthocyanidins intake and risk for oral or pharyngeal cancer did not show any significant association (Rossi et al., 2007). There was no protective effect demonstrated on the development of prostate cancer either (Bosetti et al., 2006). Optimal tumor inhibition occurs when the berry anthocyanins are added to the diet before, during and after treatment with carcinogens, suggesting that consumption of berries throughout life may maximize their chemopreventive effectiveness in humans. The fact that berry diets show a variable effect on tumorigenesis suggests that the inhibitory components of berry extract are not completely absorbed and/or that molecules housed in berry extracts do not affect certain critical signaling pathways of carcinogenesis (Stoner, 2009). Although further proves are needed, these studies open a possibility for anthocyanins to be considered for use in cancer

Antimutagenic and anticarcinogenic activity of anthocyanins is generally ascribed to their antioxidant properties conveyed by their phenolic structure. The double bonds in the ring and the hydroxyl side chains confers them potent free-radical scavenging activities (the positively charged oxygen atom in their molecule makes them more generous hydrogendonating antioxidants compared to other flavonoids), but also enables their metal chelation and protein binding properties (Kong et al., 2003). Apart from acting as direct free-radical scavengers, anthocyanins have been demonstrated to affect the activity of phase II enzymes well-known for their detoxifying and antioxidant properties and therefore important in cancer prevention. *In vivo* studies showed that the diet supplemented with freeze-dried blueberries or black raspberries, both rich in anthocyanins, led to increased glutathione Stransferase (GST) activity in rats (Boateng et al., 2007; Reen et al., 2006). On the other hand, intake of an anthocyanin-rich mixed berry juice reduced oxidative DNA damage in peripheral-blood mononuclear cells and significantly increased total glutathione (GSH) level and GSH status in whole blood in male healthy non-smoking probands (Weisel et al., 2006).

However, numerous recent studies, on the anthocyanins role in tumor growth inhibition, point at their prooxidant properties. It has been shown that the apoptotic effect of anthocyanins in malignant cells could be result of their ability to induce ROS accumulation in these cells. Moreover, the apoptotic activity was directly correlated to the number of hydroxyl groups at the B-ring (Hou et al., 2003). Interestingly, ROS generation was observed in leukemia cells treated with cyanidin-3-rutinoside, but not in normal human peripheral-

All this speaks in favor of a multi-level antioxidant activity of anthocyanins.

anthocyanin levels in colorectal tissue (Thomasset et al., 2009).

treatment in combination with other therapeutic methods.

**3.2 How anthocyanins work – The mechanisms** 

blood mononuclear cells. Parallel with the accumulation of ROS, Feng and colleagues demonstrated the increase of peroxides, but not superoxides in these cells, suggesting the reaction with the glutathione antioxidant system as one of the possible mechanisms for this prooxidant activity, together with ROS-dependent activation of p38 and JNK (Feng et al., 2007). Similarly, both delphinidin and cyanidin, showed prooxidant activity and induced apoptotic changes and cytotoxic effect in metastatic colorectal drug-resistant cells (LoVo/ADR), but not in cells originating from primary tumor site, Caco-2 (Cvorovic et al., 2010). This "inconsistent" behavior of the anthocyanidins might be influenced by cellular energy metabolism changes associated with neoplastic transformation (Warburg, 1956b). Indeed, the rate of lactate production is significantly higher in highly tumorigenic LoVo/ADR than in low tumorigenic LoVo cells (Fanciulli et al., 2000), and, presumably, in CaCo-2 as well. And even a slight decrease of pH might favor protonation of anthocyanidins, a mechanism causing loss of their free-radical scavenging activity (Borkowski et al., 2005). However, it is not clear if anthocyanidins directly promote oxidative stress in LoVo/ADR cells. One of the possible mechanisms proposed in this study is the interference with the glutathione antioxidant system. Delphinidin and cyanidin inhibited glutathione reductase (GR) activity in LoVo/ADR cells and significantly depleted their intracellular glutathione levels, while failing to induce any similar effect in CaCo-2 cells. These studies give evidence that anthocyanins preferentially kill cancer cells with high malignant characteristics and resistant to conventional treatment regimens, which could set the basis for the development of new sensitizing agents in the treatment of metastatic disease.

#### **4. Biochemical features accompanying cytotoxicity/apoptosis**

#### **4.1 Membrane transport of anthocyanins**

All the metabolic actions exerted by anthocyanins imply their cellular bioavailability. Previous *in vivo* studies have reported that anthocyanins are absorbed in the stomach and small intestine (Passamonti et al., 2003a; Talavera et al., 2003; Talavera et al., 2004). Felgines and colleagues administered an oral dose of a radiolabelled cyanidin 3-O-glucoside, demonstrating that the major site of absorption in mice is the intestine with a minimal accumulation of the radioactivity in tissues out of the gastrointestinal tract (Felgines et al., 2010).

Intestinal barrier is impermeable to most flavonoid glucosides because, based on their molecular structure, anthocyanins and their aglycones cannot cross the cell membrane passively (Dreiseitel et al., 2009). Among the influx carriers, the hexose transporters SGLT1 and GLUT 5 are expressed on apical membrane of the intestinal epithelium. Different groups suggested that anthocyanins, based on their glycosides moiety, could be transported by glucose carrier SGLT1. However, the involvement of this protein it is not completely understood (Milane et al., 2007; Talavera et al., 2004; Wolffram et al., 2002).

Recently, it was also demonstrated that GLUT2, another glucose transporter, is expressed not only at the basolateral but also at apical membranes of intestinal cells. Faria and colleagues showed that kinetic parameters of 3H-2-deoxy-D-glucose-uptake of GLUT2 are changed after acute treatment with anthocyanins, supporting a favorable use of anthocyanins in diabetic population. Interestingly they also observed an increased GLUT2 expression (not for SGLT1 or GLUT5) after a chronic exposure to anthocyanins speculating that this behavior could increase their own bioavailability (Faria et al., 2009).

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 135

the mitogen-activated protein kinases (MAPKs), phosphatidylinositol-3-kinase (PI3K)/Akt pathway, phospholipase C-g1 (PLCg1) signaling, protein kinase C, p53 signaling, ataxiatelangiectasia-mutated (ATM) kinase, nuclear factor-kappaB (NF-kB) signaling, and Jak/Stat pathway. ROS modulate the apoptotic signaling pathway through the cellular redox status by activating key protein kinases (Chan et al., 2010; Noguchi et al., 2005). Pro-oxidants such as H2O2 or other stressors, could induce apoptosis (or programmed cell death) by activating the intrinsic or "mitochondrial" apoptosis pathway that results in the damage of this subcellular compartment and the pro-apoptotic factors release (Circu and Aw, 2010; Mates et al., 2008). ROS involved in apoptosis derive both from environmental pro-oxidants or from intracellular respiratory dysfunction since mitochondria are the main site of intracellular source of ROS production. It was reported that oxidative stress plays an important role in the molecular mechanism of colorectal cancer (Keshavarzian et al., 1992) Flavones in HT-29 colon cancer cells increase the uptake of pyruvate or lactate into mitochondria, which is followed by an increase in O2- production that finally leads to apoptosis (Wenzel et al.,

The prooxidant activity of anthocyanins through the increase of intracellular ROS production has been clearly explained in several studies (Feng et al., 2007; Hou et al., 2005).

Intracellular glutathione (GSH) is a major buffer of cellular redox status due to its active SHgroup that has reducing nucleophilic properties (Meister, 1983; Meister, 1991; Meister and Anderson, 1983). It acts as reducing agent, antioxidant and free-radical scavenger against ROS generated during oxidative metabolism and/or oxidative stress (Donati et al., 1990; Hall, 1999a; Hall, 1999b; Sies, 1999) and is also involved in the metabolism of xenobiotics and some cellular molecules (Wu et al., 2004). Free glutathione is present mainly in its reduced form maintained by the action of glutathione reductase (GR), but chemical oxidation of GSH to GSSG can occur as a result of numerous enzyme-catalysed reactions that use GSH to reduce hydrogen peroxide or other peroxides to water or the corresponding alcohol (Diaz Vivancos et al., 2010). GSH is preferentially (85-90%) located in the cytosolicnuclear compartments and only a small amount is present in mitochondria and endoplasmic reticulum (Hwang et al., 1992; Meredith and Reed, 1982). The free-radical and antioxidant action of GSH depends on its involvement in different enzymatic reactions as those catalyzed by glutathione peroxidases (GPxs) (Lei, 2002), glutathione-S-transferases (GSTs), formaldehyde dehydrogenase, maleylacetoacetate isomerase, and glyoxalase I (Arrigo, 1999; Dickinson and Forman, 2002; Dickinson et al., 2002; Hayes and McLellan, 1999). Cancer cells present elevated GSH levels that generally increase antioxidant capacity and resistance to oxidative stress and regulate different mechanisms linked to carcinogenesis, sensitivity against cytotoxic drugs, ionizing radiation, and some cytokines, DNA synthesis, and cell proliferation (Estrela et al., 2006). There are yet a few reports on the possible role of avonoids, as well as other phytochemicals, in modulating the glutathione antioxidant system activity, including regulation of GSH intracellular levels through targeting its synthesis (Ramos and Aller, 2008), induction of MRP-1 mediated GSH efux (Kachadourian and Day, 2006), or inhibition of glutathione peroxidase enzyme activity (Trachootham et al., 2006). Upon grape seed extract treatment, HT29 colon cancer cells showed increased ROS production (that might result in oxidative stress in cells) and a decreased level of intracellular reduced glutathione (Kaur et al., 2011). In addition, after delphinidin and

2005).

**4.3 GSH role in apoptosis** 

Bilitranslocase (BTL) is an organic anions transporter specific for bilirubin, initially found in the membranes of hepatic sinusoidal cells, but present also at the gastric mucosa and in renal tubules (Baldini et al., 1986; Elias et al., 1990; Sottocasa et al., 1989). Some of its substrates are nicotinic acid, bromosulfophthalein, cibacron blue and some flavonoids (Passamonti et al., 2009; Passamonti et al., 2002). Among flavonoids family, 17 anthocyanins showed competitive inhibitory behavior to specific transport assay with delphinidin as the most active molecule (Passamonti et al., 2002). It was also demonstrated that the BTL is directly involved in the vasoactivity of flavonoids: vasorelaxation induced by both, cyanidin 3-glucoside and bilberry anthocyanins, was significantly decreased in aorta rings pre-treated with anti-BTL antibodies (Ziberna et al., 2011). Recent studies have revealed that bilitranslocase is also expressed at the intestinal epithelial level, in particular, at the apical domain. Caco-2 cells express BTL as well and the uptake of BSP into these cells is strongly inhibited by anti-bilitranslocase antibodies (Passamonti et al., 2009).

The results reported should be further implemented to clarify the involvement of the influx membrane transporters.

More data are available on flavonoids efflux transporters. Major interest on these proteins is linked to their involvement in cancer resistance. These proteins belong to the class of the ABC transporters (ATP-binding cassette), and their role in cancer cells is to prevent the accumulation of anticancer drugs. Some ABC transporters are MRP1 (multidrug resistanceassociated protein 1, ABCC1) (Cole et al., 1992), MRP2 (ABCC2) and MRP3 (ABCC3) (Borst et al., 1999) as well as BCRP/MXR1 (ABCG2) (Doyle et al., 1998; Miyake et al., 1999) and Breast Cancer Resistance Protein BCRP (ABCG2) however, P-glycoprotein (ABCB1) is overexpressed to the highest level and plays the major role. It was shown that flavonoids interact with these transporters but their effects are often contradictory depending on the type of cancer cells (Di Pietro et al., 2002). Moreover, a different behavior depending on the molecular structure was also demonstrated. Dreiseitel and colleagues showed that, depending on the sugar moiety, some flavonoids can act as BCRP stimulators while others act as inhibitors (Dreiseitel et al., 2009; Morris and Zhang, 2006).

However, the significance of these flavonoid–efflux transporter interactions has not been unequivocally demonstrated since it is impossible to exclude the involvement of other drugs transporters and of intracellular metabolizing enzymes that modify the substrate disposition (Morris and Zhang, 2006).

#### **4.2 Oxidative stress and apoptosis**

Reactive oxygen species (ROS) and reactive nitrogen species (RONS) are a collective term that broadly describes O2-derived free radicals such as superoxide anions (O2•−), hydroxyl radicals (HO•), peroxyl (RO2•) and alkoxyl radicals (RO•), nitrogen monoxide (NO•), peroxynitrite (ONOO-), nitrogen dioxide (NO2•) as well as O2-derived non-radical species such as hydrogen peroxide (H2O2) (Halliwell and Cross, 1994). Both reactive species are important mediators in the normal regulation of different physiological processes such as cellular proliferation or activation. On the other hand, the imbalance of cellular redox homeostasis is described at the base of many chronic diseases and is also involved in cancer development (Acharya et al., 2010).

Specic ROS such as H2O2 or superoxide have been implicated as crucial mediators of apoptotic cell death (Casado et al., 2002; Circu and Aw, 2010; Madeo et al., 1999). ROS tend to enhance survival or promote cell death by activating different factors such as members of the mitogen-activated protein kinases (MAPKs), phosphatidylinositol-3-kinase (PI3K)/Akt pathway, phospholipase C-g1 (PLCg1) signaling, protein kinase C, p53 signaling, ataxiatelangiectasia-mutated (ATM) kinase, nuclear factor-kappaB (NF-kB) signaling, and Jak/Stat pathway. ROS modulate the apoptotic signaling pathway through the cellular redox status by activating key protein kinases (Chan et al., 2010; Noguchi et al., 2005). Pro-oxidants such as H2O2 or other stressors, could induce apoptosis (or programmed cell death) by activating the intrinsic or "mitochondrial" apoptosis pathway that results in the damage of this subcellular compartment and the pro-apoptotic factors release (Circu and Aw, 2010; Mates et al., 2008). ROS involved in apoptosis derive both from environmental pro-oxidants or from intracellular respiratory dysfunction since mitochondria are the main site of intracellular source of ROS production. It was reported that oxidative stress plays an important role in the molecular mechanism of colorectal cancer (Keshavarzian et al., 1992) Flavones in HT-29 colon cancer cells increase the uptake of pyruvate or lactate into mitochondria, which is followed by an increase in O2- production that finally leads to apoptosis (Wenzel et al., 2005).

The prooxidant activity of anthocyanins through the increase of intracellular ROS production has been clearly explained in several studies (Feng et al., 2007; Hou et al., 2005).

#### **4.3 GSH role in apoptosis**

134 Colorectal Cancer – From Prevention to Patient Care

Bilitranslocase (BTL) is an organic anions transporter specific for bilirubin, initially found in the membranes of hepatic sinusoidal cells, but present also at the gastric mucosa and in renal tubules (Baldini et al., 1986; Elias et al., 1990; Sottocasa et al., 1989). Some of its substrates are nicotinic acid, bromosulfophthalein, cibacron blue and some flavonoids (Passamonti et al., 2009; Passamonti et al., 2002). Among flavonoids family, 17 anthocyanins showed competitive inhibitory behavior to specific transport assay with delphinidin as the most active molecule (Passamonti et al., 2002). It was also demonstrated that the BTL is directly involved in the vasoactivity of flavonoids: vasorelaxation induced by both, cyanidin 3-glucoside and bilberry anthocyanins, was significantly decreased in aorta rings pre-treated with anti-BTL antibodies (Ziberna et al., 2011). Recent studies have revealed that bilitranslocase is also expressed at the intestinal epithelial level, in particular, at the apical domain. Caco-2 cells express BTL as well and the uptake of BSP into these cells is strongly

The results reported should be further implemented to clarify the involvement of the influx

More data are available on flavonoids efflux transporters. Major interest on these proteins is linked to their involvement in cancer resistance. These proteins belong to the class of the ABC transporters (ATP-binding cassette), and their role in cancer cells is to prevent the accumulation of anticancer drugs. Some ABC transporters are MRP1 (multidrug resistanceassociated protein 1, ABCC1) (Cole et al., 1992), MRP2 (ABCC2) and MRP3 (ABCC3) (Borst et al., 1999) as well as BCRP/MXR1 (ABCG2) (Doyle et al., 1998; Miyake et al., 1999) and Breast Cancer Resistance Protein BCRP (ABCG2) however, P-glycoprotein (ABCB1) is overexpressed to the highest level and plays the major role. It was shown that flavonoids interact with these transporters but their effects are often contradictory depending on the type of cancer cells (Di Pietro et al., 2002). Moreover, a different behavior depending on the molecular structure was also demonstrated. Dreiseitel and colleagues showed that, depending on the sugar moiety, some flavonoids can act as BCRP stimulators while others

However, the significance of these flavonoid–efflux transporter interactions has not been unequivocally demonstrated since it is impossible to exclude the involvement of other drugs transporters and of intracellular metabolizing enzymes that modify the substrate disposition

Reactive oxygen species (ROS) and reactive nitrogen species (RONS) are a collective term that broadly describes O2-derived free radicals such as superoxide anions (O2•−), hydroxyl radicals (HO•), peroxyl (RO2•) and alkoxyl radicals (RO•), nitrogen monoxide (NO•),

such as hydrogen peroxide (H2O2) (Halliwell and Cross, 1994). Both reactive species are important mediators in the normal regulation of different physiological processes such as cellular proliferation or activation. On the other hand, the imbalance of cellular redox homeostasis is described at the base of many chronic diseases and is also involved in cancer

Specic ROS such as H2O2 or superoxide have been implicated as crucial mediators of apoptotic cell death (Casado et al., 2002; Circu and Aw, 2010; Madeo et al., 1999). ROS tend to enhance survival or promote cell death by activating different factors such as members of

), nitrogen dioxide (NO2•) as well as O2-derived non-radical species

inhibited by anti-bilitranslocase antibodies (Passamonti et al., 2009).

act as inhibitors (Dreiseitel et al., 2009; Morris and Zhang, 2006).

membrane transporters.

(Morris and Zhang, 2006).

peroxynitrite (ONOO-

**4.2 Oxidative stress and apoptosis** 

development (Acharya et al., 2010).

Intracellular glutathione (GSH) is a major buffer of cellular redox status due to its active SHgroup that has reducing nucleophilic properties (Meister, 1983; Meister, 1991; Meister and Anderson, 1983). It acts as reducing agent, antioxidant and free-radical scavenger against ROS generated during oxidative metabolism and/or oxidative stress (Donati et al., 1990; Hall, 1999a; Hall, 1999b; Sies, 1999) and is also involved in the metabolism of xenobiotics and some cellular molecules (Wu et al., 2004). Free glutathione is present mainly in its reduced form maintained by the action of glutathione reductase (GR), but chemical oxidation of GSH to GSSG can occur as a result of numerous enzyme-catalysed reactions that use GSH to reduce hydrogen peroxide or other peroxides to water or the corresponding alcohol (Diaz Vivancos et al., 2010). GSH is preferentially (85-90%) located in the cytosolicnuclear compartments and only a small amount is present in mitochondria and endoplasmic reticulum (Hwang et al., 1992; Meredith and Reed, 1982). The free-radical and antioxidant action of GSH depends on its involvement in different enzymatic reactions as those catalyzed by glutathione peroxidases (GPxs) (Lei, 2002), glutathione-S-transferases (GSTs), formaldehyde dehydrogenase, maleylacetoacetate isomerase, and glyoxalase I (Arrigo, 1999; Dickinson and Forman, 2002; Dickinson et al., 2002; Hayes and McLellan, 1999). Cancer cells present elevated GSH levels that generally increase antioxidant capacity and resistance to oxidative stress and regulate different mechanisms linked to carcinogenesis, sensitivity against cytotoxic drugs, ionizing radiation, and some cytokines, DNA synthesis, and cell proliferation (Estrela et al., 2006). There are yet a few reports on the possible role of avonoids, as well as other phytochemicals, in modulating the glutathione antioxidant system activity, including regulation of GSH intracellular levels through targeting its synthesis (Ramos and Aller, 2008), induction of MRP-1 mediated GSH efux (Kachadourian and Day, 2006), or inhibition of glutathione peroxidase enzyme activity (Trachootham et al., 2006). Upon grape seed extract treatment, HT29 colon cancer cells showed increased ROS production (that might result in oxidative stress in cells) and a decreased level of intracellular reduced glutathione (Kaur et al., 2011). In addition, after delphinidin and

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 137

vary both spatially and temporally. The elevated glycolytic pathway of cancer cells appears to be a response to hypoxia due to the growth of the tumor surpassing the available vascular supplied oxygen (Mathupala et al., 2001) and seems to be controlled directly by the antiapoptotic protein Akt that generates apoptotic resistance *in vitro* (Elstrom et al., 2004). Then, the decreased dependence on aerobic respiration becomes a selective advantage for survival and proliferation escaping from the apoptotic event. Cell metabolism is shifted toward the increased expression of glycolytic enzymes, glucose transporters, and inhibitors of mitochondrial metabolism that result in a transitional intracellular acidification (Hsu and Sabatini, 2008) and increased glucose uptake is observed coincident with the transition from colon adenomas to invasive cancer (Yasuda et al., 2001). Nevertheless, evidence that intracellular acidification is associated with the progression of apoptosis, has been steadily accumulating (Barry et al., 1993; Gottlieb et al., 1996; Li and Eastman, 1995; Rebollo et al., 1995). An important role in the intracellular acidification could be due to alterations of membrane pHi-regulating mechanisms, including the Na+/H+ exchanger (NHE) that might favor accumulation of the protons produced by energetic metabolism. NHE is ubiquitously expressed transporter in the plasma membrane with a main function to extrude H+ from the

Multidrug resistant tumor cells exhibit an altered pH gradient across different cell compartments, which favors a reduced intracellular accumulation of antineoplastic drugs and a decreased therapeutic effect. In fact, the activity and expression of NHE are increased in doxorubicin-resistant (HT29-dx) human colon carcinoma cells in comparison with doxorubicin-sensitive HT29 cells (Miraglia et al., 2005). On the other hand, it was demonstrated that activation of the NHE-1 and the resulting cellular alkalinization play a key role in oncogenic transformation (Reshkin et al., 2000). Cyanidin (10 microM), but not its glycosides, could inhibit the neurotensin- and EGF-induced increased rate of extracellular acidification in HT-29 human colon adenocarcinoma cell line probably by inhibiting cellular metabolism, rather than directly altering Na+/H+ exchange (Briviba et

The effect of anthocyanins on metabolism involved in pH modulation of apoptosis is

In humans, multistage carcinogenesis was previously considered a consequence of genetic alterations that cause activation of oncogenes and inactivation of tumor suppressor genes. In addition to genetic events, epigenetic events are another leading player in carcinogenesis (Link et al., 2010). Indeed, it is believed that majority of cancers result from changes that accumulate throughout the life due to the exposure to various endogenous factors and arguably diet and environment-mediated epigenetic perturbations play a crucial role in cancer progression in humans (Herceg, 2007). It was rst recognized more than 25 years ago that in colorectal cancer cells, global DNA methylation patterns differed considerably from

The developmental biologist Conrad H. Waddington coined the term 'epigenetics' in 1942, trying to describe reversible heritable changes in gene expression that occur without alteration in DNA sequence sufciently powerful to regulate the dynamics of gene

cytoplasm.

al., 2001).

anyway a poor-trodden path.

**5. Roadmap for further investigations 5.1 Role of flavonoids on DNA methylation** 

those in their normal counterparts (Venkatachalam et al., 2010).

expression (Waddington, 1951 as cited in (Hitchler and Domann, 2009).

cyanidin treatment in primary (Caco-2) and metastatic (LoVo and LoVo/ADR) colorectal cancer cell lines, no signicant changes in the total GSH levels were observed in Caco-2 and LoVo cells, while both were shown to deplete intracellular glutathione levels in LoVo/ADR cells. GSSG content was not measurable in Caco-2 and LoVo cells, suggesting a normal cellular GSH/GSSG ratio (30:1–300:1) (Cvorovic et al., 2010). Cells undergoing apoptosis appear to rapidly and selectively release GSH into the extracellular space (Ghibelli et al., 1995; Ghibelli et al., 1998; Hammond et al., 2007). Hammond and colleagues demonstrated that apoptotic GSH export is directly linked to MRPs. Indeed basal and apoptotic GSH releases were decreased after RNAi reduction of MRP1 expression in Jurkat cells, indicating that MRP1 is a major player in both processes (Hammond et al., 2007). MRP1-channelled GSH export from cells can be also increased by different xenobiotics, including arsenite, verapamil (VRP), and some naturally-occurring flavonoids (Leslie et al., 2003; Loe et al., 2000).

GSTs are known as a family of Phase II detoxication enzymes that catalyze the conjugation of GSH (S-glutathionylation) with different compounds as xenobiotics and drugs or their metabolites, to form mercapturates (Hayes et al., 2005).

It has been recently shown that anthocyanin fractions from selected cultivars of Georgia-Grown Blueberries at 50-150 íg/mL do induce apoptosis in HT-29 colon cancer cells but these same concentrations decrease GST activities rather than induce it (Srivastava et al., 2007).

There are several studies, in normal cells and tissues, in which it was demonstrated that anthocyanins, probably involving some protein kinases, modulate the activity of some GSHdependent enzymes, thus ameliorating the antioxidant response (Hou et al., 2010; Suda et al., 2008; Veigas et al., 2008).

GSSG formed intracellularly is continuously reduced to GSH by the activity of GR. If oxidative stress or other factors limit the GR activity (e.g., glucose-6-phosphate dehydrogenase deciency may limit NADPH supply), GSSG will accumulate (Deneke and Fanburg, 1989). In this respect, Cvorovic and colleagues showed that delphinidin and cyanidin did inhibit GR activity in LoVo/ADR cells but not in Caco2 and Lovo cells (Cvorovic et al., 2010). This has two important consequences: (i) the thiol redox status of the cell will shift, activating oxidant response transcriptional elements; and (ii) GSSG may be preferentially secreted out of the cell. (i) The protein sequences of many transcription factors contain cys residues, mainly localized in the DNA-binding domain that, when oxidized, cause a different modulation of gene expression (Arrigo, 1999). (ii) GSSG may be reduced back to GSH, but when GSSG is present in excess, it is also eliminated from the cell by export into the extracellular space. Strong evidence that this export step is mediated by MRP2 was provided by studies of GSSG transport with canalicular membrane-enriched vesicles derived from normal and EHBR (Eisai hyperbilirubinuric rats) rats (Ballatori et al., 2009).

#### **4.4 Intracellular pH and apoptosis**

Despite the genetic variability, two phenotypes common to all tumor cells are cellular alkalinization and a shift to glycolytic metabolism. In the first decade of the 20th century, Otto Warburg found that cancer cells, even in the presence of oxygen disposition and a higher request of ATP for fast growing cells, prefer to metabolize glucose via glycolysis instead of oxidative phosphorylation (Warburg, 1956a). The oxygen levels within a tumor

cyanidin treatment in primary (Caco-2) and metastatic (LoVo and LoVo/ADR) colorectal cancer cell lines, no signicant changes in the total GSH levels were observed in Caco-2 and LoVo cells, while both were shown to deplete intracellular glutathione levels in LoVo/ADR cells. GSSG content was not measurable in Caco-2 and LoVo cells, suggesting a normal cellular GSH/GSSG ratio (30:1–300:1) (Cvorovic et al., 2010). Cells undergoing apoptosis appear to rapidly and selectively release GSH into the extracellular space (Ghibelli et al., 1995; Ghibelli et al., 1998; Hammond et al., 2007). Hammond and colleagues demonstrated that apoptotic GSH export is directly linked to MRPs. Indeed basal and apoptotic GSH releases were decreased after RNAi reduction of MRP1 expression in Jurkat cells, indicating that MRP1 is a major player in both processes (Hammond et al., 2007). MRP1-channelled GSH export from cells can be also increased by different xenobiotics, including arsenite, verapamil (VRP), and some naturally-occurring flavonoids (Leslie et al., 2003; Loe et al.,

GSTs are known as a family of Phase II detoxication enzymes that catalyze the conjugation of GSH (S-glutathionylation) with different compounds as xenobiotics and drugs or their

It has been recently shown that anthocyanin fractions from selected cultivars of Georgia-Grown Blueberries at 50-150 íg/mL do induce apoptosis in HT-29 colon cancer cells but these same concentrations decrease GST activities rather than induce it (Srivastava et al.,

There are several studies, in normal cells and tissues, in which it was demonstrated that anthocyanins, probably involving some protein kinases, modulate the activity of some GSHdependent enzymes, thus ameliorating the antioxidant response (Hou et al., 2010; Suda et

GSSG formed intracellularly is continuously reduced to GSH by the activity of GR. If oxidative stress or other factors limit the GR activity (e.g., glucose-6-phosphate dehydrogenase deciency may limit NADPH supply), GSSG will accumulate (Deneke and Fanburg, 1989). In this respect, Cvorovic and colleagues showed that delphinidin and cyanidin did inhibit GR activity in LoVo/ADR cells but not in Caco2 and Lovo cells (Cvorovic et al., 2010). This has two important consequences: (i) the thiol redox status of the cell will shift, activating oxidant response transcriptional elements; and (ii) GSSG may be preferentially secreted out of the cell. (i) The protein sequences of many transcription factors contain cys residues, mainly localized in the DNA-binding domain that, when oxidized, cause a different modulation of gene expression (Arrigo, 1999). (ii) GSSG may be reduced back to GSH, but when GSSG is present in excess, it is also eliminated from the cell by export into the extracellular space. Strong evidence that this export step is mediated by MRP2 was provided by studies of GSSG transport with canalicular membrane-enriched vesicles derived from normal and EHBR (Eisai hyperbilirubinuric rats) rats (Ballatori et al.,

Despite the genetic variability, two phenotypes common to all tumor cells are cellular alkalinization and a shift to glycolytic metabolism. In the first decade of the 20th century, Otto Warburg found that cancer cells, even in the presence of oxygen disposition and a higher request of ATP for fast growing cells, prefer to metabolize glucose via glycolysis instead of oxidative phosphorylation (Warburg, 1956a). The oxygen levels within a tumor

metabolites, to form mercapturates (Hayes et al., 2005).

2000).

2007).

2009).

al., 2008; Veigas et al., 2008).

**4.4 Intracellular pH and apoptosis** 

vary both spatially and temporally. The elevated glycolytic pathway of cancer cells appears to be a response to hypoxia due to the growth of the tumor surpassing the available vascular supplied oxygen (Mathupala et al., 2001) and seems to be controlled directly by the antiapoptotic protein Akt that generates apoptotic resistance *in vitro* (Elstrom et al., 2004).

Then, the decreased dependence on aerobic respiration becomes a selective advantage for survival and proliferation escaping from the apoptotic event. Cell metabolism is shifted toward the increased expression of glycolytic enzymes, glucose transporters, and inhibitors of mitochondrial metabolism that result in a transitional intracellular acidification (Hsu and Sabatini, 2008) and increased glucose uptake is observed coincident with the transition from colon adenomas to invasive cancer (Yasuda et al., 2001). Nevertheless, evidence that intracellular acidification is associated with the progression of apoptosis, has been steadily accumulating (Barry et al., 1993; Gottlieb et al., 1996; Li and Eastman, 1995; Rebollo et al., 1995). An important role in the intracellular acidification could be due to alterations of membrane pHi-regulating mechanisms, including the Na+/H+ exchanger (NHE) that might favor accumulation of the protons produced by energetic metabolism. NHE is ubiquitously expressed transporter in the plasma membrane with a main function to extrude H+ from the cytoplasm.

Multidrug resistant tumor cells exhibit an altered pH gradient across different cell compartments, which favors a reduced intracellular accumulation of antineoplastic drugs and a decreased therapeutic effect. In fact, the activity and expression of NHE are increased in doxorubicin-resistant (HT29-dx) human colon carcinoma cells in comparison with doxorubicin-sensitive HT29 cells (Miraglia et al., 2005). On the other hand, it was demonstrated that activation of the NHE-1 and the resulting cellular alkalinization play a key role in oncogenic transformation (Reshkin et al., 2000). Cyanidin (10 microM), but not its glycosides, could inhibit the neurotensin- and EGF-induced increased rate of extracellular acidification in HT-29 human colon adenocarcinoma cell line probably by inhibiting cellular metabolism, rather than directly altering Na+/H+ exchange (Briviba et al., 2001).

The effect of anthocyanins on metabolism involved in pH modulation of apoptosis is anyway a poor-trodden path.

#### **5. Roadmap for further investigations**

#### **5.1 Role of flavonoids on DNA methylation**

In humans, multistage carcinogenesis was previously considered a consequence of genetic alterations that cause activation of oncogenes and inactivation of tumor suppressor genes. In addition to genetic events, epigenetic events are another leading player in carcinogenesis (Link et al., 2010). Indeed, it is believed that majority of cancers result from changes that accumulate throughout the life due to the exposure to various endogenous factors and arguably diet and environment-mediated epigenetic perturbations play a crucial role in cancer progression in humans (Herceg, 2007). It was rst recognized more than 25 years ago that in colorectal cancer cells, global DNA methylation patterns differed considerably from those in their normal counterparts (Venkatachalam et al., 2010).

The developmental biologist Conrad H. Waddington coined the term 'epigenetics' in 1942, trying to describe reversible heritable changes in gene expression that occur without alteration in DNA sequence sufciently powerful to regulate the dynamics of gene expression (Waddington, 1951 as cited in (Hitchler and Domann, 2009).

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 139

Although, some explanations for the 'proton leak' come from the biophysical properties of the inner membrane, much of the explanation comes from the activities of a family of mitochondrial proteins termed uncoupling proteins (UCPs) (Klingenberg, 1999; Valle et al., 2010). UCPs exploit the gap in pH concentration to transfer the proton through the inner membrane into the matrix where they are released. Consequently, the mitochondrial membrane potential decreases, reduction of O2 via the respiratory chain is no longer linked to ATP synthesis and ATP/ADP exchange is not longer maintained (Vander Heiden et al., 1999). The influence of anthocyanins on ATP/ADP ratio and on UCPs role could be the aim of

Cancer cells seem to show high glycolytic rates even when oxygen is sufficient for oxidative phosphorylation (OXPHOS). This condition leads to a survival benefit of the tumor providing protection from oxidative stress and resulting in apoptosis avoidance (Kondoh et al., 2007a; Kondoh et al., 2007b). The importance of glycolysis in the survival of cancer cells was demonstrated by Bonnet and colleagues. Their experimental approach aimed at inhibiting the anaerobic glycolisys by repressing the activity of pyruvate dehydrogenase kinase (PDK) with dichloroacetate (DCA). PDK acts as a negative modulator of pyruvate dehydrogenase, a gate-keeping mitochondrial enzyme which controls the glucose oxidative fate into the cell. DCA changes the metabolism of cancer cells from the cytoplasm-based glycolysis to the mitochondria- based glucose oxidation. This led to increased ROS production and decreased mitochondrial membrane potential, efflux of pro-apoptotic mediators from mitochondria, and induction of mitochondria-dependent apoptosis only in cancer cells (Bonnet et al., 2007). On the other hand, in the majority of mammalian cells, glycolysis is inhibited by the presence of oxygen, which allows the mitochondria to oxidize

The transcription factor p53 regulates cellular energy metabolism and antioxidant defense mechanisms. Emerging evidence has shown that these two functions of p53 contribute greatly to p53's role in tumor suppression (Bensaad and Vousden, 2007; Matoba et al., 2006; Sablina et al., 2005). Loss of p53 results in decreased oxygen consumption and impaired mitochondrial respiration and promotes a switch to high glucose utilization in aerobic

It was shown that p53 regulates the OXPHOS dependence of cell by modulating the assembly of a key complex in the mitochondrial electron chain transport: cytochrome c oxidase (COX) (Ma et al., 2007; Matoba et al., 2006). It was demonstrated, in fact, that in HCT116 cells, p53 controls the expression of SCO 2 (Synthesis of Cytochrome c Oxidase 2). SCO2 is required for the assembly of mitochondrial DNA-encoded COX II subunit (MTCO2 gene) into the COX, so, p53 directly regulates mitochondrial oxygen consumption. p53 mutations in cancer cells induce a loss in SCO2, thereby resulting in a switch from an aerobic mitochondrial respiration to anaerobic glycolysis. p53 induces SCO2 expression to enhance mitochondrial respiration and induces TIGAR expression to slow glycolysis (Won

The metabolic implications of anthocyanins through the oxidative use of glucose could be appreciated indirectly. In fact, it is known that anthocyanins induce p53 expression (Fimognari et al., 2005; Lo et al., 2007; Renis et al., 2008), but a direct involvement of this

compounds on glucose metabolic use it is not yet demonstrated.

further studies.

pyruvate to CO2 and H2O.

et al., 2011).

glycolysis in cells (Maddocks and Vousden).

**5.3 Apoptosis and oxygen consumption** 

One of the "epigenome" processes is DNA methylation, a covalent chemical modication resulting in addition of a methyl (CH3) group at the carbon 5 position of the cytosine ring in CpG dinucleotides (Kanai and Hirohashi, 2007). This process plays important roles in chromatin structure modulation, transcriptional regulation and genomic stability, and is essential for the development of mammals (Ducasse and Brown, 2006; Li, 2002). CpG dinucleotides are not uniformly distributed throughout the human genome, but are often enriched in the promoter regions of genes. Short CpG-rich regions are also called as ''CpG islands'', and these are present in more than 50% of human gene promoters and can lead to gene silencing and proliferation or to affect the metabolic processes associated with energy metabolism. (Link et al., 2010). This mechanism is an enzymatic process mediated by DNA methyltransferases (DNMT): DNMT1, also called a "maintenance methyltransferase", preserves existing methylation patterns following DNA replication; DNMT3a and DNMT3b, on the other hand, serve as *de novo* methyltransferases, which act independently of replication on both strands, altering the epigenetic information content (Yu et al., 2011).

Recent studies havedemonstrated that all three DNMTs are overexpressed in several tumor types, including tumors of the colon and rectum, bladder, and kidney. When DNMT1 and DNMT3b are knocked out in colon cancer cell lines, methylation of tumor suppressor genes such as p16 is almost entirely eliminated and the gene is re-expressed (Rhee et al., 2002), as well as it has been established that the inhibition of DNA methyltransferase activity can strongly inhibit the formation of tumors (Stresemann et al., 2006).

It is known that some nutrients like folic acid, B vitamins and SAM (S-adenosylmethionine) and anthocyanins are key components of the methyl-metabolism pathway (Vanzo et al., 2011). Their methyl-donating mechanism can rapidly alter gene expression by modulating the availability of methyl donors as well as DNMT activity (Ross, 2003). There is a growing interest in the role of polyphenols in prevention of DNA methylation. It was demonstrated that epigallocatechin-3-gallate (EGCG), a tea polyphenol, through its methylation exerted by catechol-O-methyltransferase (COMT), indirectly inhibited DNMT. Indeed, S-adenosyl-Lhomocysteine (SAH), produced by COMT reaction is a potent inhibitor of DNMT (Fang et al., 2003). On the other hand, EGCG can directly inhibit DNMT through the hydrogen bonds formation with different residues in the catalytic pocket of the enzyme (Lee et al., 2005). Moreover, Fang et al. showed that reactivation of some methylation-silenced genes by EGCG was also demonstrated in human colon cancers and prostate cancer cells (Fang et al., 2003).

#### **5.2 Apoptosis and ATP/ADP ratio**

Oxygen consumption in cells is regulated by a respiratory control system which depends on ADP and Pi. When the amount of ATP is high, the amount of ADP is limited and therefore, use of oxygen declines. In other words, oxygen consumption increases as the need for ATP arises (Valle et al., 2010). ATP generation through oxygen conversion is not a fully efficient process because a percentage of the energy of the electrochemical gradient is lost and not coupled to ATP production (Matsuyama and Reed, 2000). This situation arises due to a phenomenon called 'proton leak' which causes protons to return to the mitochondrial matrix via alternative pathways that by-pass ATP synthase (Brand, 1990; Brown and Brand, 1991; Valle et al., 2010). Lynen suggested that the increased dependence of cancer cells on glycolysis stemmed not from their inability to reduce oxygen, but rather from their inability to synthesize ATP in response to the mitochondrial proton gradient (Lynen, 1951 as cited in (Samudio et al., 2009).

Although, some explanations for the 'proton leak' come from the biophysical properties of the inner membrane, much of the explanation comes from the activities of a family of mitochondrial proteins termed uncoupling proteins (UCPs) (Klingenberg, 1999; Valle et al., 2010). UCPs exploit the gap in pH concentration to transfer the proton through the inner membrane into the matrix where they are released. Consequently, the mitochondrial membrane potential decreases, reduction of O2 via the respiratory chain is no longer linked to ATP synthesis and ATP/ADP exchange is not longer maintained (Vander Heiden et al., 1999). The influence of anthocyanins on ATP/ADP ratio and on UCPs role could be the aim of further studies.

#### **5.3 Apoptosis and oxygen consumption**

138 Colorectal Cancer – From Prevention to Patient Care

One of the "epigenome" processes is DNA methylation, a covalent chemical modication resulting in addition of a methyl (CH3) group at the carbon 5 position of the cytosine ring in CpG dinucleotides (Kanai and Hirohashi, 2007). This process plays important roles in chromatin structure modulation, transcriptional regulation and genomic stability, and is essential for the development of mammals (Ducasse and Brown, 2006; Li, 2002). CpG dinucleotides are not uniformly distributed throughout the human genome, but are often enriched in the promoter regions of genes. Short CpG-rich regions are also called as ''CpG islands'', and these are present in more than 50% of human gene promoters and can lead to gene silencing and proliferation or to affect the metabolic processes associated with energy metabolism. (Link et al., 2010). This mechanism is an enzymatic process mediated by DNA methyltransferases (DNMT): DNMT1, also called a "maintenance methyltransferase", preserves existing methylation patterns following DNA replication; DNMT3a and DNMT3b, on the other hand, serve as *de novo* methyltransferases, which act independently of replication on both strands, altering the epigenetic information content (Yu et al., 2011). Recent studies havedemonstrated that all three DNMTs are overexpressed in several tumor types, including tumors of the colon and rectum, bladder, and kidney. When DNMT1 and DNMT3b are knocked out in colon cancer cell lines, methylation of tumor suppressor genes such as p16 is almost entirely eliminated and the gene is re-expressed (Rhee et al., 2002), as well as it has been established that the inhibition of DNA methyltransferase activity can

It is known that some nutrients like folic acid, B vitamins and SAM (S-adenosylmethionine) and anthocyanins are key components of the methyl-metabolism pathway (Vanzo et al., 2011). Their methyl-donating mechanism can rapidly alter gene expression by modulating the availability of methyl donors as well as DNMT activity (Ross, 2003). There is a growing interest in the role of polyphenols in prevention of DNA methylation. It was demonstrated that epigallocatechin-3-gallate (EGCG), a tea polyphenol, through its methylation exerted by catechol-O-methyltransferase (COMT), indirectly inhibited DNMT. Indeed, S-adenosyl-Lhomocysteine (SAH), produced by COMT reaction is a potent inhibitor of DNMT (Fang et al., 2003). On the other hand, EGCG can directly inhibit DNMT through the hydrogen bonds formation with different residues in the catalytic pocket of the enzyme (Lee et al., 2005). Moreover, Fang et al. showed that reactivation of some methylation-silenced genes by EGCG was also demonstrated in human colon cancers and prostate cancer cells (Fang et al., 2003).

Oxygen consumption in cells is regulated by a respiratory control system which depends on ADP and Pi. When the amount of ATP is high, the amount of ADP is limited and therefore, use of oxygen declines. In other words, oxygen consumption increases as the need for ATP arises (Valle et al., 2010). ATP generation through oxygen conversion is not a fully efficient process because a percentage of the energy of the electrochemical gradient is lost and not coupled to ATP production (Matsuyama and Reed, 2000). This situation arises due to a phenomenon called 'proton leak' which causes protons to return to the mitochondrial matrix via alternative pathways that by-pass ATP synthase (Brand, 1990; Brown and Brand, 1991; Valle et al., 2010). Lynen suggested that the increased dependence of cancer cells on glycolysis stemmed not from their inability to reduce oxygen, but rather from their inability to synthesize ATP in response to the mitochondrial proton gradient (Lynen, 1951 as cited in

strongly inhibit the formation of tumors (Stresemann et al., 2006).

**5.2 Apoptosis and ATP/ADP ratio** 

(Samudio et al., 2009).

Cancer cells seem to show high glycolytic rates even when oxygen is sufficient for oxidative phosphorylation (OXPHOS). This condition leads to a survival benefit of the tumor providing protection from oxidative stress and resulting in apoptosis avoidance (Kondoh et al., 2007a; Kondoh et al., 2007b). The importance of glycolysis in the survival of cancer cells was demonstrated by Bonnet and colleagues. Their experimental approach aimed at inhibiting the anaerobic glycolisys by repressing the activity of pyruvate dehydrogenase kinase (PDK) with dichloroacetate (DCA). PDK acts as a negative modulator of pyruvate dehydrogenase, a gate-keeping mitochondrial enzyme which controls the glucose oxidative fate into the cell. DCA changes the metabolism of cancer cells from the cytoplasm-based glycolysis to the mitochondria- based glucose oxidation. This led to increased ROS production and decreased mitochondrial membrane potential, efflux of pro-apoptotic mediators from mitochondria, and induction of mitochondria-dependent apoptosis only in cancer cells (Bonnet et al., 2007). On the other hand, in the majority of mammalian cells, glycolysis is inhibited by the presence of oxygen, which allows the mitochondria to oxidize pyruvate to CO2 and H2O.

The transcription factor p53 regulates cellular energy metabolism and antioxidant defense mechanisms. Emerging evidence has shown that these two functions of p53 contribute greatly to p53's role in tumor suppression (Bensaad and Vousden, 2007; Matoba et al., 2006; Sablina et al., 2005). Loss of p53 results in decreased oxygen consumption and impaired mitochondrial respiration and promotes a switch to high glucose utilization in aerobic glycolysis in cells (Maddocks and Vousden).

It was shown that p53 regulates the OXPHOS dependence of cell by modulating the assembly of a key complex in the mitochondrial electron chain transport: cytochrome c oxidase (COX) (Ma et al., 2007; Matoba et al., 2006). It was demonstrated, in fact, that in HCT116 cells, p53 controls the expression of SCO 2 (Synthesis of Cytochrome c Oxidase 2). SCO2 is required for the assembly of mitochondrial DNA-encoded COX II subunit (MTCO2 gene) into the COX, so, p53 directly regulates mitochondrial oxygen consumption. p53 mutations in cancer cells induce a loss in SCO2, thereby resulting in a switch from an aerobic mitochondrial respiration to anaerobic glycolysis. p53 induces SCO2 expression to enhance mitochondrial respiration and induces TIGAR expression to slow glycolysis (Won et al., 2011).

The metabolic implications of anthocyanins through the oxidative use of glucose could be appreciated indirectly. In fact, it is known that anthocyanins induce p53 expression (Fimognari et al., 2005; Lo et al., 2007; Renis et al., 2008), but a direct involvement of this compounds on glucose metabolic use it is not yet demonstrated.

Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 141

Baldini, G., Passamonti, S., Lunazzi, G. C., Tiribelli, C., and Sottocasa, G. L. (1986): Cellular

Ballatori, N., Krance, S. M., Marchan, R., and Hammond, C. L. (2009): Plasma membrane

Barry, M. A., Reynolds, J. E., and Eastman, A. (1993): Etoposide-induced apoptosis in human HL-60 cells is associated with intracellular acidification. *Cancer Res* 53, 2349-57. Bensaad, K., and Vousden, K. H. (2007): p53: new roles in metabolism. *Trends Cell Biol* 17,

Bernardo, M. M., and Fridman, R. (2003): TIMP-2 (tissue inhibitor of metalloproteinase-2)

Bjelke, E. (1975): Dietary vitamin a and human lung cancer. *International Journal of Cancer* 15,

Block, G., Patterson, B., and Subar, A. (1992): Fruit, vegetables, and cancer prevention: a

Boateng, J., Verghese, M., Shackelford, L., Walker, L. T., Khatiwada, J., Ogutu, S., Williams,

Bobe, G., Sansbury, L. B., Albert, P. S., Cross, A. J., Kahle, L., Ashby, J., Slattery, M. L., Caan,

Bobe, G., Wang, B., Seeram, N. P., Nair, M. G., and Bourquin, L. D. (2006): Dietary

Borkowski, T., Szymusiak, H., Gliszczynska-Rwiglo, A., Rietjens, I. M., and Tyrakowska, B.

Borst, P., Evers, R., Kool, M., and Wijnholds, J. (1999): The multidrug resistance protein

Bosetti, C., Bravi, F., Talamini, R., Parpinel, M., Gnagnarella, P., Negri, E., Montella, M.,

mice fed suboptimal levels of sulindac. *J Agric Food Chem* 54, 9322-8. Bonnet, S., Archer, S. L., Allalunis-Turner, J., Haromy, A., Beaulieu, C., Thompson, R., Lee,

inhibits cancer growth. *Cancer Cell* 11, 37-51.

dependent. *J Agric Food Chem* 53, 5526-34.

family. *Biochim Biophys Acta* 1461, 347-57.

risk: a study in Italy. *Nutr Cancer* 56, 123-7.

review of the epidemiological evidence. *Nutr Cancer* 18, 1-29.

*Acta* 856, 1-10.

286-91.

561-565.

53.

*Mol Aspects Med* 30, 13-28.

*Biochem J* 374, 739-45.

*Food Chem Toxicol* 45, 725-32.

localization of sulfobromophthalein transport activity in rat liver. *Biochim Biophys* 

glutathione transporters and their roles in cell physiology and pathophysiology.

regulates MMP-2 (matrix metalloproteinase-2) activity in the extracellular environment after pro-MMP-2 activation by MT1 (membrane type 1)-MMP.

D. S., Jones, J., Guyton, M., Asiamah, D., Henderson, F., Grant, L., DeBruce, M., Johnson, A., Washington, S., and Chawan, C. B. (2007): Selected fruits reduce azoxymethane (AOM)-induced aberrant crypt foci (ACF) in Fisher 344 male rats.

B., Paskett, E., Iber, F., Kikendall, J. W., Lance, P., Daston, C., Marshall, J. R., Schatzkin, A., and Lanza, E. (2008): Dietary flavonoids and colorectal adenoma recurrence in the Polyp Prevention Trial. *Cancer Epidemiol Biomarkers Prev* 17, 1344-

anthocyanin-rich tart cherry extract inhibits intestinal tumorigenesis in APC(Min)

C. T., Lopaschuk, G. D., Puttagunta, L., Harry, G., Hashimoto, K., Porter, C. J., Andrade, M. A., Thebaud, B., and Michelakis, E. D. (2007): A mitochondria-K+ channel axis is suppressed in cancer and its normalization promotes apoptosis and

(2005): Radical scavenging capacity of wine anthocyanins is strongly pH-

Lagiou, P., Franceschi, S., and La Vecchia, C. (2006): Flavonoids and prostate cancer

#### **6. Conclusions**

The different observations found in epidemiological studies in comparison to the *in vitro* ones are linked, partly to the relatively low avonoid intake and complexity of metabolism in humans, and partly to the lack of adequate molecular biomarkers for monitoring the earliest stages of disease development in humans (Pierini et al., 2008). Moreover, the relevance of the *in vitro* studies to the *in vivo* situation needs to be confirmed in view of the high concentrations of polyphenols employed in the *in vitro* studies.

All the data recorded about the role of polyphenols and flavonoids has been obtained through the use of classical cell biology and biochemistry methods. Maybe nutrigenomics, that is the study of the effects of foods and food constituents on gene expression could deepen our understanding of these and other phytochemicals (Corthesy-Theulaz et al., 2005; Davis and Hord, 2005; Mariman, 2006).

#### **7. References**


The different observations found in epidemiological studies in comparison to the *in vitro* ones are linked, partly to the relatively low avonoid intake and complexity of metabolism in humans, and partly to the lack of adequate molecular biomarkers for monitoring the earliest stages of disease development in humans (Pierini et al., 2008). Moreover, the relevance of the *in vitro* studies to the *in vivo* situation needs to be confirmed in view of the

All the data recorded about the role of polyphenols and flavonoids has been obtained through the use of classical cell biology and biochemistry methods. Maybe nutrigenomics, that is the study of the effects of foods and food constituents on gene expression could deepen our understanding of these and other phytochemicals (Corthesy-Theulaz et al., 2005;

Acharya, A., Das, I., Chandhok, D., and Saha, T. (2010): Redox regulation in cancer: a double-edged sword with therapeutic potential. *Oxid Med Cell Longev* 3, 23-34. Aggarwal, B. B., and Shishodia, S. (2006): Molecular targets of dietary agents for prevention

AIRC (1997): World Cancer Research Fund/American Institute for Cancer Research. Food,

AIRC (2007): World Cancer Research Fund / American Institute for Cancer Research.Food,

Armstrong, B., and Doll, R. (1975): Environmental factors and cancer incidence and

Arrigo, A. P. (1999): Gene expression and the thiol redox state. *Free Radic Biol Med* 27, 936-44. Aura, A. M. (2005): In vitro digestion models for dietary phenolic compounds: *Department of Chemical Technology*, Helsinki University of Technology (Finland), Espoo. Aura, A. M., Martin-Lopez, P., O'Leary K, A., Williamson, G., Oksman-Caldentey, K. M.,

Aura, A. M., Martin-Lopez, P., O'Leary, K. A., Williamson, G., Oksman-Caldentey, K. M.,

Ávila, M., Hidalgo, M., Sánchez-Moreno, C., Pelaez, C., Requena, T., and de Pascual-Teresa,

Bagchi, D., Sen, C. K., Bagchi, M., and Atalay, M. (2004): Anti-angiogenic, antioxidant, and

Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective. ,

Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective,

mortality in different countries, with special reference to dietary practices. *Int J* 

Poutanen, K., and Santos-Buelga, C. (2005a): In vitro metabolism of anthocyanins

Poutanen, K., and Santos-Buelga, C. (2005b): In vitro metabolism of anthocyanins

S. (2009): Bioconversion of anthocyanin glycosides by Bifidobacteria and

anti-carcinogenic properties of a novel anthocyanin-rich berry extract formula.

high concentrations of polyphenols employed in the *in vitro* studies.

and therapy of cancer. *Biochem Pharmacol* 71, 1397-421.

Davis and Hord, 2005; Mariman, 2006).

Washington, DC: AIRC, 1997.

*Cancer* 15, 617-31.

pp. 216–251, Washington DC: AICR, 2007.

by human gut microflora. *Eur J Nutr* 44, 133-42.

by human gut microflora. *Eur J Nutr* 44, 133-42.

*Biochemistry (Mosc)* 69, 75-80, 1 p preceding 75.

Lactobacillus. *Food Research International* 42, 1453-1461.

**6. Conclusions** 

**7. References** 


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 143

Coates, E. M., Popa, G., Gill, C. I., McCann, M. J., McDougall, G. J., Stewart, D., and

Cole, S. P., Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E., Almquist, K. C., Stewart,

Cooke, D., Schwarz, M., Boocock, D., Winterhalter, P., Steward, W. P., Gescher, A. J., and

Corthesy-Theulaz, I., den Dunnen, J. T., Ferre, P., Geurts, J. M., Muller, M., van Belzen, N.,

Cvorovic, J., Tramer, F., Granzotto, M., Candussio, L., Decorti, G., and Passamonti, S. (2010):

D'Alessandro, A., Pieroni, L., Ronci, M., D'Aguanno, S., Federici, G., and Urbani, A. (2009):

Dallas, N. A., Xia, L., Fan, F., Gray, M. J., Gaur, P., van Buren, G., 2nd, Samuel, S., Kim, M.

Davis, C. D., and Hord, N. G. (2005): Nutritional "omics" technologies for elucidating the

Deneke, S. M., and Fanburg, B. L. (1989): Regulation of cellular glutathione. *Am J Physiol* 257,

Di Pietro, A., Conseil, G., Perez-Victoria, J. M., Dayan, G., Baubichon-Cortay, H., Trompier,

Diaz Vivancos, P., Wolff, T., Markovic, J., Pallardo, F. V., and Foyer, C. H. (2010): A nuclear

Dickinson, D. A., and Forman, H. J. (2002): Cellular glutathione and thiols metabolism.

Dickinson, D. A., Iles, K. E., Watanabe, N., Iwamoto, T., Zhang, H., Krzywanski, D. M., and

Ding, M., Feng, R., Wang, S. Y., Bowman, L., Lu, Y., Qian, Y., Castranova, V., Jiang, B. H.,

Forman, H. J. (2002): 4-hydroxynonenal induces glutamate cysteine ligase through

and Shi, X. (2006): Cyanidin-3-glucoside, a natural product derived from

glutathione cycle within the cell cycle. *Biochem J* 431, 169-78.

effects on in vitro models of colon cancer. *J Carcinog* 6, 4.

nutrition research. *Ann Nutr Metab* 49, 355-65.

cells. *Arch Biochem Biophys* 501, 151-7.

receptor inhibition. *Cancer Res* 69, 1951-7.

*Biochem Pharmacol* 64, 1019-26.

JNK in HBE1 cells. *Free Radic Biol Med* 33, 974.

1650-4.

20.

*Discov* 4, 73-82.

L163-73.

22.

Rowland, I. (2007): Colon-available raspberry polyphenols exhibit anti-cancer

A. J., Kurz, E. U., Duncan, A. M., and Deeley, R. G. (1992): Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. *Science* 258,

Marczylo, T. H. (2006): Effect of cyanidin-3-glucoside and an anthocyanin mixture from bilberry on adenoma development in the ApcMin mouse model of intestinal carcinogenesis--relationship with tissue anthocyanin levels. *Int J Cancer* 119, 2213-

and van Ommen, B. (2005): Nutrigenomics: the impact of biomics technology on

Oxidative stress-based cytotoxicity of delphinidin and cyanidin in colon cancer

Proteasome inhibitors therapeutic strategies for cancer. *Recent Pat Anticancer Drug* 

P., Lim, S. J., and Ellis, L. M. (2009): Chemoresistant colorectal cancer cells, the cancer stem cell phenotype, and increased sensitivity to insulin-like growth factor-I

role(s) of bioactive food components in colon cancer prevention. *J Nutr* 135, 2694-7.

D., Steinfels, E., Jault, J. M., de Wet, H., Maitrejean, M., Comte, G., Boumendjel, A., Mariotte, A. M., Dumontet, C., McIntosh, D. B., Goffeau, A., Castanys, S., Gamarro, F., and Barron, D. (2002): Modulation by flavonoids of cell multidrug resistance mediated by P-glycoprotein and related ABC transporters. *Cell Mol Life Sci* 59, 307-


Brand, M. D. (1990): The proton leak across the mitochondrial inner membrane. *Biochim* 

Briviba, K., Abrahamse, S. L., Pool-Zobel, B. L., and Rechkemmer, G. (2001): Neurotensin-

dietary flavonoid cyanidin but not by its glycosides. *Nutr Cancer* 41, 172-9. Brown, G. C., and Brand, M. D. (1991): On the nature of the mitochondrial proton leak.

Cai, H., Marczylo, T. H., Teller, N., Brown, K., Steward, W. P., Marko, D., and Gescher, A. J.

Candeil, L., Gourdier, I., Peyron, D., Vezzio, N., Copois, V., Bibeau, F., Orsetti, B., Scheffer,

Casado, F. J., Lostao, M. P., Aymerich, I., Larrayoz, I. M., Duflot, S., Rodriguez-Mulero, S.,

Castañeda-Ovando, A., Pacheco-Hernández, M. L., Páez-Hernández, M. E., Rodríguez, J. A.,

Center, M. M., Jemal, A., and Ward, E. (2009): International trends in colorectal cancer

Cha, H. J., Bae, S. K., Lee, H. Y., Lee, O. H., Sato, H., Seiki, M., Park, B. C., and Kim, K. W.

Chan, H. L., Chou, H. C., Duran, M., Gruenewald, J., Waterfield, M. D., Ridley, A., and

Chao, A., Thun, M. J., Connell, C. J., McCullough, M. L., Jacobs, E. J., Flanders, W. D.,

Chen, D., Daniel, K. G., Chen, M. S., Kuhn, D. J., Landis-Piwowar, K. R., and Dou, Q. P.

Chen, P. N., Kuo, W. H., Chiang, C. L., Chiou, H. L., Hsieh, Y. S., and Chu, S. C. (2006): Black

Circu, M. L., and Aw, T. Y. (2010): Reactive oxygen species, cellular redox systems, and

incidence rates. *Cancer Epidemiol Biomarkers Prev* 18, 1688-94.

irinotecan-treated metastases. *Int J Cancer* 109, 848-54.

and EGF-induced metabolic activation of colon carcinoma cells is diminished by

(2010): Anthocyanin-rich red grape extract impedes adenoma development in the Apc(Min) mouse: pharmacodynamic changes and anthocyanin levels in the murine

G. L., Ychou, M., Khan, Q. A., Pommier, Y., Pau, B., Martineau, P., and Del Rio, M. (2004): ABCG2 overexpression in colon cancer cells resistant to SN38 and in

and Pastor-Anglada, M. (2002): Nucleoside transporters in absorptive epithelia. *J* 

and Galán-Vidal, C. A. (2009): Chemical studies of anthocyanins: A review. *Food* 

(1996): Anti-invasive activity of ursolic acid correlates with the reduced expression of matrix metalloproteinase-9 (MMP-9) in HT1080 human fibrosarcoma cells.

Timms, J. F. (2010): Major role of epidermal growth factor receptor and Src kinases in promoting oxidative stress-dependent loss of adhesion and apoptosis in

Rodriguez, C., Sinha, R., and Calle, E. E. (2005): Meat consumption and risk of

(2005): Dietary flavonoids as proteasome inhibitors and apoptosis inducers in

rice anthocyanins inhibit cancer cells invasion via repressions of MMPs and u-PA

*Biophys Acta* 1018, 128-33.

*Biochim Biophys Acta* 1059, 55-62.

biophase. *Eur J Cancer* 46, 811-7.

*Physiol Biochem* 58, 207-16.

*Chemistry* 113, 859-871.

*Cancer Res* 56, 2281-4.

epithelial cells. *J Biol Chem* 285, 4307-18.

colorectal cancer. *JAMA* 293, 172-82.

expression. *Chem Biol Interact* 163, 218-29.

apoptosis. *Free Radic Biol Med* 48, 749-62.

human leukemia cells. *Biochem Pharmacol* 69, 1421-32.


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 145

Forester, S. C., and Waterhouse, A. L. (2008): Identification of Cabernet Sauvignon anthocyanin gut microflora metabolites. *J Agric Food Chem* 56, 9299-304. Fornaro, L., Masi, G., Loupakis, F., Vasile, E., and Falcone, A. (2010): Palliative treatment of unresectable metastatic colorectal cancer. *Expert Opin Pharmacother* 11, 63-77. Fujiwara, M., Okayasu, I., Takemura, T., Tanaka, I., Masuda, R., Furuhata, Y., Noji, M.,

Ghibelli, L., Coppola, S., Rotilio, G., Lafavia, E., Maresca, V., and Ciriolo, M. R. (1995): Non-

Ghibelli, L., Fanelli, C., Rotilio, G., Lafavia, E., Coppola, S., Colussi, C., Civitareale, P., and

Goldberg, R. M., Rothenberg, M. L., Van Cutsem, E., Benson, A. B., 3rd, Blanke, C. D.,

Hagiwara, A., Miyashita, K., Nakanishi, T., Sano, M., Tamano, S., Kadota, T., Koda, T.,

Hagiwara, A., Yoshino, H., Ichihara, T., Kawabe, M., Tamano, S., Aoki, H., Koda, T.,

Hall, A. G. (1999a): Glutathione and the regulation of cell death. *Adv Exp Med Biol* 457, 199-

Hall, A. G. (1999b): Review: The role of glutathione in the regulation of apoptosis. *Eur J Clin* 

Halliwell, B., and Cross, C. E. (1994): Oxygen-derived species: their relation to human disease and environmental stress. *Environ Health Perspect* 102 Suppl 10, 5-12. Hammond, C. L., Marchan, R., Krance, S. M., and Ballatori, N. (2007): Glutathione export

Harborne, J. B., and Williams, C. A. (2000): Advances in flavonoid research since 1992.

Harris, G. K., Gupta, A., Nines, R. G., Kresty, L. A., Habib, S. G., Frankel, W. L., LaPerle, K.,

during apoptosis requires functional multidrug resistance-associated proteins. *J* 

Gallaher, D. D., Schwartz, S. J., and Stoner, G. D. (2001): Effects of lyophilized black

in rats initiated with 1,2-dimethylhydrazine. *J Toxicol Sci* 27, 57-68.

lung adenocarcinomas. *Mod Pathol* 13, 723-9.

with 1,2-dimethylhydrazine. *Cancer Lett* 171, 17-25.

*Commun* 216, 313-20.

extrusion. *FASEB J* 12, 479-86.

*Acad Sci U S A* 93, 654-8.

203.

*Invest* 29, 238-45.

*Biol Chem* 282, 14337-47.

*Phytochemistry* 2000, 481-504.

Oritsu, M., Kato, M., and Oshimura, M. (2000): Telomerase activity significantly correlates with chromosome alterations, cell differentiation, and proliferation in

oxidative loss of glutathione in apoptosis via GSH extrusion. *Biochem Biophys Res* 

Ciriolo, M. R. (1998): Rescue of cells from apoptosis by inhibition of active GSH

Diasio, R. B., Grothey, A., Lenz, H. J., Meropol, N. J., Ramanathan, R. K., Becerra, C. H., Wickham, R., Armstrong, D., and Viele, C. (2007): The continuum of care: a paradigm for the management of metastatic colorectal cancer. *Oncologist* 12, 38-50. Gottlieb, R. A., Nordberg, J., Skowronski, E., and Babior, B. M. (1996): Apoptosis induced in

Jurkat cells by several agents is preceded by intracellular acidification. *Proc Natl* 

Nakamura, M., Imaida, K., Ito, N., and Shirai, T. (2001): Pronounced inhibition by a natural anthocyanin, purple corn color, of 2-amino-1-methyl-6-phenylimidazo[4,5 b]pyridine (PhIP)-associated colorectal carcinogenesis in male F344 rats pretreated

Nakamura, M., Imaida, K., Ito, N., and Shirai, T. (2002): Prevention by natural food anthocyanins, purple sweet potato color and red cabbage color, of 2-amino-1 methyl-6-phenylimidazo[4,5-b]pyridine (PhIP)-associated colorectal carcinogenesis

blackberry, exhibits chemopreventive and chemotherapeutic activity. *J Biol Chem* 281, 17359-68.


Donati, Y. R., Slosman, D. O., and Polla, B. S. (1990): Oxidative injury and the heat shock

Doyle, L. A., Yang, W., Abruzzo, L. V., Krogmann, T., Gao, Y., Rishi, A. K., and Ross, D. D.

Dreiseitel, A., Oosterhuis, B., Vukman, K. V., Schreier, P., Oehme, A., Locher, S., Hajak, G.,

Ducasse, M., and Brown, M. A. (2006): Epigenetic aberrations and cancer. *Mol Cancer* 5, 60. Elias, M. M., Lunazzi, G. C., Passamonti, S., Gazzin, B., Miccio, M., Stanta, G., Sottocasa, G.

plasma membrane of renal proximal tubule in rat. *Am J Physiol* 259, F559-64. Elstrom, R. L., Bauer, D. E., Buzzai, M., Karnauskas, R., Harris, M. H., Plas, D. R., Zhuang,

Estrela, J. M., Ortega, A., and Obrador, E. (2006): Glutathione in cancer biology and therapy.

Fanciulli, M., Bruno, T., Giovannelli, A., Gentile, F. P., Di Padova, M., Rubiu, O., and Floridi,

Felgines, C., Krisa, S., Mauray, A., Besson, C., Lamaison, J. L., Scalbert, A., Merillon, J. M.,

Felgines, C., Texier, O., Besson, C., Vitaglione, P., Lamaison, J. L., Fogliano, V., Scalbert, A.,

Feng, R., Ni, H. M., Wang, S. Y., Tourkova, I. L., Shurin, M. R., Harada, H., and Yin, X. M.

Fleschhut, J., Kratzer, F., Rechkemmer, G., and Kulling, S. E. (2006): Stability and biotransformation of various dietary anthocyanins in vitro. *Eur J Nutr* 45, 7-18. Folkman, J. (1995): Angiogenesis in cancer, vascular, rheumatoid and other disease. *Nat Med*

leukemic cells by induction of oxidative stress. *J Biol Chem* 282, 13468-76. Fimognari, C., Berti, F., Nusse, M., Cantelli-Fortii, G., and Hrelia, P. (2005): In vitro

and non-transformed T lymphocytes. *Anticancer Res* 25, 2837-40.

drug resistance and influence of lonidamine. *Clin Cancer Res* 6, 1590-7. Fang, M. Z., Wang, Y., Ai, N., Hou, Z., Sun, Y., Lu, H., Welsh, W., and Yang, C. S. (2003): Tea

Putative involvement of GLUT2. *Mol Nutr Food Res* 53, 1430-7.

stimulates aerobic glycolysis in cancer cells. *Cancer Res* 64, 3892-9.

281, 17359-68.

response. *Biochem Pharmacol* 40, 2571-7.

*Proc Natl Acad Sci U S A* 95, 15665-70.

*Crit Rev Clin Lab Sci* 43, 143-81.

the mouse. *Br J Nutr* 103, 1738-45.

1, 27-31.

absorption in rats. *Mol Nutr Food Res* 52, 959-64.

blackberry, exhibits chemopreventive and chemotherapeutic activity. *J Biol Chem*

(1998): A multidrug resistance transporter from human MCF-7 breast cancer cells.

and Sand, P. G. (2009): Berry anthocyanins and anthocyanidins exhibit distinct affinities for the efflux transporters BCRP and MDR1. *Br J Pharmacol* 158, 1942-50.

L., and Tiribelli, C. (1990): Bilitranslocase localization and function in basolateral

H., Cinalli, R. M., Alavi, A., Rudin, C. M., and Thompson, C. B. (2004): Akt

A. (2000): Energy metabolism of human LoVo colon carcinoma cells: correlation to

polyphenol (-)-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines. *Cancer Res* 63, 7563-70. Faria, A., Pestana, D., Azevedo, J., Martel, F., de Freitas, V., Azevedo, I., Mateus, N., and

Calhau, C. (2009): Absorption of anthocyanins through intestinal epithelial cells -

and Texier, O. (2010): Radiolabelled cyanidin 3-O-glucoside is poorly absorbed in

Vanella, L., and Galvano, F. (2008): Influence of glucose on cyanidin 3-glucoside

(2007): Cyanidin-3-rutinoside, a natural polyphenol antioxidant, selectively kills

anticancer activity of cyanidin-3-O-beta-glucopyranoside: effects on transformed


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 147

Hwang, C., Sinskey, A. J., and Lodish, H. F. (1992): Oxidized redox state of glutathione in

Jemal, A., Bray, F., Center, M. M., Ferlay, J., Ward, E., and Forman, D. (2011): Global cancer

Johnson, J. L., Bomser, J. A., Scheerens, J. C., and Giusti, M. M. (2011): Effect of black

Kachadourian, R., and Day, B. J. (2006): Flavonoid-induced glutathione depletion: potential

Kanai, Y., and Hirohashi, S. (2007): Alterations of DNA methylation associated with

Kang, S. Y., Seeram, N. P., Nair, M. G., and Bourquin, L. D. (2003): Tart cherry anthocyanins

Katsube, N., Iwashita, K., Tsushida, T., Yamaki, K., and Kobori, M. (2003): Induction of

Kaur, M., Tyagi, A., Singh, R. P., Sclafani, R. A., Agarwal, R., and Agarwal, C. (2011): Grape

Kay, C. D., Kroon, P. A., and Cassidy, A. (2009): The bioactivity of dietary anthocyanins is

Kazi, A., Urbizu, D. A., Kuhn, D. J., Acebo, A. L., Jackson, E. R., Greenfelder, G. P., Kumar,

Keppler, K., and Humpf, H. U. (2005): Metabolism of anthocyanins and their phenolic degradation products by the intestinal microflora. *Bioorg Med Chem* 13, 5195-205. Keshavarzian, A., Zapeda, D., List, T., and Mobarhan, S. (1992): High levels of reactive

Key, T. J. (2011): Fruit and vegetables and cancer risk. *British Journal of Cancer* 104, 6-11. Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho, P. L., Coviello,

Kondoh, H., Lleonart, M. E., Bernard, D., and Gil, J. (2007a): Protection from oxidative stress

not normal and non-transformed, cells. *Int J Mol Med* 12, 879-87.

implications for cancer treatment. *Free Radic Biol Med* 41, 65-76.

a precancerous to a malignant state. *Carcinogenesis* 28, 2434-42.

raspberry ( Rubus occidentalis L.) extract variation conditioned by cultivar, production site, and fruit maturity stage on colon cancer cell proliferation. *J Agric* 

abnormalities of DNA methyltransferases in human cancers during transition from

inhibit tumor development in Apc(Min) mice and reduce proliferation of human

apoptosis in cancer cells by Bilberry (Vaccinium myrtillus) and the anthocyanins. *J* 

seed extract upregulates p21 (Cip1) through redox-mediated activation of ERK1/2 and posttranscriptional regulation leading to cell cycle arrest in colon carcinoma

likely to be mediated by their degradation products. *Mol Nutr Food Res* 53 Suppl 1,

N. B., and Dou, Q. P. (2003): A natural musaceas plant extract inhibits proteasome activity and induces apoptosis selectively in human tumor and transformed, but

oxygen metabolites in colon cancer tissue: analysis by chemiluminescence probe.

G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. (1994): Specific association of human telomerase activity with immortal cells and cancer. *Science* 266, 2011-5. Klingenberg, M. (1999): Uncoupling protein--a useful energy dissipator. *J Bioenerg Biomembr*

by enhanced glycolysis; a possible mechanism of cellular immortalization. *Histol* 

the endoplasmic reticulum. *Science* 257, 1496-502.

statistics. *CA Cancer J Clin* 61, 69-90.

colon cancer cells. *Cancer Lett* 194, 13-9.

HT29 cells. *Mol Carcinog* 50, 553-62.

*Agric Food Chem* 51, 68-75.

*Nutr Cancer* 17, 243-9.

*Histopathol* 22, 85-90.

31, 419-30.

S92-101.

*Food Chem* 59, 1638-45.

raspberries on azoxymethane-induced colon cancer and 8-hydroxy-2' deoxyguanosine levels in the Fischer 344 rat. *Nutr Cancer* 40, 125-33.


Hayes, J. D., Flanagan, J. U., and Jowsey, I. R. (2005): Glutathione transferases. *Annu Rev* 

Hayes, J. D., and McLellan, L. I. (1999): Glutathione and glutathione-dependent enzymes

He, J., Magnuson, B. A., and Giusti, M. M. (2005): Analysis of anthocyanins in rat intestinal

He, J., Wallace, T. C., Keatley, K. E., and Failla, M. L. G., M.M. (2009): Stability of black

gastric and small intestinal tissues in the rat. *J Agric Food Chem.* 57, 3141-8. Herceg, Z. (2007): Epigenetics and cancer: towards an evaluation of the impact of

Hess, G. P., Wang, P. F., Quach, D., Barber, B., and Zhao, Z. (2010): Systemic Therapy for

Hitchler, M. J., and Domann, F. E. (2009): Metabolic defects provide a spark for the

Hiyama, E., Hiyama, K., Yokoyama, T., Matsuura, Y., Piatyszek, M. A., and Shay, J. W.

Hou, D. X. (2003): Potential mechanisms of cancer chemoprevention by anthocyanins. *Curr* 

Hou, D. X., Ose, T., Lin, S., Harazoro, K., Imamura, I., Kubo, M., Uto, T., Terahara, N.,

Hou, D. X., Tong, X., Terahara, N., Luo, D., and Fujii, M. (2005): Delphinidin 3-

Hou, Z., Qin, P., and Ren, G. (2010): Effect of anthocyanin-rich extract from black rice (Oryza

Hsu, P. P., and Sabatini, D. M. (2008): Cancer cell metabolism: Warburg and beyond. *Cell*

Huang, H. P., Shih, Y. W., Chang, Y. C., Hung, C. N., and Wang, C. J. (2008):

involved in the Ras/PI3K pathway. *J Agric Food Chem* 56, 9286-93.

environmental and dietary factors. *Mutagenesis* 22, 91-103.

in US Medical Oncology Practice. *J Oncol Pract* 6, 301-7.

epigenetic switch in cancer. *Free Radic Biol Med* 47, 115-27.

deoxyguanosine levels in the Fischer 344 rat. *Nutr Cancer* 40, 125-33. Hassimotto, N. M. A., Genovese, M. I., and Lajolo, F. M. (2008): Absorption and metabolism

(Morus nigra L.) in rats. *Nutrition Research* 28, 198-207.

*Pharmacol Toxicol* 45, 51-88.

31, 273-300.

*Chem* 53, 2859-66.

*Nat Med* 1, 249-55.

*Mol Med* 3, 149-59.

*Biophys* 440, 101-9.

*Chem* 58, 3191-6.

134, 703-7.

involved. *Int J Oncol* 23, 705-12.

raspberries on azoxymethane-induced colon cancer and 8-hydroxy-2'-

of cyanidin-3-glucoside and cyanidin-3-rutinoside extracted from wild mulberry

represent a co-ordinately regulated defence against oxidative stress. *Free Radic Res*

contents--impact of anthocyanin chemical structure on fecal excretion. *J Agric Food* 

raspberry anthocyanins in the digestive tract lumen and transport efficiency into

Metastatic Colorectal Cancer: Patterns of Chemotherapy and Biologic Therapy Use

(1995): Correlating telomerase activity levels with human neuroblastoma outcomes.

Yoshimoto, M., and Fujii, M. (2003): Anthocyanidins induce apoptosis in human promyelocytic leukemia cells: structure-activity relationship and mechanisms

sambubioside, a Hibiscus anthocyanin, induces apoptosis in human leukemia cells through reactive oxygen species-mediated mitochondrial pathway. *Arch Biochem* 

sativa L. Japonica) on chronically alcohol-induced liver damage in rats. *J Agric Food* 

Chemoinhibitory effect of mulberry anthocyanins on melanoma metastasis


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 149

Ma, W., Sung, H. J., Park, J. Y., Matoba, S., and Hwang, P. M. (2007): A pivotal role for p53: balancing aerobic respiration and glycolysis. *J Bioenerg Biomembr* 39, 243-6. Maddocks, O. D., and Vousden, K. H. Metabolic regulation by p53. *J Mol Med (Berl)* 89, 237-

Madeo, F., Frohlich, E., Ligr, M., Grey, M., Sigrist, S. J., Wolf, D. H., and Frohlich, K. U. (1999): Oxygen stress: a regulator of apoptosis in yeast. *J Cell Biol* 145, 757-67. Magnusson, B. A., Lala, G., and Kwon, Y. J. (2003): Anthocyanin-rich extracts inhibit growth

Malik, M., Zhao, C., Schoene, N., Guisti, M. M., Moyer, M. P., and Magnuson, B. A. (2003):

Manach, C., Scalbert, A., Morand, C., Remesy, C., and Jimenez, L. (2004): Polyphenols: food

Mariman, E. C. (2006): Nutrigenomics and nutrigenetics: the 'omics' revolution in nutritional

Marko, D., Puppel, N., Tjaden, Z., Jakobs, S., and Pahlke, G. (2004): The substitution pattern

Matchett, M. D., MacKinnon, S. L., Sweeney, M. I., Gottschall-Pass, K. T., and Hurta, R. A.

Matchett, M. D., MacKinnon, S. L., Sweeney, M. I., Gottschall-Pass, K. T., and Hurta, R. A.

Mates, J. M., Segura, J. A., Alonso, F. J., and Marquez, J. (2008): Intracellular redox status and

Mathupala, S. P., Rempel, A., and Pedersen, P. L. (2001): Glucose catabolism in cancer cells:

Matoba, S., Kang, J. G., Patino, W. D., Wragg, A., Boehm, M., Gavrilova, O., Hurley, P. J.,

Matsuyama, S., and Reed, J. C. (2000): Mitochondria-dependent apoptosis and cellular pH

McDougall, G. J., Fyffe, S., Dobson, P., and Stewart, D. (2005): Anthocyanins from red wine--

McGhie, T. K., and Walton, M. C. (2007): The bioavailability and absorption of anthocyanins:

hexokinase gene to hypoxic conditions. *J Biol Chem* 276, 43407-12.

towards a better understanding. *Mol Nutr Food Res* 51, 702-13.

colon cancer but not normal colonic cells. *Nutr Cancer* 46, 186-96.

sources and bioavailability. *Am J Clin Nutr* 79, 727-47.

human prostate cancer cells. *Biochem Cell Biol* 83, 637-43.

science. *Biotechnol Appl Biochem* 44, 119-28.

proliferation. *Mol Nutr Food Res* 48, 318-25.

events. *J Nutr Biochem* 17, 117-25.

regulation. *Cell Death Differ* 7, 1155-65.

*Arch Toxicol* 82, 273-99.

312, 1650-3.

of human colon cancer cells and azoxymethane-induced colon aberrant crypts in rats: Implications for colon cancer Chemoprevention. *Cancer Epidemiol Biomark Prev*

Anthocyanin-rich extract from Aronia meloncarpa E induces a cell cycle block in

of anthocyanidins affects different cellular signaling cascades regulating cell

(2005): Blueberry flavonoids inhibit matrix metalloproteinase activity in DU145

(2006): Inhibition of matrix metalloproteinase activity in DU145 human prostate cancer cells by flavonoids from lowbush blueberry (Vaccinium angustifolium): possible roles for protein kinase C and mitogen-activated protein-kinase-mediated

oxidative stress: implications for cell proliferation, apoptosis, and carcinogenesis.

identification and characterization of a marked activation response of the type II

Bunz, F., and Hwang, P. M. (2006): p53 regulates mitochondrial respiration. *Science*

their stability under simulated gastrointestinal digestion. *Phytochemistry* 66, 2540-8.

45.

12, 1323s-1324s.


Kondoh, H., Lleonart, M. E., Nakashima, Y., Yokode, M., Tanaka, M., Bernard, D., Gil, J.,

Kong, J. M., Chia, L. S., Goh, N. K., Chia, T. F., and Brouillard, R. (2003): Analysis and

Labrecque, L., Lamy, S., Chapus, A., Mihoubi, S., Durocher, Y., Cass, B., Bojanowski, M. W.,

Lala, G., Malik, M., Zhao, C., He, J., Kwon, Y., Giusti, M. M., and Magnuson, B. A. (2006):

Lamy, S., Gingras, D., and Beliveau, R. (2002): Green tea catechins inhibit vascular endothelial growth factor receptor phosphorylation. *Cancer Res* 62, 381-5. Lazze, M. C., Savio, M., Pizzala, R., Cazzalini, O., Perucca, P., Scovassi, A. I., Stivala, L. A.,

Lee, W. J., Shim, J. Y., and Zhu, B. T. (2005): Mechanisms for the inhibition of DNA methyltransferases by tea catechins and bioflavonoids. *Mol Pharmacol* 68, 1018-30. Lei, X. G. (2002): In vivo antioxidant role of glutathione peroxidase: evidence from knockout

Leslie, E. M., Deeley, R. G., and Cole, S. P. (2003): Bioflavonoid stimulation of glutathione

Li, E. (2002): Chromatin modification and epigenetic reprogramming in mammalian

Li, J., and Eastman, A. (1995): Apoptosis in an interleukin-2-dependent cytotoxic T

Liang, Y. C., Tsai, S. H., Chen, L., Lin-Shiau, S. Y., and Lin, J. K. (2003): Resveratrol-induced

Lin, M. T., Yen, M. L., Lin, C. Y., and Kuo, M. L. (2003): Inhibition of vascular endothelial

Link, A., Balaguer, F., and Goel, A. (2010): Cancer chemoprevention by dietary polyphenols:

Lo, C. W., Huang, H. P., Lin, H. M., Chien, C. T., and Wang, C. J. (2007): Effect of Hibiscus

activation of P38 MAPK and p53 pathway. *Mol Nutr Food Res* 51, 1452-60. Loe, D. W., Deeley, R. G., and Cole, S. P. (2000): Verapamil stimulates glutathione transport

promising role for epigenetics. *Biochem Pharmacol* 80, 1771-92.

murine embryonic stem cells. *Antioxid Redox Signal* 9, 293-9.

in different human cell lines. *Carcinogenesis* 25, 1427-33.

mice. *Methods Enzymol* 347, 213-25.

development. *Nat Rev Genet* 3, 662-73.

Na(+)/H(+)-antiport. *J Biol Chem* 270, 3203-11.

HT29 cells. *Biochem Pharmacol* 65, 1053-60.

821-6.

31, 11-5.

64, 1029-36.

530-8.

*Cancer* 54, 84-93.

biological activities of anthocyanins. *Phytochemistry* 64, 923-33.

and Beach, D. (2007b): A high glycolytic flux supports the proliferative potential of

Gingras, D., and Beliveau, R. (2005): Combined inhibition of PDGF and VEGF receptors by ellagic acid, a dietary-derived phenolic compound. *Carcinogenesis* 26,

Anthocyanin-rich extracts inhibit multiple biomarkers of colon cancer in rats. *Nutr* 

and Bianchi, L. (2004): Anthocyanins induce cell cycle perturbations and apoptosis

transport by the 190-kDa multidrug resistance protein 1 (MRP1). *Drug Metab Dispos*

lymphocyte cell line is associated with intracellular acidification. Role of the

G2 arrest through the inhibition of CDK7 and p34CDC2 kinases in colon carcinoma

growth factor-induced angiogenesis by resveratrol through interruption of Srcdependent vascular endothelial cadherin tyrosine phosphorylation. *Mol Pharmacol*

anthocyanins-rich extract induces apoptosis of proliferating smooth muscle cell via

by the 190-kDa multidrug resistance protein 1 (MRP1). *J Pharmacol Exp Ther* 293,


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 151

Noguchi, T., Takeda, K., Matsuzawa, A., Saegusa, K., Nakano, H., Gohda, J., Inoue, J., and

Olsson, M. E., Gustavsson, K. E., Andersson, S., Nilsson, A., and Duan, R. D. (2004):

Omura, S., Sasaki, Y., Iwai, Y., and Takeshima, H. (1995): Staurosporine, a potentially

Passamonti, S., Terdoslavich, M., Franca, R., Vanzo, A., Tramer, F., Braidot, E., Petrussa, E.,

Passamonti, S., Vrhovsek, U., and Mattivi, F. (2002): The interaction of anthocyanins with

Passamonti, S., Vrhovsek, U., Terdoslavich, M., Vanzo, A., Cocolo, A., Decorti, G., and

Passamonti, S., Vrhovsek, U., Vanzo, A., and Mattivi, F. (2003b): The stomach as a site for

Pierini, R., Gee, J. M., Belshaw, N. J., and Johnson, I. T. (2008): Flavonoids and intestinal

Prior, R. L., and Wu, X. (2006): Anthocyanins: structural characteristics that result in unique metabolic patterns and biological activities. *Free Radic Res* 40, 1014-28. Pupa, S. M., Menard, S., Forti, S., and Tagliabue, E. (2002): New insights into the role of extracellular matrix during tumor onset and progression. *J Cell Physiol* 192, 259-67. Ramos, A. M., and Aller, P. (2008): Quercetin decreases intracellular GSH content and

Rebollo, A., Gomez, J., Martinez de Aragon, A., Lastres, P., Silva, A., and Perez-Sala, D.

Reen, R. K., Nines, R., and Stoner, G. D. (2006): Modulation of N-nitrosomethylbenzylamine

Renis, M., Calandra, L., Scifo, C., Tomasello, B., Cardile, V., Vanella, L., Bei, R., La Fauci, L.,

Reshkin, S. J., Bellizzi, A., Caldeira, S., Albarani, V., Malanchi, I., Poignee, M., Alunni-

oxidative stress-induced cell death. *J Biol Chem* 280, 37033-40.

correlations with antioxidant levels. *J Agric Food Chem* 52, 7264-71.

important gift from a microorganism. *J Antibiot (Tokyo)* 48, 535-48.

bilitranslocase. *Biochem Biophys Res Commun* 296, 631-6.

anthocyanins absorption from food. *FEBS Letters* 544, 210-213.

cancers. *Br J Nutr* 99 E Suppl 1, ES53-9.

acidification. *Exp Cell Res* 218, 581-5.

leukemia cell lines. *Biochem Pharmacol* 75, 1912-23.

*Drug Metab* 10, 369-94.

France pp. 278.

*Cancer* 54, 47-57.

35.

Ichijo, H. (2005): Recruitment of tumor necrosis factor receptor-associated factor family proteins to apoptosis signal-regulating kinase 1 signalosome is essential for

Inhibition of cancer cell proliferation in vitro by fruit and berry extracts and

and Vianello, A. (2009): Bioavailability of flavonoids: a review of their membrane transport and the function of bilitranslocase in animal and plant organisms. *Curr* 

Mattivi, F. (2003a): Hepatic uptake of dietary anthocyanins and the role of bilitranslocase. 1st International Conference on Polyphenols and Health, Vichy -

potentiates the apoptotic action of the antileukemic drug arsenic trioxide in human

(1995): Apoptosis induced by IL-2 withdrawal is associated with an intracellular

metabolism by black raspberries in the esophagus and liver of Fischer 344 rats. *Nutr* 

and Galvano, F. (2008): Response of cell cycle/stress-related protein expression and DNA damage upon treatment of CaCo2 cells with anthocyanins. *Br J Nutr* 100, 27-

Fabbroni, M., Casavola, V., and Tommasino, M. (2000): Na+/H+ exchangerdependent intracellular alkalinization is an early event in malignant transformation


Meiers, S., Kemeny, M., Weyand, U., Gastpar, R., von Angerer, E., and Marko, D. (2001): The

Mennen, L. I., Sapinho, D., Ito, H., Galan, P., Hercberg, S., and Scalbert, A. (2008): Urinary

Meredith, M. J., and Reed, D. J. (1982): Status of the mitochondrial pool of glutathione in the

Milane, H. A., Al Ahmad, A., Naitchabane, M., Vandamme, T. F., Jung, L., and Ubeaud, G.

Misikangas, M., Pajari, A. M., Paivarinta, E., Oikarinen, S. I., Rajakangas, J., Marttinen, M.,

Miyake, K., Mickley, L., Litman, T., Zhan, Z., Robey, R., Cristensen, B., Brangi, M.,

demonstration of homology to ABC transport genes. *Cancer Res* 59, 8-13. Morris, M. E., and Zhang, S. (2006): Flavonoid-drug interactions: effects of flavonoids on

Naasani, I., Oh-Hashi, F., Oh-Hara, T., Feng, W. Y., Johnston, J., Chan, K., and Tsuruo, T.

Naasani, I., Seimiya, H., and Tsuruo, T. (1998): Telomerase inhibition, telomere shortening,

Nabeshima, K., Inoue, T., Shimao, Y., and Sameshima, T. (2002): Matrix metalloproteinases

Nagase, H., Sasaki, K., Kito, H., Haga, A., and Sato, T. (1998): Inhibitory effect of delphinidin

in tumor invasion: role for cell migration. *Pathol Int* 52, 255-64.

Caco-2 cell monolayer 139. *Eur J Drug Metab Pharmacokinet* 32, 139-47. Miraglia, E., Viarisio, D., Riganti, C., Costamagna, C., Ghigo, D., and Bosia, A. (2005):

Meister, A. (1983): Selective modification of glutathione metabolism. *Science* 220, 472-7. Meister, A. (1991): Glutathione deficiency produced by inhibition of its synthesis, and its reversal; applications in research and therapy. *Pharmacol Ther* 51, 155-94.

Meister, A., and Anderson, M. E. (1983): Glutathione. *Annu Rev Biochem* 52, 711-60.

growth-factor receptor. *J Agric Food Chem* 49, 958-62.

isolated hepatocyte. *J Biol Chem* 257, 3747-53.

62, 519-25.

2285-90.

30.

391-6.

*Planta Med* 64, 216-9.

*Int J Cancer* 115, 924-9.

ABC transporters. *Life Sci* 78, 2116-30.

anthocyanidins cyanidin and delphinidin are potent inhibitors of the epidermal

excretion of 13 dietary flavonoids and phenolic acids in free-living healthy subjects - variability and possible use as biomarkers of polyphenol intake. *Eur J Clin Nutr*

(2007): Transport of quercetin di-sodium salt in the human intestinal epithelial

Na+/H+ exchanger activity is increased in doxorubicin-resistant human colon cancer cells and its modulation modifies the sensitivity of the cells to doxorubicin.

Tanayama, H., Torronen, R., and Mutanen, M. (2007): Three Nordic berries inhibit intestinal tumorigenesis in multiple intestinal neoplasia/+ mice by modulating beta-catenin signaling in the tumor and transcription in the mucosa. *J Nutr* 137,

Greenberger, L., Dean, M., Fojo, T., and Bates, S. E. (1999): Molecular cloning of cDNAs which are highly overexpressed in mitoxantrone-resistant cells:

(2003): Blocking telomerase by dietary polyphenols is a major mechanism for limiting the growth of human cancer cells in vitro and in vivo. *Cancer Res* 63, 824-

and senescence of cancer cells by tea catechins. *Biochem Biophys Res Commun* 249,

from Solanum melongena on human fibrosarcoma HT-1080 invasiveness in vitro.


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 153

Singletary, K. W., Stansbury, M. J., Giusti, M., Van Breemen, R. B., Wallig, M., and Rimando,

Sottocasa, G. L., Lunazzi, G. C., and Tiribelli, C. (1989): Isolation of bilitranslocase, the anion

Srivastava, A., Akoh, C. C., Fischer, J., and Krewer, G. (2007): Effect of anthocyanin fractions

Stoner, G. D. (2009): Foodstuffs for preventing cancer: the preclinical and clinical

Stoner, G. D., Wang, L. S., and Chen, T. (2007): Chemoprevention of esophageal squamous

Stresemann, C., Brueckner, B., Musch, T., Stopper, H., and Lyko, F. (2006): Functional

Suda, I., Ishikawa, F., Hatakeyama, M., Miyawaki, M., Kudo, T., Hirano, K., Ito, A.,

Szajdek, A., and Borowska, E. J. (2008): Bioactive compounds and health-promoting properties of berry fruits: a review. *Plant Foods Hum Nutr* 63, 147-56. Takagaki, N., Sowa, Y., Oki, T., Nakanishi, R., Yogosawa, S., and Sakai, T. (2005): Apigenin

Talavera, S., Felgines, C., Texier, O., Besson, C., Lamaison, J. L., and Remesy, C. (2003):

Talavera, S., Felgines, C., Texier, O., Besson, C., Manach, C., Lamaison, J. L., and Remesy, C.

Thomasset, S., Berry, D. P., Cai, H., West, K., Marczylo, T. H., Marsden, D., Brown, K.,

Tol, J., and Punt, C. J. (2010): Monoclonal antibodies in the treatment of metastatic colorectal

Trachootham, D., Zhou, Y., Zhang, H., Demizu, Y., Chen, Z., Pelicano, H., Chiao, P. J.,

Underiner, T. L., Ruggeri, B., and Gingrich, D. E. (2004): Development of vascular

development of berries. *Cancer Prev Res (Phila)* 2, 187-94.

cell carcinoma. *Toxicol Appl Pharmacol* 224, 337-49.

constituents. *J Agric Food Chem* 51, 7280-6.

enzymes. *J Agric Food Chem* 55, 3180-5.

hepatitis. *Eur J Clin Nutr* 62, 60-7.

*Cancer Prev Res (Phila)* 2, 625-33.

cancer: a review. *Clin Ther* 32, 437-53.

phenylethyl isothiocyanate. *Cancer Cell* 10, 241-52.

agents in cancer therapy. *Curr Med Chem* 11, 731-45.

*Methods Enzymol* 174, 50-7.

*Res* 66, 2794-800.

*Int J Oncol* 26, 185-9.

*Nutr* 133, 4178-4182.

134, 2275-9.

A. (2003): Inhibition of rat mammary tumorigenesis by concord grape juice

transporter from liver plasma membrane for bilirubin and other organic anions.

from selected cultivars of Georgia-grown blueberries on apoptosis and phase II

diversity of DNA methyltransferase inhibitors in human cancer cell lines. *Cancer* 

Yamakawa, O., and Horiuchi, S. (2008): Intake of purple sweet potato beverage affects on serum hepatic biomarker levels of healthy adult men with borderline

induces cell cycle arrest and p21/WAF1 expression in a p53-independent pathway.

Anthocyanins Are Efficiently Absorbed from the Stomach in Anesthetized Rats. *J* 

(2004): Anthocyanins are efficiently absorbed from the small intestine in rats. *J Nutr*

Dennison, A., Garcea, G., Miller, A., Hemingway, D., Steward, W. P., and Gescher, A. J. (2009): Pilot study of oral anthocyanins for colorectal cancer chemoprevention.

Achanta, G., Arlinghaus, R. B., Liu, J., and Huang, P. (2006): Selective killing of oncogenically transformed cells through a ROS-mediated mechanism by beta-

endothelial growth factor receptor (VEGFR) kinase inhibitors as anti-angiogenic

and plays an essential role in the development of subsequent transformationassociated phenotypes. *FASEB J* 14, 2185-97.


Rhee, I., Bachman, K. E., Park, B. H., Jair, K. W., Yen, R. W., Schuebel, K. E., Cui, H.,

Rodriguez-Moranta, F., Salo, J., Arcusa, A., Boadas, J., Pinol, V., Bessa, X., Batiste-Alentorn,

Rose, P., Huang, Q., Ong, C. N., and Whiteman, M. (2005): Broccoli and watercress suppress

Ross, S. A. (2003): Diet and DNA methylation interactions in cancer prevention. *Ann N Y* 

Rossi, M., Garavello, W., Talamini, R., La Vecchia, C., Franceschi, S., Lagiou, P., Zambon, P.,

Sablina, A. A., Budanov, A. V., Ilyinskaya, G. V., Agapova, L. S., Kravchenko, J. E., and

Samudio, I., Fiegl, M., and Andreeff, M. (2009): Mitochondrial uncoupling and the Warburg

Seeram, N. P., Adams, L. S., Hardy, M. L., and Heber, D. (2004): Total cranberry extract

Seeram, N. P., Bourquin, L. D., and Nair, M. G. (2001): Degradation products of cyanidin glycosides from tart cherries and their bioactivities. *J Agric Food Chem* 49, 4924-9. Selma, M. V., Espin, J. C., and Tomas-Barberan, F. A. (2009): Interaction between phenolics and gut microbiota: role in human health. *J Agric Food Chem* 57, 6485-501. Shih, P. H., Yeh, C. T., and Yen, G. C. (2005): Effects of anthocyanidin on the inhibition of

Shih, P. H., Yeh, C. T., and Yen, G. C. (2007): Anthocyanins induce the activation of phase II

Shin, D. Y., Lu, J. N., Kim, G. Y., Jung, J. M., Kang, H. S., Lee, W. S., and Choi, Y. H. (2011):

Sies, H. (1999): Glutathione and its role in cellular functions. *Free Radic Biol Med* 27, 916-21.

against human tumor cell lines. *J Agric Food Chem* 52, 2512-7.

stress-induced apoptosis. *J Agric Food Chem* 55, 9427-35.

associated phenotypes. *FASEB J* 14, 2185-97.

breast cancer cells. *Toxicol Appl Pharmacol* 209, 105-13.

esophageal cancer. *Int J Cancer* 120, 1560-4.

552-6.

*Oncol* 24, 386-93.

*Acad Sci* 983, 197-207.

*Med* 11, 1306-13.

*Food Chem Toxicol* 43, 1557-66.

carcinoma cells. *Oncol Rep* 25, 567-72.

69, 2163-6.

and plays an essential role in the development of subsequent transformation-

Feinberg, A. P., Lengauer, C., Kinzler, K. W., Baylin, S. B., and Vogelstein, B. (2002): DNMT1 and DNMT3b cooperate to silence genes in human cancer cells. *Nature* 416,

E., Lacy, A. M., Delgado, S., Maurel, J., Pique, J. M., and Castells, A. (2006): Postoperative surveillance in patients with colorectal cancer who have undergone curative resection: a prospective, multicenter, randomized, controlled trial. *J Clin* 

matrix metalloproteinase-9 activity and invasiveness of human MDA-MB-231

Dal Maso, L., Bosetti, C., and Negri, E. (2007): Flavonoids and risk of squamous cell

Chumakov, P. M. (2005): The antioxidant function of the p53 tumor suppressor. *Nat* 

effect: molecular basis for the reprogramming of cancer cell metabolism. *Cancer Res*

versus its phytochemical constituents: antiproliferative and synergistic effects

proliferation and induction of apoptosis in human gastric adenocarcinoma cells.

enzymes through the antioxidant response element pathway against oxidative

Anti-invasive activities of anthocyanins through modulation of tight junctions and suppression of matrix metalloproteinase activities in HCT-116 human colon


Dietary Anthocyanins: Impact on Colorectal Cancer and Mechanisms of Action 155

Wang, S. Y., and Jiao, H. (2000): Scavenging capacity of berry crops on superoxide radicals,

Weisel, T., Baum, M., Eisenbrand, G., Dietrich, H., Will, F., Stockis, J. P., Kulling, S., Rufer,

Wenzel, U., Nickel, A., and Daniel, H. (2005): Increased mitochondrial

Wolffram, S., Block, M., and Ader, P. (2002): Quercetin-3-glucoside is transported by the

Wolter, F., Akoglu, B., Clausnitzer, A., and Stein, J. (2001): Downregulation of the cyclin

Won, K. Y., Lim, S. J., Kim, G. Y., Kim, Y. W., Han, S. A., Song, J. Y., and Lee, D. K. (2011):

Wu, G., Fang, Y. Z., Yang, S., Lupton, J. R., and Turner, N. D. (2004): Glutathione

Wu, Q. K., Koponen, J. M., Mykkanen, H. M., and Torronen, A. R. (2007): Berry phenolic

Yao, H., Xu, W., Shi, X., and Zhang, Z. (2011): Dietary flavonoids as cancer prevention agents. *J Environ Sci Health C Environ Carcinog Ecotoxicol Rev* 29, 1-31. Yasuda, S., Fujii, H., Nakahara, T., Nishiumi, N., Takahashi, W., Ide, M., and Shohtsu, A. (2001): 18F-FDG PET detection of colonic adenomas. *J Nucl Med* 42, 989-92. Yasuzawa, T., Iida, T., Yoshida, M., Hirayama, N., Takahashi, M., Shirahata, K., and Sano, H.

Yu, N. K., Baek, S. H., and Kaang, B. K. (2011): DNA methylation-mediated control of

Yun, J. W., Lee, W. S., Kim, M. J., Lu, J. N., Kang, M. H., Kim, H. G., Kim, D. C., Choi, E. J.,

Zhang, Y., Vareed, S. K., and Nair, M. G. (2005): Human tumor cell growth inhibition by

Warburg, O. (1956a): On respiratory impairment in cancer cells. *Science* 124, 269-70.

and apoptosis in colon cancer cells. *Cell Mol Life Sci* 62, 3100-5.

metabolism and its implications for health. *J Nutr* 134, 489-92.

colon cancer cells. *J Agric Food Chem* 55, 1156-63.

Warburg, O. (1956b): On the origin of cancer cells. *Science* 123, 309-14.

probands. *Biotechnol J* 1, 388-97.

cancer cell lines. *J Nutr* 131, 2197-203.

5677-84.

*Nutr* 132, 630-5.

cancer. *Hum Pathol*.

*Antibiot (Tokyo)* 39, 1072-8.

learning and memory. *Mol Brain* 4, 5.

cells. *Food Chem Toxicol* 48, 903-9.

hydrogen peroxide, hydroxyl radicals, and singlet oxygen. *J Agric Food Chem* 48,

C., Johannes, C., and Janzowski, C. (2006): An anthocyanin/polyphenolic-rich fruit juice reduces oxidative DNA damage and increases glutathione level in healthy

palmitoylcarnitine/carnitine countertransport by flavone causes oxidative stress

glucose carrier SGLT1 across the brush border membrane of rat small intestine. *J* 

D1/Cdk4 complex occurs during resveratrol-induced cell cycle arrest in colon

Regulatory role of p53 in cancer metabolism via SCO2 and TIGAR in human breast

extracts modulate the expression of p21(WAF1) and Bax but not Bcl-2 in HT-29

(1986): The structures of the novel protein kinase C inhibitors K-252a, b, c and d. *J* 

Choi, J. Y., Lee, Y. K., Ryu, C. H., Kim, G., Choi, Y. H., Park, O. J., and Shin, S. C. (2010): Characterization of a profile of the anthocyanins isolated from Vitis coignetiae Pulliat and their anti-invasive activity on HT-29 human colon cancer

nontoxic anthocyanidins, the pigments in fruits and vegetables. *Life Sci* 76, 1465-72.


Valle, A., Oliver, J., and Roca, P. (2010): Role of Uncoupling Proteins in Cancer. *Cancers* 2,

Van Cutsem, E., Kohne, C. H., Hitre, E., Zaluski, J., Chang Chien, C. R., Makhson, A.,

van Duijnhoven, F. J. B., Bueno-De-Mesquita, H. B., Ferrari, P., Jenab, M., Boshuizen, H. C.,

Investigation into Cancer and Nutrition. *Am J Clin Nutr* 89, 1441-52. Vander Heiden, M. G., Chandel, N. S., Schumacker, P. T., and Thompson, C. B. (1999): Bcl-

mitochondrial ATP/ADP exchange. *Mol Cell* 3, 159-67.

Skeels extract. *Toxicol In Vitro* 22, 1440-6.

*Cancer Genet Cytogenet* 203, 1-6.

glucosides. *J Nutr* 137, 2043-8.

Advances in Colon Cancer Research

*Biotechnol J* 2, 1214-34.

*Cancer Lett* 269, 281-90.

D'Haens, G., Pinter, T., Lim, R., Bodoky, G., Roh, J. K., Folprecht, G., Ruff, P., Stroh, C., Tejpar, S., Schlichting, M., Nippgen, J., and Rougier, P. (2009): Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. *N Engl J Med* 360,

Ros, M. M., Casagrande, C., Tjønneland, A., Olsen, A., Overvad, K., Thorlacius-Ussing, O., Clavel-Chapelon, F., Boutron-Ruault, M. C., Morois, S., Kaaks, R., Linseisen, J., Boeing, H., Nothlings, U., Trichopoulou, A., Trichopoulos, D., Misirli, G., Palli, D., Sieri, S., Panico, S., Tumino, R., Vineis, P., Peeters, P. H. M., van Gils, C. H., Ocke, M. C., Lund, E., Engeset, D., Skeie, G., Rodrguez Suarez, L., Gonzalez, C. A., Sanchez, M. J., Dorronsoro, M., Navarro, C., Barricarte, A., Berglund, G., Manjer, J., Hallmans, G., Palmqvist, R., Bingham, S. A., Khaw, K. T., Key, T. J., Allen, N. E., Boffetta, P., Slimani, N., Rinaldi, S., Gallo, V., Norat, T., and Riboli, E. (2009): Fruit, vegetables, and colorectal cancer risk: the European Prospective

xL prevents cell death following growth factor withdrawal by facilitating

uptake and metabolism of cyanidin 3-glucoside by rat kidneys and liver. *J Nat Prod*

tetrachloride induced toxicity in isolated rat hepatocytes by Syzygium cumini

Geurts van Kessel, A. (2010): The epigenetics of (hereditary) colorectal cancer.

(2007): Biotechnology of flavonoids and other phenylpropanoid-derived natural products. Part I: Chemical diversity, impacts on plant biology and human health.

Fogliano, V. (2007): Protocatechuic acid is the major human metabolite of cyanidin-

Lechner, J. F., and Stoner, G. D. (2007): Effect of freeze-dried black raspberries on human colorectal cancer lesions. AACR Special Conference in Cancer Research,

Vanzo, A., Vrhovsek, U., Tramer, F., Mattivi, F., and Passamonti, S. (2011): Exceptionally fast

Veigas, J. M., Shrivasthava, R., and Neelwarne, B. (2008): Efficient amelioration of carbon

Venkatachalam, R., Ligtenberg, M. J., Hoogerbrugge, N., de Bruijn, D. R., Kuiper, R. P., and

Ververidis, F., Trantas, E., Douglas, C., Vollmer, G., Kretzschmar, G., and Panopoulos, N.

Vitaglione, P., Donnarumma, G., Napolitano, A., Galvano, F., Gallo, A., Scalfi, L., and

Wang, L. S., Sardo, C., Rocha, C. M., McIntyre, C. M., Frankel, W., Arnold, M., Martin, E.,

Wang, L. S., and Stoner, G. D. (2008): Anthocyanins and their role in cancer prevention.

567-591.

1408-17.

74, 1049-54.


**8** 

*Brazil* 

**Polyunsaturated Fatty Acids,** 

Vera Lucia Flor Silveira1,2

Karina Vieira de Barros1, Ana Paula Cassulino1 and

**Ulcerative Colitis and Cancer Prevention** 

*1Department of Physiology, Federal University of São Paulo, São Paulo, SP, 2Department of Biological Sciences, Federal University of São Paulo, Diadema, SP,* 

Fatty acids (FA) – lipid constituents – are carboxylic acids that can be represented by the form RCO2H. Most often, the group R is a long carbon chain, unbranched and with an even number of carbon atoms and may be saturated or contain one (monounsaturated) or more double bonds (polyunsaturated) (Calder et al. 2002). Fatty acids are often referred to by their common names, but they are correctly identified by a systematic nomenclature. This nomenclature indicates first the number of carbon atoms in the hydrocarbon chain, followed by the number of double bonds, and the position of the first double bond from the terminal methyl group, which is indicated by n-9, n-7, n-6 or n -3 (Figure 1). There are two main families of polyunsaturated FA (PUFA), n-6 (or w-6) and of n-3 (or w-3) (Curi et al. 2002).

Fig. 1. Structure of some fatty acids (Sala-Vila et al. 2008).

**1. Introduction** 


### **Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention**

Karina Vieira de Barros1, Ana Paula Cassulino1 and Vera Lucia Flor Silveira1,2

*1Department of Physiology, Federal University of São Paulo, São Paulo, SP, 2Department of Biological Sciences, Federal University of São Paulo, Diadema, SP, Brazil* 

#### **1. Introduction**

156 Colorectal Cancer – From Prevention to Patient Care

Zhao, C., Giusti, M. M., Malik, M., Moyer, M. P., and Magnuson, B. A. (2004): Effects of

Ziberna, L., Lunder, M., Tramer, F., Drevensek, G., and Passamonti, S. (2011): The

Zu, X. Y., Zhang, Z. Y., Zhang, X. W., Yoshioka, M., Yang, Y. N., and Li, J. (2010):

vasodilation induced by anthocyanins. *Nutr Metab Cardiovasc Dis*.

colonic cell growth. *J Agric Food Chem* 52, 6122-8.

commercial anthocyanin-rich extracts on colonic cancer and nontumorigenic

endothelial plasma membrane transporter bilitranslocase mediates rat aortic

Anthocyanins extracted from Chinese blueberry (Vaccinium uliginosum L.) and its anticancer effects on DLD-1 and COLO205 cells. *Chin Med J (Engl)* 123, 2714-9.

> Fatty acids (FA) – lipid constituents – are carboxylic acids that can be represented by the form RCO2H. Most often, the group R is a long carbon chain, unbranched and with an even number of carbon atoms and may be saturated or contain one (monounsaturated) or more double bonds (polyunsaturated) (Calder et al. 2002). Fatty acids are often referred to by their common names, but they are correctly identified by a systematic nomenclature. This nomenclature indicates first the number of carbon atoms in the hydrocarbon chain, followed by the number of double bonds, and the position of the first double bond from the terminal methyl group, which is indicated by n-9, n-7, n-6 or n -3 (Figure 1). There are two main families of polyunsaturated FA (PUFA), n-6 (or w-6) and of n-3 (or w-3) (Curi et al. 2002).

Fig. 1. Structure of some fatty acids (Sala-Vila et al. 2008).

Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 159

FDA (Food and Drug Administration), in 1982, approving the supplementation of

parenteral nutrition with PUFA n-6 (Holman et al. 1998).

Fig. 2. Biosynthesis of some fatty acids (Sala-Vila et al. 2008)

diseases (Kinsella et al. 1990, Serhan et al. 2004).

The relationship between inflammatory response and PUFA enriched diets has been investigated in recent years. Several studies show that PUFA can modify immunological and inflammatory reactions, and that it can be used as a complementary therapy in chronic

Inflammation is a body's response to tissue injury, which can be triggered by mechanical stimuli, chemical or microbial invasion, as well as hypersensitivity reactions. This response includes complex processes that involve the immune system cells and biological mediators (Rankin et al. 2004). The acute phase response is characterised by increased blood flow and vascular permeability, increased accumulation of fluid, leukocytes and inflammatory mediators; meanwhile the chronic phase is characterised by the development of specific cellular and humoral immune responses against pathogens present at the site of injury (Saadi et al. 2002). Inflammation is characterised by redness, swelling, heat and pain. These signs occur primarily due to: vasodilatation, which allows increased blood flow to the

**2. Inflammation and PUFA** 

Triacylglycerols (TAG), formed by three FA esterified to glycerol, are the main form of fat present in the human diet. TAG of animal origin are rich in saturated fatty acids and are characterised by being solid at ambient temperature (fats), while those of vegetable origin are rich in unsaturated fatty acids and liquid at room temperature (oils). TAG act as reserve lipids found in the form of oily microdroplets, emulsified in the cytosol (Lanning 1993). In addition to TAG, other lipids are present in small amounts in the diet, such as phospholipids, cholesterol, cholesterol esters and traces of free FA. Phospholipids are the major lipid components of the cell membrane, acting as structural elements, precursors of second messengers, and affecting the activity of some enzymes, such as phospholipase A2 and protein kinase C. Thus lipids, besides being a source of energy (immediate or reserve), act as key components of our body, both in terms of structure (cellular constituents) and function (Burr & Burr 1929, 1930).

Mammals synthesise saturated fatty acids from non-lipid precursors and unsaturated n-9 series and n-7; normally the diet provides adequate amounts of these fatty acids. However, the cell membrane also needs unsaturated FA of n-3 and n-6 families to maintain their structure, fluidity and function measures. As mammals lack the enzyme delta-12 desaturase and delta-15 (found in most plants), which insert double bonds at positions 3 and 6, they do not synthesise n-3 or n-6 PUFA. As such, these FA have to be consumed in the diet and are therefore called essential fatty acids (Semplecine & Valle 1994, Burr & Burr 1929).

The PUFA most commonly consumed are linoleic acid (LA, 18:2 n-6) and -linolenic acid (ALA, 18:3 n-3). These two FA can be converted to other unsaturated derivatives. Linoleic acid can be converted to -Lilolênico (18:3 n-6), Dihomo--linolenic (20:3 n-6) and arachidonic acid (AA, 20:4 n-6) sequentially. Similarly, the -linolenic acid (18:3 n-3) is converted into eicosapentaenoic acid (EPA, 20:5 n-3) and Docosapentaenoico acid (DHA, 22:5 n-3) (Calder 2003) (Figure 2). The main dietary sources of acids LA and ALA are oils which are rich in polyunsaturated fats. The PUFA of n-6 series are derived from plants found, for example, in soybean, sunflower and evening primrose oils. The PUFA of n-3 series are predominantly found in fish oils and marine mammals, and deep cold water fish, such as mackerel, sardines, trout, salmon and tuna (Connor 1996). This occurs because many marine plants, especially phytoplankton algae, also synthesize EPA and DHA from-linolenic acid-. This synthesis of long-chain PUFA n-3 by marine algae, and their transfer through the food chain to fish, explains their abundance in some fish oils and marine mammals (Semplecine & Valle 1994).

Up until 1929, the FAs were viewed exclusively as efficient energy storage. Between 1929 and 1930, George and Mildred Burr published articles reporting the essentiality of PUFA. The authors found that the administration of diets completely devoid of fat in rats caused severe changes in relation to growth and the physiological functions of various organs, which were attributed to the lack of long-chain PUFA. Similar changes were observed in newborns undergoing a diet based on skimmed milk and then reversed by the administration of whole milk. These findings led to a systematic study being carried out by Hensen et al. In 1958, it was found that the administration of skimmed milk to infants was associated with diarrhoea and skin abnormalities, among other things. The supplementation of milk with linoleic acid reversed all symptoms. These observations therefore characterise the effects of PUFA deficiency in humans (Hensen et al. 1958, Holman et al. 1998). With the development of parenteral nutrition, which initially did not contain essential fatty acids, it became evident that a deficiency of n-type PUFA-6 caused the death of patients. This led the

Triacylglycerols (TAG), formed by three FA esterified to glycerol, are the main form of fat present in the human diet. TAG of animal origin are rich in saturated fatty acids and are characterised by being solid at ambient temperature (fats), while those of vegetable origin are rich in unsaturated fatty acids and liquid at room temperature (oils). TAG act as reserve lipids found in the form of oily microdroplets, emulsified in the cytosol (Lanning 1993). In addition to TAG, other lipids are present in small amounts in the diet, such as phospholipids, cholesterol, cholesterol esters and traces of free FA. Phospholipids are the major lipid components of the cell membrane, acting as structural elements, precursors of second messengers, and affecting the activity of some enzymes, such as phospholipase A2 and protein kinase C. Thus lipids, besides being a source of energy (immediate or reserve), act as key components of our body, both in terms of structure (cellular constituents) and

Mammals synthesise saturated fatty acids from non-lipid precursors and unsaturated n-9 series and n-7; normally the diet provides adequate amounts of these fatty acids. However, the cell membrane also needs unsaturated FA of n-3 and n-6 families to maintain their structure, fluidity and function measures. As mammals lack the enzyme delta-12 desaturase and delta-15 (found in most plants), which insert double bonds at positions 3 and 6, they do not synthesise n-3 or n-6 PUFA. As such, these FA have to be consumed in the diet and are

The PUFA most commonly consumed are linoleic acid (LA, 18:2 n-6) and -linolenic acid (ALA, 18:3 n-3). These two FA can be converted to other unsaturated derivatives. Linoleic acid can be converted to -Lilolênico (18:3 n-6), Dihomo--linolenic (20:3 n-6) and arachidonic acid (AA, 20:4 n-6) sequentially. Similarly, the -linolenic acid (18:3 n-3) is converted into eicosapentaenoic acid (EPA, 20:5 n-3) and Docosapentaenoico acid (DHA, 22:5 n-3) (Calder 2003) (Figure 2). The main dietary sources of acids LA and ALA are oils which are rich in polyunsaturated fats. The PUFA of n-6 series are derived from plants found, for example, in soybean, sunflower and evening primrose oils. The PUFA of n-3 series are predominantly found in fish oils and marine mammals, and deep cold water fish, such as mackerel, sardines, trout, salmon and tuna (Connor 1996). This occurs because many marine plants, especially phytoplankton algae, also synthesize EPA and DHA from-linolenic acid-. This synthesis of long-chain PUFA n-3 by marine algae, and their transfer through the food chain to fish, explains their abundance in some fish oils and marine mammals

Up until 1929, the FAs were viewed exclusively as efficient energy storage. Between 1929 and 1930, George and Mildred Burr published articles reporting the essentiality of PUFA. The authors found that the administration of diets completely devoid of fat in rats caused severe changes in relation to growth and the physiological functions of various organs, which were attributed to the lack of long-chain PUFA. Similar changes were observed in newborns undergoing a diet based on skimmed milk and then reversed by the administration of whole milk. These findings led to a systematic study being carried out by Hensen et al. In 1958, it was found that the administration of skimmed milk to infants was associated with diarrhoea and skin abnormalities, among other things. The supplementation of milk with linoleic acid reversed all symptoms. These observations therefore characterise the effects of PUFA deficiency in humans (Hensen et al. 1958, Holman et al. 1998). With the development of parenteral nutrition, which initially did not contain essential fatty acids, it became evident that a deficiency of n-type PUFA-6 caused the death of patients. This led the

therefore called essential fatty acids (Semplecine & Valle 1994, Burr & Burr 1929).

function (Burr & Burr 1929, 1930).

(Semplecine & Valle 1994).

FDA (Food and Drug Administration), in 1982, approving the supplementation of parenteral nutrition with PUFA n-6 (Holman et al. 1998).

Fig. 2. Biosynthesis of some fatty acids (Sala-Vila et al. 2008)

#### **2. Inflammation and PUFA**

The relationship between inflammatory response and PUFA enriched diets has been investigated in recent years. Several studies show that PUFA can modify immunological and inflammatory reactions, and that it can be used as a complementary therapy in chronic diseases (Kinsella et al. 1990, Serhan et al. 2004).

Inflammation is a body's response to tissue injury, which can be triggered by mechanical stimuli, chemical or microbial invasion, as well as hypersensitivity reactions. This response includes complex processes that involve the immune system cells and biological mediators (Rankin et al. 2004). The acute phase response is characterised by increased blood flow and vascular permeability, increased accumulation of fluid, leukocytes and inflammatory mediators; meanwhile the chronic phase is characterised by the development of specific cellular and humoral immune responses against pathogens present at the site of injury (Saadi et al. 2002). Inflammation is characterised by redness, swelling, heat and pain. These signs occur primarily due to: vasodilatation, which allows increased blood flow to the

Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 161

nutritional therapy in their treatment (Ferguson et al. 2007, Pizato et al. 2005, Razack et al. 2007). Malnutrition is common in these patients, and interventions through adequate nutritional therapy so as to restore the nutritional status have been associated with an improved recovery process involving the improvement of the immune system during periods of the exacerbation of the disease (Razack et al. 2007). Several characteristics contribute to the malnutrition observed in patients: 1) there is a decrease in the oral intake of nutrients associated with abdominal pain and anorexia; 2) the mucosal inflammation associated with diarrhoea leads to a loss of protein, minerals, blood, electrolytes and trace elements. In addition, multiple resections or bacterial overgrowth in the colon can cause adverse effects, such as the poor nutritional absorption of micronutrients: 3) drug therapies can lead to malnutrition. For example, sulfasalazine reduces the absorption of folic acid, and corticosteroids reduce calcium absorption and adversely affect the protein metabolism (Wild

Although much progress has been made in understanding IBD, its aetiology is not fully elucidated. However, it is believed that there is involvement of immune factors, both genetic and environmental (Laroux et al. 2001, Cheon et al. 2006, Sainathan et al. 2008). Some studies have suggested that IBDs represent an inappropriate and exaggerated response of the intestinal mucosal immune system to normal intestinal microflora – in genetically susceptible individuals – which can be attributed in part to an imbalance between effector T cells (T eff) cells and T regulatory cells (T reg). (Sanchez-Muñoz et al. 2008, Ma et al. 2007). Effector T cells are helper T lymphocytes (lymph CD4 +) and cytolytic T lymphocytes (lymph CD8 +) that are activated during the adaptive or acquired immune response. The helper T cells secrete cytokines, whose function is to stimulate the proliferation and differentiation of T cells, as well as other cells including B lymphocytes, macrophages and other leukocytes (Sanchez-Muñoz et al. 2008, Sainathan et al. 2008). Cytolytic T lymphocytes destroy cells that produce antigens, such as cells infected by viruses or other intracellular microbes. Since regulatory T cells are cells capable of blocking the activation and effector function of T lymphocytes (Abbas & Lichtman 2005), some studies indicate that the suppressive action of these cells is linked to the secretion of immunosuppressive cytokines, such as IL-10 and Transforming Growth Factor Beta (TGF-β). TGF-β inhibits the proliferation of T and B cells, whereas IL-10 inhibits macrophage activation and is the main antagonist of Macrophage Activating Factor and Interferon Gamma (IFN-γ) (Sanchez-

The innate immune response in IBDs also plays an important role. This response is the first line of defence of the immune system, attended by phagocytic cells, natural killer cells, blood proteins, and including fractions of complements and other mediators of inflammation such as cytokines (Abbas & Lichtman 2005). Cytokines are polypeptides – produced mainly by immune cells – that facilitate communication between cells, stimulate the proliferation of antigen-specific effector cells, and mediate systemic inflammation and local roads in the endocrine, paracrine and autocrine (Muños-Sanchez et al. 2008). Dendritic cells and activated macrophages secrete various cytokines that regulate the inflammatory response. Once secreted, these cytokines promote the differentiation of T cells, activating the adaptive immune response (Abbas & Lichtman 2005). The T-helper cells or CD4 + T cells can differentiate into subpopulations of effector T cells that produce different sets of cytokines and, therefore, play different effector functions. The most well-defined subpopulations of effector T cells are T helper cells type 1 (Th1) and type 2 (Th2) (Abbas &

et al. 2007).

Muñoz et al. 2008).

affected area; increased vascular permeability, which facilitates the diffusion of molecules such as antibodies; cytokines and other plasma proteins to the site of injury and cellular infiltration, which occur through chemotaxis and diapedesis and the direct movement of inflammatory cells through the vessel wall towards the site of inflammation. In addition, during the inflammatory response catabolic and metabolic changes may occur, as well as biosynthetic activation in various organs and enzyme systems and cells of the immune system.

The inflammatory response begins the process of immune elimination of invading pathogens and toxins for the repair of damaged tissue (Rang & Dale 1995). The nonspecific inflammatory response can be seen, for example, in the phagocytosis of bacteria or leftover tissue, the secretion of proteolytic enzymes, the production of reactive oxygen species and the secretion of molecular modulators. It can also be immune-mediated, where there is the participation of lymphocytes and antigen-presenting cells. This second type is closely associated with the onset and maintenance of chronic inflammation (Pompei et al. 1999).

The inflammatory process is controlled by cellular and molecular components. Among the cellular components are neutrophils, monocytes, lymphocytes, macrophages fixed, dendritic cells, mast cells and eosinophils. These cells accumulate in inflamed tissues and interact with the endothelial cells of the microcirculation. Different adhesion molecules participate in these interactions, including selectins, integrins and intercellular adhesion molecules (ICAM) (Rang & Dale 1995). Neutrophils constitute 60% of circulating leukocytes and act as the first line of cellular defence, and they may participate in both reactions as a nonspecific defence and as specific antigen reactions (Curi et al. 1997). Monocytes represent approximately 3-6% of circulating leukocytes in human blood, and they migrate to different tissues where they differentiate into macrophages in response to different stimuli. These cells participate in a variety of functions related to the host's defence, the most well known being the phagocytosis of microorganisms and cell debris, and cytotoxic activity against microorganisms, virus-infected cells and tumour cells (Curi et al. 2002).

The molecular components of inflammation include vasoactive substances (kinins, histamine), proinflammatory cytokines (such as Tumour Necrose Factor (TNF), Interleukin (IL)-1 and IL-6), anti-inflammatory cytokines (such as IL-4, IL-10 and IL-13), chemokines, acute phase proteins, bioactive lipids (such as eicosanoids derived from AA), Platelet Activating Factor, diacylglycerol, ceramides, cAMP, and inositol triphosphate, amongst others.

#### **3. Inflammatory bowel disease and carcinogenesis**

Inflammatory bowel diseases (IBDs) are chronic disorders of the gastrointestinal (GI), which generally refer to two conditions, namely ulcerative colitis and Crohn's disease (Galvez et al. 2006). IBDs are characterised by chronic diarrhoea, malabsorption, mucosal barrier dysfunction and inflammatory intestinal process, being incurable clinically (Benedetti & Plum 1996). Ulcerative colitis encompasses a spectrum of diffuse inflammation and the continuous surface of the colon, which begins in the rectum and may extend to the proximal level. Crohn's disease is characterised by transmural inflammation affecting any asymmetric portion of the GI tract, from the mouth to the anus (Benedetti & Plum 1996).

IBDs cause nutritional deficiencies, such as calorie and protein malnutrition, and deficiencies in vitamins, minerals and trace elements. This underscores the importance of

affected area; increased vascular permeability, which facilitates the diffusion of molecules such as antibodies; cytokines and other plasma proteins to the site of injury and cellular infiltration, which occur through chemotaxis and diapedesis and the direct movement of inflammatory cells through the vessel wall towards the site of inflammation. In addition, during the inflammatory response catabolic and metabolic changes may occur, as well as biosynthetic activation in various organs and enzyme systems and cells of the immune

The inflammatory response begins the process of immune elimination of invading pathogens and toxins for the repair of damaged tissue (Rang & Dale 1995). The nonspecific inflammatory response can be seen, for example, in the phagocytosis of bacteria or leftover tissue, the secretion of proteolytic enzymes, the production of reactive oxygen species and the secretion of molecular modulators. It can also be immune-mediated, where there is the participation of lymphocytes and antigen-presenting cells. This second type is closely associated with the onset and maintenance of chronic inflammation (Pompei et al. 1999). The inflammatory process is controlled by cellular and molecular components. Among the cellular components are neutrophils, monocytes, lymphocytes, macrophages fixed, dendritic cells, mast cells and eosinophils. These cells accumulate in inflamed tissues and interact with the endothelial cells of the microcirculation. Different adhesion molecules participate in these interactions, including selectins, integrins and intercellular adhesion molecules (ICAM) (Rang & Dale 1995). Neutrophils constitute 60% of circulating leukocytes and act as the first line of cellular defence, and they may participate in both reactions as a nonspecific defence and as specific antigen reactions (Curi et al. 1997). Monocytes represent approximately 3-6% of circulating leukocytes in human blood, and they migrate to different tissues where they differentiate into macrophages in response to different stimuli. These cells participate in a variety of functions related to the host's defence, the most well known being the phagocytosis of microorganisms and cell debris, and cytotoxic activity against

microorganisms, virus-infected cells and tumour cells (Curi et al. 2002).

portion of the GI tract, from the mouth to the anus (Benedetti & Plum 1996).

**3. Inflammatory bowel disease and carcinogenesis** 

The molecular components of inflammation include vasoactive substances (kinins, histamine), proinflammatory cytokines (such as Tumour Necrose Factor (TNF), Interleukin (IL)-1 and IL-6), anti-inflammatory cytokines (such as IL-4, IL-10 and IL-13), chemokines, acute phase proteins, bioactive lipids (such as eicosanoids derived from AA), Platelet Activating Factor, diacylglycerol, ceramides, cAMP, and inositol triphosphate, amongst

Inflammatory bowel diseases (IBDs) are chronic disorders of the gastrointestinal (GI), which generally refer to two conditions, namely ulcerative colitis and Crohn's disease (Galvez et al. 2006). IBDs are characterised by chronic diarrhoea, malabsorption, mucosal barrier dysfunction and inflammatory intestinal process, being incurable clinically (Benedetti & Plum 1996). Ulcerative colitis encompasses a spectrum of diffuse inflammation and the continuous surface of the colon, which begins in the rectum and may extend to the proximal level. Crohn's disease is characterised by transmural inflammation affecting any asymmetric

IBDs cause nutritional deficiencies, such as calorie and protein malnutrition, and deficiencies in vitamins, minerals and trace elements. This underscores the importance of

system.

others.

nutritional therapy in their treatment (Ferguson et al. 2007, Pizato et al. 2005, Razack et al. 2007). Malnutrition is common in these patients, and interventions through adequate nutritional therapy so as to restore the nutritional status have been associated with an improved recovery process involving the improvement of the immune system during periods of the exacerbation of the disease (Razack et al. 2007). Several characteristics contribute to the malnutrition observed in patients: 1) there is a decrease in the oral intake of nutrients associated with abdominal pain and anorexia; 2) the mucosal inflammation associated with diarrhoea leads to a loss of protein, minerals, blood, electrolytes and trace elements. In addition, multiple resections or bacterial overgrowth in the colon can cause adverse effects, such as the poor nutritional absorption of micronutrients: 3) drug therapies can lead to malnutrition. For example, sulfasalazine reduces the absorption of folic acid, and corticosteroids reduce calcium absorption and adversely affect the protein metabolism (Wild et al. 2007).

Although much progress has been made in understanding IBD, its aetiology is not fully elucidated. However, it is believed that there is involvement of immune factors, both genetic and environmental (Laroux et al. 2001, Cheon et al. 2006, Sainathan et al. 2008). Some studies have suggested that IBDs represent an inappropriate and exaggerated response of the intestinal mucosal immune system to normal intestinal microflora – in genetically susceptible individuals – which can be attributed in part to an imbalance between effector T cells (T eff) cells and T regulatory cells (T reg). (Sanchez-Muñoz et al. 2008, Ma et al. 2007). Effector T cells are helper T lymphocytes (lymph CD4 +) and cytolytic T lymphocytes (lymph CD8 +) that are activated during the adaptive or acquired immune response. The helper T cells secrete cytokines, whose function is to stimulate the proliferation and differentiation of T cells, as well as other cells including B lymphocytes, macrophages and other leukocytes (Sanchez-Muñoz et al. 2008, Sainathan et al. 2008). Cytolytic T lymphocytes destroy cells that produce antigens, such as cells infected by viruses or other intracellular microbes. Since regulatory T cells are cells capable of blocking the activation and effector function of T lymphocytes (Abbas & Lichtman 2005), some studies indicate that the suppressive action of these cells is linked to the secretion of immunosuppressive cytokines, such as IL-10 and Transforming Growth Factor Beta (TGF-β). TGF-β inhibits the proliferation of T and B cells, whereas IL-10 inhibits macrophage activation and is the main antagonist of Macrophage Activating Factor and Interferon Gamma (IFN-γ) (Sanchez-Muñoz et al. 2008).

The innate immune response in IBDs also plays an important role. This response is the first line of defence of the immune system, attended by phagocytic cells, natural killer cells, blood proteins, and including fractions of complements and other mediators of inflammation such as cytokines (Abbas & Lichtman 2005). Cytokines are polypeptides – produced mainly by immune cells – that facilitate communication between cells, stimulate the proliferation of antigen-specific effector cells, and mediate systemic inflammation and local roads in the endocrine, paracrine and autocrine (Muños-Sanchez et al. 2008). Dendritic cells and activated macrophages secrete various cytokines that regulate the inflammatory response. Once secreted, these cytokines promote the differentiation of T cells, activating the adaptive immune response (Abbas & Lichtman 2005). The T-helper cells or CD4 + T cells can differentiate into subpopulations of effector T cells that produce different sets of cytokines and, therefore, play different effector functions. The most well-defined subpopulations of effector T cells are T helper cells type 1 (Th1) and type 2 (Th2) (Abbas &

Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 163

ROS – which are the cellular consequences of oxidative stress – may cause DNA oxidation, resulting in damage to all four bases and in the deoxy-ribose-molecule triggering the appearance of genetic mutations and initiating colorectal carcinogenesis (Chapkin et al.

With the large number of cytokines and growth factors released during inflammation, the immune cells and nonimmune cells may influence the process of carcinogenesis (Fantini et al. 2008). These mediators activate NF-kB, inducible nitric oxide synthase, and cyclooxygenase-2-related signalling pathways, which are associated with the delay or suppression of the apoptosis of intestinal epithelial cells and the modulation of angiogenesis (Chapkin et al. 2007, Fantini et al. 2008). Apoptosis – programmed cell death – is the mechanism by which the intestine eliminates cells with irreparable DNA damage, and the

The integrity of DNA is vital for cell division, and oxidative changes may interfere with transcription, translation and DNA replication, and may also increase mutations, senescence

Epidemiological studies have been conducted in an attempt to correlate nutritional factors with chronic diseases and carcinogenesis on set. In this context, we can observe in recent years a drastic alteration in dietetic habits, mainly in lipids' composition and contents (Wild et al. 2007), leading to an association with the type and amount of fatty acid intake by diet, and the development of diseases (Figler et al. 2007). Asian countries that have changed from a traditional diet (i.e. high in fish and cruciferous vegetables) to a Western diet lifestyle (i.e. high in red meat and saturated fat), such as Singaporean Chinese (who have had a historically low risk for colorectal cancer), have doubled this risk in the past three decades,

Linoleic acid intake, in western countries, increased considerably in the 20th century, followed by vegetable oil and margarine introduction, which resulted in a significant rise in the n-6:n-3 PUFA ratio in the diet (Calder 2008). The incidence of IBDs is higher in western populations and has increased in developing countries which have adopted industrialised urban lifestyles associated with changes in dietetic habits, including an increased fast food

PUFA n-6 and n-3 are incorporated in cell membrane phospholipids and can influence immunological and inflammatory responses by modifying fluidity, the antioxidant defence system and the inflammatory mediators (Calder 2008, Kinsella et al. 1990,

N-3 PUFA, EPA and DHA competitively inhibit AA oxygenation by cyclooxigenase, decreasing the synthesis of eicosanoids from series 2 and 4 from AA, with a concomitant increase in prostaglandin (PG), tromboxanes (TX) from 3 series and leukotrienes from 5 series (Yaqoob & Calder 1995). On the other hand, an excessive amount of n-6 PUFA, in diet poor in n-3 PUFA, can contribute to PGE2, TXA2 and LTB4 overproduction – potent inflammatory mediators. Eicosanoids produced from EPA (n-3 PUFA) are, in general, less active in inflammatory process than derived AA eicosanoids (Calder 1996, 1998, Kikuchi et

The inflammatory response is designed to remove the inciting stimulus and resolve tissue damage. However, excessive inflammatory response can cause local tissue damage and

inhibition of this response is a characteristic of colon cancer (Bancroft et al. 2003).

**4. Inflammatory bowel disease and dietary fatty acids** 

2007).

and cell death (Miranda et al. 2008).

after dietetic modification (Stern et al. 2009).

intake with high lipids content (Wild et al. 2007).

Simopoulos 2003).

al. 1998).

Lichtman 2005, Fuss et al. 2004). IFN- is associated with Th1 cells, while IL-4 and IL-5 are associated with Th2 cells. Today it is clear that individual cells can express various mixtures of cytokines, and that there may be many sub-populations with heterogeneous patterns of cytokine production. However, chronic immune reactions are often dominated by either Th1 or Th2 populations (Kampen et al. 2005). These sub-populations show differences in the expression of several cytokine receptors, and these differences may reflect the activation state of the cell, determine their effectors' functions, and participate in the development and expansion of their sub-populations (Abbas & Lichtman 2005). IBDs can cause an imbalance between regulatory T cells and T effector cells Th1/ Th2. The lack of appropriate regulation of T cells and the overproduction of effector T cells are related to the development and exacerbation of IBDs (Muños-Sanchez et al. 2008, Zhang et al. 2005).

Patients with IBDs, particularly ulcerative colitis, are at risk of developing cancer that is 10 times higher than that of the general population, indicating that chronic intestinal inflammation is an important risk factor for developing colon cancer (Gommeaux et al. 2007). Some studies have shown that the risk of developing cancer increases exponentially with the duration of the illness, and the extent and intensity of inflammation in the intestinal mucosa (Burstein & Fearon 2008).

The process of carcinogenesis seems to involve a sequence of events, where the chronically inflamed and hyperplastic epithelium progresses to initially flat foci of dysplasia, adenoma and finally to adenocarcinoma. Uncontrolled inflammation is associated with oxidative stress and oxidative cell damage. During cell proliferation, oxidative DNA lesions induce mutations that are commonly observed in oncogenesis and tumour suppressor genes, such as p53 (Gommeaux et al. 2007, Seril et al. 2003). It is likely that the cells of the colonic mucosa, persistently subjected to oxidizing agents, suffer progressive oxidative damage in their DNA, which can cause mutations in tumour suppressor genes (p53), oncogenes (k-ras) and genes that encode the repair of proteins (MSH2 and MLH1) (Gommeaux et al. 2007). The initiation of carcinogenesis is caused by an irreversible alteration of the DNA through the reaction of this molecule with carcinogenic substances. Thus, mechanisms of carcinogen detoxification, DNA repair, and the elimination of cells that have modified DNA (apoptosis, for example), are important for protection against cancer initiation (Brown et al. 1994). For initiation to occur requires not only the modification of DNA, but also its replication and cell proliferation, so that the original mutation can be fixed. Most human cancers originate from epithelial cells (carcinoma), as these are exposed to carcinogens (in the air or in food) and they are rapidly proliferating (Bartsch et al. 1996). In general, electrophilic substances are carcinogens or are metabolised to carcinogens substancesduring the process of detoxification. Such substances are attracted to molecules with high electron densities – such as DNA bases – which end up calling and leading to the formation of adducts (Bartsch et al., 2006).

The basis of the DNA which is more susceptible to this type of attack is guanine, but the adducts thereby formed have been reported in other bases. Being formed in DNA by specific chemical mechanisms, such adducts may lead to mutations in proto-oncogenesis or tumour suppressor genes, and they start the process of carcinogenesis (Lehman et al. 1994, Kinzler et al. 1996).

It is well established that inflammation facilitates the progression of normal cells to malignant cells, the production of pro-inflammatory cytokines such as TNF, IL-1, IL-6, IL-23 and reactive oxygen species (ROS) and nitrogen (Bartsch et al. 2006, Roessner et. al. 2008).

Lichtman 2005, Fuss et al. 2004). IFN- is associated with Th1 cells, while IL-4 and IL-5 are associated with Th2 cells. Today it is clear that individual cells can express various mixtures of cytokines, and that there may be many sub-populations with heterogeneous patterns of cytokine production. However, chronic immune reactions are often dominated by either Th1 or Th2 populations (Kampen et al. 2005). These sub-populations show differences in the expression of several cytokine receptors, and these differences may reflect the activation state of the cell, determine their effectors' functions, and participate in the development and expansion of their sub-populations (Abbas & Lichtman 2005). IBDs can cause an imbalance between regulatory T cells and T effector cells Th1/ Th2. The lack of appropriate regulation of T cells and the overproduction of effector T cells are related to the development and

Patients with IBDs, particularly ulcerative colitis, are at risk of developing cancer that is 10 times higher than that of the general population, indicating that chronic intestinal inflammation is an important risk factor for developing colon cancer (Gommeaux et al. 2007). Some studies have shown that the risk of developing cancer increases exponentially with the duration of the illness, and the extent and intensity of inflammation in the intestinal

The process of carcinogenesis seems to involve a sequence of events, where the chronically inflamed and hyperplastic epithelium progresses to initially flat foci of dysplasia, adenoma and finally to adenocarcinoma. Uncontrolled inflammation is associated with oxidative stress and oxidative cell damage. During cell proliferation, oxidative DNA lesions induce mutations that are commonly observed in oncogenesis and tumour suppressor genes, such as p53 (Gommeaux et al. 2007, Seril et al. 2003). It is likely that the cells of the colonic mucosa, persistently subjected to oxidizing agents, suffer progressive oxidative damage in their DNA, which can cause mutations in tumour suppressor genes (p53), oncogenes (k-ras) and genes that encode the repair of proteins (MSH2 and MLH1) (Gommeaux et al. 2007). The initiation of carcinogenesis is caused by an irreversible alteration of the DNA through the reaction of this molecule with carcinogenic substances. Thus, mechanisms of carcinogen detoxification, DNA repair, and the elimination of cells that have modified DNA (apoptosis, for example), are important for protection against cancer initiation (Brown et al. 1994). For initiation to occur requires not only the modification of DNA, but also its replication and cell proliferation, so that the original mutation can be fixed. Most human cancers originate from epithelial cells (carcinoma), as these are exposed to carcinogens (in the air or in food) and they are rapidly proliferating (Bartsch et al. 1996). In general, electrophilic substances are carcinogens or are metabolised to carcinogens substancesduring the process of detoxification. Such substances are attracted to molecules with high electron densities – such as DNA bases – which end up calling and leading to the formation of adducts (Bartsch et al.,

The basis of the DNA which is more susceptible to this type of attack is guanine, but the adducts thereby formed have been reported in other bases. Being formed in DNA by specific chemical mechanisms, such adducts may lead to mutations in proto-oncogenesis or tumour suppressor genes, and they start the process of carcinogenesis (Lehman et al. 1994,

It is well established that inflammation facilitates the progression of normal cells to malignant cells, the production of pro-inflammatory cytokines such as TNF, IL-1, IL-6, IL-23 and reactive oxygen species (ROS) and nitrogen (Bartsch et al. 2006, Roessner et. al. 2008).

exacerbation of IBDs (Muños-Sanchez et al. 2008, Zhang et al. 2005).

mucosa (Burstein & Fearon 2008).

2006).

Kinzler et al. 1996).

ROS – which are the cellular consequences of oxidative stress – may cause DNA oxidation, resulting in damage to all four bases and in the deoxy-ribose-molecule triggering the appearance of genetic mutations and initiating colorectal carcinogenesis (Chapkin et al. 2007).

With the large number of cytokines and growth factors released during inflammation, the immune cells and nonimmune cells may influence the process of carcinogenesis (Fantini et al. 2008). These mediators activate NF-kB, inducible nitric oxide synthase, and cyclooxygenase-2-related signalling pathways, which are associated with the delay or suppression of the apoptosis of intestinal epithelial cells and the modulation of angiogenesis (Chapkin et al. 2007, Fantini et al. 2008). Apoptosis – programmed cell death – is the mechanism by which the intestine eliminates cells with irreparable DNA damage, and the inhibition of this response is a characteristic of colon cancer (Bancroft et al. 2003).

The integrity of DNA is vital for cell division, and oxidative changes may interfere with transcription, translation and DNA replication, and may also increase mutations, senescence and cell death (Miranda et al. 2008).

#### **4. Inflammatory bowel disease and dietary fatty acids**

Epidemiological studies have been conducted in an attempt to correlate nutritional factors with chronic diseases and carcinogenesis on set. In this context, we can observe in recent years a drastic alteration in dietetic habits, mainly in lipids' composition and contents (Wild et al. 2007), leading to an association with the type and amount of fatty acid intake by diet, and the development of diseases (Figler et al. 2007). Asian countries that have changed from a traditional diet (i.e. high in fish and cruciferous vegetables) to a Western diet lifestyle (i.e. high in red meat and saturated fat), such as Singaporean Chinese (who have had a historically low risk for colorectal cancer), have doubled this risk in the past three decades, after dietetic modification (Stern et al. 2009).

Linoleic acid intake, in western countries, increased considerably in the 20th century, followed by vegetable oil and margarine introduction, which resulted in a significant rise in the n-6:n-3 PUFA ratio in the diet (Calder 2008). The incidence of IBDs is higher in western populations and has increased in developing countries which have adopted industrialised urban lifestyles associated with changes in dietetic habits, including an increased fast food intake with high lipids content (Wild et al. 2007).

PUFA n-6 and n-3 are incorporated in cell membrane phospholipids and can influence immunological and inflammatory responses by modifying fluidity, the antioxidant defence system and the inflammatory mediators (Calder 2008, Kinsella et al. 1990, Simopoulos 2003).

N-3 PUFA, EPA and DHA competitively inhibit AA oxygenation by cyclooxigenase, decreasing the synthesis of eicosanoids from series 2 and 4 from AA, with a concomitant increase in prostaglandin (PG), tromboxanes (TX) from 3 series and leukotrienes from 5 series (Yaqoob & Calder 1995). On the other hand, an excessive amount of n-6 PUFA, in diet poor in n-3 PUFA, can contribute to PGE2, TXA2 and LTB4 overproduction – potent inflammatory mediators. Eicosanoids produced from EPA (n-3 PUFA) are, in general, less active in inflammatory process than derived AA eicosanoids (Calder 1996, 1998, Kikuchi et al. 1998).

The inflammatory response is designed to remove the inciting stimulus and resolve tissue damage. However, excessive inflammatory response can cause local tissue damage and

Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 165

et al. 2002, Harbige 1998, MacLean 2005). Fish oil supplementation seems to increase apoptosis on top of colonic crypts, where tumours and polyps are usually developed (Paulsen et al. 1997; Courtney et al. 2006, Hong et al. 2005). Bégin et al. (1991) showed that under some specific conditions, long chain PUFA – mainly GLA, AA, EPA, and DHA – are the most effective for inducing tumour cell death. However, this effect depends upon the

The role of n-3 and n-6 PUFA on cancer development has been extensively investigated in epidemiological and experimental studies. The contrasting role of these fatty acids in carcinogenesis – n-3 as protectors and n-6 as promoters – remains as an intriguing question

In rats with colitis induced by Dextran Sulphate Sodium (DSS), our group showed that a normal fat PUFA rich diet, with a balance on the n-6:n-3 ratio, can increase IL-10 cytokine – an immunoregulatory cytokine that influences the immunological system – both on the innate and cell-mediated response, reduce disease activity and the loss of weight, improve the histological score and protect against DNA damage (Barros et al., 2010). IL-10 is considered an immunoregulatory cytokine which exerts effects in both the innate immune response and in the adaptive immune response. IL-10 Knochout animals, for example, develop colitis spontaneously, and 30 to 60% of these animals show invasive carcinoma of the colon between 3 and 6 months of age (Hegazi et al. 2006, McCafferty et al. 2000). These animals have two important characteristics: 1) an increased intestinal permeability in an early age, and before the onset of the disease; 2) the development of colitis, dependent on the microbiological presence in the intestinal lumen. These characteristics suggest that the colitis observed in these animals can develop as a consequence of the high intestinal permeability that increases in the luminal agent's mucosal immune system (Arrieta et al. 2008). Some studies have demonstrated the role of IL-10 on gastrointestinal mucosal homeostasis maintenance. The mechanism by which this cytokine regulates mucosal inflammation is probably multifactorial; however, it is associated with reduced antigen presentation (Hegazi et al. 2006, McCafferty et al. 2000), an increased release of IFN-� and IL-12 – a cytokine that inhibits the differentiation of T lymphocytes into Th1 lymphocytes (Rennick & Fort 2000). There is strong evidence that IL-10 promotes the differentiation and the increase of the activity of the regulatory T cells (Hegazi et al. 2006). *In vitro* studies have demonstrated that the administration of IL-10 reduces the release of pro-inflammatory cytokines in lamina propria mononuclear cells amongst patients with Crohn's Disease. In addition, high doses of IL-10 administered intraperitoneally into mice with colitis, induced by Trinitrobenzenesulphonic acid (TNBS), are able to restore the tolerance of the lamina

Considering the abundance of fatty acids in cells and its susceptibility to oxidation, PUFA are – for the oxidants – more likely targets than DNA (Shimizu et al., 2001, Wagner et al., 1994). It is estimated that approximately 60 molecules of linoleic acid and 200 of arachidonic acid are consumed by oxidants that react with the lipid bilayer. Autocatalytic oxidation triggers a cascade that generates numerous genotoxic substances, and such damage to lipids has important implications for the integrity of DNA (Wagner et al., 1994). The peroxidation of membrane lipids initiates autocatalytic breaks with the consequent formation of cytotoxic and genotoxic metabolites, such as malondialdehyde and hidroxinomenal. The degradation of these products can interfere with intracellular signalling cascades, involving replication

type of cancer cells tested and the concentration of the fatty acid used.

in the fields of nutritional and cancer research (Eder et al. 2008).

propria mononuclear cells (Duchmann et al. 1996).

and cell death (Eder et al. 2008).

remodelling, which may lead to a significant and chronic injury. Therefore, acute inflammation in healthy individuals is self-limited and has an active termination program (Seki et al. 2009). In the past, it was believed that this termination program was a passive mechanism but, nowadays, it is known that the process of the resolution of inflammation is an active and well controlled event. In part, this is due to the formation of newly endogenous mediators that act as local autacoids stimulating proresolving mechanisms (Serhan 2007, Gilroy et al. 2004). These proresolving mediators are derived from essential fatty acids, and include lipoxins (LX) from AA and resolvines (Rv) and protectins from EPA and DHA (Gilroy et al. 2004), that are biosynthesised in inflammatory exudates during spontaneous resolution (Figure 3).

Fig. 3. News inflammatory mediators (Galli & Calder 2009)

The process of the resolution of inflammation has become a topic of interest because of expanding views of their action, particularly in chronic disorders where unresolved inflammation is a key factor leading to colon carcinogenesis. These newly identified LXs and Rvs have proven to be potent regulators of both leukocyte and cytokine production, thereby regulating the events of interest in inflammation and resolution. In light of the existing knowledge of the interconnected pathways of pro-inflammatory mediators (leukotrienes, chemokines (IL8, SDF-1α, MIP-1α, MCP-1,2 etc), and cytokines (IL3, IL6, IL12, IL-1β, GM-CSF, B94, TNF-α etc)), the anti-inflammatory properties of pro-resolving mediators in preventing the chronic inflammation which leads to carcinogenesis requires further study. Clinical trials have demonstrated the beneficial effects of fish oil supplementation – rich in EPA and DHA – in chronic and acute inflammatory conditions (Innis et al. 2006, Simopoulos

remodelling, which may lead to a significant and chronic injury. Therefore, acute inflammation in healthy individuals is self-limited and has an active termination program (Seki et al. 2009). In the past, it was believed that this termination program was a passive mechanism but, nowadays, it is known that the process of the resolution of inflammation is an active and well controlled event. In part, this is due to the formation of newly endogenous mediators that act as local autacoids stimulating proresolving mechanisms (Serhan 2007, Gilroy et al. 2004). These proresolving mediators are derived from essential fatty acids, and include lipoxins (LX) from AA and resolvines (Rv) and protectins from EPA and DHA (Gilroy et al. 2004), that are biosynthesised in inflammatory exudates during

spontaneous resolution (Figure 3).

Fig. 3. News inflammatory mediators (Galli & Calder 2009)

The process of the resolution of inflammation has become a topic of interest because of expanding views of their action, particularly in chronic disorders where unresolved inflammation is a key factor leading to colon carcinogenesis. These newly identified LXs and Rvs have proven to be potent regulators of both leukocyte and cytokine production, thereby regulating the events of interest in inflammation and resolution. In light of the existing knowledge of the interconnected pathways of pro-inflammatory mediators (leukotrienes, chemokines (IL8, SDF-1α, MIP-1α, MCP-1,2 etc), and cytokines (IL3, IL6, IL12, IL-1β, GM-CSF, B94, TNF-α etc)), the anti-inflammatory properties of pro-resolving mediators in preventing the chronic inflammation which leads to carcinogenesis requires further study. Clinical trials have demonstrated the beneficial effects of fish oil supplementation – rich in EPA and DHA – in chronic and acute inflammatory conditions (Innis et al. 2006, Simopoulos et al. 2002, Harbige 1998, MacLean 2005). Fish oil supplementation seems to increase apoptosis on top of colonic crypts, where tumours and polyps are usually developed (Paulsen et al. 1997; Courtney et al. 2006, Hong et al. 2005). Bégin et al. (1991) showed that under some specific conditions, long chain PUFA – mainly GLA, AA, EPA, and DHA – are the most effective for inducing tumour cell death. However, this effect depends upon the type of cancer cells tested and the concentration of the fatty acid used.

The role of n-3 and n-6 PUFA on cancer development has been extensively investigated in epidemiological and experimental studies. The contrasting role of these fatty acids in carcinogenesis – n-3 as protectors and n-6 as promoters – remains as an intriguing question in the fields of nutritional and cancer research (Eder et al. 2008).

In rats with colitis induced by Dextran Sulphate Sodium (DSS), our group showed that a normal fat PUFA rich diet, with a balance on the n-6:n-3 ratio, can increase IL-10 cytokine – an immunoregulatory cytokine that influences the immunological system – both on the innate and cell-mediated response, reduce disease activity and the loss of weight, improve the histological score and protect against DNA damage (Barros et al., 2010). IL-10 is considered an immunoregulatory cytokine which exerts effects in both the innate immune response and in the adaptive immune response. IL-10 Knochout animals, for example, develop colitis spontaneously, and 30 to 60% of these animals show invasive carcinoma of the colon between 3 and 6 months of age (Hegazi et al. 2006, McCafferty et al. 2000). These animals have two important characteristics: 1) an increased intestinal permeability in an early age, and before the onset of the disease; 2) the development of colitis, dependent on the microbiological presence in the intestinal lumen. These characteristics suggest that the colitis observed in these animals can develop as a consequence of the high intestinal permeability that increases in the luminal agent's mucosal immune system (Arrieta et al. 2008). Some studies have demonstrated the role of IL-10 on gastrointestinal mucosal homeostasis maintenance. The mechanism by which this cytokine regulates mucosal inflammation is probably multifactorial; however, it is associated with reduced antigen presentation (Hegazi et al. 2006, McCafferty et al. 2000), an increased release of IFN-� and IL-12 – a cytokine that inhibits the differentiation of T lymphocytes into Th1 lymphocytes (Rennick & Fort 2000). There is strong evidence that IL-10 promotes the differentiation and the increase of the activity of the regulatory T cells (Hegazi et al. 2006). *In vitro* studies have demonstrated that the administration of IL-10 reduces the release of pro-inflammatory cytokines in lamina propria mononuclear cells amongst patients with Crohn's Disease. In addition, high doses of IL-10 administered intraperitoneally into mice with colitis, induced by Trinitrobenzenesulphonic acid (TNBS), are able to restore the tolerance of the lamina propria mononuclear cells (Duchmann et al. 1996).

Considering the abundance of fatty acids in cells and its susceptibility to oxidation, PUFA are – for the oxidants – more likely targets than DNA (Shimizu et al., 2001, Wagner et al., 1994). It is estimated that approximately 60 molecules of linoleic acid and 200 of arachidonic acid are consumed by oxidants that react with the lipid bilayer. Autocatalytic oxidation triggers a cascade that generates numerous genotoxic substances, and such damage to lipids has important implications for the integrity of DNA (Wagner et al., 1994). The peroxidation of membrane lipids initiates autocatalytic breaks with the consequent formation of cytotoxic and genotoxic metabolites, such as malondialdehyde and hidroxinomenal. The degradation of these products can interfere with intracellular signalling cascades, involving replication and cell death (Eder et al. 2008).

Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 167

(Stern et. al. 2009). However, using a subset of this prospective cohort, Stern et al. (2009) reported that the marine n–3 PUFA association with rectal cancer is confined to those who carry the PARP codon 762 Ala allele. The PARP protein plays an important role in

In this regard, some studies have shown that genetic variability in the FADS1-FADS-2 gene cluster, and the encoding delta-5 (D5D) and delta-6 (D6D) desaturases, have been associated with plasma long-chain PUFA and lipid levels in adults (Bokor et al. 2010). Desaturases and elongases catalyse the conversion of PUFAs in humans. The D5D and D6D desaturases are known to be the key enzyme of this pathway. Both desaturases are expressed in a majority of human tissue, with the highest levels in liver, but also with major amounts in the brain, the heart and the lungs. The hypothesis that they play a key role in inflammatory diseases is strengthened by functional studies in mice, where selective D5D and D6D inhibitors showed

Several single nucleotide polymorphisms (SNP) in FADS genes were reported in humans, and some showed association between FADS SNPs and fatty acids in serum or plasma phospholipids, and erythrocyte membrane and adipose tissue (Schaeffer et al. 2006, Malerba et al. 2008, Rzehak et al. 2009), demonstrating that these concentrations are influenced not only by diet, but also to a large extent by genetic variants common in the world population

Abbas AK, Lichtman AH, Pillai S. Celular and Molecular Immunology, Elsevier, 6th edition,

Arrieta MC, Madsen K, Doyle J, Meddings J. Reducing small intestinal permeability attenuates colitis in the IL-10 gene deficient mouse. Gut. 58:41-8, 2008. Barros KV, Xavier RA, Abreu GG, Martinez CA, Ribero ML Gambero A, Carvalho PO,

Barstch H. DNA adducts in human carcinogenesis: etiological relevance and structure

Bartsch H and Nair J. Chronic inflammation and oxidative stress in the genesis and

Bégin ME, Ells G, Das UN, Horrobin DF. Differential killing of human carcinoma cells

Benedett JC & Plum F. Doença inflamatória intestinal. In: Cecil: Tratado de medicina

Bokor S, Dumont J, Spinneker A, Gonzalez-Gross M, et al., Single nucleotide

Brown K et al. Distribution and reactivity of inhaled 14C-labeled toluene diisocyanate (TDI)

Nascimento CM, Silveira VL. Soybean and fish oil mixture protects against DNA damage and decrease colonic inflammation in rats with dextran sulfate sodium

perpetuation of cancer: role of lipid peroxidation, DNA damage and repair. Arch

supplemented with n-3 and n-6 polyunsaturated fatty acids. J. Natl.Cancer Inst.,

polymorphisms in the FADS gene cluster are associated with delta-5 and delta-6 desaturase activities estimated by serum fatty acid ratios. J Lipid Res. 51(8):2325-33,

maintaining genomic stability, apoptosis, and in regulating transcription.

an anti-inflammatory response.

(Koletzko et al. 2011).

2007.

2010.

(DSS). Lipids Health Dis. 68:8-14, 2010.

Surg. 391:499-510, 2006.

77(5):1053-1062, 1986.

activity relationship. Mutat. Res. 340: 67-79, 1996.

interna, 20º ed, V1, Guanabara Koogan, 1996.

in rats.Arch toxicol. 16 (suppl);68:434-43, 1994.

**6. References** 

The dietary lipids that are related to the pro-oxidative attack of the colonic epithelial cells may be an important contributor to carcinogenesis (Nowak et al. 2007, Udilova et al. 2003). So far, there is still no specific treatment for IBDs and the best strategy to regulate the exacerbated inflammatory response is to interfere with the multiple phases of the inflammatory cascade with anti-inflammatory and immunosuppressive drugs. These drugs, however, have serious side-effects that limit their use (Stein et al., 2000). Dietary treatment may be an alternative to drug therapy (Camuesco et al., 2005, Nowak et al., 2007).

Although the high intake of PUFA has been related to colorectal cancer, several studies show that, besides the genotoxic effects of lipid peroxidation, epigenetic factors may also be responsible for an increased cancer risk after excessive PUFA intake (Nystrom et al. 2009). Using a model of DSS colitis and a high fat diet (20%), in our laboratory, we did not observe an exacerbation of experimental ulcerative colitis in relation to the diet control group (5%) Besides, the great balance in the n-6:n-3 PUFA ratio (2:1) caused beneficial effects on both pro- and anti-inflammatory cytokine balance and protected the DNA against damage (Barros et al. 2010).

Sasasuki et al. (2010) in an epidemiological study where it was inquired as to whether the intake of n-3 and n-6 PUFA are related to a decreased risk of colorectal cancer development. They found that, in a population with high fish consumption and a wide range of n-3 PUFA intakes, the PUFAs originating with marine consumption may be inversely related to the risk of cancer in proximal sites of the large bowel. On the other hand, Dahm et al. (2010), in a case-control study nested within seven prospective UK cohort studies, comprising 579 cases of the incidence of colorectal cancer and 1996 matched controls, did not find any association between total dietary fat, saturated, monounsaturated and PUFA intakes, and colorectal cancer risk.

#### **5. Polymorphisms**

Conclusive evidence between colorectal cancer and PUFA in epidemiological studies may be related to genetic influence. The relationship between genes and the environment has been recognised as central to knowledge of disease and health. During the last two decades, advances in molecular biology have demonstrated that genetic factors determine disease susceptibility, while environmental factors determine whether or not genetically susceptible individuals will be affected (Simopoulos et al. 2008, Paolini-Giacobini et al. 2003). In this context, nutritional aspects are beginning to be considered as one of the most important environmental factors (Simopoulos et al. 2008). Several studies have shown the mechanisms by which genes may influence the metabolism of nutrients, as well as the mechanisms by which nutrients can influence gene expression (Simopoulos et al. 2008, Paolini-Giacobini et al. 2003, Calder 2007). With advances in science, and emphasis on the study of nutrigenomics and nutrigenetics, it has been shown that certain nutrients can influence the inflammatory response, accelerating or regressing the development of many diseases (Heller et al. 2002, Weiss et al. 2002, Mayer et al. 2003, Paolini-Giacobini et al. 2003, Simopoulos et al. 2008).

Stern et al. (2009), from the Singapore Chinese Health Study, through analyses taking into account variants in genes that are relevant for the proposed PUFAs mechanism of action – hypothesised that the genes which play key roles in the pathways that repair PUFA-induced damage might modify the effect of these FA on colorectal cancer. This study also showed that diets high in marine n–3 PUFA were positively associated with colorectal cancer risk (Stern et. al. 2009). However, using a subset of this prospective cohort, Stern et al. (2009) reported that the marine n–3 PUFA association with rectal cancer is confined to those who carry the PARP codon 762 Ala allele. The PARP protein plays an important role in maintaining genomic stability, apoptosis, and in regulating transcription.

In this regard, some studies have shown that genetic variability in the FADS1-FADS-2 gene cluster, and the encoding delta-5 (D5D) and delta-6 (D6D) desaturases, have been associated with plasma long-chain PUFA and lipid levels in adults (Bokor et al. 2010). Desaturases and elongases catalyse the conversion of PUFAs in humans. The D5D and D6D desaturases are known to be the key enzyme of this pathway. Both desaturases are expressed in a majority of human tissue, with the highest levels in liver, but also with major amounts in the brain, the heart and the lungs. The hypothesis that they play a key role in inflammatory diseases is strengthened by functional studies in mice, where selective D5D and D6D inhibitors showed an anti-inflammatory response.

Several single nucleotide polymorphisms (SNP) in FADS genes were reported in humans, and some showed association between FADS SNPs and fatty acids in serum or plasma phospholipids, and erythrocyte membrane and adipose tissue (Schaeffer et al. 2006, Malerba et al. 2008, Rzehak et al. 2009), demonstrating that these concentrations are influenced not only by diet, but also to a large extent by genetic variants common in the world population (Koletzko et al. 2011).

#### **6. References**

166 Colorectal Cancer – From Prevention to Patient Care

The dietary lipids that are related to the pro-oxidative attack of the colonic epithelial cells may be an important contributor to carcinogenesis (Nowak et al. 2007, Udilova et al. 2003). So far, there is still no specific treatment for IBDs and the best strategy to regulate the exacerbated inflammatory response is to interfere with the multiple phases of the inflammatory cascade with anti-inflammatory and immunosuppressive drugs. These drugs, however, have serious side-effects that limit their use (Stein et al., 2000). Dietary treatment

Although the high intake of PUFA has been related to colorectal cancer, several studies show that, besides the genotoxic effects of lipid peroxidation, epigenetic factors may also be responsible for an increased cancer risk after excessive PUFA intake (Nystrom et al. 2009). Using a model of DSS colitis and a high fat diet (20%), in our laboratory, we did not observe an exacerbation of experimental ulcerative colitis in relation to the diet control group (5%) Besides, the great balance in the n-6:n-3 PUFA ratio (2:1) caused beneficial effects on both pro- and anti-inflammatory cytokine balance and protected the DNA against damage

Sasasuki et al. (2010) in an epidemiological study where it was inquired as to whether the intake of n-3 and n-6 PUFA are related to a decreased risk of colorectal cancer development. They found that, in a population with high fish consumption and a wide range of n-3 PUFA intakes, the PUFAs originating with marine consumption may be inversely related to the risk of cancer in proximal sites of the large bowel. On the other hand, Dahm et al. (2010), in a case-control study nested within seven prospective UK cohort studies, comprising 579 cases of the incidence of colorectal cancer and 1996 matched controls, did not find any association between total dietary fat, saturated, monounsaturated and PUFA intakes, and

Conclusive evidence between colorectal cancer and PUFA in epidemiological studies may be related to genetic influence. The relationship between genes and the environment has been recognised as central to knowledge of disease and health. During the last two decades, advances in molecular biology have demonstrated that genetic factors determine disease susceptibility, while environmental factors determine whether or not genetically susceptible individuals will be affected (Simopoulos et al. 2008, Paolini-Giacobini et al. 2003). In this context, nutritional aspects are beginning to be considered as one of the most important environmental factors (Simopoulos et al. 2008). Several studies have shown the mechanisms by which genes may influence the metabolism of nutrients, as well as the mechanisms by which nutrients can influence gene expression (Simopoulos et al. 2008, Paolini-Giacobini et al. 2003, Calder 2007). With advances in science, and emphasis on the study of nutrigenomics and nutrigenetics, it has been shown that certain nutrients can influence the inflammatory response, accelerating or regressing the development of many diseases (Heller et al. 2002, Weiss et al. 2002, Mayer et al. 2003, Paolini-Giacobini et al. 2003,

Stern et al. (2009), from the Singapore Chinese Health Study, through analyses taking into account variants in genes that are relevant for the proposed PUFAs mechanism of action – hypothesised that the genes which play key roles in the pathways that repair PUFA-induced damage might modify the effect of these FA on colorectal cancer. This study also showed that diets high in marine n–3 PUFA were positively associated with colorectal cancer risk

may be an alternative to drug therapy (Camuesco et al., 2005, Nowak et al., 2007).

(Barros et al. 2010).

colorectal cancer risk.

**5. Polymorphisms** 

Simopoulos et al. 2008).


Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 169

Ferguson LR, Peterman I, Hubner C, Philpott M, Shellin AN. Uncoupling gene-diet interactions in inflammatory bowel disease (IBD). Genes Nutr. 2:71-73, 2007. Figler M, Gasztonyi B, Cseh J et al. Association of n-3 and n-6 long-chain polyunsaturated

Fuss IJ, Heller F, Boirivant M, Leon F, Yoshida M et al. Nonclassical CD1 d-restricted NK T

Galvez J, Gracioso JS, Camuesco D et al. Intestinal anti-inflammatory activity of a

Galli C & Calder PC. Effects of fat and fatty acid intake on inflammatory and immune

Gilroy DW, Newson J, Sawmynaden P, Willoughby DA, and Croxtall JD. A novel role for

Gommeaux J, Cano C, Garcia S, Gironella M, Petri S et al. Colitis and colitis-associated

Harbidgel LS. Dietary w-6 e w-3 fatty acids in immunity and autoimmune disease. Proc

Hegazi RAF, Saad RS , Mady H et al., Dietary fatty acids modulate chronic colitis, colitis-

Heller AR, Fischer S, Rossel T et al., Impact of n-3 fatty acid supplemented parenteral

Hensen AE, Haggard ME, et al. Essential fatty acids in infant nutrition. J Nutr, 66: 565-76,

Holman RT. The slow discovery of the importance of n-3 essential fatty acids in human

Hong MY, Bancroft LK, Turner ND et al., Fish oil decreases oxidative stress DNA damage

Kampen CV, Gauldie J, Collins SM. Proinflammatry proprieties of IL-4 in the intestinal microenvironment. Am J Physiol Gastrointest Liver Physiol. 288:G111-G117, 2005. Kinsella JE, Lokesh B, Brougton KS, Whelan J. Dietary polyunsaturated fatty acids and

Kinzler KW, Vogelstein B. Life (and death) in a malignant tumour. Nature 4;379(6560):19-20,

eicosanoids potencial effects on the modulation of inflammatory and immune cells:

by enhancing apoptosis in rat colon. Nutr Cancer. 52:166-175, 2005. Innis SM. Essential fatty acids in growth and development. Prog. Lip. Res., 30:39-108, 1991. Innis SM, Pinsk V, Jacobson K. Dietary lipids and intestinal inflammatory disease. J Pediatr.

responses: A critical review. Ann Nutr Metab 55:123–139, 2009.

protein 1. Molecular and cellular biology. 27:2215-228, 2007.

Hang HP, Dale MM. Farmacologia. 3 ed. Guanabara Koogan, 1995.

Journal of Nutrition. 97:1154-1161, 2007.

Clin Invest. 113:1490-1497, 2004.

2006.

J. 18:489–498, 2004.

Nutr Soc. 57:555-62,1998.

Nutrition. 22:275-282, 2006.

health. J. Nutr. 128:427S-433S, 1998.

an overview. Nutrition. 5:24-44, 1990.

S101; 2002.

149:S89-S96, 2006.

1958.

1996.

fatty acids in plasma lipid classes with inflammatory bowel diseases. British

cells that produce IL-13 characterize an atypical Th2 response in ulcerative colitis. J

lyophilized infusion of Turnera ulmifolia in TNBS rat colits. Fitoterapia. 77:515-20,

phospholipase A2 isoforms in the checkpoint control of acute inflammation. FASEB

cancer are exacerbated in mice deficient for tumour protein 53-induced nuclear

associated colon neoplasia and COX-2 expression in IL-10 Knockout mice.

nutrition on haemostasis patters after major abdominal surgery. Brit J Nutr. 87:S95 -


Burr GO, Burr MM. A new deficiency disease produced by the rigid exclusion of fat from

Burstein E and Fearon ER. Colitis and cancer: a tale of inflammatory cells and theirs

Calder PC. Effects of fatty acids and dietary lipids on cell of the immune system. Proc. Nutr

Calder PC. Long-chain n-3 fatty acids and inflammation: potential application in surgical

Calder PC. Immunoregulatory and anti-inflammatory effects of n-3 polyunsaturated fatty acids. Brazilian Journal of Medical and Biological Research. 31:67-490, 1998. Calder PC. Immunomodulation by omega-3 fatty acids. Prostaglandins Leukot Essent Fatty

Calder PC. Polyunsaturated fatty acids, inflammatory processes and inflammatory bowel

Camuesco D, Comalada M, Concha A, Nieto A, Sierra S, Xaus J, Zarzuelo A, Gálvez J.

Chapkin RS, Davidson LA, Weeks BR, Lupton JR, McMurray DN. Immunomodulatory

Cheon JH, Kim JS, Kim JM, Kim N, Jung HC, Song S. Plant sterol guggulsterone inhibits

Courtney ED, Matthews S, Finlayson C et al. Eicosapentaenoic acid (EPA) reduces crypt cell

Curi TP, Melo MP, Azevedo RB et al. Glutamine utilization by rat neutrophils: presence of phosphate-dependent glutaminase. Am. J. Physiol, 273:C1124-9, 1997. Curi R, Pompeia C, Miyasaka CK, Procopio J. Entendendo a gordura – os ácidos graxos.

Dahm CC, Keogh RH, Lentjes MAH, et al. Intake of dietary fats and colorectal cancer risk:

Duchmann R, Schimitt E, Knolle P et al. Tolerance towards resident intestinal flora in mice is

Eder E, Wacker M, Wanek P. Lipid peroxidation-related 1, N2-propanodeoxyguanosine-

Fantini MC, Palloni F. Cytokines: from gut inflammation to colorectal cancer. Curr Drug

history of colorectal adenomas. J Colorectal Dis. 22:765-76, 2007.

antibodies to interleukin-12. Eur J Immuno. 26:934-938, 1996.

ameliorates acute murine colitis. Inflamm Bowel Dis. 12:1152-1161, 2006. Connor WE. Omega-3 essential fatty acids in infant neurological development. PUFA

acids, in rats with DSS-induced colitis. Clin Nutr. 25:466-76, 2006.

Intestinal anti-inflammatory activity of combined quercitrin and dietary olive oil supplemented with fish oil, rich in EPA and DHA (n-3) polyunsaturated fatty

effects of (n-3) fatty acids: Putative link to inflammation and colon cancer. J Nutr.

nuclear factor-kb sinaling in intestinal epithelial cells by blocking IkB kinase and

proliferation and increases apoptosis In normal colonic mucosa in subjects with a

Prospective findings from the UK Dietary Cohort Consortium. Cancer

abrogated in experimental colitis and restored by treatment with interleukin-10 or

DNA adducts induced by endogenously formed 4-hydroxy-2-nonenal in organs of female rats fed diets supplemented with sunflower, rapeseed, olive or coconut oil.

and trauma patients. Braz J Med Biol Res. 36: 433-446, 2003.

the diet. Nutr Rev. 31:248–9, 1973.

Soc. 55:127-150, 1996.

Acids. 77:327-35, 2007.

137:200S-204S, 2007.

Manole, 1ed., 2002.

Information. 1 (1):1-6, 1996.

Epidemiology 34:562–567, 2010.

Mutat Res. 654:101-107, 2008.

Targets. 29:375-80, 2008.

cytokines. J Clin Invest. 118:464-467, 2008.

disease. Mol Nutr Food Res. 52:885-897, 2008.


Polyunsaturated Fatty Acids, Ulcerative Colitis and Cancer Prevention 171

Pizato N, Bonatto S, Yamazaki RK, Aikawa J, et al., Ratio of n-6 to n-3 fatty acids in the diet

Pompéia C, Lopes LR, Miyasaka CK, Procópio J, Sannomiya P. Effect of fatty acids on

Razack R, Seidner DL. Nutrition in inflammatory bowel disease. Curr Opin Gastroenterol.

Rankin Ja. Biological mediators of acute inflammation. AACN Clin Issues. 15(1): 3-17, 2004. Rennick DM, Fort MM. Lessons from genetically engineered animal models. XII. IL-10-

Ribeiro ML, Priolli DG, Miranda DC, Arçari DP, Pedrazzoli J, Martinez CAR. Analysis of

Roessner A, Kuester D, Malfertheiner P, Schneider-Stock. Oxidative stress in ulcerative

Rzehak P, Heinrich J, Klopp N, et al., Evidence for an association between genetic variants

Saadi S, Wrenshall LE, Platt JL. Regional manifestations and control of the immune system.

Sainathan SK, Hanna EM, Gong Q et al., Granulocyte macrophage colony-stimulating factor ameliorates DSS-induced experimental colitis. Inflamm Bowel Dis. 14:88-98, 2008. Sala-Vila A, Miles E.A, Calder PC. Fatty acid composition abnormalities in atopic disease:

Sanchez-Muñoz F, Dominguez-Lopes A, Yamamoto-Furusho JK. Role of cytokines in inflammatory bowel disease. World J Gastroenterol. 21:4280-4288, 2008. Sasazuki S, Inoue M, Iwasaki M, et al., Intake of n-3 and n-6 polyunsaturated fatty acids and

Schaeffer L, Gohlke H, Muller M, et al., Common genetic variants of the FADS1 FADS2 gene

Seki H, Tani Y, Arita M. Omega-3 PUFA derived anti-inflammatory lipid mediator resolving

Semplecine A and Valle R. Fish oils and their possible role in the treatment of cardiovascular

Serhan, C.N. A search for endogenous mechanisms of anti-inflammation uncovers novel

Serhan CN. Resolution phases of inflammation: novel endogenous anti-inflammatory and

leukocyte function. Med. Res. 33 (11): 1255-68, 1999.

and cancer. 53:194-201, 2005.

Liver Physiol. 278:829-33, 2000.

Cancer. 267-272, 2008.

FASEB J. 16(8):849-56. 2002.

Experimental Allergy, 38:1432–1450, 2008.

diseases., Pharmac Ther. 61: 385-387, 1994.

prospective study. Int. J. Cancer. 000: 000–000, 2010.

E1. Prostaglandins Other Lipid Mediat. 89(3-4):126-30, 2009.

2008.

2004.

2007.

23:400-405, 2007.

affects tumour growth and cachexia in Walker 256 tumour-bearing rats. Nutrition

deficient (IL-10(-/-) mice and intestinal inflammation. Am J Physiol Gastrointest

oxidative DNA damage in patients with colorectal cancer. Clinical Colorectal

colitis-associated carcinogenesis. Pathology - Research and Practices. 204:511-524,

of the fatty acid desaturase 1 fatty acid desaturase 2 (FADS1 FADS2) gene cluster and the fatty acid composition of erythrocyte membranes. Br J Nutr. 101:20–6, 2009.

evidence explored and role in the disease process examined. Clinical and

development of colorectal cancer by subsite: Japan Public Health Center–based

cluster and their reconstructed haplotypes are associated with the fatty acid composition in phospholipids. Human Molecular Genetics. 15 (11):1745–1756, 2006.

chemical mediators: missing links to resolution. Histochem Cell Biol. 122:305-321,

pro-resolving lipid mediators and pathways. Annu. Rev. Immunol. 25:101–137,


KIkuchi S, Sakomoto T, Ishikawa C, Yazawa K, Torii S. Modulation of eosinophil chemotatic

Koletzko B, Lattka E, Zeilinger S, LlIig T, Steer C. Genetic variants of the fatty acid

Lehman TA and Harris CC. Mutational spectra of protooncogenes and tumour suppressor genes: clues in predicting cancer aetiology. IARC Sci. 125:399-412, 1994. Lehninger AL, Nelson DL, Cox MM. Princíples of Biochemistry, 4th edition, New York, NY:

Ma X, Torbenson M, Hamad ARA, Soloski MJ, Li Z. High-fat diet modulates non-CD1d-

MacLean C, Mojica WA, Newberry SJ et al., Systematic review of the effects of n-3 fatty acids in inflammatory bowel disease. Am J Clin Nutr. 82:611-9, 2005. McCafferty DM, Sihota E, Muscara M, Wallace JL, Sharkey KA, Kubes P. Spontaneously

Malerba G, Schaeffer L, Xumerle L, et al. SNPs of the FADS gene cluster are associated with

Mayer K, Gokorsc S, Fegbeutel C, Hattar K et al. Parenteral nutrition with fish oil modulates

McCafferty DM, Sihota E, Muscara M, Wallace JL, Sharkey KA, Kubes P. Spontaneously

Miranda DC, Arçari DP, Pedrazolli Jr, Carvalho PO, Cerutti SM, Bastos DHM, Ribeiro ML.

Nowak J, W, Weylandt KH, Habbel P, Wang J, Dignass A, Glickman J, Kang J. Colitis-

Nystrom M & Mutanen M. Diet and epigenetics in colon cancer. World J Gastroenterol,

Paoloni-Giacobino A, Grimble R, and Pichard C. Genomic interactions with disease and

Paulsen JE, Elvsaas IK, Steffensen IL, Alexander J. A fish oil derived concentrate enriched in

Essential Fatty Acids. 58 (3):243-248, 1998.

W.H. Freeman and Company, 2005.

Gastrointest Liver Physiol. 279:G90-G99, 2000.

Gastrointest Liver Physiol. 279:G90-G99, 2000.

nutrition. Clin Nutr. 22:507-514, 2003.

and DNA repais in mice. Mutagenesis. 23:261-5, 2008.

endogenous n-3 fatty acids. Carcinogenesis. 28:1991-1995, 2007.

277, 2001.

138, 2007.

Lipids, 2008.

15:257-63, 2009.

2003.

activities to leukotriene B4 by n-3 polyunsaturated fatty acids. Prost, LK and

desaturase gene cluster predict amounts of red blood cell docosahexaenoic and other polyunsaturated fatty acids in pregnant women: findings from the Avon Longitudinal Study of Parents and Children. Am J Clin Nutr. 93(1):211-9, 2011. Laroux FS, Pavlick KP, Wolf RE, Grisham MB. Dysregulation of intestinal mucosal

immunity: implications in inflammatory bowel disease. News Physiol Sci. 16:272-

restricted natural killer T cells and regulatory T cells in mouse colon and exacerbates experimental colitis. Clinical and Experimental Immunology. 151:130-

developing chronic colitis in IL-10/iNOS double-deficient mice. Am J Physiol

polyunsaturated fatty acids in a cohort of patients with cardiovascular disease.

cytokine response in patients with sepsis. Am J Resp Crit Care Med. 167:1321-1328,

developing chronic colitis in IL-10/iNOS double-deficient mice. Am J Physiol

Protective effects of mate tea (Ilex paraguariensis) on H2O2-induced DNA damage

associated colon tumorigenesis is suppressed in transgenic mice rich in

eicosapentaenoic and docosahexaenoic acid as ethyl ester suppresses the formation and growth of intestinal polyps in the min mouse. Carcinogenesis 18:1905-10, 1997.


**0**

**9**

<sup>1</sup>*USA* 2,3,4*UK*

**The Molecular Genetic Events in Colorectal Cancer and Diet**

<sup>3</sup>*Medical Research Council Centre for Nutritional Epidemiology*

<sup>4</sup>*Department of Pathology, Addenbrooke's Hospital, University of Cambridge*

<sup>1</sup>*Cold Spring Harbor Laboratory* <sup>2</sup>*Queen Mary, University of London*

*in Cancer Prevention and Survival*

Adam Naguib1, Laura J Gay2, Panagiota N Mitrou3 and Mark J Arends4

Compelling evidence suggests that dietary intakes directly influence colorectal cancer (CRC) risk. Initial observations that CRC incidence is not ubiquitous worldwide, with incidence rates varying up to twenty-five fold between populations (Parkin et al., 2005), indicate the large degree to which this cancer type is influenced by diet and environment. Additionally, observations that migration of individuals confers rapid (within one generation) adoption of the CRC incidence of the host population (Boyle & Langman, 2000; McMichael & Giles, 1988), suggest that dietary and environmental factors determine the risk of colorectal neoplasia to a

As diagnosis and treatment of CRC have improved, the study of the pathogenesis of colorectal neoplasia has increased. The most frequent precursor of CRC is the adenoma. As a proportion of adenomas, those of large size, with villous architecture and high grade dysplasia often progress to invasive adenocarcinoma, and this progression is associated with accumulation of mutations and other genetic and epigenetic changes. In the effort to understand the mechanisms and causes of colorectal cancer development, molecular genetic analyses have identified a variety of molecular changes and protein targets involved in colorectal tumourigenesis. The greater understanding of genetic, epigenetic and expression changes that occur during the development and progression of CRC has shown that these neoplasms do not comprise a single disease. Instead, colorectal cancers comprise a collection of distinct and independent neoplastic pathways, such as those pathways displaying chromosomal instability (CIN), microsatellite instability (MSI) or gene promoter activity changes due to the epigenetic phenomenon of methylation at CG dinucleotides (referred to as CIMP: CpG island methylation phenotype, whereby CpG describes dinoculeotides of cytosine and guanosine, separated by the characteristic phosphate group in the DNA structure). Each pathway subtype is characterised by individual genetic and molecular characteristics (Poulogiannis,

degree similar to, or in excess of, genetic predisposition.

**1. Introduction**


### **The Molecular Genetic Events in Colorectal Cancer and Diet**

Adam Naguib1, Laura J Gay2, Panagiota N Mitrou3 and Mark J Arends4

<sup>1</sup>*Cold Spring Harbor Laboratory* <sup>2</sup>*Queen Mary, University of London* <sup>3</sup>*Medical Research Council Centre for Nutritional Epidemiology in Cancer Prevention and Survival* <sup>4</sup>*Department of Pathology, Addenbrooke's Hospital, University of Cambridge* <sup>1</sup>*USA* 2,3,4*UK*

#### **1. Introduction**

172 Colorectal Cancer – From Prevention to Patient Care

Serhan CN, Arita M, Hong S, Gotlinger K. Resolvins, docosatrienes, and neuroprotectins,

Seril DN, Liao J, Yang GY, Yang CS. Oxidative stress and ulcerative colitis-associated

Shimizu T, Igarashi J, Ohtuka Y, Oguchi S, Kaneko K, Yamashiro Y. Effects of n-3

Simopoulos AP. Importance of the ratio of omega-6/omega-3 essential fatty acids:

Simopoulos AP. Omega-3 fatty acids in inflammation and autoimmune diseases. J Am Coll

Simopoulos AP. The importance of the omega-6 omega-3 fatty acid ratio in cardiovascular disease and other chronic diseases. ExpBiolMed(Maywood). 233:674–88, 2008. Stein RB, Hanauer SB. Comparative tolerability of treatments for inflammatory bowel

Stern MC, Butler LM, Corral R, et al., Polyunsaturated Fatty Acids, DNA Repair Single

Udilova N, Jurck D, Marian B et al., Induction of peroxidation in biomebranes by dietary oil

Wagner BA; Buettner GR and Burns. Free radical-mediated lipid peroxidation in cells:

Weiss G, Meyer F, Mathies B, Pross M, et al., Immunomodulation by perioperative

Wild GE, Drozdowski L et al., Nutritional modulation of the inflammatory response in

YaqooB P & Calder PC. Effects of dietary lipid manipulation upon inflammatory mediator production murine macrophages. Cellular Immunology. 163: 120-128, 1996. Zhang P, Smith R, Chapkin RS, McMurray DN. Dietary (n-3) polyunsaturated fatty acids

components. Food and chemical toxicology. 41:1481- 1489, 2003.

administration of n-3 fatty acids. Brit J Nutr. 87:S89-S94, 2002.

Nucleotide Polymorphisms and Colorectal Cancer in the Singapore Chinese Health

oxidizability is a function of cell lipid bis-allylic hydrogen content. Biochemistry.

inflammatory bowel disease – from the molecular to the integrative to the clinical.

modulates murine Th1/Th2 balance toward the pole by supression of Th1

evolutionary aspects. World Rev Nutr Diet. 92:1-22, 2003.

Study. J Nutrigenet Nutrigenomics. 2:273–279, 2009.

Lipids. 39:1125–32, 2004.

colitis. Digestion. 63:49-54, 2001.

disease. Drug Saf. 23:429-48, 2000.

World J Gastroenterol. 13:1-7, 2007.

development. J Nutr. 135:1745-1751, 2005.

Nutr. 21:495-505, 2002.

33:4449, 1994.

2003.

novel omega-3-derived mediators, and their endogenous aspirin-triggered epimers.

carcinogenesis: studies in humans and animal models. Carcinogenesis. 24:353-62,

polyunsaturated fatty acids and vitamin E on colonic mucosal leukotriene generation, lipid peroxidation, and microcirculation in rats with experimental

> Compelling evidence suggests that dietary intakes directly influence colorectal cancer (CRC) risk. Initial observations that CRC incidence is not ubiquitous worldwide, with incidence rates varying up to twenty-five fold between populations (Parkin et al., 2005), indicate the large degree to which this cancer type is influenced by diet and environment. Additionally, observations that migration of individuals confers rapid (within one generation) adoption of the CRC incidence of the host population (Boyle & Langman, 2000; McMichael & Giles, 1988), suggest that dietary and environmental factors determine the risk of colorectal neoplasia to a degree similar to, or in excess of, genetic predisposition.

> As diagnosis and treatment of CRC have improved, the study of the pathogenesis of colorectal neoplasia has increased. The most frequent precursor of CRC is the adenoma. As a proportion of adenomas, those of large size, with villous architecture and high grade dysplasia often progress to invasive adenocarcinoma, and this progression is associated with accumulation of mutations and other genetic and epigenetic changes. In the effort to understand the mechanisms and causes of colorectal cancer development, molecular genetic analyses have identified a variety of molecular changes and protein targets involved in colorectal tumourigenesis. The greater understanding of genetic, epigenetic and expression changes that occur during the development and progression of CRC has shown that these neoplasms do not comprise a single disease. Instead, colorectal cancers comprise a collection of distinct and independent neoplastic pathways, such as those pathways displaying chromosomal instability (CIN), microsatellite instability (MSI) or gene promoter activity changes due to the epigenetic phenomenon of methylation at CG dinucleotides (referred to as CIMP: CpG island methylation phenotype, whereby CpG describes dinoculeotides of cytosine and guanosine, separated by the characteristic phosphate group in the DNA structure). Each pathway subtype is characterised by individual genetic and molecular characteristics (Poulogiannis,

in Colorectal Cancer and Diet 3

The Molecular Genetic Events in Colorectal Cancer and Diet 175

Fig. 1. **RAS, RAF and the MAPK signalling pathway: frequently perturbed in colorectal neoplasms.** Initially, RAS is inactive in a RAS-GTP bound state. **I:** Initiation of signalling through the MAPK pathway occurs at the plasma membrane. Upon extracellular ligand binding to membrane receptor tyrosine kinases (RTK), such as epidermal growth factor binding to the epidermal growth factor receptor (IA), receptor conformational change gives rise to receptor autophosphorylation. Subsequently, src-homology 2 (SH2) domains present in the GRB2 adaptor protein bind the phosphate moieties on the activated receptor (IB). Src-homology 3 (SH3) domains in GRB2 bind proline-rich motifs present in son of sevenless

(SOS), localising SOS to the inner surface of the plasma membrane. SOS, a guanine nucleotide exchange factor (GEF) interacts with RAS proteins, catalysing the exchange of GDP for GTP, thus activating RAS to a RAS-GTP state (IC). **II:** Upon activation, RAS phosphorylates cytosolic RAF. The resulting activation of RAF in turn phosphorylates cytosolic MEK, which then phosphorylates ERK, leading to induction and repression of distinct transcription programmes, promoting cell proliferation and modulating cell death by apoptosis, among other processes. The vast majority of mutations in the *K-RAS* or *BRAF* genes are in distinct hotspot regions: *K-RAS* at codons 12 and 13, and also, but much more infrequently at codons 61 and 146 (Forbes et al., 2008). Additionally, mutations observed at lower prevalences at other sites in the gene have been described and their functional

significance determined (Naguib, Wilson, Adams & Arends, 2011). Mutations in *BRAF* occur most frequently at codons 463-468 and codon 600 (Forbes et al., 2008). Activating mutations in the *K-RAS* and *BRAF* genes render their protein products constitutively active, leading to increased transduction through this signalling axis. Additionally, mutationally active *K-RAS* can also propogate signalling through other pathways, including the PI3K/AKT axis.

Luo, Poulogiannis, Ye, Hamoudi, Zhang, Dong & Arends, 2011). *K-RAS* mutations are observed 20-50% of sporadic human CRC and *BRAF* mutations are observed in 5-15% of CRC (Forbes et al., 2008). The high frequencies at which *K-RAS* and *BRAF* mutations are observed in CRC has prompted several analyses of dietary intakes in relation to these genetic lesions.

Specific types of meat consumption have been identified as associated with general CRC incidence (Norat et al., 2005; Santarelli et al., 2008) with plausible mechanisms postulated as to the manner in which these consumptions may influence colorectal carcinogenesis (Kuhnle & Bingham, 2007; Kuhnle et al., 2007). Consequently, several studies have attempted to identify

**2.1.1** *K-RAS* **mutation and meat consumption**

Ichimura, Hamoudi, Luo, Leung, Yuen, Harrison, Wyllie & Arends, 2010; Poulogiannis, McIntyre, Dimitriadi, Apps, Wilson, Ichimura, Luo, Cantley, Wyllie, Adams & Arends, 2010). Dietary constituents have been studied in relation to the major genetic and molecular changes occurring in CRC development, including alterations in the proto-oncogenes, *K-RAS* and *BRAF* and the tumour suppressor genes *p53* and *APC*. Many studies have analysed a wide variety of dietary components in an effort to elucidate which, if any, dietary constituents may contribute to their mutation in CRC progression. Furthermore, in addition to these genetic lesions, the epigenetic phenomenon of CIMP and MSI have similarly been analysed in relation to dietary constituents.

This review is intended to summarise the currently available literature describing the associations between the molecular genetic changes seen most prevalently in colorectal cancer and dietary intakes. This report does not attempt to assess dietary associations with total CRC incidence. The objective is to highlight consensus observations, where several sources of data exist, suggestive of causative or protective effects of dietary constituents regarding specific molecular genetic changes frequently observed in colorectal neoplasia. Throughout, an emphasis is placed on the number of cases analysed in individual studies, but notably absent are descriptions of odds ratios, hazard ratios or p-values. Throughout, all associations discussed are statistically significant (all p0.05). However, due to the varying methodology of data collection and statistical analysis across studies, the inclusion of differing variables in adjusted models and the lack of consensus regarding the degree to which analyses should be adjusted following multiple statistical tests, detailed statistical aspects are not discussed. In order for an assessment to be made of the potential statistical power of each analysis, the number of cases involved in each study is instead highlighted when a statistically significant association is discussed. Full details of all statistical analyses can be found in the original reports, referenced in the text and listed at the end of the chapter.

#### **2. Dietary influences on the major genetic and epigenetic perturbations leading to colorectal cancer development and progression**

#### **2.1** *K-RAS* **and** *BRAF* **in colorectal cancer: the MAPK signalling pathway**

Mitogen activated protein kinase (MAPK) signal transduction pathways are present in all eukaryotes, six versions of which have been distinguished in mammals (Robinson & Cobb, 1997). MAPK signal propagation is responsible for regulating a variety of cellular processes, which include potentially pro-tumourignenic properties such as proliferation, apoptosis and transformation (Arends et al., 1993; 1994; Peyssonnaux & Eychene, 2001; Robinson & Cobb, 1997). The best characterised of these pathways is the LIGAND RECEPTOR-RAS-RAF-MEK-ERK pathway (Figure 1), which consists of core modules including the RAS and RAF proteins. Although three *RAS* genes have been identified (*H-RAS*, *N-RAS* and *K-RAS*), the *K-RAS* gene is the only one mutated at significant frequency in CRC (Bos, 1989). Similarly, of the three *RAF* genes identified (*ARAF*, *BRAF* and *CRAF*/*RAF-1*), only the *BRAF* gene is mutated at significant frequencies in human cancers (Fransen et al., 2004). Experimental mouse models have provided direct evidence that mutated *K-RAS* genes expressed in the intestinal epithelium do not significantly initiate intestinal adenoma growth, but they can cooperate either with other mutant genes or carcinogens to accelerate intestinal tumour formation (Luo et al., 2007; 2009; Luo, Poulogiannis, Ye, Hamoudi & Arends, 2011; 2 Colorectal Cancer

Ichimura, Hamoudi, Luo, Leung, Yuen, Harrison, Wyllie & Arends, 2010; Poulogiannis, McIntyre, Dimitriadi, Apps, Wilson, Ichimura, Luo, Cantley, Wyllie, Adams & Arends, 2010). Dietary constituents have been studied in relation to the major genetic and molecular changes occurring in CRC development, including alterations in the proto-oncogenes, *K-RAS* and *BRAF* and the tumour suppressor genes *p53* and *APC*. Many studies have analysed a wide variety of dietary components in an effort to elucidate which, if any, dietary constituents may contribute to their mutation in CRC progression. Furthermore, in addition to these genetic lesions, the epigenetic phenomenon of CIMP and MSI have similarly been analysed in relation

This review is intended to summarise the currently available literature describing the associations between the molecular genetic changes seen most prevalently in colorectal cancer and dietary intakes. This report does not attempt to assess dietary associations with total CRC incidence. The objective is to highlight consensus observations, where several sources of data exist, suggestive of causative or protective effects of dietary constituents regarding specific molecular genetic changes frequently observed in colorectal neoplasia. Throughout, an emphasis is placed on the number of cases analysed in individual studies, but notably absent are descriptions of odds ratios, hazard ratios or p-values. Throughout, all associations discussed are statistically significant (all p0.05). However, due to the varying methodology of data collection and statistical analysis across studies, the inclusion of differing variables in adjusted models and the lack of consensus regarding the degree to which analyses should be adjusted following multiple statistical tests, detailed statistical aspects are not discussed. In order for an assessment to be made of the potential statistical power of each analysis, the number of cases involved in each study is instead highlighted when a statistically significant association is discussed. Full details of all statistical analyses can be found in the original

**2. Dietary influences on the major genetic and epigenetic perturbations leading to**

Mitogen activated protein kinase (MAPK) signal transduction pathways are present in all eukaryotes, six versions of which have been distinguished in mammals (Robinson & Cobb, 1997). MAPK signal propagation is responsible for regulating a variety of cellular processes, which include potentially pro-tumourignenic properties such as proliferation, apoptosis and transformation (Arends et al., 1993; 1994; Peyssonnaux & Eychene, 2001; Robinson & Cobb, 1997). The best characterised of these pathways is the LIGAND RECEPTOR-RAS-RAF-MEK-ERK pathway (Figure 1), which consists of core modules including the RAS and RAF proteins. Although three *RAS* genes have been identified (*H-RAS*, *N-RAS* and *K-RAS*), the *K-RAS* gene is the only one mutated at significant frequency in CRC (Bos, 1989). Similarly, of the three *RAF* genes identified (*ARAF*, *BRAF* and *CRAF*/*RAF-1*), only the *BRAF* gene is mutated at significant frequencies in human cancers (Fransen et al., 2004). Experimental mouse models have provided direct evidence that mutated *K-RAS* genes expressed in the intestinal epithelium do not significantly initiate intestinal adenoma growth, but they can cooperate either with other mutant genes or carcinogens to accelerate intestinal tumour formation (Luo et al., 2007; 2009; Luo, Poulogiannis, Ye, Hamoudi & Arends, 2011;

reports, referenced in the text and listed at the end of the chapter.

**2.1** *K-RAS* **and** *BRAF* **in colorectal cancer: the MAPK signalling pathway**

**colorectal cancer development and progression**

to dietary constituents.

Fig. 1. **RAS, RAF and the MAPK signalling pathway: frequently perturbed in colorectal neoplasms.** Initially, RAS is inactive in a RAS-GTP bound state. **I:** Initiation of signalling through the MAPK pathway occurs at the plasma membrane. Upon extracellular ligand binding to membrane receptor tyrosine kinases (RTK), such as epidermal growth factor binding to the epidermal growth factor receptor (IA), receptor conformational change gives rise to receptor autophosphorylation. Subsequently, src-homology 2 (SH2) domains present in the GRB2 adaptor protein bind the phosphate moieties on the activated receptor (IB). Src-homology 3 (SH3) domains in GRB2 bind proline-rich motifs present in son of sevenless (SOS), localising SOS to the inner surface of the plasma membrane. SOS, a guanine nucleotide exchange factor (GEF) interacts with RAS proteins, catalysing the exchange of GDP for GTP, thus activating RAS to a RAS-GTP state (IC). **II:** Upon activation, RAS phosphorylates cytosolic RAF. The resulting activation of RAF in turn phosphorylates cytosolic MEK, which then phosphorylates ERK, leading to induction and repression of distinct transcription programmes, promoting cell proliferation and modulating cell death by apoptosis, among other processes. The vast majority of mutations in the *K-RAS* or *BRAF* genes are in distinct hotspot regions: *K-RAS* at codons 12 and 13, and also, but much more infrequently at codons 61 and 146 (Forbes et al., 2008). Additionally, mutations observed at lower prevalences at other sites in the gene have been described and their functional significance determined (Naguib, Wilson, Adams & Arends, 2011). Mutations in *BRAF* occur most frequently at codons 463-468 and codon 600 (Forbes et al., 2008). Activating mutations in the *K-RAS* and *BRAF* genes render their protein products constitutively active, leading to increased transduction through this signalling axis. Additionally, mutationally active *K-RAS* can also propogate signalling through other pathways, including the PI3K/AKT axis.

Luo, Poulogiannis, Ye, Hamoudi, Zhang, Dong & Arends, 2011). *K-RAS* mutations are observed 20-50% of sporadic human CRC and *BRAF* mutations are observed in 5-15% of CRC (Forbes et al., 2008). The high frequencies at which *K-RAS* and *BRAF* mutations are observed in CRC has prompted several analyses of dietary intakes in relation to these genetic lesions.

#### **2.1.1** *K-RAS* **mutation and meat consumption**

Specific types of meat consumption have been identified as associated with general CRC incidence (Norat et al., 2005; Santarelli et al., 2008) with plausible mechanisms postulated as to the manner in which these consumptions may influence colorectal carcinogenesis (Kuhnle & Bingham, 2007; Kuhnle et al., 2007). Consequently, several studies have attempted to identify

in Colorectal Cancer and Diet 5

The Molecular Genetic Events in Colorectal Cancer and Diet 177

plausible that if specific meat types, as suggested in at least one study (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, Goldbohm & van den Brandt, 2005), are linked to the incidence of *K-RAS* mutated CRC, that grouping of meat types together may have failed to identify associations where they existed. However, in practical terms, it should be noted that similarities in the composition of meat types, such as in terms of haem content, a postulated carcinogen intermediate (Kuhnle & Bingham, 2007), justify a grouping of types in order to minimise multiple statistical testing and to test consumption levels large enough to be likely

Several reports have analysed the relationship between base changes at specific positions in the *K-RAS* gene, types of mutations (i.e. transition *versus* transversion) or specific types of base changes (i.e. G→A) in relation to meat intakes. It is entirely plausible that the nature of the mutation, not the gene in which it arises, is linked to dietary constituents. However, due to the very limited number of studies instigated with objectives of such an analysis, and the often low numbers of different mutation subgroups existent in the studies which do attempt such an assessment rendering lower statistical power, such analyses are not discussed in this

Several studies have described an association between folate intake and the prevalence of *K-RAS* mutations in CRC. A report analysing 390 *K-RAS* wildtype and 218 *K-RAS* mutated CRC identified an increased consumption of folate associated with a reduced prevalence of *K-RAS* mutated rectal, but not colonic, cancers in males only (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, van Engeland, Goldbohm & van den Brandt, 2005). Testing in this study demonstrated that in the male participants of this cohort, increased intake of folate was linked to reduced prevalence of rectal cancer incidence, however, this link, when considering mutation status, seemed only to reduce the risk of *K-RAS* mutated rectal cancers. A large analysis of colorectal adenomas (558 wildtype, 120 *K-RAS* mutated) also identified increased folate intake associated with a reduced incidence of *K-RAS* mutation (Martínez et al., 1999). However, in addition to these positive associations in relatively large cohorts, several other studies have failed to identify a link between folate intake and *K-RAS* mutation status in

**Study** *K-RAS* **WT** *K-RAS* **mutated dietary**

increase or decrease in consumption respectively.

**2.1.2** *K-RAS* **mutation and folate consumption**

to affect bowel carcinogenesis.

review.

**CRC/RC/adenomas CRC/RC/adenomas association** Bautista *et al* 1997 CRC: 67 CRC: 39 ↑ MUFA with *K-RAS* mutation, ↓ calcium with *K-RAS* mutation Bongaerts *et al* 2006 CRC: 385 CRC: 193 no association between alcohol and *K-RAS* mutated or wildtype cancers Brink *et al* 2004 CRC: 390 CRC: 218 ↑ PUFA (specifically linoleic acid) with *K-RAS* mutated colonic, but not rectal, cancers Brink *et al* 2005 CRC: 390 CRC: 218 ↑ folate reduced risk of *K-RAS* mutated rectal, not colonic, cancer in men only Brink *et al* 2005 CRC: 390 CRC: 218 ↑ beef, ↓ pork with *K-RAS* wildtype colonic tumours, ↓ pork with *K-RAS* wildtype rectal tumours Laso *et al* 2004 CRC: 68 CRC: 49 *K-RAS* codon 12 mutation was associated with ↓ vitamin A, B1, D and iron Martinez *et al* 1999 Adenomas: 558 Adenomas: 120 ↑ folate reduced risk of developing *K-RAS* mutated adenomas Naguib *et al* 2010 CRC: 155 CRC: 41 ↑ white meat consumption with *K-RAS* mutation O'Brien *et al* 2000 RC: 28 RC: 15 no association between red meat consumption and *K-RAS* mutation Schernhammer *et al* 2008 CRC: 427 CRC: 242 no association between folate intake and prevalence of *K-RAS* mutated or wildtype cancers Slattery *et al* 2000 CRC: 971 CRC: 457 ↓ cruciferous vegetables with reduced risk of *K-RAS* mutation Slattery *et al* 2010 RC: 535 RC: 215 no association between calcium and vitamin D and *K-RAS* mutation Slattery *et al* 2010 RC: 535 RC: 215 ↑ vegetables and dietary fibre with a reduced risk of *K-RAS* mutations Wark *et al* 2006 Adenomas: 453 Adenomas: 81 ↓ MUFA and ↑ vitamin B2 associated with *K-RAS* mutation Table 1. Summarised description of literature analysing *K-RAS* mutations in colorectal neoplasms (case numbers provided) in relation to dietary intakes and the statistically significant findings described. *WT*: wildtype, *CRC*: colorectal cancer, *RC*: rectal cancer, *MUFA*: monunsaturated fatty acid, *PUFA*: polyunsaturated fatty acid, ↑ and ↓ denote an

the nature of these associations in relation to *K-RAS* mutations. Some reports have identified associations with meat consumption and *K-RAS* mutation, although not all.

A single study analysing 390 *K-RAS* wildtype and 218 *K-RAS* mutated CRC identified an increased consumption of beef with higher incidence of *K-RAS* wildtype colonic cancers (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, Goldbohm & van den Brandt, 2005). In this same report, a reduction in pork consumption was found to be linked to reduced frequency of both colonic and rectal cancers harbouring mutated *K-RAS*. Another report, assessing *K-RAS* mutations and diet in 155 *K-RAS* wildtype and 41 *K-RAS* mutated CRC, identified an increased white meat consumption associated with higher incidence of *K-RAS* mutated CRC (Naguib et al., 2010). Although positive associations were identified in these two analyses, there appears to be little consistency between these independent findings. The report by Naguib and colleagues also analysed red and processed meat consumption in relation to mutation status and found no statistically significant association between the two, although, beef consumption was not tested independently of other meat types, as in the report by Brink and co-workers. The study by Naguib and colleagues did not test pork consumption in isolation: this meat type was included in the 'red' or 'processed' meat categories. Similarly, Brink and coworkers did not identify an association between white meat and increased incidence of *K-RAS* mutations. This analysis tested the consumption of chicken in isolation, not in a combined 'white meat' category containing other meat types, such as turkey etc.

Notwithstanding the identified statistically significant associations between meat consumption and *K-RAS* mutation status described above, the majority of studies which have attempted to address this question have failed to identify any link between meat intake and the mutation status of this gene. An analysis testing 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC assessed animal protein intake and found no link between this and *K-RAS* mutation status (Bautista et al., 1997) although clearly, 'animal protein' as a variable makes no distinction between meat types and is an assessment of protein, not animal product, consumption. A large analysis testing 971 *K-RAS* wildtype and 457 *K-RAS* mutated CRC (Slattery et al., 2000) identified no assocaition between total *K-RAS* mutations and meat intake. A small study (28 wildtype, 15 mutated rectal cancers) failed to identify an association between red meat intake and *K-RAS* mutation (O'Brien et al., 2000). An assessment of a larger cohort of rectal cancers (535 *K-RAS* wildtype and 215 *K-RAS* mutated) corroborated this observation of lack of association with red meat intake and rectal cancer (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010).

In addition to colorectal cancers, pre-cancerous adenomas have also been tested in order to identify dietary assocaitions with *K-RAS* mutation status in the early stages of colorectal neoplasia. An assessment of 558 *K-RAS* wildtype and 120 *K-RAS* mutated adenomas failed to identify an association between red meat intake and mutation status (Martínez et al., 1999). Another study, testing 453 *K-RAS* wildtype and 81 *K-RAS* mutated adenomas also failed to identify a statistically significant association between red meat, processed meat or poultry and *K-RAS* mutation status (Wark et al., 2006).

Published reports assessing *K-RAS* mutation status in CRC in relation to meat intakes provide limited evidence to suggest that total *K-RAS* mutations are either positively or negatively associated with meat consumption. The majority of studies have categorised meat types according to shared properties (such as haem content or preservation methods) and have generally failed to identify links between these groups and *K-RAS* mutation status. It is

#### 176 Colorectal Cancer – From Prevention to Patient Care The Molecular Genetic Events in Colorectal Cancer and Diet 5 The Molecular Genetic Events in Colorectal Cancer and Diet 177

4 Colorectal Cancer

the nature of these associations in relation to *K-RAS* mutations. Some reports have identified

A single study analysing 390 *K-RAS* wildtype and 218 *K-RAS* mutated CRC identified an increased consumption of beef with higher incidence of *K-RAS* wildtype colonic cancers (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, Goldbohm & van den Brandt, 2005). In this same report, a reduction in pork consumption was found to be linked to reduced frequency of both colonic and rectal cancers harbouring mutated *K-RAS*. Another report, assessing *K-RAS* mutations and diet in 155 *K-RAS* wildtype and 41 *K-RAS* mutated CRC, identified an increased white meat consumption associated with higher incidence of *K-RAS* mutated CRC (Naguib et al., 2010). Although positive associations were identified in these two analyses, there appears to be little consistency between these independent findings. The report by Naguib and colleagues also analysed red and processed meat consumption in relation to mutation status and found no statistically significant association between the two, although, beef consumption was not tested independently of other meat types, as in the report by Brink and co-workers. The study by Naguib and colleagues did not test pork consumption in isolation: this meat type was included in the 'red' or 'processed' meat categories. Similarly, Brink and coworkers did not identify an association between white meat and increased incidence of *K-RAS* mutations. This analysis tested the consumption of chicken in isolation, not in a combined 'white meat' category containing other meat types, such as turkey etc. Notwithstanding the identified statistically significant associations between meat consumption and *K-RAS* mutation status described above, the majority of studies which have attempted to address this question have failed to identify any link between meat intake and the mutation status of this gene. An analysis testing 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC assessed animal protein intake and found no link between this and *K-RAS* mutation status (Bautista et al., 1997) although clearly, 'animal protein' as a variable makes no distinction between meat types and is an assessment of protein, not animal product, consumption. A large analysis testing 971 *K-RAS* wildtype and 457 *K-RAS* mutated CRC (Slattery et al., 2000) identified no assocaition between total *K-RAS* mutations and meat intake. A small study (28 wildtype, 15 mutated rectal cancers) failed to identify an association between red meat intake and *K-RAS* mutation (O'Brien et al., 2000). An assessment of a larger cohort of rectal cancers (535 *K-RAS* wildtype and 215 *K-RAS* mutated) corroborated this observation of lack of association with red meat intake and rectal cancer (Slattery, Curtin,

In addition to colorectal cancers, pre-cancerous adenomas have also been tested in order to identify dietary assocaitions with *K-RAS* mutation status in the early stages of colorectal neoplasia. An assessment of 558 *K-RAS* wildtype and 120 *K-RAS* mutated adenomas failed to identify an association between red meat intake and mutation status (Martínez et al., 1999). Another study, testing 453 *K-RAS* wildtype and 81 *K-RAS* mutated adenomas also failed to identify a statistically significant association between red meat, processed meat or poultry

Published reports assessing *K-RAS* mutation status in CRC in relation to meat intakes provide limited evidence to suggest that total *K-RAS* mutations are either positively or negatively associated with meat consumption. The majority of studies have categorised meat types according to shared properties (such as haem content or preservation methods) and have generally failed to identify links between these groups and *K-RAS* mutation status. It is

associations with meat consumption and *K-RAS* mutation, although not all.

Wolff, Herrick, Caan & Samowitz, 2010).

and *K-RAS* mutation status (Wark et al., 2006).


Table 1. Summarised description of literature analysing *K-RAS* mutations in colorectal neoplasms (case numbers provided) in relation to dietary intakes and the statistically significant findings described. *WT*: wildtype, *CRC*: colorectal cancer, *RC*: rectal cancer, *MUFA*: monunsaturated fatty acid, *PUFA*: polyunsaturated fatty acid, ↑ and ↓ denote an increase or decrease in consumption respectively.

plausible that if specific meat types, as suggested in at least one study (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, Goldbohm & van den Brandt, 2005), are linked to the incidence of *K-RAS* mutated CRC, that grouping of meat types together may have failed to identify associations where they existed. However, in practical terms, it should be noted that similarities in the composition of meat types, such as in terms of haem content, a postulated carcinogen intermediate (Kuhnle & Bingham, 2007), justify a grouping of types in order to minimise multiple statistical testing and to test consumption levels large enough to be likely to affect bowel carcinogenesis.

Several reports have analysed the relationship between base changes at specific positions in the *K-RAS* gene, types of mutations (i.e. transition *versus* transversion) or specific types of base changes (i.e. G→A) in relation to meat intakes. It is entirely plausible that the nature of the mutation, not the gene in which it arises, is linked to dietary constituents. However, due to the very limited number of studies instigated with objectives of such an analysis, and the often low numbers of different mutation subgroups existent in the studies which do attempt such an assessment rendering lower statistical power, such analyses are not discussed in this review.

#### **2.1.2** *K-RAS* **mutation and folate consumption**

Several studies have described an association between folate intake and the prevalence of *K-RAS* mutations in CRC. A report analysing 390 *K-RAS* wildtype and 218 *K-RAS* mutated CRC identified an increased consumption of folate associated with a reduced prevalence of *K-RAS* mutated rectal, but not colonic, cancers in males only (Brink, Weijenberg, de Goeij, Roemen, Lentjes, de Bruïne, van Engeland, Goldbohm & van den Brandt, 2005). Testing in this study demonstrated that in the male participants of this cohort, increased intake of folate was linked to reduced prevalence of rectal cancer incidence, however, this link, when considering mutation status, seemed only to reduce the risk of *K-RAS* mutated rectal cancers. A large analysis of colorectal adenomas (558 wildtype, 120 *K-RAS* mutated) also identified increased folate intake associated with a reduced incidence of *K-RAS* mutation (Martínez et al., 1999).

However, in addition to these positive associations in relatively large cohorts, several other studies have failed to identify a link between folate intake and *K-RAS* mutation status in

in Colorectal Cancer and Diet 7

The Molecular Genetic Events in Colorectal Cancer and Diet 179

In addition to these observations, one report describes an increase in polyunsaturated fatty acids (PUFA), specifically linoleic acid, as associated with increased prevalence of *K-RAS* mutated colonic, but not rectal, cancers (Brink et al., 2004). However, this association with PUFA has not been identifed in any other report (Bautista et al., 1997; Laso et al., 2004; Naguib et al., 2010; Slattery et al., 2000; Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010; Wark

Taken together, the published data describing the association of dietary fats with *K-RAS* mutations have failed to identify a convincing association, and have generated conflicting results. Presently, the evidence suggesting that the *K-RAS* mutation status of colorectal neoplasia may be affected by fat intakes is weak: the limited data available suggest that the mutation status of this gene is largely independent of this dietary consumption. It should be noted however, that although fat intake itself is probably not associated with this mutation type, increased body mass index (BMI), which may be associated with fat intake, is associated

The mutation status of *K-RAS* in CRC has also been linked to several other dietary variables in addition to meat, folate and fat. Testing of 971 *K-RAS* wildtype and 457 *K-RAS* mutated CRC identified an increased risk of *K-RAS* mutations with reduced consumption of cruciferous vegetables (Slattery et al., 2000). Another analysis of rectal cancers (535 wildtype, 215 mutated) identified a reduced incidence of *K-RAS* mutated rectal cancers with increased vegetable and fibre intake (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). Although corroborative, these two analyses were performed on the same test cohort and are yet to be identified in other independent populations. In this cohort at least, the data of Slattery and colleagues suggest that increased vegetable intake reduced the prevalence of *K-RAS* mutations in CRC, with an

Increased vitamin B2 intake has been identified to reduce the prevalence of adenomas harbouring *K-RAS* mutations. In an analysis of 453 *K-RAS* mutated and 81 *K-RAS* wildtype pre-cancerous adenomas an inverse association suggested a protective effect against *K-RAS* mutated adenomas. This protection was not found in relation to the prevalence of *K-RAS* wildtype adenomas (Wark et al., 2006). This association has not been identified in cohorts

Some dietary intakes have been repeatedly tested with no link to the prevalence of *K-RAS* mutation in CRC having been identified, notably alcohol. Many studies have included assessment of this dietary factor, with some studies analysing alcohol consumption

In summary, current literature describing the assessment of *K-RAS* mutation status in colorectal neoplasia has identified many associations with dietary intakes (summarised in Table 1). Very few of these associations have been repeatedly identified in independent cohorts, making assessment of their general validity challenging. Presently, few dietary components seem to be strongly linked to *K-RAS* mutation status in CRC across many populations, environments and genetic backgrounds. Furthermore, it is problematic to directly compare different studies. Other than folate, which has been described by the World Cancer Research Fund as having a 'limited' protective effect against CRC, which may impart this limited protection through reduced prevalence of *K-RAS* mutation, there is a lack of strong

et al., 2006).

with overall CRC risk.

testing colorectal cancers.

**2.1.4** *K-RAS* **mutation and other dietary constituents**

overt association identified in rectally located neoplasia.

independent of any other dietary factors (Bongaerts et al., 2006).

colorectal neoplasms. Reports describing the testing of 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC (Bautista et al., 1997), 68 *K-RAS* wildtype, 49 *K-RAS* mutated CRC (Laso et al., 2004), 155 *K-RAS* wildtype, 41 *K-RAS* mutated CRC (Naguib et al., 2010), 427 *K-RAS* wildtype, 242 *K-RAS* mutated CRC (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008), 971 *K-RAS* wildtype 457 *K-RAS* mutated CRC (Slattery et al., 2000) and 453 *K-RAS* wildtype, 81 *K-RAS* mutated adenomas (Wark et al., 2006) failed to identify folate intake as associated with *K-RAS* mutation status.

Increased consumption of folate offering some degree of protection against *K-RAS* mutation was observed in two independent studies. The failure to confirm this link in many other reports may potentially be explained several ways. Firstly, many of the studies described which identified no link between *K-RAS* mutation and folate intake contained relatively few mutated samples (*<*100). It is plausible that in these instances too few cases were analysed to detect any association, although this does not explain the studies which failed to identify a link using relatively large sample sets (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008; Slattery et al., 2000). Secondly, some dietary constituents have been described to affect folate utilisation, such as alcohol (Eichholzer et al., 2001; Freudenheim et al., 1991). It may be possible that the protective effect of folate against *K-RAS* mutation is only prevalent in the context of certain dietary patterns, possibly explaining why associations are not observed in all epidemiological studies. Finally, Martinez and colleagues identified an increased protective effect against *K-RAS* mutation as provided by supplement derived intake relative to natural dietary intake of this macronutrient (Martínez et al., 1999). The nature of folate consumption, i.e. bioavailablilty, may also determine the degree to which it offers a protective effect in colorectal carcinogenesis.

Although not observed in every analysis, increased intake of folate is associated with a reduced prevalence of total CRC incidence, which is observed in approximately half of the analyses testing this link (Eichholzer et al., 2001). It is probable, that at least to a limited degree and in certain circumstances, that this may be due to the ability of folate to protect against *K-RAS* mutation during development of colorectal neoplasia.

#### **2.1.3** *K-RAS* **mutation and fat consumption**

Consumption of several forms of fat intake have been described to affect the prevalence of *K-RAS* mutations in CRC. However, there is no consensus in the literature to date, regarding both the manner of the association and type of fat involved. Independent studies have identified monounsaturated fatty acid (MUFA) consumption as associated with the prevalence of *K-RAS* mutations in CRC. One report, analysing 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC, identified an increased MUFA consumption as linked to an increased prevalence of *K-RAS* mutated CRC (Bautista et al., 1997). MUFA, mostly derived from olive oil in this population, reduced the risk of CRC harbouring wildtype *K-RAS*, but offered no protection against *K-RAS* mutated cancers. However, contradictory findings of an increased MUFA consumption being associated with a higher prevalence of *K-RAS* wildtype neoplasia in a study assessing adenomas (453 wildtype, 81 mutated) (Wark et al., 2006) challenges the observation made by Bautista and co-workers. Other published reports have failed to identify any link between *K-RAS* mutation status and MUFA intake (Brink et al., 2004; Laso et al., 2004; Naguib et al., 2010; Slattery et al., 2000; Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010).

6 Colorectal Cancer

colorectal neoplasms. Reports describing the testing of 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC (Bautista et al., 1997), 68 *K-RAS* wildtype, 49 *K-RAS* mutated CRC (Laso et al., 2004), 155 *K-RAS* wildtype, 41 *K-RAS* mutated CRC (Naguib et al., 2010), 427 *K-RAS* wildtype, 242 *K-RAS* mutated CRC (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008), 971 *K-RAS* wildtype 457 *K-RAS* mutated CRC (Slattery et al., 2000) and 453 *K-RAS* wildtype, 81 *K-RAS* mutated adenomas (Wark et al., 2006) failed to identify folate intake as associated with *K-RAS*

Increased consumption of folate offering some degree of protection against *K-RAS* mutation was observed in two independent studies. The failure to confirm this link in many other reports may potentially be explained several ways. Firstly, many of the studies described which identified no link between *K-RAS* mutation and folate intake contained relatively few mutated samples (*<*100). It is plausible that in these instances too few cases were analysed to detect any association, although this does not explain the studies which failed to identify a link using relatively large sample sets (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008; Slattery et al., 2000). Secondly, some dietary constituents have been described to affect folate utilisation, such as alcohol (Eichholzer et al., 2001; Freudenheim et al., 1991). It may be possible that the protective effect of folate against *K-RAS* mutation is only prevalent in the context of certain dietary patterns, possibly explaining why associations are not observed in all epidemiological studies. Finally, Martinez and colleagues identified an increased protective effect against *K-RAS* mutation as provided by supplement derived intake relative to natural dietary intake of this macronutrient (Martínez et al., 1999). The nature of folate consumption, i.e. bioavailablilty, may also determine the degree to which it offers a protective

Although not observed in every analysis, increased intake of folate is associated with a reduced prevalence of total CRC incidence, which is observed in approximately half of the analyses testing this link (Eichholzer et al., 2001). It is probable, that at least to a limited degree and in certain circumstances, that this may be due to the ability of folate to protect

Consumption of several forms of fat intake have been described to affect the prevalence of *K-RAS* mutations in CRC. However, there is no consensus in the literature to date, regarding both the manner of the association and type of fat involved. Independent studies have identified monounsaturated fatty acid (MUFA) consumption as associated with the prevalence of *K-RAS* mutations in CRC. One report, analysing 67 *K-RAS* wildtype and 39 *K-RAS* mutated CRC, identified an increased MUFA consumption as linked to an increased prevalence of *K-RAS* mutated CRC (Bautista et al., 1997). MUFA, mostly derived from olive oil in this population, reduced the risk of CRC harbouring wildtype *K-RAS*, but offered no protection against *K-RAS* mutated cancers. However, contradictory findings of an increased MUFA consumption being associated with a higher prevalence of *K-RAS* wildtype neoplasia in a study assessing adenomas (453 wildtype, 81 mutated) (Wark et al., 2006) challenges the observation made by Bautista and co-workers. Other published reports have failed to identify any link between *K-RAS* mutation status and MUFA intake (Brink et al., 2004; Laso et al., 2004; Naguib et al., 2010; Slattery et al., 2000; Slattery, Curtin, Wolff, Herrick, Caan & Samowitz,

against *K-RAS* mutation during development of colorectal neoplasia.

mutation status.

effect in colorectal carcinogenesis.

2010).

**2.1.3** *K-RAS* **mutation and fat consumption**

In addition to these observations, one report describes an increase in polyunsaturated fatty acids (PUFA), specifically linoleic acid, as associated with increased prevalence of *K-RAS* mutated colonic, but not rectal, cancers (Brink et al., 2004). However, this association with PUFA has not been identifed in any other report (Bautista et al., 1997; Laso et al., 2004; Naguib et al., 2010; Slattery et al., 2000; Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010; Wark et al., 2006).

Taken together, the published data describing the association of dietary fats with *K-RAS* mutations have failed to identify a convincing association, and have generated conflicting results. Presently, the evidence suggesting that the *K-RAS* mutation status of colorectal neoplasia may be affected by fat intakes is weak: the limited data available suggest that the mutation status of this gene is largely independent of this dietary consumption. It should be noted however, that although fat intake itself is probably not associated with this mutation type, increased body mass index (BMI), which may be associated with fat intake, is associated with overall CRC risk.

#### **2.1.4** *K-RAS* **mutation and other dietary constituents**

The mutation status of *K-RAS* in CRC has also been linked to several other dietary variables in addition to meat, folate and fat. Testing of 971 *K-RAS* wildtype and 457 *K-RAS* mutated CRC identified an increased risk of *K-RAS* mutations with reduced consumption of cruciferous vegetables (Slattery et al., 2000). Another analysis of rectal cancers (535 wildtype, 215 mutated) identified a reduced incidence of *K-RAS* mutated rectal cancers with increased vegetable and fibre intake (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). Although corroborative, these two analyses were performed on the same test cohort and are yet to be identified in other independent populations. In this cohort at least, the data of Slattery and colleagues suggest that increased vegetable intake reduced the prevalence of *K-RAS* mutations in CRC, with an overt association identified in rectally located neoplasia.

Increased vitamin B2 intake has been identified to reduce the prevalence of adenomas harbouring *K-RAS* mutations. In an analysis of 453 *K-RAS* mutated and 81 *K-RAS* wildtype pre-cancerous adenomas an inverse association suggested a protective effect against *K-RAS* mutated adenomas. This protection was not found in relation to the prevalence of *K-RAS* wildtype adenomas (Wark et al., 2006). This association has not been identified in cohorts testing colorectal cancers.

Some dietary intakes have been repeatedly tested with no link to the prevalence of *K-RAS* mutation in CRC having been identified, notably alcohol. Many studies have included assessment of this dietary factor, with some studies analysing alcohol consumption independent of any other dietary factors (Bongaerts et al., 2006).

In summary, current literature describing the assessment of *K-RAS* mutation status in colorectal neoplasia has identified many associations with dietary intakes (summarised in Table 1). Very few of these associations have been repeatedly identified in independent cohorts, making assessment of their general validity challenging. Presently, few dietary components seem to be strongly linked to *K-RAS* mutation status in CRC across many populations, environments and genetic backgrounds. Furthermore, it is problematic to directly compare different studies. Other than folate, which has been described by the World Cancer Research Fund as having a 'limited' protective effect against CRC, which may impart this limited protection through reduced prevalence of *K-RAS* mutation, there is a lack of strong

in Colorectal Cancer and Diet 9

The Molecular Genetic Events in Colorectal Cancer and Diet 181

dietary factors may be associated with *BRAF* mutated CRC. Secondly, *BRAF* is identified at higher frequency in CRC demonstrating CIMP and MSI. Definitive evidence is yet to be provided describing the order in which tumours displaying CIMP and MSI acquire these instabilities and when *BRAF* mutations are acquired during progression. It is plausible that mutation in the *BRAF* gene is secondary to the acquisition of these global genomic alterations. As such, the question of diet and any relationships with this mutation may be redundant, if following the acquisition of CIMP and MSI status, *BRAF* mutation may arise independent of dietary influences. Thirdly, the limited number of studies available addressing the question of dietary associations and *BRAF* mutation may be too few in number to identify any dietary associations with this lesion, or, the majority of the studies performed are correct and that in this instance, dietary components do not affect the prevalence of *BRAF* mutations in CRC.

The *p53* tumour suppressor gene is the most commonly mutated gene in all human cancers, mutated in approximately 50% of human malignancies, including 50-60% of CRC (Forbes et al., 2008). Subsequent to its activation following DNA damage, oxidative stress or other cellular insults, wildtype p53 protein accumulates in the cell nucleus and acts as a transcription factor, capable of activating and suppressing transcription programmes leading to cell cycle arrest, DNA damage repair and apoptosis (Aylon & Oren, 2011; Bourdon et al., 2003). As such, perturbation of the normal role of p53 is highly selected for in cancer cells. The high prevalence of *p53* mutation in CRC, notably in later stage cancers, has led to various

Mutations in the *p53* gene have been linked to a variety of dietary intakes. Low folate and vitamin B6 intakes have been linked to p53 over-expressing cancers (Schernhammer, Ogino & Fuchs, 2008). This report, analysing 143 p53 over-expressing and 256 colonic cancers demonstrating low or absent p53 expression used an immunohistochemical (IHC) analysis to assess p53 accumulation following mutation. p53 over-expression or accumulation is the result of reduced protein degradation, mostly due to point mutations in the *p53* gene, greatly increasing the half-life of the gene's protein product (Melhem et al., 1995). This fast method of assessment of a range of activating *p53* mutations should be interpreted with some caution however, as less commonly observed mutations giving rise to truncated protein, such as those introducing premature *stop* codons, are not identified using this method. The observation linking low folate intake to an increased prevalence of cancers of the colon exhibiting over-expression of p53 is yet to be corroborated. Two reports using DNA sequencing, testing 62 *p53* mutated and 123 *p53* wildtype CRC (Park et al., 2010) and 686 *p53* mutated and 772 *p53* wildtype colonic cancers (Slattery et al., 2002), identified no link between p53 status and folate intakes. Little or no apparent other data exist describing vitamin B6 intakes and possible

Specific meat intakes have been linked to *p53* mutation status in several independent studies. One report by Park and colleagues identified an increased consumption of red and total meat (all types, including poultry) as associated with increased prevalence of *p53* mutations in CRC, however, this was only present in advanced stage CRC (Dukes' C or D), not in those of less advanced stages (Dukes' A or B) (Park et al., 2010). In addition to this, an assessment by

studies of mutations of this gene in the context of dietary consumptions.

**2.2** *p53* **mutations in colorectal cancer**

**2.2.1** *p53* **mutations and dietary associations**

relationships with *p53* mutation status.

evidence to firmly suggest any other dietary intakes affect the prevalence of *K-RAS* mutations in CRC.

#### **2.1.5** *BRAF* **mutations and dietary associations**

Relative to *K-RAS*, far fewer data exist describing the association between *BRAF* mutations in CRC and dietary influences. A prospective study involving 186 colorectal cancers, of which 29 harboured *BRAF* mutations, analysing meat, fruit and vegetable, fat, vitamin and macronutrient intakes identified no potential dietary associations with *BRAF* mutation in CRC (Naguib et al., 2010).

Other analyses have centred on analysing dietary constituents which may act as methyl group donors, such as folate, or vitamins, such as B6 and B2, which function as co-factors in the pathway responsible for DNA methylation (de Vogel et al., 2008; Kim, 2005). Based on observations that *BRAF* mutation has been linked previously to the CIMP phenotype (Lee et al., 2008; Samowitz et al., 2005; Velho et al., 2008) and has been linked to 60-80% of CRC demonstrating the highest levels of CIMP with concurrent MSI (Kambara et al., 2004; Samowitz et al., 2005), this mutation type may be influenced by dietary factors thought to influence DNA methylation. One such analysis used data and tissue samples from 648 individuals, of which 101 harboured CRC with mutations in the *BRAF* gene. This report identified a positive association between *BRAF* mutation in males and the highest tertile of folate consumption (de Vogel et al., 2008). This same report also identified an inverse correlation between methionine intake, as well as no association between vitamin B2 and alcohol consumption and *BRAF* mutations in the male portion of the cohort. In the female cohort members, no dietary consumptions were identified which were associated with *BRAF* mutations. An additional assessment of 86 *BRAF* mutated and 300 *BRAF* wildtype colonic cancers failed to identify an association between alcohol, folate, vitamins B6 and B12 or methionine consumption and *BRAF* mutation status (Schernhammer et al., 2011).

Another study population, of which 1108 cases of CRC were assessed for the presence of *BRAF* mutations, identified no associations between the 114 cancers harbouring this genetic lesion and intake of either vitamins B6, B12, folate, methionine or fibre consumptions, when compared with non-cancerous controls (Slattery et al., 2007). Similarly, the determination of *BRAF* mutation status in 189 CRC cases in another study cohort identified no associations between mutations in this gene and plasma levels of folate, vitamin B12 and homocysteine (Van Guelpen et al., 2010).

At present, few analyses of dietary intake in relation to the incidence of *BRAF* mutations in CRC have been attempted, and the majority of the limited data which do exist generally fail to identify strong associations between CRC harbouring *BRAF* mutations and any dietary constituent. In only one study to date, limited, sex specific dietary associations with *BRAF* mutation have been identified (de Vogel et al., 2008), but these observations are yet to be validated and corroborated in other studies.

The lack of identification of any of dietary component associated with *BRAF* mutation in CRC may have several causes. Primarily, only one study, analysing a very limited number of *BRAF* mutated tumours (n=29) has attempted a broad analysis of many dietary factors (Naguib et al., 2010). The remaining limited data has involved anlaysis of only a selected spectrum of dietary components hypothesised to be involved in the DNA methylation process. The limited scope of these studies in terms of dietary factors tested does not exclude the possibility that other 8 Colorectal Cancer

evidence to firmly suggest any other dietary intakes affect the prevalence of *K-RAS* mutations

Relative to *K-RAS*, far fewer data exist describing the association between *BRAF* mutations in CRC and dietary influences. A prospective study involving 186 colorectal cancers, of which 29 harboured *BRAF* mutations, analysing meat, fruit and vegetable, fat, vitamin and macronutrient intakes identified no potential dietary associations with *BRAF* mutation in CRC

Other analyses have centred on analysing dietary constituents which may act as methyl group donors, such as folate, or vitamins, such as B6 and B2, which function as co-factors in the pathway responsible for DNA methylation (de Vogel et al., 2008; Kim, 2005). Based on observations that *BRAF* mutation has been linked previously to the CIMP phenotype (Lee et al., 2008; Samowitz et al., 2005; Velho et al., 2008) and has been linked to 60-80% of CRC demonstrating the highest levels of CIMP with concurrent MSI (Kambara et al., 2004; Samowitz et al., 2005), this mutation type may be influenced by dietary factors thought to influence DNA methylation. One such analysis used data and tissue samples from 648 individuals, of which 101 harboured CRC with mutations in the *BRAF* gene. This report identified a positive association between *BRAF* mutation in males and the highest tertile of folate consumption (de Vogel et al., 2008). This same report also identified an inverse correlation between methionine intake, as well as no association between vitamin B2 and alcohol consumption and *BRAF* mutations in the male portion of the cohort. In the female cohort members, no dietary consumptions were identified which were associated with *BRAF* mutations. An additional assessment of 86 *BRAF* mutated and 300 *BRAF* wildtype colonic cancers failed to identify an association between alcohol, folate, vitamins B6 and B12 or

methionine consumption and *BRAF* mutation status (Schernhammer et al., 2011).

Another study population, of which 1108 cases of CRC were assessed for the presence of *BRAF* mutations, identified no associations between the 114 cancers harbouring this genetic lesion and intake of either vitamins B6, B12, folate, methionine or fibre consumptions, when compared with non-cancerous controls (Slattery et al., 2007). Similarly, the determination of *BRAF* mutation status in 189 CRC cases in another study cohort identified no associations between mutations in this gene and plasma levels of folate, vitamin B12 and homocysteine

At present, few analyses of dietary intake in relation to the incidence of *BRAF* mutations in CRC have been attempted, and the majority of the limited data which do exist generally fail to identify strong associations between CRC harbouring *BRAF* mutations and any dietary constituent. In only one study to date, limited, sex specific dietary associations with *BRAF* mutation have been identified (de Vogel et al., 2008), but these observations are yet to be

The lack of identification of any of dietary component associated with *BRAF* mutation in CRC may have several causes. Primarily, only one study, analysing a very limited number of *BRAF* mutated tumours (n=29) has attempted a broad analysis of many dietary factors (Naguib et al., 2010). The remaining limited data has involved anlaysis of only a selected spectrum of dietary components hypothesised to be involved in the DNA methylation process. The limited scope of these studies in terms of dietary factors tested does not exclude the possibility that other

in CRC.

(Naguib et al., 2010).

(Van Guelpen et al., 2010).

validated and corroborated in other studies.

**2.1.5** *BRAF* **mutations and dietary associations**

dietary factors may be associated with *BRAF* mutated CRC. Secondly, *BRAF* is identified at higher frequency in CRC demonstrating CIMP and MSI. Definitive evidence is yet to be provided describing the order in which tumours displaying CIMP and MSI acquire these instabilities and when *BRAF* mutations are acquired during progression. It is plausible that mutation in the *BRAF* gene is secondary to the acquisition of these global genomic alterations. As such, the question of diet and any relationships with this mutation may be redundant, if following the acquisition of CIMP and MSI status, *BRAF* mutation may arise independent of dietary influences. Thirdly, the limited number of studies available addressing the question of dietary associations and *BRAF* mutation may be too few in number to identify any dietary associations with this lesion, or, the majority of the studies performed are correct and that in this instance, dietary components do not affect the prevalence of *BRAF* mutations in CRC.

#### **2.2** *p53* **mutations in colorectal cancer**

The *p53* tumour suppressor gene is the most commonly mutated gene in all human cancers, mutated in approximately 50% of human malignancies, including 50-60% of CRC (Forbes et al., 2008). Subsequent to its activation following DNA damage, oxidative stress or other cellular insults, wildtype p53 protein accumulates in the cell nucleus and acts as a transcription factor, capable of activating and suppressing transcription programmes leading to cell cycle arrest, DNA damage repair and apoptosis (Aylon & Oren, 2011; Bourdon et al., 2003). As such, perturbation of the normal role of p53 is highly selected for in cancer cells. The high prevalence of *p53* mutation in CRC, notably in later stage cancers, has led to various studies of mutations of this gene in the context of dietary consumptions.

#### **2.2.1** *p53* **mutations and dietary associations**

Mutations in the *p53* gene have been linked to a variety of dietary intakes. Low folate and vitamin B6 intakes have been linked to p53 over-expressing cancers (Schernhammer, Ogino & Fuchs, 2008). This report, analysing 143 p53 over-expressing and 256 colonic cancers demonstrating low or absent p53 expression used an immunohistochemical (IHC) analysis to assess p53 accumulation following mutation. p53 over-expression or accumulation is the result of reduced protein degradation, mostly due to point mutations in the *p53* gene, greatly increasing the half-life of the gene's protein product (Melhem et al., 1995). This fast method of assessment of a range of activating *p53* mutations should be interpreted with some caution however, as less commonly observed mutations giving rise to truncated protein, such as those introducing premature *stop* codons, are not identified using this method. The observation linking low folate intake to an increased prevalence of cancers of the colon exhibiting over-expression of p53 is yet to be corroborated. Two reports using DNA sequencing, testing 62 *p53* mutated and 123 *p53* wildtype CRC (Park et al., 2010) and 686 *p53* mutated and 772 *p53* wildtype colonic cancers (Slattery et al., 2002), identified no link between p53 status and folate intakes. Little or no apparent other data exist describing vitamin B6 intakes and possible relationships with *p53* mutation status.

Specific meat intakes have been linked to *p53* mutation status in several independent studies. One report by Park and colleagues identified an increased consumption of red and total meat (all types, including poultry) as associated with increased prevalence of *p53* mutations in CRC, however, this was only present in advanced stage CRC (Dukes' C or D), not in those of less advanced stages (Dukes' A or B) (Park et al., 2010). In addition to this, an assessment by

in Colorectal Cancer and Diet 11

The Molecular Genetic Events in Colorectal Cancer and Diet 183

Further evidence is needed to substantiate these isolated observations. Future studies should focus on the analysis of the potential association of vegetable and meat intakes in relation to p53 status as several data exist suggesting a possible link between these intakes and p53

The *adenomatous polyposis coli* (*APC*) gene is one of the most frequently mutated genes in colorectal cancer (Sjöblom et al., 2006; Wood et al., 2007), with some studies reporting 50-80% of CRC harbouring mutations in this gene (Forbes et al., 2008). The majority of mutations identified in CRC in the *APC* gene are located in exon 15 in the central third of the coding sequence, the *mutation cluster region*, which corresponds to the *β*-catenin-binding region of the protein (Goss & Groden, 2000). Mutations in *APC* most frequently result in truncation of the protein, corresponding with a reduction in the ability of APC to bind *β*-catenin (Figure 2). In addition to its role as a modulator of WNT pathway signalling, APC also has a role in mitosis and cytokinesis: cells harbouring truncated APC undergo abnormal chromosomal segregation and may develop aneuploidy (Ceol et al., 2007). Wildtype APC functions as a regulator of apoptosis, differentiation and migration and functions during cell division (Ceol

Although mutations in other genes, such as *p53*, may be almost as frequent as those in *APC* in CRC, *APC* mutations seem to be particularly prevalent from the earliest stages of CRC initiation and progression. Dysplastic aberrant crypt foci (ACF), monocryptal or oligocryptal adenomas, which are the lesions considered to be the earliest forms of colorectal neoplasia, frequently display *APC* mutations (Jen et al., 1994) and can develop into CRC through the adenoma-carcinoma sequence (Suehiro & Hinoda, 2008; Takayama et al., 1998). Intriguingly, the more frequently occurring heteroplastic ACF, which possess limited, if any, potential to develop to malignancy, very rarely harbour *APC* mutations but frequently exhibit *K-RAS* mutations (Jen et al., 1994). These data suggest that initiating genetic lesions in CRC determine malignant potential, and that if the initial mutations occur in the *APC* gene, there is a high probability of subsequent adenoma formation. In concordance with observations in dysplastic ACF, *APC* mutations are very frequently observed in colorectal adenomas (Kinzler & Vogelstein, 1996) and when inherited as germline *APC* mutations allow formation of hundreds of colorectal adenomas in the Familial Adenomatous Polyposis Coli syndrome. Hence, there have been several analyses of APC mutations in CRC relation to dietary intakes, with the purpose of identifying links between this early genetic lesion and

*APC* mutations have been linked to several dietary constituents. One report, analysing 121 *APC* wildtype and 63 *APC* mutated colonic cancers, identified alcohol as inversely associated with *APC* mutated and positively associated with *APC* wildtype cancers (Diergaarde, van Geloof, van Muijen, Kok & Kampman, 2003). Additionally, red meat, fish and fat, notably unsaturated fat, were shown to be associated with development of *APC* mutated colonic cancers. Conversely, another report assessing 347 *APC* wildtype CRC and 184 *APC* mutated CRC identified increased consumption of saturated fat, but not unsaturated fats, as associated with *APC* mutated rectal cancers (Weijenberg et al., 2007). Furthermore, the analysis by

aberrations, although contrary observations have been published.

**2.3** *APC* **mutations in colorectal cancer**

et al., 2007; Fodde et al., 2001; Goss & Groden, 2000).

**2.3.1** *APC* **mutations and dietary associations**

dietary carcinogens.

Slattery and co-workers identified high glycaemic load, increased red meat, increased fast food and increased trans fatty acid intakes as associated with increased prevalence of *p53* mutations in colonic cancers (Slattery et al., 2002). These two independent studies suggest that red meat in particular may promote mutations in *p53* in neoplasia of the large intestine. However, these data do not completely overlap: the study by Park and colleagues only found this association in advanced stage cancers and the report by Slattery and co-workers assessed only colonic, not rectal cancers. Opposed to the above observations of meat intakes promoting *p53* mutations in CRC, an IHC based analysis (73 p53 over-expressing, 90 p53 absent CRC) identified increased beef consumption as associated with reduced prevalence of p53 over-expressing cancers (Freedman et al., 1996). Further data are needed to evaluate the potential association of meat, and meat types, with *p53* mutations in CRC, with particular emphasis on cancer location and stage.

In a study of colonic cancers assessing both p53 expression and *p53* gene mutations, total and saturated fats were identified as linked to tumours not over-expressing p53 or harbouring gene mutations (Voskuil et al., 1999). Of the 185 colonic cancers tested in this study, 81 displayed p53 overexpression by IHC, of which 59 were found to harbour mutations in the sequenced region (exons 5-8) of these cancers. Mutations in *p53* were not found to be linked to total fat intake in other reports assessing either CRC as a general subgroup (Park et al., 2010) or rectal cancers in particular (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). An analysis of 340 *p53* mutated and 410 *p53* wildtype rectal cancers reported an increased consumption of vegetables, whole grains and fibre associated with reduced prevalence of *p53* mutation (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). Conversely, a high intake of refined grains was found to increase the prevalence of rectal cancer harbouring *p53* mutations. Increased intakes of cruciferous vegetables have also been described to be associated with reduced prevalence of p53 over-expressing CRC (73 p53 over-expressing CRC, 90 p53 absent CRC) (Freedman et al., 1996). The observation of increased vegetable intakes associated with reduced frequency of *p53* mutations in CRC was not observed in another study analysing general CRC (Park et al., 2010). Fibre was not observed to be associated with a protective effect in analyses combining colonic and rectal cancers (Park et al., 2010) or assessing colonic cancers in isolation (Slattery et al., 2002; Voskuil et al., 1999).

Alcohol intakes and *p53* mutation status in CRC have been assessed in several reports. A study analysing 340 *p53* mutated and 410 *p53* wildtype rectal cancers identified increased beer consumption as being associated with higher prevalence of *p53* mutations when compared with non-beer drinkers (Slattery, Wolff, Herrick, Curtin, Caan & Samowitz, 2010). No associations between alcohol intakes and *p53* mutation status have been identified in several analyses of colonic cancers (Schernhammer, Ogino & Fuchs, 2008; Voskuil et al., 1999), however, neither of these studies assessed specific alcoholic beverages, just total alcohol intake. Total alcohol intake was found to be linked to increased prevalence of *p53* mutations in CRC of advanced Dukes' stage (C and D), but not in CRC of less advanced stage (Dukes' A or B) (Park et al., 2010). Another report analysing Dukes' stage C cancers by IHC (42 p53 over-expressing CRC, 65 p53 absent CRC) did not identify total alcohol intake as linked to p53 expression status (Zhang et al., 1995).

Presently, the limited data on *p53* mutation status in CRC and dietary intakes are inconsistent. As a result, several consumptions have been linked to *p53* mutation status but none have been corroborated by other studies performing a similar assessment in an independent cohort. Further evidence is needed to substantiate these isolated observations. Future studies should focus on the analysis of the potential association of vegetable and meat intakes in relation to p53 status as several data exist suggesting a possible link between these intakes and p53 aberrations, although contrary observations have been published.

#### **2.3** *APC* **mutations in colorectal cancer**

10 Colorectal Cancer

Slattery and co-workers identified high glycaemic load, increased red meat, increased fast food and increased trans fatty acid intakes as associated with increased prevalence of *p53* mutations in colonic cancers (Slattery et al., 2002). These two independent studies suggest that red meat in particular may promote mutations in *p53* in neoplasia of the large intestine. However, these data do not completely overlap: the study by Park and colleagues only found this association in advanced stage cancers and the report by Slattery and co-workers assessed only colonic, not rectal cancers. Opposed to the above observations of meat intakes promoting *p53* mutations in CRC, an IHC based analysis (73 p53 over-expressing, 90 p53 absent CRC) identified increased beef consumption as associated with reduced prevalence of p53 over-expressing cancers (Freedman et al., 1996). Further data are needed to evaluate the potential association of meat, and meat types, with *p53* mutations in CRC, with particular

In a study of colonic cancers assessing both p53 expression and *p53* gene mutations, total and saturated fats were identified as linked to tumours not over-expressing p53 or harbouring gene mutations (Voskuil et al., 1999). Of the 185 colonic cancers tested in this study, 81 displayed p53 overexpression by IHC, of which 59 were found to harbour mutations in the sequenced region (exons 5-8) of these cancers. Mutations in *p53* were not found to be linked to total fat intake in other reports assessing either CRC as a general subgroup (Park et al., 2010) or rectal cancers in particular (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). An analysis of 340 *p53* mutated and 410 *p53* wildtype rectal cancers reported an increased consumption of vegetables, whole grains and fibre associated with reduced prevalence of *p53* mutation (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). Conversely, a high intake of refined grains was found to increase the prevalence of rectal cancer harbouring *p53* mutations. Increased intakes of cruciferous vegetables have also been described to be associated with reduced prevalence of p53 over-expressing CRC (73 p53 over-expressing CRC, 90 p53 absent CRC) (Freedman et al., 1996). The observation of increased vegetable intakes associated with reduced frequency of *p53* mutations in CRC was not observed in another study analysing general CRC (Park et al., 2010). Fibre was not observed to be associated with a protective effect in analyses combining colonic and rectal cancers (Park et al., 2010) or

assessing colonic cancers in isolation (Slattery et al., 2002; Voskuil et al., 1999).

Alcohol intakes and *p53* mutation status in CRC have been assessed in several reports. A study analysing 340 *p53* mutated and 410 *p53* wildtype rectal cancers identified increased beer consumption as being associated with higher prevalence of *p53* mutations when compared with non-beer drinkers (Slattery, Wolff, Herrick, Curtin, Caan & Samowitz, 2010). No associations between alcohol intakes and *p53* mutation status have been identified in several analyses of colonic cancers (Schernhammer, Ogino & Fuchs, 2008; Voskuil et al., 1999), however, neither of these studies assessed specific alcoholic beverages, just total alcohol intake. Total alcohol intake was found to be linked to increased prevalence of *p53* mutations in CRC of advanced Dukes' stage (C and D), but not in CRC of less advanced stage (Dukes' A or B) (Park et al., 2010). Another report analysing Dukes' stage C cancers by IHC (42 p53 over-expressing CRC, 65 p53 absent CRC) did not identify total alcohol intake as linked to p53

Presently, the limited data on *p53* mutation status in CRC and dietary intakes are inconsistent. As a result, several consumptions have been linked to *p53* mutation status but none have been corroborated by other studies performing a similar assessment in an independent cohort.

emphasis on cancer location and stage.

expression status (Zhang et al., 1995).

The *adenomatous polyposis coli* (*APC*) gene is one of the most frequently mutated genes in colorectal cancer (Sjöblom et al., 2006; Wood et al., 2007), with some studies reporting 50-80% of CRC harbouring mutations in this gene (Forbes et al., 2008). The majority of mutations identified in CRC in the *APC* gene are located in exon 15 in the central third of the coding sequence, the *mutation cluster region*, which corresponds to the *β*-catenin-binding region of the protein (Goss & Groden, 2000). Mutations in *APC* most frequently result in truncation of the protein, corresponding with a reduction in the ability of APC to bind *β*-catenin (Figure 2). In addition to its role as a modulator of WNT pathway signalling, APC also has a role in mitosis and cytokinesis: cells harbouring truncated APC undergo abnormal chromosomal segregation and may develop aneuploidy (Ceol et al., 2007). Wildtype APC functions as a regulator of apoptosis, differentiation and migration and functions during cell division (Ceol et al., 2007; Fodde et al., 2001; Goss & Groden, 2000).

Although mutations in other genes, such as *p53*, may be almost as frequent as those in *APC* in CRC, *APC* mutations seem to be particularly prevalent from the earliest stages of CRC initiation and progression. Dysplastic aberrant crypt foci (ACF), monocryptal or oligocryptal adenomas, which are the lesions considered to be the earliest forms of colorectal neoplasia, frequently display *APC* mutations (Jen et al., 1994) and can develop into CRC through the adenoma-carcinoma sequence (Suehiro & Hinoda, 2008; Takayama et al., 1998). Intriguingly, the more frequently occurring heteroplastic ACF, which possess limited, if any, potential to develop to malignancy, very rarely harbour *APC* mutations but frequently exhibit *K-RAS* mutations (Jen et al., 1994). These data suggest that initiating genetic lesions in CRC determine malignant potential, and that if the initial mutations occur in the *APC* gene, there is a high probability of subsequent adenoma formation. In concordance with observations in dysplastic ACF, *APC* mutations are very frequently observed in colorectal adenomas (Kinzler & Vogelstein, 1996) and when inherited as germline *APC* mutations allow formation of hundreds of colorectal adenomas in the Familial Adenomatous Polyposis Coli syndrome. Hence, there have been several analyses of APC mutations in CRC relation to dietary intakes, with the purpose of identifying links between this early genetic lesion and dietary carcinogens.

#### **2.3.1** *APC* **mutations and dietary associations**

*APC* mutations have been linked to several dietary constituents. One report, analysing 121 *APC* wildtype and 63 *APC* mutated colonic cancers, identified alcohol as inversely associated with *APC* mutated and positively associated with *APC* wildtype cancers (Diergaarde, van Geloof, van Muijen, Kok & Kampman, 2003). Additionally, red meat, fish and fat, notably unsaturated fat, were shown to be associated with development of *APC* mutated colonic cancers. Conversely, another report assessing 347 *APC* wildtype CRC and 184 *APC* mutated CRC identified increased consumption of saturated fat, but not unsaturated fats, as associated with *APC* mutated rectal cancers (Weijenberg et al., 2007). Furthermore, the analysis by

in Colorectal Cancer and Diet 13

The Molecular Genetic Events in Colorectal Cancer and Diet 185

context, these observations partially confirm the increased consumption of general red meat that was observed to be associated with an increased risk of *APC* mutated CRC in the report

In addition to reports assessing *APC* mutation status relative to dietary intakes in CRC, a single study has assessed these relationships in colorectal adenomas (Diergaarde et al., 2005). This analysis of 117 *APC* wildtype adenomas and 161 *APC* mutated colorectal adenomas identified a high intake of red meat and fat as associated with increased prevalence of *APC* wildtype adenomas. These observations are intriguing as identification of increased consumptions of certain red meat types being specifically associated with certain *APC*

Taken together, the available data describing *APC* mutations in CRC in relation to dietary intakes are too few and inconsistent to draw any strong conclusions. However, several analyses have identified certain meat consumptions as linked to either colonic or rectal cancers with a particular *APC* mutation status. These observations, although not in full agreement, indicate that certain red meat types, determined by both animal origin and preparation method, may affect the prevalence of mutation in *APC* in CRC. Further assessment of these particular dietary associations are warranted to determine the relationship between *APC* mutation status and specific red meat consumptions. Based on these somewhat conflicting

**2.4 Microsatellite instability (MSI) and CpG island methylator phenotype (CIMP) in**

Acquired variation in length of repetitive DNA sequences (microsatellites) can be detected as microsatellite instability (MSI) and is prevalent in approximately 15% of sporadic CRC and in almost all CRC in Lynch/Hereditary Non-Polyposis Colorectal Cancer syndrome (Soreide et al., 2006). MSI arises as a result of DNA replication errors that produce a change in length of repetitive sequences, which if not repaired (by the DNA mismatch repair (MMR) system), accumulate with increasing frequency (Martin et al., 2010; Soreide et al., 2006). In sporadic CRC, the most frequent inactivating cause of MMR is the methylation of the *MLH1* promoter

The MMR process is responsible for the correction of DNA replication errors which result in small insertions or deletions in the genome; these are especially prevalent at microsatellites due to increased frequency of DNA polymerase slippage at repetitive sequences. In humans, two major components comprise the MMR pathway: MutS (which is present in two heterodimers of MSH2/MSH6 and MSH2/MSH3) and MutL (which is also present in several heterodimer forms:- MLH1/PMS2, MLH1/PMS1 and MLH1/MLH3) (Martin et al., 2010). Disruption of the formation of the MutS and MutL dimers (by abrogation of the component proteins due to acquired promoter methylation or mutation) leads to a limited or defective MMR pathway, giving rise to genomic instability whereby DNA regions, most frequently repetitive sequences, increase or decrease in length (MSI). Such instability can lead to gene

mutations, frequently of frameshift type, which can contribute to cancer progression.

CIMP is observed in 30-40% of proximal colonic and 3-12% of distal/rectal cancers (Curtin et al., 2011; Ibrahim et al., 2011). The exact causes of excessive methylation in DNA regions harbouring high levels of adjacent cytosine and guanine bases (CpG islands) are

**2.4.1 MSI and CIMP as genomic instabilities in colorectal cancer**

on one or both alleles (Herman et al., 1998; Wheeler et al., 2000).

wildtype CRC has been described previously (Lüchtenborg et al., 2005).

by Diergaarde and co-workers.

data, some associations do seem plausible.

**colorectal cancer**

Fig. 2. **APC and the WNT signalling pathway**. **A:** In the absence of WNT signal, free *β*-catenin is bound by APC, in a complex with axin/conductin and glycogen synthase kinase 3*β* (GSK3*β*) and this complex acts as a scaffold, bringing *β*-catenin into close proximity with GSK3*β*. This results in GSK3*β* mediated phosphorylation of *β* catenin. **B:** Phosphorylated *β*-catenin is recognised by the SCF complex and is polyubiquitinated. **C:** Polyubiquitinated *β*-catenin is recognised by the proteasome and degraded. In the absence of WNT signalling, *β*-catenin is largely degraded, thus preventing *β*-catenin nuclear accumulation and subsequent co-activation of transcription programs. Upon binding of WNT ligand to membrane-located receptors, a subsequent signalling cascade prevents formation of the APC-axin-conductin-GSK3*β* complex. As a result, *β*-catenin avoids degradation and can translocate to the nucleus where it co-activates transcription of target genes, such as *c-myc*.

Weijenberg and co-workers identified specific types of *APC* wildtype CRC (i.e. those harbouring *K-RAS* mutations and showing no loss of MLH1 expression [see 2.4.2]) as being linked to increased intake of linoleic acid, a polyunsaturated fatty acid.

A further study has identified increased consumption of folate associated with reduced prevalence of *APC* wildtype colonic cancer, but increased prevalence of *APC* mutated colonic cancers in males (de Vogel et al., 2006). These associations were not observed in rectal cancers of men or in either colonic or rectal female cancer cases. This analysis, studying 347 *APC* wildtype CRC and 182 *APC* mutated CRC, also identified increased vitamin B2 and iron intakes in men associated with colonic cancers harbouring *APC* mutations compared with those men with colonic cancer not harbouring *APC* mutations.

These analyses are difficult to compare, notably as Diergaarde and colleagues did not stratify cases by sex or cancer location, which may possibly explain the lack of association between folate intake and *APC* mutation status in their report. The study by Diergaarde and co-workers did not analyse iron or vitamin B2, and de Vogel and colleagues did not assess meat and fish intakes. Alcohol association with *APC* mutation status was not observed in the testing by de Vogel and co-workers. Assessed in conjunction, these studies do not corroborate each other as direct comparisons are difficult to make.

Further analysis of *APC* mutation status has been performed in the context of specific meat intakes. In a study of 347 *APC* wildtype CRC and 184 *APC* mutated CRC, increased processed meat consumption was linked to an increased prevalence of *APC* mutated colonic cancers (Lüchtenborg et al., 2005). Additionally, increased beef consumption was linked to increased frequency of *APC* wildtype colonic cancers. Rectal cancers without *APC* mutations were found to be more prevalent amongst those with increased consumption of other meat types, which included horsemeats, lamb and mutton among other products. This detailed analysis of *APC* mutation status in the context of very specific meat types, with both positive and negative associations having been identified, is yet to be corroborated by similarly detailed meat-type subgroups testing in additional studies. This report does suggest however, that meat classification is important when testing for associations with *APC* mutations. In this 12 Colorectal Cancer

Fig. 2. **APC and the WNT signalling pathway**. **A:** In the absence of WNT signal, free

*β*-catenin is largely degraded, thus preventing *β*-catenin nuclear accumulation and subsequent co-activation of transcription programs. Upon binding of WNT ligand to membrane-located receptors, a subsequent signalling cascade prevents formation of the APC-axin-conductin-GSK3*β* complex. As a result, *β*-catenin avoids degradation and can translocate to the nucleus where it co-activates transcription of target genes, such as *c-myc*.

linked to increased intake of linoleic acid, a polyunsaturated fatty acid.

those men with colonic cancer not harbouring *APC* mutations.

each other as direct comparisons are difficult to make.

*β*-catenin is bound by APC, in a complex with axin/conductin and glycogen synthase kinase 3*β* (GSK3*β*) and this complex acts as a scaffold, bringing *β*-catenin into close proximity with GSK3*β*. This results in GSK3*β* mediated phosphorylation of *β* catenin. **B:** Phosphorylated *β*-catenin is recognised by the SCF complex and is polyubiquitinated. **C:** Polyubiquitinated *β*-catenin is recognised by the proteasome and degraded. In the absence of WNT signalling,

Weijenberg and co-workers identified specific types of *APC* wildtype CRC (i.e. those harbouring *K-RAS* mutations and showing no loss of MLH1 expression [see 2.4.2]) as being

A further study has identified increased consumption of folate associated with reduced prevalence of *APC* wildtype colonic cancer, but increased prevalence of *APC* mutated colonic cancers in males (de Vogel et al., 2006). These associations were not observed in rectal cancers of men or in either colonic or rectal female cancer cases. This analysis, studying 347 *APC* wildtype CRC and 182 *APC* mutated CRC, also identified increased vitamin B2 and iron intakes in men associated with colonic cancers harbouring *APC* mutations compared with

These analyses are difficult to compare, notably as Diergaarde and colleagues did not stratify cases by sex or cancer location, which may possibly explain the lack of association between folate intake and *APC* mutation status in their report. The study by Diergaarde and co-workers did not analyse iron or vitamin B2, and de Vogel and colleagues did not assess meat and fish intakes. Alcohol association with *APC* mutation status was not observed in the testing by de Vogel and co-workers. Assessed in conjunction, these studies do not corroborate

Further analysis of *APC* mutation status has been performed in the context of specific meat intakes. In a study of 347 *APC* wildtype CRC and 184 *APC* mutated CRC, increased processed meat consumption was linked to an increased prevalence of *APC* mutated colonic cancers (Lüchtenborg et al., 2005). Additionally, increased beef consumption was linked to increased frequency of *APC* wildtype colonic cancers. Rectal cancers without *APC* mutations were found to be more prevalent amongst those with increased consumption of other meat types, which included horsemeats, lamb and mutton among other products. This detailed analysis of *APC* mutation status in the context of very specific meat types, with both positive and negative associations having been identified, is yet to be corroborated by similarly detailed meat-type subgroups testing in additional studies. This report does suggest however, that meat classification is important when testing for associations with *APC* mutations. In this context, these observations partially confirm the increased consumption of general red meat that was observed to be associated with an increased risk of *APC* mutated CRC in the report by Diergaarde and co-workers.

In addition to reports assessing *APC* mutation status relative to dietary intakes in CRC, a single study has assessed these relationships in colorectal adenomas (Diergaarde et al., 2005). This analysis of 117 *APC* wildtype adenomas and 161 *APC* mutated colorectal adenomas identified a high intake of red meat and fat as associated with increased prevalence of *APC* wildtype adenomas. These observations are intriguing as identification of increased consumptions of certain red meat types being specifically associated with certain *APC* wildtype CRC has been described previously (Lüchtenborg et al., 2005).

Taken together, the available data describing *APC* mutations in CRC in relation to dietary intakes are too few and inconsistent to draw any strong conclusions. However, several analyses have identified certain meat consumptions as linked to either colonic or rectal cancers with a particular *APC* mutation status. These observations, although not in full agreement, indicate that certain red meat types, determined by both animal origin and preparation method, may affect the prevalence of mutation in *APC* in CRC. Further assessment of these particular dietary associations are warranted to determine the relationship between *APC* mutation status and specific red meat consumptions. Based on these somewhat conflicting data, some associations do seem plausible.

#### **2.4 Microsatellite instability (MSI) and CpG island methylator phenotype (CIMP) in colorectal cancer**

#### **2.4.1 MSI and CIMP as genomic instabilities in colorectal cancer**

Acquired variation in length of repetitive DNA sequences (microsatellites) can be detected as microsatellite instability (MSI) and is prevalent in approximately 15% of sporadic CRC and in almost all CRC in Lynch/Hereditary Non-Polyposis Colorectal Cancer syndrome (Soreide et al., 2006). MSI arises as a result of DNA replication errors that produce a change in length of repetitive sequences, which if not repaired (by the DNA mismatch repair (MMR) system), accumulate with increasing frequency (Martin et al., 2010; Soreide et al., 2006). In sporadic CRC, the most frequent inactivating cause of MMR is the methylation of the *MLH1* promoter on one or both alleles (Herman et al., 1998; Wheeler et al., 2000).

The MMR process is responsible for the correction of DNA replication errors which result in small insertions or deletions in the genome; these are especially prevalent at microsatellites due to increased frequency of DNA polymerase slippage at repetitive sequences. In humans, two major components comprise the MMR pathway: MutS (which is present in two heterodimers of MSH2/MSH6 and MSH2/MSH3) and MutL (which is also present in several heterodimer forms:- MLH1/PMS2, MLH1/PMS1 and MLH1/MLH3) (Martin et al., 2010). Disruption of the formation of the MutS and MutL dimers (by abrogation of the component proteins due to acquired promoter methylation or mutation) leads to a limited or defective MMR pathway, giving rise to genomic instability whereby DNA regions, most frequently repetitive sequences, increase or decrease in length (MSI). Such instability can lead to gene mutations, frequently of frameshift type, which can contribute to cancer progression.

CIMP is observed in 30-40% of proximal colonic and 3-12% of distal/rectal cancers (Curtin et al., 2011; Ibrahim et al., 2011). The exact causes of excessive methylation in DNA regions harbouring high levels of adjacent cytosine and guanine bases (CpG islands) are

in Colorectal Cancer and Diet 15

The Molecular Genetic Events in Colorectal Cancer and Diet 187

Satia *et al* 2005 CC: 437 CC: 49 no association between diet and MSI status [some associations comparing MSI/MSS cases vs controls] Schernhammer *et al* 2008 CC: 542 CC: 127 no statistically significant association between folate, vitamin B6, B12, methionine or alcohol and MSI status

Table 2. Summarised description of literature analysing microsatellite instability (MSI) in colorectal neoplasia in relation to dietary intakes with the statistically significant associations described. *MSS*: microsatellite stability, *WT*: wildtype, *CRC*: colorectal cancer, *CC*: colonic

may also be plausible that such relationships exist and are particularly subtle. Methylation of the *MLH1* promoter, leading to gene silencing and subsequent DNA MMR deficiency, occurs in the vast majority, but not all, of MSI CRC (Kuismanen et al., 2000); suggesting that other components of the MMR system can be disrupted, such as mutations to the *MSH2* or *MSH6* genes, and that MSI may develop from a group of distinct initial aberrations in a small proportion of CRC. Furthermore, subsequent instability at microsatellites as a result may depend on other promoting factors. As such, it appears that a series of molecular events takes place leading to the MSI phenotype, which may arise from different epigenetic silencing or mutational events in different cancers. The multiple causes of MSI, and the different associated factors, may explain, at least in part, the lack of consistently identified dietary constituents which have been associated with this type of genomic instability. Alternatively, age-related susceptibility to promoter methylation, including the MLH1 promoter, may be the

Studies assessing dietary associations with CIMP in CRC have centred largely on testing intakes of those compounds which may act as methyl group donors, or which function in the biochemical pathways responsible for methylation processes. Vitamin B6 has been described as associated with an increased prevalence of CIMP in CRC in one study assessing 496 CIMP-low/absent and 152 CIMP-high cancers (de Vogel et al., 2008). However, several other reports, assessing 288 CIMP-low/absent and 87 CIMP-high (Schernhammer et al., 2011) and 824 CIMP-low/absent and 330 CIMP-high (Slattery et al., 2007) colonic cancers failed to

A similar lack of consensus has been observed when assessing vitamin B12. A single study assessing 107 CIMP-low/absent and 44 CIMP-high colonic cancers described an increased serum vitamin B12 concentration as associated with CIMP in this cohort (Mokarram et al., 2008). Schernhammer and colleagues (Schernhammer et al., 2011) and Slattery and co-workers (Slattery et al., 2007) did not identify a similar association in their studies. A report assessing 163 CIMP-low/absent and 27 CIMP-high CRC also identified no association between vitamin B12 intakes and CIMP status (Van Guelpen et al., 2010). Assessment of folate intake in relation to CIMP status has consistently failed to identify associations between the two in both colorectal and colonic cancer studies (Schernhammer et al., 2011; Slattery et al., 2007;

**Study MSS/MSI-low MSI/MSI-high dietary**

Diergaarde *et al* 2003 CC: 144 CC: 40 ↑ red meat associated with MSS cancers Jensen *et al* 2008 CRC: 111 CRC: 19 no association between MSI and folate or vitamin B12 Poynter *et al* 2009 CRC: 1337 CRC: 227 ↑ alcohol associated with MSS cancers

Slattery *et al* 2001 CC: 1244 CC: 266 ↑ alcohol associated with MSI cancers Van Guelpen *et al* 2010 CRC: 166 CRC: 24 Increased levels of plasma folate associated with MSI cancers Wu *et al* 2001 CC: 238 CC: 35 ↑ heterocyclic aromatic amines associated with MSI cancers

cancer, ↑ and ↓ denote an increase or decrease in consumption respectively.

predominant risk factor for MSI in CRC rather than dietary factors.

**2.4.3 CIMP and dietary associations**

identify a similar association.

Van Guelpen et al., 2010).

**CRC/CC CRC/CC association** Chang *et al* 2007 CRC: 179 CRC: 16 no statistically significant association between folate or vitamin B12 and MSI status de Vogel *et al* 2008 CRC: 572 CRC: 76 no statistically significant association between folate, vitamin B2, methionine or alcohol and MSI status

unknown, although some evidence exists which suggest that such an increase in methyl group incorporation at these sites occurs during ageing in normal epithelial cells in the gut, and this is elevated in cancer (Toyota et al., 1999). Hypermethylation of gene promotors, in addition to or independent of methylation of other local DNA sequences, leads to transcriptional silencing of those genes. Such transcriptional silencing can be considered as one mechanism by which genes can be 'knocked out', in addition to mutation and deletion, in Knudson's model of tumour suppressor gene inactivation (Kondo & Issa, 2004). In this way, the aberrant methylation of genes can contribute to their inactivation in cancer epigenetically, such that in the absence of inactivating genetic changes tumour suppressor gene activity can be lost, leading to cancer progression.

#### **2.4.2 MSI and dietary associations**

MSI in CRC has been assessed in relation to dietary intakes in several reports, many of which did not identify a link between this type of genomic instability in CRC and specific dietary intakes (Chang et al., 2007; de Vogel et al., 2008; Jensen et al., 2008; Schernhammer, Giovannuccci, Fuchs & Ogino, 2008) (Table 2). However, a limited number of studies have described links between dietary intakes and MSI in colorectal neoplasms. An analysis of 144 microsatellite stable (MSS) and 40 MSI colonic cancers described an increased intake of red meat as associated with increased prevalence of MSS cancers (Diergaarde, Braam, van Muijen, Ligtenberg, Kok & Kampman, 2003). However, an assessment of 437 MSS and 49 MSI colonic cancers, failed to identify a similar association with red meat and MSS status (Satia et al., 2005). Additionally, a further report, testing 238 MSS and 35 MSI colonic cancers also failed to identify red meat intake as associated with MSI or MSS status (Wu et al., 2001). However, in the study performed by Wu and colleagues, heterocyclic amines were found to be associated with increased prevalence of MSI CRC. Heterocyclic amines can be produced during certain high-temperature methods of cooking of meats (Santarelli et al., 2008). Consequently, it is plausible that cooking method, independent of, or in conjunction with, certain meat types, may be associated with MSI status in CRC, potentially explaining the inconsistent observations between MSI and meat intakes.

Alcohol intake has been described as associated with MSI status in CRC. One report, analysing 1337 MSS and 227 MSI CRC identified increased alcohol intake as associated with a higher prevalence of MSS cancers (Poynter et al., 2009). Discordantly, a second analysis of 1244 MSS and 266 MSI colonic cancers identified increased alcohol consumption as linked to increased prevalence of MSI cancers (Slattery et al., 2001).

Folate intake has also been assessed relative to MSI status in CRC. Increased levels of plasma folate were associated with MSI cancer prevalence in a report assessing 166 MSS and 24 MSI CRC (Van Guelpen et al., 2010). However, assessment of dietary intake of folate in studies testing 179 MSS and 16 MSI CRC (Chang et al., 2007), 572 MSS and 76 MSI CRC (de Vogel et al., 2008), 111 MSS and 19 MSI CRC (Jensen et al., 2008) and 542 MSS and 127 MSI colonic cancers (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008) all identified no association between folate intake and MSI status in CRC.

Presently, the data describing dietary associations and MSI status in CRC are contradictory and difficult to interpret. No strong associations have been identified and corroborated in independent cohorts. The difficulty in identification of plausible dietary constituents which may affect MSI prevalence in CRC may be due to the lack of such a relationship existing. It 14 Colorectal Cancer

unknown, although some evidence exists which suggest that such an increase in methyl group incorporation at these sites occurs during ageing in normal epithelial cells in the gut, and this is elevated in cancer (Toyota et al., 1999). Hypermethylation of gene promotors, in addition to or independent of methylation of other local DNA sequences, leads to transcriptional silencing of those genes. Such transcriptional silencing can be considered as one mechanism by which genes can be 'knocked out', in addition to mutation and deletion, in Knudson's model of tumour suppressor gene inactivation (Kondo & Issa, 2004). In this way, the aberrant methylation of genes can contribute to their inactivation in cancer epigenetically, such that in the absence of inactivating genetic changes tumour suppressor gene activity can be lost,

MSI in CRC has been assessed in relation to dietary intakes in several reports, many of which did not identify a link between this type of genomic instability in CRC and specific dietary intakes (Chang et al., 2007; de Vogel et al., 2008; Jensen et al., 2008; Schernhammer, Giovannuccci, Fuchs & Ogino, 2008) (Table 2). However, a limited number of studies have described links between dietary intakes and MSI in colorectal neoplasms. An analysis of 144 microsatellite stable (MSS) and 40 MSI colonic cancers described an increased intake of red meat as associated with increased prevalence of MSS cancers (Diergaarde, Braam, van Muijen, Ligtenberg, Kok & Kampman, 2003). However, an assessment of 437 MSS and 49 MSI colonic cancers, failed to identify a similar association with red meat and MSS status (Satia et al., 2005). Additionally, a further report, testing 238 MSS and 35 MSI colonic cancers also failed to identify red meat intake as associated with MSI or MSS status (Wu et al., 2001). However, in the study performed by Wu and colleagues, heterocyclic amines were found to be associated with increased prevalence of MSI CRC. Heterocyclic amines can be produced during certain high-temperature methods of cooking of meats (Santarelli et al., 2008). Consequently, it is plausible that cooking method, independent of, or in conjunction with, certain meat types, may be associated with MSI status in CRC, potentially explaining

Alcohol intake has been described as associated with MSI status in CRC. One report, analysing 1337 MSS and 227 MSI CRC identified increased alcohol intake as associated with a higher prevalence of MSS cancers (Poynter et al., 2009). Discordantly, a second analysis of 1244 MSS and 266 MSI colonic cancers identified increased alcohol consumption as linked to increased

Folate intake has also been assessed relative to MSI status in CRC. Increased levels of plasma folate were associated with MSI cancer prevalence in a report assessing 166 MSS and 24 MSI CRC (Van Guelpen et al., 2010). However, assessment of dietary intake of folate in studies testing 179 MSS and 16 MSI CRC (Chang et al., 2007), 572 MSS and 76 MSI CRC (de Vogel et al., 2008), 111 MSS and 19 MSI CRC (Jensen et al., 2008) and 542 MSS and 127 MSI colonic cancers (Schernhammer, Giovannuccci, Fuchs & Ogino, 2008) all identified no association between

Presently, the data describing dietary associations and MSI status in CRC are contradictory and difficult to interpret. No strong associations have been identified and corroborated in independent cohorts. The difficulty in identification of plausible dietary constituents which may affect MSI prevalence in CRC may be due to the lack of such a relationship existing. It

leading to cancer progression.

**2.4.2 MSI and dietary associations**

the inconsistent observations between MSI and meat intakes.

prevalence of MSI cancers (Slattery et al., 2001).

folate intake and MSI status in CRC.


Table 2. Summarised description of literature analysing microsatellite instability (MSI) in colorectal neoplasia in relation to dietary intakes with the statistically significant associations described. *MSS*: microsatellite stability, *WT*: wildtype, *CRC*: colorectal cancer, *CC*: colonic cancer, ↑ and ↓ denote an increase or decrease in consumption respectively.

may also be plausible that such relationships exist and are particularly subtle. Methylation of the *MLH1* promoter, leading to gene silencing and subsequent DNA MMR deficiency, occurs in the vast majority, but not all, of MSI CRC (Kuismanen et al., 2000); suggesting that other components of the MMR system can be disrupted, such as mutations to the *MSH2* or *MSH6* genes, and that MSI may develop from a group of distinct initial aberrations in a small proportion of CRC. Furthermore, subsequent instability at microsatellites as a result may depend on other promoting factors. As such, it appears that a series of molecular events takes place leading to the MSI phenotype, which may arise from different epigenetic silencing or mutational events in different cancers. The multiple causes of MSI, and the different associated factors, may explain, at least in part, the lack of consistently identified dietary constituents which have been associated with this type of genomic instability. Alternatively, age-related susceptibility to promoter methylation, including the MLH1 promoter, may be the predominant risk factor for MSI in CRC rather than dietary factors.

#### **2.4.3 CIMP and dietary associations**

Studies assessing dietary associations with CIMP in CRC have centred largely on testing intakes of those compounds which may act as methyl group donors, or which function in the biochemical pathways responsible for methylation processes. Vitamin B6 has been described as associated with an increased prevalence of CIMP in CRC in one study assessing 496 CIMP-low/absent and 152 CIMP-high cancers (de Vogel et al., 2008). However, several other reports, assessing 288 CIMP-low/absent and 87 CIMP-high (Schernhammer et al., 2011) and 824 CIMP-low/absent and 330 CIMP-high (Slattery et al., 2007) colonic cancers failed to identify a similar association.

A similar lack of consensus has been observed when assessing vitamin B12. A single study assessing 107 CIMP-low/absent and 44 CIMP-high colonic cancers described an increased serum vitamin B12 concentration as associated with CIMP in this cohort (Mokarram et al., 2008). Schernhammer and colleagues (Schernhammer et al., 2011) and Slattery and co-workers (Slattery et al., 2007) did not identify a similar association in their studies. A report assessing 163 CIMP-low/absent and 27 CIMP-high CRC also identified no association between vitamin B12 intakes and CIMP status (Van Guelpen et al., 2010). Assessment of folate intake in relation to CIMP status has consistently failed to identify associations between the two in both colorectal and colonic cancer studies (Schernhammer et al., 2011; Slattery et al., 2007; Van Guelpen et al., 2010).

in Colorectal Cancer and Diet 17

The Molecular Genetic Events in Colorectal Cancer and Diet 189

Schernhammer *et al* 2011 CC: 288† CC: 87† no association between folate, vitamin B6, B12, methionine or alcohol consumption and CIMP Slattery *et al* 2007 CC: 824†† CC: 330†† no association between folate, vitamin B6, B12, methionine or alcohol consumption and CIMP Slattery *et al* 2010 RC: 776†† RC: 74†† no association between calcium and vitamin D consumption and CIMP Slattery *et al* 2010 RC: 776†† RC: 74†† ↑ fat-rich dairy products and ↑ omega-3 fatty acids associated with CIMP

immunohistochemistry. ∓ CIMP positive status defined by methylation of one or more of the *p16*, *MLH1* or *MSH2* promoters. † CIMP positive status defined by methylation at 11 of 16 tested markers. †† CIMP positive status defined by methylation at 2 of 5 tested markers. ‡

ratios should be further analysed in order to interpret the relative 'strength' of the associations

In addition to statistical methods, the methodology of dietary assessment in each individual report has not been discussed. Dietary intakes can be measured in a variety of ways, including person-to-person interview, food frequency questionnaires, food diaries and biomarker assessment. Such an assessment is beyond the scope of this chapter. Outside of this review, several specific reports have been published describing the merits, limits and practicality of some of the available options for dietary assessment (Bingham et al., 1995; Day et al., 2001). To fully interpret dietary associations identified in different studies, although not discussed herein, an appreciation of dietary assessment methodology, and the relative accuracy of such

Further to the limits inherent in the compilation of this review, consideration of the nature of assessment of dietary intakes relative to characteristics of colorectal cancers is required. For example, considerably more data exist describing the relationship between mutations in *K-RAS* and diet than for *APC*. An 'assessment bias' exists, presumably due to the significantly simpler task of examining hotspot mutation regions of the *K-RAS* proto-oncogene compared with the longer lengths of sequencing required for mutational assessment of tumour suppressor genes. As a result, the molecular genetic changes which occur during CRC development have not been assessed at equal frequencies. Such 'assessment bias' should be noted when considering such a broad view, as presented in this chapter. This should be particularly considered when trying to interpret the genetic or molecular changes which have

It should also be understood that in many reports assessing dietary associations in CRC broad definitions are employed, in order to maintain the practicality and feasibility of studies. For example, often reports describing mutations in *BRAF* are actually describing mutations only in exons 11 and 15; reports describing *p53* mutations are frequently only describing mutational analyses of exons 5-8. Such limited analyses of coding regions is justified, with the significant majority of mutations in these examples being present in the regions described.

**CRC/CC/RC CRC/CC/RC association** de Vogel *et al* 2008 CRC: 496\* CRC: 152\* ↑ vitamin B6 associated with *MLH1* promoter methylation in males only Diergaarde *et al* 2003 CC: 167\*\* CC: 17\*\* ↓ fruit associated with *MLH1* promoter methylation and concurrent absence of MLH1 protein Mas *et al* 2007 CRC: 98\* CRC: 22\* ↓ vitamin A associated with *MLH1* promoter methylation Mokarram *et al* 2008 CC: 107∓ CC: 44∓ increased levels of serum B12 associated with CIMP

**Study CIMP-low/absent CIMP-high dietary**

Slattery *et al* 2010 RC: 776†† RC: 74†† long-term ↑ spirits/liquor with CIMP Van Guelpen *et al* 2010 CRC: 163‡ CRC: 27‡ no association between plasma folate or plasma vitamin B12 and CIMP Table 3. Summarised description of literature analysing CpG island methylator phenotype (CIMP) in colorectal neoplasia in relation to dietary intakes with the statistically significant associations described. *WT*: wildtype, *CRC*: colorectal cancer, *CC*: colonic cancer, *RC*: rectal cancer, ↑ and ↓ denote an increase or decrease in consumption respectively. \* CIMP positive status defined by *MLH1* promoter methylation. \*\* CIMP positive status defined by *MLH1*

promoter methylation and concurrent loss of MLH1 expression as determined by

CIMP positive status defined by methylation at 6 of 8 tested markers.

been tested in relation to diet in only a small number of studies.

highlighted here.

techniques, should be taken into account.

A single report, assessing 167 CIMP-low/negative and 17 CIMP-high colonic cancers identified reduced fruit intake as associated with an increased prevalence of CIMP-high colonic cancer (Diergaarde, Braam, van Muijen, Ligtenberg, Kok & Kampman, 2003). In an independent study, reduced consumption of vitamin A was identified as associated with increased prevalence of CIMP-high CRC (98 CIMP-low/absent CRC and 22 CIMP-high CRC) (Mas et al., 2007). These observations are yet to be corroborated in other studies. An additional report, assessing 776 CIMP-low/absent and 74 CIMP-high rectal cancers (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010) failed to identify fruit intakes as associated with CIMP-high rectal cancer prevalence. Little additional data exists describing vitamin A intakes relative to CIMP status in CRC.

A limited number of additional associations have been observed relating CIMP status to certain dietary patterns. One report, assessing broad dietary patterns in addition to specific nutrient and foodstuff intakes identified increased fat-rich dairy products and omega-3 fatty acid consumption as associated with increased frequency of CIMP-high rectal cancers (776 CIMP-low/absent cancers and 74 CIMP-high cancers) (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). In an additional analysis, using this same patient cohort, long-term liquor/spirit intake was also found to be associated with an increased prevalence of CIMP-high status (Slattery, Wolff, Herrick, Curtin, Caan & Samowitz, 2010). Very few studies have assessed alcohol intake in terms of beverage consumed, as such, this observation awaits confirmation in an independent study. Additional data do not exist at present which validate the observed associations between increased consumption of fat-rich dairy products and omega-3 fatty acid with CIMP-high status.

There is no dietary intake which has been identified in several cohorts as associated with CIMP-high colorectal neoplasia. This may be due to the variety of methodologies used to assess CIMP status and the different criteria used to define CIMP-high status in these cancers, with no consensus method and definition having been used across studies (see Table 3). Furthermore, CIMP itself is the resulting phenotype of precursor genetic and epigenetic aberrations. As such, it may be plausible that this CRC subtype may not be linked to dietary risk factors, but instead diet may be linked to the causative precursor events, such as *MLH1* promoter methylation and MSI. Assessment of large study cohorts, in which CIMP-high cancers are categorised by causative lesions or processes, would in part help to understand dietary intakes and causation in the context of this phenotype.

#### **3. Review limitations**

This review has attempted to assess the available data describing the relationship between dietary factors and the molecular genetic events occurring during the development and progression of CRC. Published analyses have been summarised and where consensus between studies exists, this has been highlighted. Although providing a synopsis of the available information, several limitations are inherent in such a general discussion.

No detailed analysis or discussion of the methods of statistical analysis in each report has been provided. The wide range of methodology employed for this purpose across studies makes such a discussion in the present chapter impractical. Opinions on statistical methods vary across reports in terms of adjustment for multiple testing, inclusion of confounding variables in statistical models and the requirement for power calculations. In this context, no discussion or comparison of statistical methods has been attempted; notably, hazard and odds 16 Colorectal Cancer

A single report, assessing 167 CIMP-low/negative and 17 CIMP-high colonic cancers identified reduced fruit intake as associated with an increased prevalence of CIMP-high colonic cancer (Diergaarde, Braam, van Muijen, Ligtenberg, Kok & Kampman, 2003). In an independent study, reduced consumption of vitamin A was identified as associated with increased prevalence of CIMP-high CRC (98 CIMP-low/absent CRC and 22 CIMP-high CRC) (Mas et al., 2007). These observations are yet to be corroborated in other studies. An additional report, assessing 776 CIMP-low/absent and 74 CIMP-high rectal cancers (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010) failed to identify fruit intakes as associated with CIMP-high rectal cancer prevalence. Little additional data exists describing vitamin A intakes

A limited number of additional associations have been observed relating CIMP status to certain dietary patterns. One report, assessing broad dietary patterns in addition to specific nutrient and foodstuff intakes identified increased fat-rich dairy products and omega-3 fatty acid consumption as associated with increased frequency of CIMP-high rectal cancers (776 CIMP-low/absent cancers and 74 CIMP-high cancers) (Slattery, Curtin, Wolff, Herrick, Caan & Samowitz, 2010). In an additional analysis, using this same patient cohort, long-term liquor/spirit intake was also found to be associated with an increased prevalence of CIMP-high status (Slattery, Wolff, Herrick, Curtin, Caan & Samowitz, 2010). Very few studies have assessed alcohol intake in terms of beverage consumed, as such, this observation awaits confirmation in an independent study. Additional data do not exist at present which validate the observed associations between increased consumption of fat-rich dairy products and

There is no dietary intake which has been identified in several cohorts as associated with CIMP-high colorectal neoplasia. This may be due to the variety of methodologies used to assess CIMP status and the different criteria used to define CIMP-high status in these cancers, with no consensus method and definition having been used across studies (see Table 3). Furthermore, CIMP itself is the resulting phenotype of precursor genetic and epigenetic aberrations. As such, it may be plausible that this CRC subtype may not be linked to dietary risk factors, but instead diet may be linked to the causative precursor events, such as *MLH1* promoter methylation and MSI. Assessment of large study cohorts, in which CIMP-high cancers are categorised by causative lesions or processes, would in part help to understand

This review has attempted to assess the available data describing the relationship between dietary factors and the molecular genetic events occurring during the development and progression of CRC. Published analyses have been summarised and where consensus between studies exists, this has been highlighted. Although providing a synopsis of the available

No detailed analysis or discussion of the methods of statistical analysis in each report has been provided. The wide range of methodology employed for this purpose across studies makes such a discussion in the present chapter impractical. Opinions on statistical methods vary across reports in terms of adjustment for multiple testing, inclusion of confounding variables in statistical models and the requirement for power calculations. In this context, no discussion or comparison of statistical methods has been attempted; notably, hazard and odds

relative to CIMP status in CRC.

omega-3 fatty acid with CIMP-high status.

**3. Review limitations**

dietary intakes and causation in the context of this phenotype.

information, several limitations are inherent in such a general discussion.


Table 3. Summarised description of literature analysing CpG island methylator phenotype (CIMP) in colorectal neoplasia in relation to dietary intakes with the statistically significant associations described. *WT*: wildtype, *CRC*: colorectal cancer, *CC*: colonic cancer, *RC*: rectal cancer, ↑ and ↓ denote an increase or decrease in consumption respectively. \* CIMP positive status defined by *MLH1* promoter methylation. \*\* CIMP positive status defined by *MLH1* promoter methylation and concurrent loss of MLH1 expression as determined by immunohistochemistry. ∓ CIMP positive status defined by methylation of one or more of the *p16*, *MLH1* or *MSH2* promoters. † CIMP positive status defined by methylation at 11 of 16 tested markers. †† CIMP positive status defined by methylation at 2 of 5 tested markers. ‡ CIMP positive status defined by methylation at 6 of 8 tested markers.

ratios should be further analysed in order to interpret the relative 'strength' of the associations highlighted here.

In addition to statistical methods, the methodology of dietary assessment in each individual report has not been discussed. Dietary intakes can be measured in a variety of ways, including person-to-person interview, food frequency questionnaires, food diaries and biomarker assessment. Such an assessment is beyond the scope of this chapter. Outside of this review, several specific reports have been published describing the merits, limits and practicality of some of the available options for dietary assessment (Bingham et al., 1995; Day et al., 2001). To fully interpret dietary associations identified in different studies, although not discussed herein, an appreciation of dietary assessment methodology, and the relative accuracy of such techniques, should be taken into account.

Further to the limits inherent in the compilation of this review, consideration of the nature of assessment of dietary intakes relative to characteristics of colorectal cancers is required. For example, considerably more data exist describing the relationship between mutations in *K-RAS* and diet than for *APC*. An 'assessment bias' exists, presumably due to the significantly simpler task of examining hotspot mutation regions of the *K-RAS* proto-oncogene compared with the longer lengths of sequencing required for mutational assessment of tumour suppressor genes. As a result, the molecular genetic changes which occur during CRC development have not been assessed at equal frequencies. Such 'assessment bias' should be noted when considering such a broad view, as presented in this chapter. This should be particularly considered when trying to interpret the genetic or molecular changes which have been tested in relation to diet in only a small number of studies.

It should also be understood that in many reports assessing dietary associations in CRC broad definitions are employed, in order to maintain the practicality and feasibility of studies. For example, often reports describing mutations in *BRAF* are actually describing mutations only in exons 11 and 15; reports describing *p53* mutations are frequently only describing mutational analyses of exons 5-8. Such limited analyses of coding regions is justified, with the significant majority of mutations in these examples being present in the regions described.

in Colorectal Cancer and Diet 19

The Molecular Genetic Events in Colorectal Cancer and Diet 191

prognosis. Some have suggested that excessive administration of dietary advice may prove to be counterproductive: advice should be administered sparsely and where the greatest potential to reduce cancer-related deaths exists. It is in this context that the understanding of diet and the molecular subtypes of CRC has the greatest potential and will provide the

At present, although data exist describing the association of particular dietary factors with the specific molecular genetic changes in CRC, very few consistently reproducible associations have been described. Several factors may contribute to this, including variations in study methodologies (dietary intake assessment, sequencing strategies), statistical assessment (variation in the statistical power/number of samples, inclusion of different confounding

Assessment of the presently available data do describe some associations which warrant further study: *K-RAS* mutation appears to be less frequent in CRC in individuals consuming a high folate diet. Furthermore, *APC* mutation appears to be associated with meat intakes to

At present, the study of diet in relation to the specific subtypes of CRC is at an exciting stage. Sequencing technology advancements now provide an avenue by which the total genetic composition of CRC, and the specific molecular subtypes, can be assessed. Using such tools, detailed understanding of genomewide events can be correlated with dietary intakes. Such modern approaches, coupled with renewed efforts to improve, validate and employ the most reliable and accurate methods of dietary intake assessment, provide the keys to the success of this field, which will help to provide the sought after end goal of a reduction in the number of

This work was supported by EPIC Norfolk and the Medical Research Council Centre for Nutritional Epidemiology in Cancer Prevention and Survival. Furthermore, the dedication and support of Professor Kay-Tee Khaw and the late Professor Sheila Bingham were essential

Arends, M. J., McGregor, A. H., Toft, N. J., Brown, E. J. & Wyllie, A. H. (1993). Susceptibility

is associated with endonuclease availability, *Br J Cancer* 68(6): 1127–1133. Arends, M. J., McGregor, A. H. & Wyllie, A. H. (1994). Apoptosis is inversely related to

Aylon, Y. & Oren, M. (2011). New plays in the p53 theater, *Curr Opin Genet Dev* 21(1): 86–92. Bautista, D., Obrador, A., Moreno, V., Cabeza, E., Canet, R., Benito, E., Bosch, X. & Costa, J.

ras, and HPV oncogenes, *Am J Pathol* 144(5): 1045–1057.

to apoptosis is differentially regulated by c-myc and mutated Ha-ras oncogenes and

necrosis and determines net growth in tumors bearing constitutively expressed myc,

(1997). Ki-ras mutation modifies the protective effect of dietary monounsaturated

greatest impact in the effort to reduce the number of CRC-related deaths.

**5. Conclusions and summary**

CRC-related deaths.

**7. References**

**6. Acknowledgments**

in the completion of this chapter.

variables in models) and features of study design.

some degree, although this exact relationship is unclear.

Furthermore, such limitations increase the practicality of these studies, in terms of both financial support and time investments required. Additionally, these limited regions of analyses are frequently selected based on biological evidence. Although justified, the limited extent to which genes are searched for the presence of mutations should be appreciated, and such variability between studies may in part explain inconsistent observations. In conjunction with this, different methods of mutational assessment provide different levels of sensitivity. For example, hotspot mutational assessment has been demonstrated to be more sensitively performed using pyrosequencing compared with dideoxysequencing (Naguib et al., 2010; Ogino et al., 2005). Such discrepancies between different reports were not discussed in this chapter, but should be considered when making side-by-side comparisons of studies.

In addition to the genetic and epigenetic changes giving rise to CRC development and progression described in this chapter, additional events occur during progression of these neoplasms. Furthermore, these events may be associated with dietary intakes, and data exist describing their associations with dietary consumptions; for example, loss of PTEN expression has also been tested for association with dietary intakes in CRC (Naguib, Cooke, Happerfield, Kerr, Gay, Luben, Ball, Mitrou, McTaggart & Arends, 2011). Studies of genetic and epigentic events beyond those discussed here were omitted due to the current low number of studies assessing their relationship with diet.

#### **4. Future directions of the field**

Next generation sequencing technology now affords the practical and accurate sequencing of entire genomes, with such strategies being employed to assess the genetic changes in several cancer types (Stratton et al., 2009). Furthermore, genomewide single nucleotide polymorphism analyses are being employed in a variety of settings. With these tools it is now possible to ask different questions relating diet to cancer. Are certain chemicals in the diet associated with an increased prevalence of any type of base change across the genome? Are transitions or transversions associated with intakes of specific compounds? The prospect of such investigations greatly expand the potential to understand the biochemical implications of certain dietary intakes, and provide an attractive avenue by which the identification of initiating factors in colorectal carcinogenesis might be pursued.

At present, a moderate number of studies have attempted to assess what impact, if any, dietary factors may have on CRC and the molecular subtypes of tumours which comprise this disease. With new technologies becoming available which have the power to expand this field of study, the underlying question of the purpose of such analyses should be clarified. Simply identifying dietary links to disease is only of limited use: how can this understanding be employed to reduce cancer-related mortality? It may be unrealistic to expect that if dietary constituents can be shown to be associated with increased prevalence of any particular molecular subtypes of colorectal cancer that these may be eliminated from the diet. The overwhelming evidence describing the strong association between tobacco use and cancer mortality fails to deter a significant number of smokers; although, the identification of such a link has undoubtedly provided individuals with knowledge upon which informed decisions have been made to refrain from tobacco use. Instead, a more 'protective' approach might be endorsed, such that dietary constituents which are found to confer protection against certain types of CRC might be promoted. This may be particularly useful in the attempt to lower the number of diagnoses of the molecular subtypes of CRC which confer a poor prognosis. Some have suggested that excessive administration of dietary advice may prove to be counterproductive: advice should be administered sparsely and where the greatest potential to reduce cancer-related deaths exists. It is in this context that the understanding of diet and the molecular subtypes of CRC has the greatest potential and will provide the greatest impact in the effort to reduce the number of CRC-related deaths.

#### **5. Conclusions and summary**

18 Colorectal Cancer

Furthermore, such limitations increase the practicality of these studies, in terms of both financial support and time investments required. Additionally, these limited regions of analyses are frequently selected based on biological evidence. Although justified, the limited extent to which genes are searched for the presence of mutations should be appreciated, and such variability between studies may in part explain inconsistent observations. In conjunction with this, different methods of mutational assessment provide different levels of sensitivity. For example, hotspot mutational assessment has been demonstrated to be more sensitively performed using pyrosequencing compared with dideoxysequencing (Naguib et al., 2010; Ogino et al., 2005). Such discrepancies between different reports were not discussed in this

chapter, but should be considered when making side-by-side comparisons of studies.

assessing their relationship with diet.

**4. Future directions of the field**

In addition to the genetic and epigenetic changes giving rise to CRC development and progression described in this chapter, additional events occur during progression of these neoplasms. Furthermore, these events may be associated with dietary intakes, and data exist describing their associations with dietary consumptions; for example, loss of PTEN expression has also been tested for association with dietary intakes in CRC (Naguib, Cooke, Happerfield, Kerr, Gay, Luben, Ball, Mitrou, McTaggart & Arends, 2011). Studies of genetic and epigentic events beyond those discussed here were omitted due to the current low number of studies

Next generation sequencing technology now affords the practical and accurate sequencing of entire genomes, with such strategies being employed to assess the genetic changes in several cancer types (Stratton et al., 2009). Furthermore, genomewide single nucleotide polymorphism analyses are being employed in a variety of settings. With these tools it is now possible to ask different questions relating diet to cancer. Are certain chemicals in the diet associated with an increased prevalence of any type of base change across the genome? Are transitions or transversions associated with intakes of specific compounds? The prospect of such investigations greatly expand the potential to understand the biochemical implications of certain dietary intakes, and provide an attractive avenue by which the identification of

At present, a moderate number of studies have attempted to assess what impact, if any, dietary factors may have on CRC and the molecular subtypes of tumours which comprise this disease. With new technologies becoming available which have the power to expand this field of study, the underlying question of the purpose of such analyses should be clarified. Simply identifying dietary links to disease is only of limited use: how can this understanding be employed to reduce cancer-related mortality? It may be unrealistic to expect that if dietary constituents can be shown to be associated with increased prevalence of any particular molecular subtypes of colorectal cancer that these may be eliminated from the diet. The overwhelming evidence describing the strong association between tobacco use and cancer mortality fails to deter a significant number of smokers; although, the identification of such a link has undoubtedly provided individuals with knowledge upon which informed decisions have been made to refrain from tobacco use. Instead, a more 'protective' approach might be endorsed, such that dietary constituents which are found to confer protection against certain types of CRC might be promoted. This may be particularly useful in the attempt to lower the number of diagnoses of the molecular subtypes of CRC which confer a poor

initiating factors in colorectal carcinogenesis might be pursued.

At present, although data exist describing the association of particular dietary factors with the specific molecular genetic changes in CRC, very few consistently reproducible associations have been described. Several factors may contribute to this, including variations in study methodologies (dietary intake assessment, sequencing strategies), statistical assessment (variation in the statistical power/number of samples, inclusion of different confounding variables in models) and features of study design.

Assessment of the presently available data do describe some associations which warrant further study: *K-RAS* mutation appears to be less frequent in CRC in individuals consuming a high folate diet. Furthermore, *APC* mutation appears to be associated with meat intakes to some degree, although this exact relationship is unclear.

At present, the study of diet in relation to the specific subtypes of CRC is at an exciting stage. Sequencing technology advancements now provide an avenue by which the total genetic composition of CRC, and the specific molecular subtypes, can be assessed. Using such tools, detailed understanding of genomewide events can be correlated with dietary intakes. Such modern approaches, coupled with renewed efforts to improve, validate and employ the most reliable and accurate methods of dietary intake assessment, provide the keys to the success of this field, which will help to provide the sought after end goal of a reduction in the number of CRC-related deaths.

#### **6. Acknowledgments**

This work was supported by EPIC Norfolk and the Medical Research Council Centre for Nutritional Epidemiology in Cancer Prevention and Survival. Furthermore, the dedication and support of Professor Kay-Tee Khaw and the late Professor Sheila Bingham were essential in the completion of this chapter.

#### **7. References**


in Colorectal Cancer and Diet 21

The Molecular Genetic Events in Colorectal Cancer and Diet 193

Diergaarde, B., Braam, H., van Muijen, G. N. P., Ligtenberg, M. J. L., Kok, F. J. & Kampman,

Diergaarde, B., Tiemersma, E. W., Braam, H., van Muijen, G. N. P., Nagengast, F. M., Kok, F. J.

Diergaarde, B., van Geloof, W. L., van Muijen, G. N. P., Kok, F. J. & Kampman, E. (2003).

carcinomas: a Dutch population-based study, *Carcinogenesis* 24(2): 283–290. Eichholzer, M., Luthy, J., Moser, U. & Fowler, B. (2001). Folate and the risk of

Fodde, R., Smits, R. & Clevers, H. (2001). APC, signal transduction and genetic instability in

Forbes, S. A., Bhamra, G., Bamford, S., Dawson, E., Kok, C., Clements, J., Menzies, A., Teague,

Fransen, K., Klintenas, M., Osterstrom, A., Dimberg, J., Monstein, H.-J. & Soderkvist, P.

Freedman, A. N., Michalek, A. M., Marshall, J. R., Mettlin, C. J., Petrelli, N. J., Black, J. D.,

Freudenheim, J. L., Graham, S., Marshall, J. R., Haughey, B. P., Cholewinski, S. & Wilkinson,

Goss, K. H. & Groden, J. (2000). Biology of the adenomatous polyposis coli tumor suppressor,

Herman, J. G., Umar, A., Polyak, K., Graff, J. R., Ahuja, N., Issa, J. P., Markowitz, S., Willson,

Ibrahim, A. E. K., Arends, M. J., Silva, A.-L., Wyllie, A. H., Greger, L., Ito, Y., Vowler,

Jen, J., Powell, S. M., Papadopoulos, N., Smith, K. J., Hamilton, S. R., Vogelstein, B. & Kinzler,

Cancer (COSMIC), *Curr Protoc Hum Genet* Chapter 10: Unit 10.11.

*Cancer Epidemiol Biomarkers Prev* 12(11 Pt 1): 1130–1136.

colorectal adenomas, *Int J Cancer* 113(1): 126–132.

colorectal cancer., *Nat Rev Cancer* 1(1): 55–67.

adenocarcinomas., *Carcinogenesis* 25(4): 527–533.

cancer, *J Nutr* 136(12): 3015–3021.

131(37-38): 539–549.

*Prev* 5(4): 285–291.

*J Clin Oncol* 18(9): 1967–1979.

neoplastic progression, *Gut* 60(4): 499–508.

*S A* 95(12): 6870–6875.

54(21): 5523–5526.

20(2): 368–374.

& Weijenberg, M. P. (2006). Dietary folate and APC mutations in sporadic colorectal

E. (2003). Dietary factors and microsatellite instability in sporadic colon carcinomas,

& Kampman, E. (2005). Dietary factors and truncating APC mutations in sporadic

Dietary factors and the occurrence of truncating APC mutations in sporadic colon

colorectal, breast and cervix cancer: the epidemiological evidence., *Swiss Med Wkly*

J. W., Futreal, P. A. & Stratton, M. R. (2008). The Catalogue of Somatic Mutations in

(2004). Mutation analysis of the BRAF, ARAF and RAF-1 genes in human colorectal

Zhang, Z. F., Satchidanand, S. & Asirwatham, J. E. (1996). Familial and nutritional risk factors for p53 overexpression in colorectal cancer, *Cancer Epidemiol Biomarkers*

G. (1991). Folate intake and carcinogenesis of the colon and rectum., *Int J Epidemiol*

J. K., Hamilton, S. R., Kinzler, K. W., Kane, M. F., Kolodner, R. D., Vogelstein, B., Kunkel, T. A. & Baylin, S. B. (1998). Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma, *Proc Natl Acad Sci U*

S. L., Huang, T. H.-M., Tavaré, S., Murrell, A. & Brenton, J. D. (2011). Sequential DNA methylation changes are associated with DNMT3B overexpression in colorectal

K. W. (1994). Molecular determinants of dysplasia in colorectal lesions., *Cancer Res*

fat and calcium on sporadic colorectal cancer, *Cancer Epidemiol Biomarkers Prev* 6(1): 57–61.


20 Colorectal Cancer

Bingham, S. A., Cassidy, A., Cole, T. J., Welch, A., Runswick, S. A., Black, A. E., Thurnham,

Bongaerts, B. W. C., de Goeij, A. F. P. M., van den Brandt, P. A. & Weijenberg, M. P. (2006).

Boyle, P. & Langman, J. S. (2000). ABC of colorectal cancer: Epidemiology., *BMJ*

Brink, M., Weijenberg, M. P., de Goeij, A. F. P. M., Roemen, G. M. J. M., Lentjes, M. H. F. M.,

Brink, M., Weijenberg, M. P., de Goeij, A. F. P. M., Roemen, G. M. J. M., Lentjes, M. H. F. M.,

Brink, M., Weijenberg, M. P., De Goeij, A. F. P. M., Schouten, L. J., Koedijk, F. D. H., Roemen, G.

Ceol, C. J., Pellman, D. & Zon, L. I. (2007). APC and colon cancer: two hits for one., *Nat Med*

Chang, S.-C., Lin, P.-C., Lin, J.-K., Yang, S.-H., Wang, H.-S. & Li, A. F.-Y. (2007). Role

Curtin, K., Slattery, M. L. & Samowitz, W. S. (2011). CpG island methylation in colorectal

Day, N., McKeown, N., Wong, M., Welch, A. & Bingham, S. (2001). Epidemiological

de Vogel, S., Bongaerts, B. W. C., Wouters, K. A. D., Kester, A. D. M., Schouten, L. J., de Goeij,

de Vogel, S., van Engeland, M., Lüchtenborg, M., de Bruïne, A. P., Roemen, G. M. J. M.,

Bos, J. L. (1989). Ras oncogenes in human cancer: a review., *Cancer Res* 49(17): 4682–4689. Bourdon, J. C., Laurenzi, V. D., Melino, G. & Lane, D. (2003). p53: 25 years of research and

more questions to answer., *Cell Death Differ* 10(4): 397–399.

Netherlands Cohort Study, *Int J Cancer* 114(5): 824–830.

Cohort Study, *Carcinogenesis* 25(9): 1619–1628.

colorectal cancers, *Int J Colorectal Dis* 22(5): 483–489.

cancer: past, present and future, *Patholog Res Int* 2011: 902674.

biological markers, *Br J Nutr* 73(4): 531–550.

6(1): 57–61.

*Alcohol* 38(3): 147–154.

321(7264): 805–808.

13(11): 1286–1287.

*Epidemiol* 30(2): 309–317.

cancer, *Carcinogenesis* 29(9): 1765–1773.

Study, *Br J Cancer* 92(7): 1310–1320.

fat and calcium on sporadic colorectal cancer, *Cancer Epidemiol Biomarkers Prev*

D., Bates, C., Khaw, K. T. & Key, T. J. (1995). Validation of weighed records and other methods of dietary assessment using the 24 h urine nitrogen technique and other

Alcohol and the risk of colon and rectal cancer with mutations in the K-ras gene,

de Bruïne, A. P., Goldbohm, R. A. & van den Brandt, P. A. (2005). Meat consumption and K-ras mutations in sporadic colon and rectal cancer in The Netherlands Cohort

de Bruïne, A. P., van Engeland, M., Goldbohm, R. A. & van den Brandt, P. A. (2005). Dietary folate intake and k-ras mutations in sporadic colon and rectal cancer in The

M. J. M., Lentjes, M. H. F. M., De Bruïne, A. P., Goldbohm, R. A. & Van Den Brandt, P. A. (2004). Fat and K-ras mutations in sporadic colorectal cancer in The Netherlands

of MTHFR polymorphisms and folate levels in different phenotypes of sporadic

assessment of diet: a comparison of a 7-day diary with a food frequency questionnaire using urinary markers of nitrogen, potassium and sodium, *Int J*

A. F. P. M., de Bruïne, A. P., Goldbohm, R. A., van den Brandt, P. A., van Engeland, M. & Weijenberg, M. P. (2008). Associations of dietary methyl donor intake with MLH1 promoter hypermethylation and related molecular phenotypes in sporadic colorectal

Lentjes, M. H. F. M., Goldbohm, R. A., van den Brandt, P. A., de Goeij, A. F. P. M.

& Weijenberg, M. P. (2006). Dietary folate and APC mutations in sporadic colorectal cancer, *J Nutr* 136(12): 3015–3021.


in Colorectal Cancer and Diet 23

The Molecular Genetic Events in Colorectal Cancer and Diet 195

Luo, F., Poulogiannis, G., Ye, H., Hamoudi, R., Zhang, W., Dong, G. & Arends, M. J. (2011).

Martínez, M. E., Maltzman, T., Marshall, J. R., Einspahr, J., Reid, M. E., Sampliner, R., Ahnen,

McMichael, A. J. & Giles, G. G. (1988). Cancer in migrants to Australia: extending the

Melhem, M. F., Law, J. C., el Ashmawy, L., Johnson, J. T., Landreneau, R. J., Srivastava,

Mokarram, P., Naghibalhossaini, F., Saberi Firoozi, M., Hosseini, S. V., Izadpanah, A., Salahi,

Naguib, A., Cooke, J. C., Happerfield, L., Kerr, L., Gay, L. J., Luben, R. N., Ball, R. Y., Mitrou,

Naguib, A., Mitrou, P. N., Gay, L. J., Cooke, J. C., Luben, R. N., Ball, R. Y., McTaggart, A.,

Naguib, A., Wilson, C. H., Adams, D. J. & Arends, M. J. (2011). Activation of K-RAS by

Norat, T., Bingham, S., Ferrari, P., Slimani, N., Jenab, M., Mazuir, M., Overvad, K., Olsen, A.,

Investigation into cancer and nutrition., *J Natl Cancer Inst* 97(12): 906–916. O'Brien, H., Matthew, J. A., Gee, J. M., Watson, M., Rhodes, M., Speakman, C. T., Stebbings,

co-mutation of codons 19 and 20 is transforming, *J Mol Signal* 6: 2.

descriptive epidemiological data., *Cancer Res* 48(3): 751–756.

mutation in sporadic colorectal adenomas, *Cancer Res* 59(20): 5181–5185. Mas, S., Lafuente, M. J., Crescenti, A., Trias, M., Ballesta, A., Molina, R., Zheng, S., Wiencke,

does not alter tumour chromosome stability, *J Pathol* 223(3): 390–399. Martin, S. A., Lord, C. J. & Ashworth, A. (2010). Therapeutic targeting of the DNA mismatch

repair pathway, *Clin Cancer Res* 16(21): 5107–5113.

and hMLH1, *Anticancer Res* 27(2): 1151–1156.

*World J Gastroenterol* 14(23): 3662–3671.

and dietary factors, *BMC Cancer* 11: 123.

cancers in the EPIC Norfolk study, *BMC Cancer* 10: 99.

146(5): 1170–1177.

Mutant K-ras promotes carcinogen-induced murine colorectal tumourigenesis, but

D. J., Hamilton, S. R. & Alberts, D. S. (1999). Risk factors for Ki-ras protooncogene

J. K. & Lafuente, A. (2007). Lower specific micronutrient intake in colorectal cancer patients with tumors presenting promoter hypermethylation in p16(INK4a), p4(ARF)

S. & Whiteside, T. L. (1995). Assessment of sensitivity and specificity of immunohistochemical staining of p53 in lung and head and neck cancers, *Am J Pathol*

H., Malek-Hosseini, S. A., Talei, A. & Mojallal, M. (2008). Methylenetetrahydrofolate reductase C677T genotype affects promoter methylation of tumor-specific genes in sporadic colorectal cancer through an interaction with folate/vitamin B12 status,

P. N., McTaggart, A. & Arends, M. J. (2011). Alterations in PTEN and PIK3CA in colorectal cancers in the EPIC Norfolk study: associations with clinicopathological

Arends, M. J. & Rodwell, S. A. (2010). Dietary, lifestyle and clinicopathological factors associated with BRAF and K-ras mutations arising in distinct subsets of colorectal

Tjonneland, A., Clavel, F., Boutron-Ruault, M.-C., Kesse, E., Boeing, H., Bergmann, M. M., Nieters, A., Linseisen, J., Trichopoulou, A., Trichopoulos, D., Tountas, Y., Berrino, F., Palli, D., Panico, S., Tumino, R., Vineis, P., Bueno-de Mesquita, H. B., Peeters, P. H. M., Engeset, D., Lund, E., Skeie, G., Ardanaz, E., Gonzalez, C., Navarro, C., Quiros, J. R., Sanchez, M.-J., Berglund, G., Mattisson, I., Hallmans, G., Palmqvist, R., Day, N. E., Khaw, K.-T., Key, T. J., San Joaquin, M., Hemon, B., Saracci, R., Kaaks, R. & Riboli, E. (2005). Meat, fish, and colorectal cancer risk: the European Prospective

W. S., Kennedy, H. J. & Johnson, I. T. (2000). K-ras mutations, rectal crypt cells


22 Colorectal Cancer

Jensen, L. H., Lindebjerg, J., Crüger, D. G., Brandslund, I., Jakobsen, A., Kolvraa, S. & Nielsen,

and thymidylate synthase in colorectal cancer, *Cancer Invest* 26(6): 583–589. Kambara, T., Simms, L. A., Whitehall, V. L., Spring, K. J., Wynter, C. V., Walsh, M. D., Barker,

in serrated polyps and cancers of the colorectum, *Gut* 53(8): 1137–1144. Kim, Y.-I. (2005). Nutritional epigenetics: impact of folate deficiency on DNA methylation

Kinzler, K. W. & Vogelstein, B. (1996). Lessons from hereditary colorectal cancer, *Cell*

Kondo, Y. & Issa, J.-P. J. (2004). Epigenetic changes in colorectal cancer, *Cancer Metastasis Rev*

Kuhnle, G. G. & Bingham, S. A. (2007). Dietary meat, endogenous nitrosation and colorectal

Kuhnle, G. G. C., Story, G. W., Reda, T., Mani, A. R., Moore, K. P., Lunn, J. C. & Bingham, S. A.

Kuismanen, S. A., Holmberg, M. T., Salovaara, R., de la Chapelle, A. & Peltomäki, P.

of microsatellite-unstable colorectal cancers, *Am J Pathol* 156(5): 1773–1779. Laso, N., Mas, S., Jose Lafuente, M., Casterad, X., Trias, M., Ballesta, A., Molina, R., Salas,

Lee, S., Cho, N. Y., Choi, M., Yoo, E. J., Kim, J. H. & Kang, G. H. (2008). Clinicopathological

Luo, F., Brooks, D. G., Ye, H., Hamoudi, R., Poulogiannis, G., Patek, C. E., Winton, D. J. &

Luo, F., Brooks, D. G., Ye, H., Hamoudi, R., Poulogiannis, G., Patek, C. E., Winton, D. J. &

Luo, F., Poulogiannis, G., Ye, H., Hamoudi, R. & Arends, M. J. (2011). Synergism between

tumorigenesis in Msh2-deficient mice, *Oncogene* 26(30): 4415–4427.

K-ras and Wnt pathways, *Int J Exp Pathol* 90(5): 558–574.

(2007). Diet-induced endogenous formation of nitroso compounds in the GI tract.,

(2000). Genetic and epigenetic modification of MLH1 accounts for a major share

J., Ascaso, C., Zheng, S., Wiencke, J. K. & Lafuente, A. (2004). Decrease in specific micronutrient intake in colorectal cancer patients with tumors presenting Ki-ras

features of CpG island methylator phenotype-positive colorectal cancer and its adverse prognosis in relation to KRAS/BRAF mutation, *Pathol Int* 58(2): 104–113. Lüchtenborg, M., Weijenberg, M. P., de Goeij, A. F. P. M., Wark, P. A., Brink, M., Roemen, G. M.

J. M., Lentjes, M. H. F. M., de Bruïne, A. P., Goldbohm, R. A., van 't Veer, P. & van den Brandt, P. A. (2005). Meat and fish consumption, APC gene mutations and hMLH1 expression in colon and rectal cancer: a prospective cohort study (The Netherlands),

Arends, M. J. (2007). Conditional expression of mutated K-ras accelerates intestinal

Arends, M. J. (2009). Mutated K-ras(Asp12) promotes tumourigenesis in Apc(Min) mice more in the large than the small intestines, with synergistic effects between

K-rasVal12 and mutant Apc accelerates murine large intestinal tumourigenesis, *Oncol*

and colon cancer susceptibility, *J Nutr* 135(11): 2703–2709.

cancer, *Biochem Soc Trans* 35(Pt 5): 1355–1357.

mutation, *Anticancer Res* 24(3b): 2011–2020.

*Cancer Causes Control* 16(9): 1041–1054.

*Rep* 26(1): 125–133.

*Free Radic Biol Med* 43(7): 1040–1047.

87(2): 159–170.

23(1-2): 29–39.

J. N. (2008). Microsatellite instability and the association with plasma homocysteine

M. A., Arnold, S., McGivern, A., Matsubara, N., Tanaka, N., Higuchi, T., Young, J., Jass, J. R. & Leggett, B. A. (2004). BRAF mutation is associated with DNA methylation


in Colorectal Cancer and Diet 25

The Molecular Genetic Events in Colorectal Cancer and Diet 197

Sjöblom, T., Jones, S., Wood, L. D., Parsons, D. W., Lin, J., Barber, T. D., Mandelker, D., Leary,

Slattery, M. L., Anderson, K., Curtin, K., Ma, K. N., Schaffer, D. & Samowitz, W. (2001). Dietary intake and microsatellite instability in colon tumors, *Int J Cancer* 93(4): 601–607. Slattery, M. L., Curtin, K., Anderson, K., Ma, K. N., Edwards, S., Leppert, M., Potter, J.,

Slattery, M. L., Curtin, K., Ma, K., Edwards, S., Schaffer, D., Anderson, K. & Samowitz, W.

Slattery, M. L., Curtin, K., Sweeney, C., Levin, T. R., Potter, J., Wolff, R. K., Albertsen,

Slattery, M. L., Curtin, K., Wolff, R. K., Herrick, J. S., Caan, B. J. & Samowitz, W. (2010).

Slattery, M. L., Wolff, R. K., Herrick, J. S., Curtin, K., Caan, B. J. & Samowitz, W. (2010). Alcohol

Soreide, K., Janssen, E. A., Soiland, H., Korner, H. & Baak, J. P. (2006). Microsatellite instability

Stratton, M. R., Campbell, P. J. & Futreal, P. A. (2009). The cancer genome, *Nature*

Suehiro, Y. & Hinoda, Y. (2008). Genetic and epigenetic changes in aberrant crypt foci and

Takayama, T., Katsuki, S., Takahashi, Y., Ohi, M., Nojiri, S., Sakamaki, S., Kato, J., Kogawa,

Toyota, M., Ahuja, N., Ohe-Toyota, M., Herman, J. G., Baylin, S. B. & Issa, J. P. (1999).

Van Guelpen, B., Dahlin, A. M., Hultdin, J., Eklöf, V., Johansson, I., Henriksson, M. L.,

Velho, S., Moutinho, C., Cirnes, L., Albuquerque, C., Hamelin, R., Schmitt, F., Carneiro, F.,

K., Miyake, H. & Niitsu, Y. (1998). Aberrant crypt foci of the colon as precursors of

CpG island methylator phenotype in colorectal cancer, *Proc Natl Acad SciUSA*

Cullman, I., Hallmans, G. & Palmqvist, R. (2010). One-carbon metabolism and CpG island methylator phenotype status in incident colorectal cancer: a nested

Oliveira, C. & Seruca, R. (2008). BRAF, KRAS and PIK3CA mutations in colorectal serrated polyps and cancer: primary or secondary genetic events in colorectal

epigenetic changes, *Cancer Causes Control* 21(8): 1237–1245.

breast and colorectal cancers, *Science* 314(5797): 268–274.

*Cancer Epidemiol Biomarkers Prev* 11(6): 541–548.

in colorectal cancer, *Br J Surg* 93(4): 395–406.

serrated polyps., *Cancer Sci* 99(6): 1071–1076.

adenoma and cancer., *N Engl J Med* 339(18): 1277–1284.

case-referent study, *Cancer Causes Control* 21(4): 557–566.

carcinogenesis?, *BMC Cancer* 8: 255.

120(3): 656–663.

53(8): 1182–1189.

458(7239): 719–724.

96(15): 8681–8686.

R. J., Ptak, J., Silliman, N., Szabo, S., Buckhaults, P., Farrell, C., Meeh, P., Markowitz, S. D., Willis, J., Dawson, D., Willson, J. K. V., Gazdar, A. F., Hartigan, J., Wu, L., Liu, C., Parmigiani, G., Park, B. H., Bachman, K. E., Papadopoulos, N., Vogelstein, B., Kinzler, K. W. & Velculescu, V. E. (2006). The consensus coding sequences of human

Schaffer, D. & Samowitz, W. S. (2000). Associations between dietary intake and Ki-ras mutations in colon tumors: a population-based study, *Cancer Res* 60(24): 6935–6941.

(2002). Diet activity, and lifestyle associations with p53 mutations in colon tumors.,

H. & Samowitz, W. S. (2007). Diet and lifestyle factor associations with CpG island methylator phenotype and BRAF mutations in colon cancer, *Int J Cancer*

Diet, physical activity, and body size associations with rectal tumor mutations and

consumption and rectal tumor mutations and epigenetic changes, *Dis Colon Rectum*

proliferation, and meat consumption in patients with left-sided colorectal carcinoma, *Eur J Cancer Prev* 9(1): 41–47.


24 Colorectal Cancer

Ogino, S., Kawasaki, T., Brahmandam, M., Yan, L., Cantor, M., Namgyal, C., Mino-Kenudson,

Parkin, D. M., Bray, F., Ferlay, J. & Pisani, P. (2005). Global cancer statistics, 2002., *CA Cancer J*

Peyssonnaux, C. & Eychene, A. (2001). The Raf/MEK/ERK pathway: new concepts of

Poulogiannis, G., Ichimura, K., Hamoudi, R. A., Luo, F., Leung, S. Y., Yuen, S. T., Harrison,

Poulogiannis, G., McIntyre, R. E., Dimitriadi, M., Apps, J. R., Wilson, C. H., Ichimura, K., Luo,

Poynter, J. N., Haile, R. W., Siegmund, K. D., Campbell, P. T., Figueiredo, J. C., Limburg,

Robinson, M. J. & Cobb, M. H. (1997). Mitogen-activated protein kinase pathways, *Curr Opin*

Samowitz, W. S., Albertsen, H., Herrick, J., Levin, T. R., Sweeney, C., Murtaugh, M. A.,

Santarelli, R. L., Pierre, F. & Corpet, D. E. (2008). Processed meat and colorectal cancer: a review of epidemiologic and experimental evidence, *Nutr Cancer* 60(2): 131–144. Satia, J. A., Keku, T., Galanko, J. A., Martin, C., Doctolero, R. T., Tajima, A., Sandler, R. S. &

Colon Cancer Study, *Cancer Epidemiol Biomarkers Prev* 14(2): 429–436. Schernhammer, E. S., Giovannuccci, E., Fuchs, C. S. & Ogino, S. (2008). A prospective study of

and CpG island methylator phenotype (CIMP), *PLoS One* 6(6): e21102. Schernhammer, E. S., Ogino, S. & Fuchs, C. S. (2008). Folate and vitamin B6 intake and risk of colon cancer in relation to p53 expression, *Gastroenterology* 135(3): 770–780.

in Apc mutant mice, *Proc Natl Acad SciUSA* 107(34): 15145–15150.

KRAS mutation detection by Pyrosequencing., *J Mol Diagn* 7(3): 413–421. Park, J. Y., Mitrou, P. N., Keen, J., Dahm, C. C., Gay, L. J., Luben, R. N., McTaggart, A.,

*Eur J Cancer Prev* 9(1): 41–47.

*Mutagenesis* 25(4): 351–358.

activation, *Biol Cell* 93(1-2): 53–62.

changes in colorectal cancer, *J Pathol* 220(3): 338–347.

*Epidemiol Biomarkers Prev* 18(10): 2745–2750.

*Cell Biol* 9(2): 180–186.

129(3): 837–845.

*Clin* 55(2): 74–108.

proliferation, and meat consumption in patients with left-sided colorectal carcinoma,

M., Lauwers, G. Y., Loda, M. & Fuchs, C. S. (2005). Sensitive sequencing method for

Khaw, K.-T., Ball, R. Y., Arends, M. J. & Rodwell, S. A. (2010). Lifestyle factors and p53 mutation patterns in colorectal cancer patients in the EPIC-Norfolk study,

D. J., Wyllie, A. H. & Arends, M. J. (2010). Prognostic relevance of DNA copy number

F., Cantley, L. C., Wyllie, A. H., Adams, D. J. & Arends, M. J. (2010). PARK2 deletions occur frequently in sporadic colorectal cancer and accelerate adenoma development

P., Young, J., Le Marchand, L., Potter, J. D., Cotterchio, M., Casey, G., Hopper, J. L., Jenkins, M. A., Thibodeau, S. N., Newcomb, P. A., Baron, J. A. & Colon Cancer Family Registry (2009). Associations between smoking, alcohol consumption, and colorectal cancer, overall and by tumor microsatellite instability status, *Cancer*

Wolff, R. K. & Slattery, M. L. (2005). Evaluation of a large, population-based sample supports a CpG island methylator phenotype in colon cancer, *Gastroenterology*

Carethers, J. M. (2005). Diet, lifestyle, and genomic instability in the North Carolina

dietary folate and vitamin B and colon cancer according to microsatellite instability and KRAS mutational status, *Cancer Epidemiol Biomarkers Prev* 17(10): 2895–2898. Schernhammer, E. S., Giovannucci, E., Baba, Y., Fuchs, C. S. & Ogino, S. (2011). B vitamins,

methionine and alcohol intake and risk of colon cancer in relation to BRAF mutation


**10** 

*Germany* 

**Colorectal Cancer and Alcohol** 

*Centre of Alcohol Research, University of Heidelberg,* 

*Department of Medicine (Gastroenterology & Hepatology), Salem Medical Centre,* 

Chronic alcohol consumption may lead to a variety of diseases and may deteriorate a great number of existing health problems. Among all these diseases the development of certain types of cancer is of major concern. Since decades it has been known that chronic alcohol consumption is a risk factor for cancer of the upper aerodigestive tract (oral cavity, pharynx, larynx and oesophagus), the liver and the female breast. Data with respect to alcohol and cancer concerning other organs do not show such clear correlations. In February 2007 an international group of epidemiologists and alcohol researchers met at the International Agency for Research on Cancer (IARC) in Lyon, France, to evaluate the role of alcohol and its first metabolite acetaldehyde as potential carcinogens in experimental animals and humans. The working group has concluded from the epidemiological data available that the occurrence of malignant tumours mentioned above is related to the consumption of alcoholic beverages. In addition, at this time epidemiologic and experimental data showed

Worldwide a total of approximately 389,000 cases of cancer presenting 3.6 % of oral cancers (5.2 % in men and 1.7 % in women) derive from chronic alcohol consumption (Rehm et al., 2004). Besides the fact that alcohol is a co-carcinogen and may act as a promoter alcohol can also accelerate tumour spread as exemplified for liver metastases of colorectal cancer possibly due to immune suppression and induction of angiogenesis by the expression of vascular endothelial growth factor (VEGF) (Seitz & Stickel, 2010). In addition, it is important to know that ethanol also interacts with the metabolism of chemo-therapeutic drugs which can result in

This review focuses solely on the effect of chronic alcohol consumption on colorectal cancer, a cancer which is wide spread in Western societies and is No. 3 cancer in men and women in Germany. The present review will, therefore, discuss epidemiology of alcohol and colorectal cancer, will briefly address possible mechanism by which alcohol stimulates colorectal carcinogenesis and may finally give some suggestions and recommendations with

An increased risk for the development of colorectal cancer associated with chronic alcohol ingestion has been considered for decades. In 1974 Breslow and Enstrom emphasized a

a decreased respond to medication and increased side-effects (De Bruijn & Slee, 1992).

that alcohol is also a risk factor for colorectal cancer (Baan et al., 2007).

respective to earlier detection and identification of high risk groups.

**1. Introduction** 

**2. Epidemiology** 

*Department of Gastroenterology, City Hospital Wolfsburg, Wolfsburg,* 

Seitz K. Helmut and Nils Homann


### **Colorectal Cancer and Alcohol**

#### Seitz K. Helmut and Nils Homann

*Department of Medicine (Gastroenterology & Hepatology), Salem Medical Centre, Centre of Alcohol Research, University of Heidelberg, Department of Gastroenterology, City Hospital Wolfsburg, Wolfsburg, Germany* 

#### **1. Introduction**

26 Colorectal Cancer

198 Colorectal Cancer – From Prevention to Patient Care

Voskuil, D. W., Kampman, E., van Kraats, A. A., Balder, H. F., van Muijen, G. N., Goldbohm,

Wheeler, J. M., Loukola, A., Aaltonen, L. A., Mortensen, N. J. & Bodmer, W. F. (2000). The

Wood, L. D., Parsons, D. W., Jones, S., Lin, J., Sjoblom, T., Leary, R. J., Shen, D., Boca, S. M.,

Wu, A. H., Shibata, D., Yu, M. C., Lai, M. Y. & Ross, R. K. (2001). Dietary heterocyclic

Zhang, Z. F., Zeng, Z. S., Sarkis, A. S., Klimstra, D. S., Charytonowicz, E., Pollack, D., Vena,

human breast and colorectal cancers, *Science* 318(5853): 1108–1113.

sporadic colorectal cancers, *J Med Genet* 37(8): 588–592.

colorectal cancer, *Br J Cancer* 71(4): 888–893.

carcinomas: relation to dietary risk factors, *Int J Cancer* 81(5): 675–681. Wark, P. A., Van der Kuil, W., Ploemacher, J., Van Muijen, G. N. P., Mulder, C. J. J.,

*Control* 18(8): 865–879.

22(10): 1681–1684.

R. A. & van't Veer, P. (1999). p53 over-expression and p53 mutations in colon

Weijenberg, M. P., Kok, F. J. & Kampman, E. (2006). Diet, lifestyle and risk of K-ras mutation-positive and -negative colorectal adenomas, *Int J Cancer* 119(2): 398–405. Weijenberg, M. P., Luchtenborg, M., de Goeij, A. F., Brink, M., van Muijen, G. N., de Bruine,

A. P., Goldbohm, R. A. & van den Brandt, P. A. (2007). Dietary fat and risk of colon and rectal cancer with aberrant MLH1 expression, APC or KRAS genes, *Cancer Causes*

role of hypermethylation of the hMLH1 promoter region in HNPCC versus MSI+

Barber, T., Ptak, J., Silliman, N., Szabo, S., Dezso, Z., Ustyanksky, V., Nikolskaya, T., Nikolsky, Y., Karchin, R., Wilson, P. A., Kaminker, J. S., Zhang, Z., Croshaw, R., Willis, J., Dawson, D., Shipitsin, M., Willson, J. K., Sukumar, S., Polyak, K., Park, B. H., Pethiyagoda, C. L., Pant, P. V., Ballinger, D. G., Sparks, A. B., Hartigan, J., Smith, D. R., Suh, E., Papadopoulos, N., Buckhaults, P., Markowitz, S. D., Parmigiani, G., Kinzler, K. W., Velculescu, V. E. & Vogelstein, B. (2007). The genomic landscapes of

amines and microsatellite instability in colon adenocarcinomas, *Carcinogenesis*

J., Guillem, J., Marshall, J. R., Cordon-Cardo, C., Cohen, A. M. & Begg, C. B. (1995). Family history of cancer, body weight, and p53 nuclear overexpression in Duke's C Chronic alcohol consumption may lead to a variety of diseases and may deteriorate a great number of existing health problems. Among all these diseases the development of certain types of cancer is of major concern. Since decades it has been known that chronic alcohol consumption is a risk factor for cancer of the upper aerodigestive tract (oral cavity, pharynx, larynx and oesophagus), the liver and the female breast. Data with respect to alcohol and cancer concerning other organs do not show such clear correlations. In February 2007 an international group of epidemiologists and alcohol researchers met at the International Agency for Research on Cancer (IARC) in Lyon, France, to evaluate the role of alcohol and its first metabolite acetaldehyde as potential carcinogens in experimental animals and humans. The working group has concluded from the epidemiological data available that the occurrence of malignant tumours mentioned above is related to the consumption of alcoholic beverages. In addition, at this time epidemiologic and experimental data showed that alcohol is also a risk factor for colorectal cancer (Baan et al., 2007).

Worldwide a total of approximately 389,000 cases of cancer presenting 3.6 % of oral cancers (5.2 % in men and 1.7 % in women) derive from chronic alcohol consumption (Rehm et al., 2004). Besides the fact that alcohol is a co-carcinogen and may act as a promoter alcohol can also accelerate tumour spread as exemplified for liver metastases of colorectal cancer possibly due to immune suppression and induction of angiogenesis by the expression of vascular endothelial growth factor (VEGF) (Seitz & Stickel, 2010). In addition, it is important to know that ethanol also interacts with the metabolism of chemo-therapeutic drugs which can result in a decreased respond to medication and increased side-effects (De Bruijn & Slee, 1992).

This review focuses solely on the effect of chronic alcohol consumption on colorectal cancer, a cancer which is wide spread in Western societies and is No. 3 cancer in men and women in Germany. The present review will, therefore, discuss epidemiology of alcohol and colorectal cancer, will briefly address possible mechanism by which alcohol stimulates colorectal carcinogenesis and may finally give some suggestions and recommendations with respective to earlier detection and identification of high risk groups.

#### **2. Epidemiology**

An increased risk for the development of colorectal cancer associated with chronic alcohol ingestion has been considered for decades. In 1974 Breslow and Enstrom emphasized a

Colorectal Cancer and Alcohol 201

This is especially relevant when a carcinogen such as acetoxymethylmethylnitrosamine (AMMN) is locally applied to the rectal mucosa (Seitz et al., 1990). Some evidence exist that acetaldehyde (AA) is an important factor since inhibition of its degradation stimulates

For a detailed summary of the animal experiments performed so far we refer to the

Various risk factors for ethanol-mediated colorectal carcinogenesis exist. Five out of 6 studies of the effect of alcohol on the occurrence of adenoma polyps in large intestine showed a positive association with alcohol (Kune et al., 1992; Seitz et al., 1998). In addition, also the occurrence of hyperplastic polyps is enhanced when more than 30 grams of alcohol per day were consumed. The relative risk for men was 1.8 and for women 2.5 (Kearney et

Alcohol affects the adenoma/carcinoma sequence at the different early steps (Boutron et al., 1995). High alcohol intake favours high risk polyps or colorectal cancer occurrence among patients with adenoma (Bardou et al., 2002). It has also been reported that a reduction in alcohol consumption for individuals with genetic predisposition for colorectal cancer had large beneficial effects on tumour incidence (Le Marchand et al., 1999). Thus, patients with tendency towards colorectal polyps have an increased risk to develop carcinoma when they

Another additional risk is possibly the presence of ulcerative colitis, although the data are not completely clear. Alcohol by itself may additionally enhance inflammation and may

Another important factor is that alcohol reduces the availability of folate which results in a decrease of methylation and thus, a decrease of thymidine generation, DNA synthesis and cellular regeneration, in a situation of enhanced need. Therefore, folate, methionine and vitamin B6 deficiency are risk factors for ethanol mediated colorectal carcinogenesis (Giovannucci et al., 1995; Larsson et al., 2005; Schernhammer et al., 2008; Weinstein et al.,

It is well known that tobacco smoking is associated with a higher risk for colonic adenoma and hyperplastic polyp formation as well as increased incidence of colorectal carcinoma

Age, another risk factor for colorectal cancer may also affect ethanol mediated cancer development in the large intestine. It has been shown in animal experiments that mucosal damage induced by chronic alcohol ingestion is more pronounced with advanced age

Finally, genetic risk factor with respect to alcohol metabolism and colorectal cancer has to be taken into consideration. Alcohol is metabolised by alcohol dehydrogenase (ADH) to AA. Seven different ADHs exist and two of them (ADH1B and ADH1C) reveal polymorphism. Among the two ADH1C is of considerable interest (see below) (Edenberg, 1997). Individuals with increase metabolism of ethanol via ADH1C due to homozygosity of the ADH1C\*1 allele seem to have a significantly increased risk for colorectal cancer when they consume

2008; Yamaji et al.; 2009; Figueiredo et al., 2009; Lee et al., 2010).

more than 30 grams alcohol per day (Homann et al., 2006).

compared to youth (Simanowski et al., 1994).

**4. Risk factors in alcohol mediated colorectal carcinogenesis** 

colorectal cancer (Seitz et al., 1990).

al., 1995).

consume additional alcohol.

thus favour carcinogenesis.

(Seitz & Cho, 2009).

following review article (Seitz et al., 2006).

correlation between beer consumption and rectal cancer occurrence (Breslow & Enstrom 1974). In 1992 Kune and Vitetta summarized the results of more than 50 epidemiologic studies between 1957 and 1991 including 7 correlational studies, more than 40 case control studies and 17 prospective cohort studies on the role of alcohol on the development of colorectal tumours (Kune & Vitetta 1992). Most of these studies reported a positive association of large bowel cancer with alcohol consumption. In addition a positive trend with respect to dose-response was found in 5 out of 10 case control studies and in all prospective cohort studies in which a dose-response analysis has been performed.

In the nineties of the last century another 12 epidemiological studies have been published with inconsistent results (Seitz et al., 2006). Most importantly a prospective cohort study in Japan reported a positive dose-response relationship between alcohol intake and colon cancer risk in men and women (N. Shimizu et al., 2003), while a Danish population based cohort study showed no association (Pedersen et al., 2003).

A panel of experts at a WHO consensus conference on nutrition and colorectal cancer reviewed in 1999 the epidemiology on alcohol and colorectal cancer and it was concluded that alcohol ingestion even in a low dose-intake between 10 and 40 grams especially consumed as beer resulted in a 1.5-fold increased risk for colorectal cancer and to a lesser extend for colonic cancer in both sexes but predominantly in men (Scheppach et al., 1999).

Cho et al showed in a meta-analysis of 8 cohort studies from the US and Europe a trend between the increased amount of alcohol intake and the risk for colorectal cancer. In this meta-analysis a consumption of more than 45 grams per day was associated with an increased risk of 45 % (Cho et al., 2004).

A huge prospective follow-up study of more than 10,000 US citizens concluded that alcohol consumption of one or more alcoholic beverages per day is associated with a 70 % greater risk of colonic cancer with a strong positive dose-response relationship (Su & Arab, 2004).

Most recently it was proposed that the alcohol colorectal cancer association is more apparent in Japanese than in Western populations. A pooled analysis of results from 5 cohort studies from Japan showed a strong and highly significant association between alcohol intake and colorectal cancer not only in men but also in women (Mizoue et al., 2008). Twenty five per cent of colorectal cancer cases in men were attributable to an alcohol intake of more than 23 grams per day. A recent meta analysis from the IARC of 34 case control and 7 cohort studies provides strong evidence for an association between alcohol consumption of more than 1 drink per day and the risk for colorectal cancer (Fedirko et al., 2011). Similar results were reported from the Netherlands (Bongaerts et al., 2008, 2010) and the US (Thygesen et al., 2008) but not from Great Britain (Park et al., 2009, 2010).

The accumulation of all these convincing epidemiologic data on alcohol and colorectal cancer resulted by the IARC that chronic alcohol consumption is a risk factor for colorectal cancer (Baan et al., 2007).

#### **3. Animal experiments**

Various animal experiments have been performed to study the effect of alcohol on chemically induced colorectal cancer. The results of these experiments depend on the experimental design, the type of carcinogen used, its time duration of exposure and dosage as well as the route of alcohol administration. While alcohol alone does not induce colorectal tumours, the administration of alcohol together with a colorectal carcinogen does under certain experimental conditions result in a stimulation of carcinogenesis (Seitz et al., 2006).

correlation between beer consumption and rectal cancer occurrence (Breslow & Enstrom 1974). In 1992 Kune and Vitetta summarized the results of more than 50 epidemiologic studies between 1957 and 1991 including 7 correlational studies, more than 40 case control studies and 17 prospective cohort studies on the role of alcohol on the development of colorectal tumours (Kune & Vitetta 1992). Most of these studies reported a positive association of large bowel cancer with alcohol consumption. In addition a positive trend with respect to dose-response was found in 5 out of 10 case control studies and in all

In the nineties of the last century another 12 epidemiological studies have been published with inconsistent results (Seitz et al., 2006). Most importantly a prospective cohort study in Japan reported a positive dose-response relationship between alcohol intake and colon cancer risk in men and women (N. Shimizu et al., 2003), while a Danish population based

A panel of experts at a WHO consensus conference on nutrition and colorectal cancer reviewed in 1999 the epidemiology on alcohol and colorectal cancer and it was concluded that alcohol ingestion even in a low dose-intake between 10 and 40 grams especially consumed as beer resulted in a 1.5-fold increased risk for colorectal cancer and to a lesser extend for colonic cancer in both sexes but predominantly in men (Scheppach et al., 1999). Cho et al showed in a meta-analysis of 8 cohort studies from the US and Europe a trend between the increased amount of alcohol intake and the risk for colorectal cancer. In this meta-analysis a consumption of more than 45 grams per day was associated with an

A huge prospective follow-up study of more than 10,000 US citizens concluded that alcohol consumption of one or more alcoholic beverages per day is associated with a 70 % greater risk of colonic cancer with a strong positive dose-response relationship (Su & Arab, 2004). Most recently it was proposed that the alcohol colorectal cancer association is more apparent in Japanese than in Western populations. A pooled analysis of results from 5 cohort studies from Japan showed a strong and highly significant association between alcohol intake and colorectal cancer not only in men but also in women (Mizoue et al., 2008). Twenty five per cent of colorectal cancer cases in men were attributable to an alcohol intake of more than 23 grams per day. A recent meta analysis from the IARC of 34 case control and 7 cohort studies provides strong evidence for an association between alcohol consumption of more than 1 drink per day and the risk for colorectal cancer (Fedirko et al., 2011). Similar results were reported from the Netherlands (Bongaerts et al., 2008, 2010) and the US (Thygesen et al.,

The accumulation of all these convincing epidemiologic data on alcohol and colorectal cancer resulted by the IARC that chronic alcohol consumption is a risk factor for colorectal

Various animal experiments have been performed to study the effect of alcohol on chemically induced colorectal cancer. The results of these experiments depend on the experimental design, the type of carcinogen used, its time duration of exposure and dosage as well as the route of alcohol administration. While alcohol alone does not induce colorectal tumours, the administration of alcohol together with a colorectal carcinogen does under certain experimental conditions result in a stimulation of carcinogenesis (Seitz et al., 2006).

prospective cohort studies in which a dose-response analysis has been performed.

cohort study showed no association (Pedersen et al., 2003).

2008) but not from Great Britain (Park et al., 2009, 2010).

increased risk of 45 % (Cho et al., 2004).

cancer (Baan et al., 2007).

**3. Animal experiments** 

This is especially relevant when a carcinogen such as acetoxymethylmethylnitrosamine (AMMN) is locally applied to the rectal mucosa (Seitz et al., 1990). Some evidence exist that acetaldehyde (AA) is an important factor since inhibition of its degradation stimulates colorectal cancer (Seitz et al., 1990).

For a detailed summary of the animal experiments performed so far we refer to the following review article (Seitz et al., 2006).

#### **4. Risk factors in alcohol mediated colorectal carcinogenesis**

Various risk factors for ethanol-mediated colorectal carcinogenesis exist. Five out of 6 studies of the effect of alcohol on the occurrence of adenoma polyps in large intestine showed a positive association with alcohol (Kune et al., 1992; Seitz et al., 1998). In addition, also the occurrence of hyperplastic polyps is enhanced when more than 30 grams of alcohol per day were consumed. The relative risk for men was 1.8 and for women 2.5 (Kearney et al., 1995).

Alcohol affects the adenoma/carcinoma sequence at the different early steps (Boutron et al., 1995). High alcohol intake favours high risk polyps or colorectal cancer occurrence among patients with adenoma (Bardou et al., 2002). It has also been reported that a reduction in alcohol consumption for individuals with genetic predisposition for colorectal cancer had large beneficial effects on tumour incidence (Le Marchand et al., 1999). Thus, patients with tendency towards colorectal polyps have an increased risk to develop carcinoma when they consume additional alcohol.

Another additional risk is possibly the presence of ulcerative colitis, although the data are not completely clear. Alcohol by itself may additionally enhance inflammation and may thus favour carcinogenesis.

Another important factor is that alcohol reduces the availability of folate which results in a decrease of methylation and thus, a decrease of thymidine generation, DNA synthesis and cellular regeneration, in a situation of enhanced need. Therefore, folate, methionine and vitamin B6 deficiency are risk factors for ethanol mediated colorectal carcinogenesis (Giovannucci et al., 1995; Larsson et al., 2005; Schernhammer et al., 2008; Weinstein et al., 2008; Yamaji et al.; 2009; Figueiredo et al., 2009; Lee et al., 2010).

It is well known that tobacco smoking is associated with a higher risk for colonic adenoma and hyperplastic polyp formation as well as increased incidence of colorectal carcinoma (Seitz & Cho, 2009).

Age, another risk factor for colorectal cancer may also affect ethanol mediated cancer development in the large intestine. It has been shown in animal experiments that mucosal damage induced by chronic alcohol ingestion is more pronounced with advanced age compared to youth (Simanowski et al., 1994).

Finally, genetic risk factor with respect to alcohol metabolism and colorectal cancer has to be taken into consideration. Alcohol is metabolised by alcohol dehydrogenase (ADH) to AA. Seven different ADHs exist and two of them (ADH1B and ADH1C) reveal polymorphism. Among the two ADH1C is of considerable interest (see below) (Edenberg, 1997). Individuals with increase metabolism of ethanol via ADH1C due to homozygosity of the ADH1C\*1 allele seem to have a significantly increased risk for colorectal cancer when they consume more than 30 grams alcohol per day (Homann et al., 2006).

Colorectal Cancer and Alcohol 203

degradation is detectable at relative low activities only (Seitz & Oneta, 1998). The net amount of AA present in the tissue may determine its toxic and carcinogenic action. Thus, individuals with an increased production and decreased degradation of AA are especially at risk for colorectal cancer development. It has been proposed that the ALDH activity of colonic mucosa may be sufficient for the removal of AA produced by colonic mucosal ADH during ethanol oxidation but it is insufficient for the removal of AA produced by

The most striking evidence of the causal role of AA in ethanol-associated carcinogenesis derives from genetic linkage studies in alcoholics and / or heavy drinkers. Individuals who accumulate AA due to polymorphism and / or mutation in the gene coding for enzymes responsible for AA generation and detoxification have been shown to have an increased cancer risk. In Japan as well as in other Asian countries a high percentage of individuals carry a mutation of the ALDH2 gene which codes for an enzyme with low activity leading to elevated AA levels after alcohol consumption. While homozygotes are completely protected against alcoholism and alcohol associated diseases due to the fact that they cannot tolerate alcohol even at very small doses, heterozygotes (ALDH2\*1,2) have an increased risk for alimentary

In addition, polymorphism of the ADH1C gene may also modulate AA levels. ADH1C\*1 transcription leads to an ADH isoenzyme 2.5 times more active than that from ADH1C\*2 (Figure1). We evaluated whether the associated between alcohol consumption and colorectal cancer development is modified by ADH1C polymorphism. We recruited 173 individuals with colorectal tumours diagnosed by coloscopy and 788 control individuals without colorectal tumours and determined their genotypes. Genotype ADH1C\*1/1 was more frequent in patient with alcohol associated colorectal neoplasia compared to patients without cancer (Homann et al., 2006). In addition, only individuals drinking more than 30 grams ethanol per day with a genotype ADH1C\*1/1 had an increased risk for colorectal tumours. These data identify ADH1C homozygosity as a genetic risk marker for colorectal tumours in individuals consuming more than 30 grams alcohol per day and emphasize further the role of AA as a carcinogenic agent in alcohol mediated colorectal

It has been shown that after alcohol administration, the amount of AA per gram of tissue is highest for the colonic mucosa compared to all other tissues in the body (Seitz et al., 1987). This is primarily due to the production of AA by the faecal bacteria. AA has toxic effects on the colon mucosa resulting in secondary compensatory hyperregeneration with increased crypt cell production rates and an extension of the proliferative compartment towards the lumen of the crypt (Simanowski et al., 1994, 2001). This change in crypt cell dynamics

The alcohol associated hyperregeneration of the colonic mucosa is especially pronounced with increasing age (Simanowski et al., 1994). As already pointed out, this may have

High AA levels have been found after alcohol administration in the colon of rats and these concentrations were significantly lower in germ free animals as compared to the conventional rats suggesting that faecal bacteria are capable of producing AA (Seitz et al., 1990). Indeed, the reversed microbial ADH reaction produces under aerobic or microaerobic conditions striking amounts of AA when human colonic contents or some microbes representing normal colonic flora are incubated in vitro at 37°C with increasing ethanol

represents a condition associating with increased risk for colorectal cancer.

practical implications since age by itself is a risk factor for CRC.

tract cancer including the colon and the rectum (Yokoyama et al., 1998) (Figure.1).

intracolonic bacteria.

carcinogenesis.

#### **5. Possible mechanisms of alcohol mediated colorectal carcinogenesis**

#### **5.1 Acetaldehyde (AA)**

Most recently the IARC has identified AA as an important carcinogen for humans (Secretan et al., 2009). AA is produced from ethanol via ADH. In the gastrointestinal mucosa various ADHs are present and capable to produce AA from alcohol. In addition, gastrointestinal bacteria of the upper gastrointestinal tract and of the large intestine can metabolize ethanol to AA (Salaspuro, 2003) (Figure 1).

AA is highly toxic and carcinogenic and causes point mutations in the hypoxanthineguanine phosphoribosyltransferase localized in human lymphocytes, induces sister chromatid exchanges and cross-chromosomal aberration (Seitz & Stickel, 2010). AA forms stable adducts with DNA. One of these adducts is especially generated in hyperregenerative tissues (in the presence of spermine and spermidine) such as the upper gastrointestinal tract and the colon where chronic alcohol consumption results in tissue hyperregenerativity. This propane DNA adduct is highly carcinogenic (Brooks & Thiruvathu, 2005). There is significant evidence that AA is responsible for the carcinogenic effect of alcohol in the upper gastrointestinal tract, oesophagus, larynx, pharynx and oral cavity (Baan et al., 2007; Seitz & Stickel, 2010). For more details about the role of AA on upper gastrointestinal cancer we refer to the following review article (Baan et al., 2007; Stickel et al., 2006).

Fig. 1. Ethanol metabolism by mucosal enzymes and gastrointestinal bacteria. Ethanol is first metabolized by alcohol dehydrogenase (ADH) to acetaldehyde (AA) which is toxic and carcinogenic. AA is then further converted by acetaldehyde dehydrogenase (ALDH) to acetate which is non-toxic and is channelled into the intermediary metabolism of the cell. Accumulation of AA may either occur with rapid generation through ADH or slow degradation through ALDH. ADH1B and ADH1C are polymorphic. While the ADH1B\*2 allele encodes for ADH enzymes approximately 40-fold more active as compared to the enzymes encoded by the ADH1B\*2 allele, the ADH1C\*1 allele encodes for an enzyme approximately 2.5-fold more active than the enzyme encoded by the ADH1C\*2 allele. Thus, ADH1B\*2,2 and ADH1C\*1,1 homozygotes may accumulate AA. In addition, 40 % of Asians are heterozygote for the ALDH2 gene (ALDH2\*1,2) encoding for an enzyme with very low ALDH activity resulting in AA accumulation.

In the colonic mucosa ADH1 and ADH3 are present and are involved in AA generation. On the other hand acetaldehyde dehydrogenase (ALDH), the enzyme responsible for AA

Most recently the IARC has identified AA as an important carcinogen for humans (Secretan et al., 2009). AA is produced from ethanol via ADH. In the gastrointestinal mucosa various ADHs are present and capable to produce AA from alcohol. In addition, gastrointestinal bacteria of the upper gastrointestinal tract and of the large intestine can metabolize ethanol

AA is highly toxic and carcinogenic and causes point mutations in the hypoxanthineguanine phosphoribosyltransferase localized in human lymphocytes, induces sister chromatid exchanges and cross-chromosomal aberration (Seitz & Stickel, 2010). AA forms stable adducts with DNA. One of these adducts is especially generated in hyperregenerative tissues (in the presence of spermine and spermidine) such as the upper gastrointestinal tract and the colon where chronic alcohol consumption results in tissue hyperregenerativity. This propane DNA adduct is highly carcinogenic (Brooks & Thiruvathu, 2005). There is significant evidence that AA is responsible for the carcinogenic effect of alcohol in the upper gastrointestinal tract, oesophagus, larynx, pharynx and oral cavity (Baan et al., 2007; Seitz & Stickel, 2010). For more details about the role of AA on upper gastrointestinal cancer we

Fig. 1. Ethanol metabolism by mucosal enzymes and gastrointestinal bacteria. Ethanol is first metabolized by alcohol dehydrogenase (ADH) to acetaldehyde (AA) which is toxic and carcinogenic. AA is then further converted by acetaldehyde dehydrogenase (ALDH) to acetate which is non-toxic and is channelled into the intermediary metabolism of the cell. Accumulation of AA may either occur with rapid generation through ADH or slow degradation through ALDH. ADH1B and ADH1C are polymorphic. While the ADH1B\*2 allele encodes for ADH enzymes approximately 40-fold more active as compared to the enzymes encoded by the ADH1B\*2 allele, the ADH1C\*1 allele encodes for an enzyme approximately 2.5-fold more active than the enzyme encoded by the ADH1C\*2 allele. Thus, ADH1B\*2,2 and ADH1C\*1,1 homozygotes may accumulate AA. In addition, 40 % of Asians are heterozygote for the ALDH2 gene (ALDH2\*1,2) encoding for an enzyme with very low

In the colonic mucosa ADH1 and ADH3 are present and are involved in AA generation. On the other hand acetaldehyde dehydrogenase (ALDH), the enzyme responsible for AA

refer to the following review article (Baan et al., 2007; Stickel et al., 2006).

**5. Possible mechanisms of alcohol mediated colorectal carcinogenesis** 

**5.1 Acetaldehyde (AA)** 

to AA (Salaspuro, 2003) (Figure 1).

ALDH activity resulting in AA accumulation.

degradation is detectable at relative low activities only (Seitz & Oneta, 1998). The net amount of AA present in the tissue may determine its toxic and carcinogenic action. Thus, individuals with an increased production and decreased degradation of AA are especially at risk for colorectal cancer development. It has been proposed that the ALDH activity of colonic mucosa may be sufficient for the removal of AA produced by colonic mucosal ADH during ethanol oxidation but it is insufficient for the removal of AA produced by intracolonic bacteria.

The most striking evidence of the causal role of AA in ethanol-associated carcinogenesis derives from genetic linkage studies in alcoholics and / or heavy drinkers. Individuals who accumulate AA due to polymorphism and / or mutation in the gene coding for enzymes responsible for AA generation and detoxification have been shown to have an increased cancer risk. In Japan as well as in other Asian countries a high percentage of individuals carry a mutation of the ALDH2 gene which codes for an enzyme with low activity leading to elevated AA levels after alcohol consumption. While homozygotes are completely protected against alcoholism and alcohol associated diseases due to the fact that they cannot tolerate alcohol even at very small doses, heterozygotes (ALDH2\*1,2) have an increased risk for alimentary tract cancer including the colon and the rectum (Yokoyama et al., 1998) (Figure.1).

In addition, polymorphism of the ADH1C gene may also modulate AA levels. ADH1C\*1 transcription leads to an ADH isoenzyme 2.5 times more active than that from ADH1C\*2 (Figure1). We evaluated whether the associated between alcohol consumption and colorectal cancer development is modified by ADH1C polymorphism. We recruited 173 individuals with colorectal tumours diagnosed by coloscopy and 788 control individuals without colorectal tumours and determined their genotypes. Genotype ADH1C\*1/1 was more frequent in patient with alcohol associated colorectal neoplasia compared to patients without cancer (Homann et al., 2006). In addition, only individuals drinking more than 30 grams ethanol per day with a genotype ADH1C\*1/1 had an increased risk for colorectal tumours. These data identify ADH1C homozygosity as a genetic risk marker for colorectal tumours in individuals consuming more than 30 grams alcohol per day and emphasize further the role of AA as a carcinogenic agent in alcohol mediated colorectal carcinogenesis.

It has been shown that after alcohol administration, the amount of AA per gram of tissue is highest for the colonic mucosa compared to all other tissues in the body (Seitz et al., 1987). This is primarily due to the production of AA by the faecal bacteria. AA has toxic effects on the colon mucosa resulting in secondary compensatory hyperregeneration with increased crypt cell production rates and an extension of the proliferative compartment towards the lumen of the crypt (Simanowski et al., 1994, 2001). This change in crypt cell dynamics represents a condition associating with increased risk for colorectal cancer.

The alcohol associated hyperregeneration of the colonic mucosa is especially pronounced with increasing age (Simanowski et al., 1994). As already pointed out, this may have practical implications since age by itself is a risk factor for CRC.

High AA levels have been found after alcohol administration in the colon of rats and these concentrations were significantly lower in germ free animals as compared to the conventional rats suggesting that faecal bacteria are capable of producing AA (Seitz et al., 1990). Indeed, the reversed microbial ADH reaction produces under aerobic or microaerobic conditions striking amounts of AA when human colonic contents or some microbes representing normal colonic flora are incubated in vitro at 37°C with increasing ethanol

Colorectal Cancer and Alcohol 205

There is increasing evidence that alcohol related epigenetic changes of DNA methylation and histone acetylation do occur which may potentially modulate tumour development (Stickel et al., 2006). Epidemiologic data have clearly shown that folate deficiency alone or together with methionine deficiency increases the risk for ethanol mediated colorectal cancer (Giovannucci et al., 1995; Larsson et al., 2005; Schernhammer et al., 2008; Weinstein et al., 2008; Yamaji et al., 2009; Figueiredo et al., 2009; Lee et al., 2010). Similarly, vitamin B6 deficiency also enhances tumour risk. All these factors are involved in methyl transfer. Their deficiency results in DNA hypomethylation, a condition relevant in carcinogenesis. In addition, histone acetylation is also favoured by chronic ethanol consumption (Kim & Shukla, 2006; Choudhury & Shukla, 2008) since ethanol metabolism leads to the accumulation of acetate on one hand and to a change in the intracellular redox potential with increasing concentrations of NADH and decreasing concentrations of NAD on the other hand. This change in redox potential also favours histone acetylation. In addition, histone deacetylation is blocked by ethanol through its inhibitory effect on histone

Indeed, animal experiments have shown DNA hypomethylation in the colon following chronic ethanol ingestion (Choi et al., 1999). However, site specific hypomethylation have

Other mechanisms of ethanol mediated carcinogenesis in the colon may including deficiency of retinoic acid (Wang X.D. & Seitz, 2004), effect of ethanol on intracellular signalling (Dunty, 2010) including various pathways such as Mitogenic signals (MAPK, RAS), Insensitivity to anti-growth signals (Rb and Cell cycle control, TGFβ), Apoptosis (p53, PTEN), Angiogenesis, Metastasis (ECM, Osteopontin, Wnt) and an ethanol effect on

Chronic ethanol consumption at a dose of more than 30 grams ethanol per day is a risk factor for colorectal cancer. Chronic ethanol consumption also increases the risk for colorectal cancer in individuals with polyps and colorectal inflammation as well as in those with an ALDH mutation (ALDH2\* 1,2, only Asians) and ADH1C\*1 homozygosity since they

The mechanisms of ethanol mediated colorectal carcinogenesis may involve AA produced by mucosal ADH and intestinal bacteria. In addition, oxidative stress induced by ethanol

As a clinical consequence, chronic alcohol consumers should be screened earlier as the general population for colorectal cancer either by faecal blood test or colonoscopy

The authors wish to thank Mrs. Heike Grönebaum for typing the manuscript. Original research was supported by grants of the Manfred Lautenschläger- and Dietmar Hopp

**5.3 Epigenetics** 

deacetylase (HDA).

not been demonstrated so far.

inflammation (Wang, J., 2010).

may also play a role.

**7. Acknowledgements** 

Foundation.

**6. Summary, conclusion and recommendations** 

accumulate AA following ethanol ingestion.

depending on the methods available.

**5.4 Other mechanisms** 

concentrations (Jokelainen et al., 1996, 1994; Salaspuro et al., 1999). This reaction is already active at comparatively low ethanol concentrations (10-100 mg %) which exist in the colon following social drinking (Jokelainen et al., 1996). AA formation catalysed by microbial ADH takes place at a pH normally found in the colon and is rapidly reduced with decreasing pH (Jokelainen et al., 1994).

The administration of antibiotics to animals has significantly decreased colonic bacteria and colonic AA production (Jokelainen et al., 1997).


#### **5.2 Oxidative stress**

Chronic ethanol consumption results in the induction of cytochrome P4502E1 (CYP2E1) predominantly in the liver (Seitz & Stickel, 2010) but also in other tissues including the colorectal mucosa (Shimizu et al., 1990). This CYP2E1 induction is associated with an increased metabolism of ethanol through CYP2E1 and the generation of AA but also of reactive oxygen species (ROS). For more details we refer to the following review article (Seitz & Stickel, 2007). ROS may attack lipids and result in lipidperoxidation with the generation of lipidperoxidation products such as 4-hydroxy-nonenal or malondialdehyde. These lipidperoxidation products may bind to proteins but also to DNA and form exocyclic etheno DNA adducts with a high carcinogenic potency (Wang et al., 2009).

The effect of chronic ethanol consumption on the induction of CYP2E1 and the activation of procarcinogens has been convincingly demonstrated by the use of azoxymethane (AOM), a procarcinogen for the colon. The metabolism of AOM to its ultimate carcinogen has been inhibited in the presence of ethanol in the body since ethanol competes for the binding site at CYP2E1 but significantly enhanced when ethanol is withdrawn in a condition where CYP2E1 is induced and completely available for the metabolism of AOM (Sohn et al., 1987).

The induction of CYP2E1 in the colonic mucosa may lead to an enhanced activation of dietary (nitrosamines) and cigarette smoke derived (polycyclic hydrocarbons) procarcinogens and may be one mechanism by which ethanol enhances colorectal carcinogenesis (Seitz & Osswald, 1994).

In this context it is interesting that alpha tocopherol, a radical scavenger, prevents mucosal cell hyperproliferation induced by ethanol suggesting that ROS may be involved in this process (Vincon et al., 2003).

#### **5.3 Epigenetics**

204 Colorectal Cancer – From Prevention to Patient Care

concentrations (Jokelainen et al., 1996, 1994; Salaspuro et al., 1999). This reaction is already active at comparatively low ethanol concentrations (10-100 mg %) which exist in the colon following social drinking (Jokelainen et al., 1996). AA formation catalysed by microbial ADH takes place at a pH normally found in the colon and is rapidly reduced with

The administration of antibiotics to animals has significantly decreased colonic bacteria and

c. crypt cell production rate correlates significantly with AA levels in the colonic mucosa, (d) colonic AA levels show a significant inverse relationship with mucosal folate concentration which supports in vitro data showing a destruction of folate by AA, d. individuals with the inactive form of ALDH2 resulting in elevated AA concentrations

e. individuals homozygous for the ADH1C\*1 allele coding for an enzyme with a 2.5 times higher AA production have also an increased risk for colorectal cancer, the action of AA seems the major mechanism of ethanol-mediated colorectal cancer

Chronic ethanol consumption results in the induction of cytochrome P4502E1 (CYP2E1) predominantly in the liver (Seitz & Stickel, 2010) but also in other tissues including the colorectal mucosa (Shimizu et al., 1990). This CYP2E1 induction is associated with an increased metabolism of ethanol through CYP2E1 and the generation of AA but also of reactive oxygen species (ROS). For more details we refer to the following review article (Seitz & Stickel, 2007). ROS may attack lipids and result in lipidperoxidation with the generation of lipidperoxidation products such as 4-hydroxy-nonenal or malondialdehyde. These lipidperoxidation products may bind to proteins but also to DNA and form exocyclic

The effect of chronic ethanol consumption on the induction of CYP2E1 and the activation of procarcinogens has been convincingly demonstrated by the use of azoxymethane (AOM), a procarcinogen for the colon. The metabolism of AOM to its ultimate carcinogen has been inhibited in the presence of ethanol in the body since ethanol competes for the binding site at CYP2E1 but significantly enhanced when ethanol is withdrawn in a condition where CYP2E1 is induced and completely available for the metabolism of AOM

The induction of CYP2E1 in the colonic mucosa may lead to an enhanced activation of dietary (nitrosamines) and cigarette smoke derived (polycyclic hydrocarbons) procarcinogens and may be one mechanism by which ethanol enhances colorectal

In this context it is interesting that alpha tocopherol, a radical scavenger, prevents mucosal cell hyperproliferation induced by ethanol suggesting that ROS may be involved in this

exhibit an increased risk for CRC when they consume alcohol,

etheno DNA adducts with a high carcinogenic potency (Wang et al., 2009).

a. high AA levels occur in the colon due to bacterial and mucosal ethanol oxidation b. animal experiments show an increased occurrence of colorectal tumours induced by the specific locally acting carcinogen AMMN when cyanamide, an ALDH inhibitor, is

decreasing pH (Jokelainen et al., 1994).

colonic AA production (Jokelainen et al., 1997).

applied and AA levels are elevated,

development.

**5.2 Oxidative stress** 

(Sohn et al., 1987).

carcinogenesis (Seitz & Osswald, 1994).

process (Vincon et al., 2003).

There is increasing evidence that alcohol related epigenetic changes of DNA methylation and histone acetylation do occur which may potentially modulate tumour development (Stickel et al., 2006). Epidemiologic data have clearly shown that folate deficiency alone or together with methionine deficiency increases the risk for ethanol mediated colorectal cancer (Giovannucci et al., 1995; Larsson et al., 2005; Schernhammer et al., 2008; Weinstein et al., 2008; Yamaji et al., 2009; Figueiredo et al., 2009; Lee et al., 2010). Similarly, vitamin B6 deficiency also enhances tumour risk. All these factors are involved in methyl transfer. Their deficiency results in DNA hypomethylation, a condition relevant in carcinogenesis. In addition, histone acetylation is also favoured by chronic ethanol consumption (Kim & Shukla, 2006; Choudhury & Shukla, 2008) since ethanol metabolism leads to the accumulation of acetate on one hand and to a change in the intracellular redox potential with increasing concentrations of NADH and decreasing concentrations of NAD on the other hand. This change in redox potential also favours histone acetylation. In addition, histone deacetylation is blocked by ethanol through its inhibitory effect on histone deacetylase (HDA).

Indeed, animal experiments have shown DNA hypomethylation in the colon following chronic ethanol ingestion (Choi et al., 1999). However, site specific hypomethylation have not been demonstrated so far.

#### **5.4 Other mechanisms**

Other mechanisms of ethanol mediated carcinogenesis in the colon may including deficiency of retinoic acid (Wang X.D. & Seitz, 2004), effect of ethanol on intracellular signalling (Dunty, 2010) including various pathways such as Mitogenic signals (MAPK, RAS), Insensitivity to anti-growth signals (Rb and Cell cycle control, TGFβ), Apoptosis (p53, PTEN), Angiogenesis, Metastasis (ECM, Osteopontin, Wnt) and an ethanol effect on inflammation (Wang, J., 2010).

#### **6. Summary, conclusion and recommendations**

Chronic ethanol consumption at a dose of more than 30 grams ethanol per day is a risk factor for colorectal cancer. Chronic ethanol consumption also increases the risk for colorectal cancer in individuals with polyps and colorectal inflammation as well as in those with an ALDH mutation (ALDH2\* 1,2, only Asians) and ADH1C\*1 homozygosity since they accumulate AA following ethanol ingestion.

The mechanisms of ethanol mediated colorectal carcinogenesis may involve AA produced by mucosal ADH and intestinal bacteria. In addition, oxidative stress induced by ethanol may also play a role.

As a clinical consequence, chronic alcohol consumers should be screened earlier as the general population for colorectal cancer either by faecal blood test or colonoscopy depending on the methods available.

#### **7. Acknowledgements**

The authors wish to thank Mrs. Heike Grönebaum for typing the manuscript. Original research was supported by grants of the Manfred Lautenschläger- and Dietmar Hopp Foundation.

Colorectal Cancer and Alcohol 207

Giovannucci, E.; Rimm, E.B.; Ascherio, A.; Stampfer, M.J.; Colditz, G.A. & Willett, W.C.

Homann N; Stickel F; König IR; Jacobs A; Junghanns K; Benesova M; Schuppan D; Himsel S;

Jokelainen, K.; Roine, R.; Väänänen, H.; Salaspuro, M. (1994). In vitro acetaldehyde

Jokelainen, K.; Siitonen, A.; Jousimies-Somer, H. (1996). In vitro alcohol dehydrogenase-

Jokelainen, K.; Matysiak-Budnik, T.; Mäkisalo, H.; Höckerstedt, K.; Salaspuro, M. (1996).

Jokelainen K.; Nosova T.; Koivisto T.; Väkeväinen S.; Jousimies-Somer H.; Heine R.;

Kearney, J.; Giovannucci, E.; Rimm, E.B.; Stampfer, M.J.; Colditz, G.A.; Ascherio, A.; Bleday,

Kim, J.S. & Shukla, S.D. (2006) Acute in vivo effect of ethanol (binge drinking) on histone H3

Kune, G.A. & Vitetta, L. (1992). Alcohol consumption and the etiology of colorectal cancer: a

Larsson, S.C.; Giovannucci, E. & Wolk, A. (2005). Vitamin B6 intake, alcohol consumption,

Lee, J.E.; Giovannucci, E.; Fuchs, C.S.; Willett, W.C.; Zeisel, S.H. & Cho, E. (2010). Choline

Le Marchand, L.; Wilkens, L.R.; Hankin, J.H.; Kolonel, L.N. & Lyu, L.C. (1999). Independent

Park, J.Y.; Mitrou, P.N.; Dahm, C.C.; Luben, R.N.; Wareham, N.J.; Khaw, K.T. & Rodwell,

Nutrition-Norfolk study. *Cancer Epidemiol* Vol. 33(5): pp. 347-354.

modifications in rat tissues. *Alcohol Alcohol*. Vol. 41: pp. 126-32.

formation by human colonic bacteria. *Gut* Vol. 35: pp. 1271-1274.

colonic flora in man. *Alcohol Clin Exp Res* Vol. 20: pp. 967-972.

18(4): pp. 1041-1049.

Vol. 6: pp. 45-56.

*J Natl Cancer Inst* Vol. 87: pp. 265-273.

heavy drinkers. *Int J Cancer* Vol. 118(8):pp. 1998-2002.

ethanol oxidation in piglets. *Gut* Vol. 39: pp. 100-104.

ciprofloxacin in rats. *Life Sci* Vol. 61: pp. 1755-1762.

*Gastroenterology* Vol. 128: pp. 1830-1837.

*Biomarkers Prev* Vol. 19(3): pp. 884-887.

*Am J Epidemiol* Vol. 167: pp. 1397-1406.

characteristics and dietary and genetic factors. *Cancer Epidemiol Biomarkers Prev* Vol.

(1995). Alcohol, low-methionine – low folate diets, and risk of colon cancer in men.

Zuber-Jerger I; Hellerbrand C; Ludwig D; Caselmann WH; Seitz HK (2006). Alcohol dehydrogenase 1C\*1 allele is a genetic marker for alcohol-associated cancer in

mediated acetaldehyde production by aerobic bacteria representing the normal

High intracolonic acetaldehyde values produced by a bacteriocolonic pathway for

Salaspuro M. (1997). Inhibition of bacteriocolonic pathway for ethanol oxidation by

R. & Willett, W.C (1995). Diet, alcohol, and smoking and the occurrence of hyperplastic polyps of the colon and rectum (United States). *Cancer Causes Control*

review of the scientific evidence from 1957 to 1991. *Nutr Cancer* Vol. 18: pp. 97-111.

and colorectal cancer: a longitudinal population-based cohort of women.

and betaine intake and the risk of colorectal cancer in men. *Cancer Epidemiol* 

and joint effects of family history and lifestyle on colorectal cancer risk: implications for prevention. *Cancer Epidemiol Biomarkers Prev* Vol. 8: pp. 45-851. Mizoue, T.; Inoue, M.; Wakai, K.; Nagata, C.; Shimazu, T.; Tsuji, I.; Otani, T.; Tanaka, K.;

Matsuo, K.; Tamakoshi, A.; Sasazuki, S. & Tsugane S. (2008). Alcohol drinking and colorectal cancer in Japanese: a pooled analysis of results from five cohort studies.

S.A. (2009). Baseline alcohol consumption, type of alcoholic beverage and risk of colorectal cancer in the European Prospective Investigation into Cancer and

#### **8. References**


Baan, R.; Straif, K.; Grosse, Y.; Secretan, B.; El Ghissassi, F.; Bouvard, V.; et al. (2007) Carcinogenicity of alcoholic beverages. *Lancet Oncol.* Vol. 8: pp. 292-3 Bardou, M.; Montembault, S.; Giraud V.; Balian A.; Borotto E.; Houdayer C.; Capron F.;

Bongaerts, B.W.; van den Brandt, P.A.; Goldbohm, R.A.; de Goeij, A.F. & Weijenberg, M.P.

Bongaerts, B.W.; de Goeij, A.F.; Wouters, K.A.; van Engeland, M.; Gottschalk, R.W.; Van

Brooks, P.J. & Thiruvathu, J.A. (2005). DNA adducts from acetaldehyde: implications for

Cho, E.; Smith-Warner, S.A.; Ritz, J.; van den Brandt, P.A.; Colditz, G.A.; Folsom, A.R.;

Choudhury, M. & Shukla, S.D. (2008). Surrogate alcohols and their metabolites modify

DeBruijn, E.A.; Slee, P.H.T.J. (1992) Alcohol effects on anticancer drug metabolism. In:

Dunty B. (2010) Alcohol, cancer genes, and signaling pathways. NIAAA Extramural Advisory Board 2010, *Alcohol and Cancer*, NIH June 8-9, 2010, pp. 69-93. Edenberg, H.J. (2007). Role of alcohol dehydrogenase and aldehyde dehydrogenase variants.

Fedirko, V.; Tramacere, I.; Bagnardi, V.; Rota, M.; Scotti, L.; Islami, F.; Negri, E.; Straif, K.;

Figueiredo, J.C.; Grau, M.V.; Wallace, K.; Levine, A.J.; Shen, L.; Hamdan, R.; Chen, X.;

alcohol-related carcinogenesis. *Alcohol* Vol. 35: pp. 187-193.

deacetylase. *Alcohol Clin Exp Res*. Vol. 32: pp. 829-39.

cancer at specific subsites. *Int J Cancer* Vol. 123(10): pp. 2411-2417.

control study. *Gut* Vol. 50: pp. 38-42.

colon. *J Nutr*. Vol. 129: pp. 1945-50.

Alcohol Res Health Vol. 30: pp. 5-37.

studies. *Ann Oncol* (Epub ahead of print)

Tokyo, pp. 135-150.

Chaput J.C. & Naveau S (2002). Excessive alcohol consumption favours high risk polyp or colorectal cancer occurrence among patients with adenomas: a case

(2008). Alcohol consumption, type of alcoholic beverage and risk of colorectal

Schooten, F.J., et al.(2010) Alcohol consumption, alcohol dehydrogenase 1C (ADH1C) genotype, and risk of colorectal cancer in the Netherlands Cohort Study on diet and cancer. Alcohol (Epub ahead of printing) URL: *www.alcoholjournal.org* Boutron, M.C.; Faivre, J.; Dop, M.C.; Quipourt, V. & Senesse, P. (1995). Tobacco, alcohol, and colorectal tumors: a multistep process. *Am J Epidemiol* Vol. 141: pp. 1038-1046. Breslow, N.E. & Enstrom, J.E. (1974). Geographic correlations between mortality rates and

alcohol, tobacco consumption in the United States. *J Natl Cancer Inst* Vol. 53: pp.

Freudenheim, J.L.; Giovannucci, E.; Goldbohm, R.A.; Graham, S.; Holmberg, L.; Kim, D.H.; Malila, N.; Miller, A.B.; Pietinen, P.; Rohan, T.E.; Sellers; T.A., Speizer, F.E.; Willett, W.C.; Wolk, A. & Hunter, D.J. (2004). Alcohol intake and colorectal cancer: a pooled analysis of 8 cohort studies. *Ann Intern Med* Vol. 140: pp. 603-613. Choi, S.W.; Stickel, F.; Baik, H.W.; Kim, Y.I.; Seitz, H.K. & Mason, J.B. (1999). Chronic alcohol

consumption induces genomic but not p53-specific DNA hypomethylation in rat

histone H3 acetylation: involvement of histone acetyl transferase and histone

*Alcohol and Cancer* (ed. R.R. Watson), CRC Press Boca Raton, Ann Arbor, London,

Romieu, I.; La Vecchia, C. ; Boffetta, P. & Jenab, M. (2011). Alcohol drinking and colorectal cancer risk: an overall and dose-response meta-analysis of published

Bresalier, R.S.; McKeown-Eyssen, G.; Haile, R.W.; Baron, J.A. & Issa, J.P.J. (2009). Global DNA hypomethylation (LINE-1) in the normal colon and lifestyle

**8. References** 

631-639.

characteristics and dietary and genetic factors. *Cancer Epidemiol Biomarkers Prev* Vol. 18(4): pp. 1041-1049.


Colorectal Cancer and Alcohol 209

Seitz, H.K.; Simanowski, U.A.; Garzon, F.T.; Rideout, J.M.; Peters, T.J.; Koch, A.; Berger,

Shimizu, N.; Nagata, C.; Shimizu, H.; Kametani, M.; Takeyama, N.; Ohnuma, T. &

Simanowski, U.A.; Suter, P.; Russel, R.M.; Heller, M.; Waldherr, R.; Ward, R.; Peters, T.J.;

Sohn, O.A.; Fiala, E.S. & Puz, C. (1987). Enhancement of rat liver microsomal metabolism of

Stickel, F.; Herold, C. & Seitz, H.K. (2006). Alcohol and Methyl transfer: Implications for

Su, L.J. & Arab, L. (2004). Alcohol consumption and risk of colon cancer: evidence from the

Thygesen, L.C.; Wu, K.; Gronbaek, M.; Fuchs C.S.; Willett, W.C. & Giovannucci, E. (2008).

Wang J. (2010) Alcohol and inflammation in cancer development. NIAAA Extramural Advisory Board 2010, Alcohol and Cancer, NIH June 8-9,2010, pp. 134-148. Wang, Y.; Millonig, G.; Nair, J.; Patsenker, E.; Stickel, F. & Mueller, S. (2009) Ethanol-

Wang, X.D. & Seitz, H.K. (2004) Alcohol and Retinoid Interaction. In: *Nutrition and Alcohol:* 

Weinstein, S.J.; Albanes, D.; Selhub, J.; Graubard, B.; Lim, U.; Taylor, P.R.; Virtamo, J. &

Yamaji, T.; Iwasaki, M.; Sasazuki, S.; Sakamoto, H.; Yoshida, T. & Tsugane, S. (2009).

313-321, CRC Press Boca Raton, London, New York, Washington.

hyperregeneration with age in F244 rats. *Gut* Vol. 35: pp. 1102-1106. Simanowski, U.A.; Homann, N.; Knühl, M.; Arce, C.; Waldherr, R.; Conradt, C.; Bosch, F.X.&

*Gastroenterology* Vol. 93: pp. 1044-1050.

Alcohol-related Hepatocarcinogenesis

liver disease. *Hepatology*. Vol. 50: pp. 453-61.

*Nutr Cancer* Vol. 50: pp. 111-119.

1043.

Vol. 49: pp. 418-422.

47: pp. 3123-3129.

27: pp. 100-106.

3240.

M.R.; Einecke, H. & Maiwald, M. (1990). Possible role of acetaldehyde in ethanolrelated rectal cocarcinogenesis in the rat. *Gastroenterology* Vol. 98: pp. 406-413. Shimizu, M.; Lasker, J.M.; Tsutsumi, M.; Lieber, C.S. (1990). Immunohistochemical

localization of ethanol inducible cytochrome P4502E1 in rat alimentary tract.

Matsushita, S. (2003). Height, weight, and alcohol consumption in relation to the risk of colorectal cancer in Japan: a prospective study. *Br J Cancer* Vol. 88: pp. 1038-

Smith, D. & Seitz, H.K. (1994). Enhancement of ethanol induced rectal mucosal

Seitz, H.K. (2001). Increased rectal cell proliferation following alcohol abuse. *Gut*

azoxymethane to methylazoxymethanol by chronic ethanol administration: similarity to the microsomal metabolism of N-nitrosomethylamine. *Cancer Res* Vol.

national health and nutrition examination survey I epidemiologic follow-up study.

Alcohol intake and colorectal cancer: a comparison of approaches for including repeated measures of alcohol consumption. *Epidemiology* Vol. 19(2): pp. 258-264. Vincon, P.; Wunderer, J.; Simanowski, U.A.; Koll, M.; Preedy, V.R.; Peters,T.J.; Werner, J.;

Waldherr, R.; Seitz, H.K. (2003). Effect of ethanol and vitamin E on cell regeneration and BCL-2 expression in the colorectal mucosa of rats. *Alcoholism Clin Exp Res* Vol.

induced cytochrome P4502E1 causes carcinogenic etheno-DNA lesions in alcoholic

*Linking nutrient interactions and dietary intake*. Watson RR, Preedy, V.R. (Ed(s).) pp.

Stolzenberg-Solomon, R. (2008). One-carbon metabolism Biomarkers and risk of colon and rectal cancers. Cancer *Epidemiol Biomarkers Prev* Vol. 17(11): pp. 3233-

Methionine synthase A2756G polymorphism interacts with alcohol and folate


Park, J.Y.; Dahm, C.C.; Keogh, R.H.; Mitrou, P.N.; Cairns B.J.; Greenwood, D.C.; Spencer,

Pedersen, A., Johansen, C. & Gronbaek, M. (2003). Relations between amount and type of

Rehm, J.; Room, R.; Monteiro, R.; Gmel, G.; Graham, K.; Rehn, T. (2004) Global and Regional

Salaspuro, V.; Nyfors, S.; Heine, R.; Siitonen, A.; Salaspuro, M.; Jousimies-Somer, H. (1999).

Salaspuro, M.; (2003). Acetaldehyde, microbes, and cancer of the digestive tract. *Critical* 

Schernhammer, E.S.; Ogino, S. & Fuchs, C.S. (2008). Folate and vitamin B6 intake and risk of

Scheppach W.; Bingham, S.; Boutron-Ruault, M.C.; Gerhardsson de Verdier, M.; Moreno, V.;

Secretan B., Streif K., Baan R., Grosse Y., El Ghissassi F., Bouvard V., Benbrahin-Tallaa L,

Seitz, H.K. & Osswald, B. (1992) Effect of ethanol on procarcinogen activation. In: *Alcohol and* 

Seitz, H.K. & Stickel, F. (2007) Molecular mechanisms of alcohol-mediated carcinogenesis.

Seitz, H.K. & Stickel, F. (2010) Acetaldehyde as an underestimated risk factor for cancer

Seitz, H.K., Cho, C.H. (2009) Contribution of Alcohol and Tobacco Use in Gastrointestinal

Seitz, H.K.; Pöschl, G.; Salaspuro, M. (2006) Alcohol and Cancer of the large intestine. In: *Alcohol Tobacco and Cancer*, Cho, C.H.; Purohit, V. pp. 63-77, Karger, Basel Seitz, H.K.; Simanowski, U.A.; Garzon, F.T.; Peters, T.J. (1987). Alcohol and cancer (Letter to

Seitz, H.K.; Oneta, C.M. (1998). Gastrointestinal alcohol dehydrogenases. *Nutr Rev* Vol. 56:

Seitz, H.K.; Pöschl, G. & Simanowski, U.A. (1998). Alcohol and cancer. *Recent Dev Alcohol*

UK Dietary Cohort Consortium. *Br J Cancer* Vol. 24: pp.

Murray, C.; (Ed(s).), World Health Organisation , Geneva.

*reviews in Clinical Laboratory Science* Vol. 40: pp. 183-208.

*Cancer* ,Watson, R. pp. 55-72, CRC Press, Boca Raton:;

Vol. 52: pp. 861-867.

770-780.

Vol. 8: pp. 57-62.

10, pp. 1033-34.

pp. 52-60.

Vol. 14: pp. 67-95.

*Gastroenterol* Vol. 34: pp. 967-973.

*Nat Rev Cancer* Vol. 7: pp. 599-612.

pp. 217-241, Humana Press, New York

the Editor). *Hepatology* Vol. 7: pp. 616.

E.A.; Fentiman, I.S.; Shipley, M.J.; Brunner, E.J.; Cade, J.E.; Burley, V.J.; Mishra, G.D.; Kuh, D.; Stephen, A.M.; White, I.R.; Luben, R.N.; Mulligan, A.A,; Khaw, K.T. & Rodwell, S.A. (2010). Alcohol intake and risk of colorectal cancer: results from the

alcohol and colon and rectal cancer in a Danish population based cohort study. *Gut*

Burden of Disease Attributable to Selected Major Risk Factors. In: *Comparative Quantification of Health Risks* Ezatti, M.; Murray, C.; Lopez, AD.; Rodgers, A.;

Ethanol oxidation and acetaldehyde production in vitro by human intestinal strains of Eschericia coli under aerobic, microaerobic; and anaerobic conditions. *Scand J* 

colon cancer in relation to p53 mutational status. *Gastroenterology* Vol. 135(3): pp

Nagengast, F.M.; Reifen, R.; Riboli, E.; Seitz, H.K. & Wahrendorf, J. (1999). WHO consensus statement on the role of nutrition in colorectal cancer. *Eur J Cancer Prev*

Guha N., Freeman C., Galichet L., et al. (2009) A review of human carcinogens – Part E: tobacco, areca nut, alcohol, coal smoke, and salted fish. Lancet Oncol. Vol.

development: role of genetics in ethanol metabolism. *Genes Nutr.* Vol. 5: pp. 121-128

Cancer Development . In: *Cancer Epidemiology, Vol. 2, Modifiable Factors*, Verma, M.


**11** 

**Effects of Dietary Counseling on** 

**Patients with Colorectal Cancer**

Renata Dobrila-Dintinjana1, Dragan Trivanović2,

*2General Hospital Pula, Department of Internal Medicine,* 

*University of Rijeka, School of Medicine,* 

*Clinical Hospital Center Rijeka,* 

*5Bosnia and Herzegovina* 

*1,2,3,4Croatia* 

*3Clinic of General Medicine Dr Dintinjana,* 

*Faculty of Medicine, University of Sarajevo,* 

Marijan Dintinjana3, Jelena Vukelic4 and Nenad Vanis5 *1Department of Radiation Oncology, Clinical Hospital Center Rijeka,* 

*4Department of Speech and Hearing Disorders Diagnostics and Rehabilitation,* 

*5Division for Gastroenterology and Hepatology, Clinical Center University of Sarajevo,* 

Cancers of the colon and rectum together are second most common tumor type worldwide. The prognosis for the survival after disease progression is usually poor (1). Cancer anorexiacachexia syndrome is highly prevalent among patients with colorectal cancer, and has a large impact on morbidity and mortality, and on patient quality of life. Early intervention with nutritional supplementation has been shown to halt malnutrition, and may improve

The etiology of cancer-associated malnutrition appears to be related to the pathological loss of inhibitory control of catabolic pathways, whose increased activities are not

The goals of nutritional support in patients with colorectal cancer are to improve nutritional status to allow initiation and completion of active anticancer therapies (chemotherapy and

Cancer growth and dissemination but also cancer treatments, including surgery, chemotherapy, and radiation therapy, interfere with taste, ingestion, swallowing, and digest food which leads to hypophagia. Also, chemotherapy agents may cause nausea and diarrhea (3, 4). Although many new agents are on the market to combat these symptoms,

We studied the influence of nutritional support (counseling, nutritional supplements, megestrol acetate) on physical status and symptoms in patients with colorectal cancer during chemotherapy. The study was designed to investigate whether dietary counseling or oral nutrition commercial supplements during chemotherapy and/or BSC affected nutritional status and influence survival status prevalence in patients with colorectal cancer.

counterbalanced by the increased central and peripheral anabolic drive (3).

**1. Introduction** 

outcome in some patients (2).

or radiotherapy) and improve quality of life (3, 4).

prevalence of colorectal cancer is still high (1).

intake to influence the risk of colorectal adenoma. *Cancer Epidemiol Biomarkers Prev* Vol. 18(1): pp. 267-274.

Yokoyama, A.; Muramatsu, T.; Ohmori, T.; Yokoyama, T.; Okuyama, K.; Takahashi, H.; Hasegawa, Y.; Higuchi, S.; Maruyama, K.; Shirakura, K.; Ishii, H. (1998). Alcoholrelated cancers and aldehydrogenase-2 in Japanese alcoholics. *Carcinogenesis* Vol. 19: pp. 1383-1387.

### **Effects of Dietary Counseling on Patients with Colorectal Cancer**

Renata Dobrila-Dintinjana1, Dragan Trivanović2, Marijan Dintinjana3, Jelena Vukelic4 and Nenad Vanis5 *1Department of Radiation Oncology, Clinical Hospital Center Rijeka, University of Rijeka, School of Medicine, 2General Hospital Pula, Department of Internal Medicine, 3Clinic of General Medicine Dr Dintinjana, 4Department of Speech and Hearing Disorders Diagnostics and Rehabilitation, Clinical Hospital Center Rijeka, 5Division for Gastroenterology and Hepatology, Clinical Center University of Sarajevo, Faculty of Medicine, University of Sarajevo, 1,2,3,4Croatia* 

*5Bosnia and Herzegovina* 

#### **1. Introduction**

210 Colorectal Cancer – From Prevention to Patient Care

Yokoyama, A.; Muramatsu, T.; Ohmori, T.; Yokoyama, T.; Okuyama, K.; Takahashi, H.;

Vol. 18(1): pp. 267-274.

19: pp. 1383-1387.

intake to influence the risk of colorectal adenoma. *Cancer Epidemiol Biomarkers Prev*

Hasegawa, Y.; Higuchi, S.; Maruyama, K.; Shirakura, K.; Ishii, H. (1998). Alcoholrelated cancers and aldehydrogenase-2 in Japanese alcoholics. *Carcinogenesis* Vol.

> Cancers of the colon and rectum together are second most common tumor type worldwide. The prognosis for the survival after disease progression is usually poor (1). Cancer anorexiacachexia syndrome is highly prevalent among patients with colorectal cancer, and has a large impact on morbidity and mortality, and on patient quality of life. Early intervention with nutritional supplementation has been shown to halt malnutrition, and may improve outcome in some patients (2).

> The etiology of cancer-associated malnutrition appears to be related to the pathological loss of inhibitory control of catabolic pathways, whose increased activities are not counterbalanced by the increased central and peripheral anabolic drive (3).

> The goals of nutritional support in patients with colorectal cancer are to improve nutritional status to allow initiation and completion of active anticancer therapies (chemotherapy and or radiotherapy) and improve quality of life (3, 4).

> Cancer growth and dissemination but also cancer treatments, including surgery, chemotherapy, and radiation therapy, interfere with taste, ingestion, swallowing, and digest food which leads to hypophagia. Also, chemotherapy agents may cause nausea and diarrhea (3, 4). Although many new agents are on the market to combat these symptoms, prevalence of colorectal cancer is still high (1).

> We studied the influence of nutritional support (counseling, nutritional supplements, megestrol acetate) on physical status and symptoms in patients with colorectal cancer during chemotherapy. The study was designed to investigate whether dietary counseling or oral nutrition commercial supplements during chemotherapy and/or BSC affected nutritional status and influence survival status prevalence in patients with colorectal cancer.

Effects of Dietary Counseling on Patients with Colorectal Cancer 213

polyposis, hereditary nonpolyposis colorectal cancer, Peutz-Jeghers disease, juvenile

Ulcerative colitis, among other diseases in the medical history, is the top risk factor. The longer the disease and the segment of the colon affected, the higher the risk. The risk is also increased in individuals with Crohn's disease. The patient undergoing surgery for colorectal

The inheritance determines individual susceptibility to sporadic cancer but the lifestyle and environmental exposures are necessary for cancer expression. Colorectal cancer incidence varies between different geographic regions and incidence and mortality rates have been highest in developed western nations (10, 11). The basic argument that environment plays a huge role in colorectal cancer expression we get from observational studies in migrant populations. Migrants from low-incident regions to the high-incident regions of North America within one generation accept the incidence of the host country. Yet, studies with migrants also suggest that geographic variation in colorectal cancer incidences is due to environmental exposures and not due to the inherent predisposition (racial and ethnic

Population based investigations have found many dietary and other environmental factors associated with colorectal cancer incidence (2). Most of these studies have methodologic limitations and therefore the interpretation of such studies has to be made with caution. Many studies have been conducted to investigate external factors that may increase or diminish the incidence of colorectal cancer. Many authors recognize four risk factor categories: epidemiological, intestinal, dietetic and mixed. The most frequently reported factors, among these shown to increase the risk of developing the disease, are diets rich in meat and animal fats (bile salts), physical inactivity, smoking and alcohol consumption (1, 2, 12). Between other diets ingested, consumption of red meat has the strongest correlation with colorectal cancer; over 30 case control studies report increased risk of colorectal cancer with higher red meat intake. Especially fried, barbecued and well-done meat is associated with colorectal cancer risk. Obesity and high caloric intake are the independent risk factors for colorectal cancer, excessive body mass gives a two fold increase in colorectal cancer, and this association in more expressed in men than women (12). Although studies carried out in humans and animal models have shown a positive correlation between the saturated fats/red meat consumption and the development of colorectal cancer, only a few of them are confirmed to be statistically significant. The total amount of fat in terms of daily caloric intake (>40%) and their form appears to have special significance. The conversion of dietary phospholipids to diacylglycerol by intestinal bacteria is assumed to be a potential mechanism of carcinogenesis. Diacylglycerol can enter the epithelial cell directly and by stimulating protein kinase C, it evokes intracellular signal transduction or mucosal proliferation. Another important mechanism is the formation of free radicals during fat metabolism and the mucosal damage induced by the secondary bile acids (lithocholic acid). Nitroso compounds, heat-generated heterocyclic amines and high protein intake (accelerated epithelial proliferation) have potential carcinogenic consequencies (12, 13). Among the factors mentioned to reduce the risk are diet high in plant fibers and calcium, antioxidants (vitamin E, selenium etc.), menopausal hormone replacement therapy and administration of nonsteroidal anti-inflammatory drugs (12). In 1990 more than 13 studies showed significant reductions in colorectal cancer risk comparing the group with higher vs group with lower fibre intake. Some of the potential benefit mechanisms are: increased stool

polyposis, Cowden disease are rare conditions (6, 7).

cancer has three times the risk of cancer recurrence (9).

group) (1, 10, 11).

**Results:** Three hundred and eighty-eight colorectal cancer patients were included in the study. Nottingham Screening Tool Questionnaire, Appetite Loss Scale and Karnofsky Performance Status were taken to evaluate the nutritive status of patients. Group I consisted of 215 patients who were monitored prospectively and were given nutritional support and in this group weight gain of 1,5 kg (0,6-2,8 kg) and appetite improvement was observed in patients with colorectal cancer. In both groups Karnofsky Performance Status didn't change significantly reflecting the impact of the disease itself.

Nutritional counseling, supplemental feeding and pharmacological support do temporarily stop weight loss and improve appetite, QoL and social life, but this improvement has no implications on patients KPS and course of their disease.

**Conclusion**: These results encourage further studies with more specific nutritional supplementation in patients with colorectal cancer and probably in gastrointestinal oncology.

#### **2. Colorectal cancer**

The incidence and mortality rates for cancers of the colon and rectum are among the highest of all malignancies worldwide (1, 2). Colorectal cancer is second in global cancer incidence and it is the most common cause of cancer death among non-smokers. US and EU incidence figures exceed global averages, which is consistent with an increased risk in industrialized nations (2). Factors associated with increased risk of colorectal cancer are host susceptibility and a sequence of different carcinogenic exposures. Specific etiology for sporadic colorectal cancer is still elusive but predisposing hereditary and environmental factors have been clearly identified (5).

#### **2.1 Etiology of colorectal cancer**

Important causes of colorectal cancers are uncommon genetic syndromes. A small percentage of "sporadic" colon cancers cluster in families. Relatives of people with colorectal cancer have increased risk for colorectal cancer, and risk varies depending on the number of relatives affected and the age at which cancer occurred (5).

Colorectal cancer is a heterogeneous disease that can develop through only partly known complex series of molecular changes. It is a long-term, gradual process which, besides external factors (carcinogens), also involves ever more recognized hereditary factors that cause genetic changes capable of triggering the uncontrolled mucosal (epithelial) growth (1, 5). The sequence of events that leads to the development of disease are passage from normal mucosa to adenoma – malignantly transformed adenoma and invasive carcinoma is associated with a series of genetic events occurring over long periods (5-7 years), the knowledge of which keeps expanding for the past ten years (1). In other words, the malignant transformation of cells requires various types of genetic damage in the form of gene mutation, deletion, amplification or expression disorder (1). In the 1990-ies, Fearon and Vogelstein first developed an algorithm for genetic events in colorectal cancer. According to their model, sporadic colon cancer arises as a result of a series of genetic changes that affect the process progression from enhanced epithelial proliferation to metastatic disease. The ultimate outcome of the process depends more on the number of accumulated changes than on their chronology.

The syndromes of colorectal cancer are inherited in an autosomal dominant fashion and are categorized by phenotypic, histological and genetic findings. Familial adenomatous

**Results:** Three hundred and eighty-eight colorectal cancer patients were included in the study. Nottingham Screening Tool Questionnaire, Appetite Loss Scale and Karnofsky Performance Status were taken to evaluate the nutritive status of patients. Group I consisted of 215 patients who were monitored prospectively and were given nutritional support and in this group weight gain of 1,5 kg (0,6-2,8 kg) and appetite improvement was observed in patients with colorectal cancer. In both groups Karnofsky Performance Status didn't change

Nutritional counseling, supplemental feeding and pharmacological support do temporarily stop weight loss and improve appetite, QoL and social life, but this improvement has no

**Conclusion**: These results encourage further studies with more specific nutritional supplementation in patients with colorectal cancer and probably in gastrointestinal

The incidence and mortality rates for cancers of the colon and rectum are among the highest of all malignancies worldwide (1, 2). Colorectal cancer is second in global cancer incidence and it is the most common cause of cancer death among non-smokers. US and EU incidence figures exceed global averages, which is consistent with an increased risk in industrialized nations (2). Factors associated with increased risk of colorectal cancer are host susceptibility and a sequence of different carcinogenic exposures. Specific etiology for sporadic colorectal cancer is still elusive but predisposing hereditary and environmental factors have been

Important causes of colorectal cancers are uncommon genetic syndromes. A small percentage of "sporadic" colon cancers cluster in families. Relatives of people with colorectal cancer have increased risk for colorectal cancer, and risk varies depending on the

Colorectal cancer is a heterogeneous disease that can develop through only partly known complex series of molecular changes. It is a long-term, gradual process which, besides external factors (carcinogens), also involves ever more recognized hereditary factors that cause genetic changes capable of triggering the uncontrolled mucosal (epithelial) growth (1, 5). The sequence of events that leads to the development of disease are passage from normal mucosa to adenoma – malignantly transformed adenoma and invasive carcinoma is associated with a series of genetic events occurring over long periods (5-7 years), the knowledge of which keeps expanding for the past ten years (1). In other words, the malignant transformation of cells requires various types of genetic damage in the form of gene mutation, deletion, amplification or expression disorder (1). In the 1990-ies, Fearon and Vogelstein first developed an algorithm for genetic events in colorectal cancer. According to their model, sporadic colon cancer arises as a result of a series of genetic changes that affect the process progression from enhanced epithelial proliferation to metastatic disease. The ultimate outcome of the process depends more on the number of accumulated changes than

The syndromes of colorectal cancer are inherited in an autosomal dominant fashion and are categorized by phenotypic, histological and genetic findings. Familial adenomatous

number of relatives affected and the age at which cancer occurred (5).

significantly reflecting the impact of the disease itself.

implications on patients KPS and course of their disease.

oncology.

**2. Colorectal cancer** 

clearly identified (5).

on their chronology.

**2.1 Etiology of colorectal cancer** 

polyposis, hereditary nonpolyposis colorectal cancer, Peutz-Jeghers disease, juvenile polyposis, Cowden disease are rare conditions (6, 7).

Ulcerative colitis, among other diseases in the medical history, is the top risk factor. The longer the disease and the segment of the colon affected, the higher the risk. The risk is also increased in individuals with Crohn's disease. The patient undergoing surgery for colorectal cancer has three times the risk of cancer recurrence (9).

The inheritance determines individual susceptibility to sporadic cancer but the lifestyle and environmental exposures are necessary for cancer expression. Colorectal cancer incidence varies between different geographic regions and incidence and mortality rates have been highest in developed western nations (10, 11). The basic argument that environment plays a huge role in colorectal cancer expression we get from observational studies in migrant populations. Migrants from low-incident regions to the high-incident regions of North America within one generation accept the incidence of the host country. Yet, studies with migrants also suggest that geographic variation in colorectal cancer incidences is due to environmental exposures and not due to the inherent predisposition (racial and ethnic group) (1, 10, 11).

Population based investigations have found many dietary and other environmental factors associated with colorectal cancer incidence (2). Most of these studies have methodologic limitations and therefore the interpretation of such studies has to be made with caution. Many studies have been conducted to investigate external factors that may increase or diminish the incidence of colorectal cancer. Many authors recognize four risk factor categories: epidemiological, intestinal, dietetic and mixed. The most frequently reported factors, among these shown to increase the risk of developing the disease, are diets rich in meat and animal fats (bile salts), physical inactivity, smoking and alcohol consumption (1, 2, 12). Between other diets ingested, consumption of red meat has the strongest correlation with colorectal cancer; over 30 case control studies report increased risk of colorectal cancer with higher red meat intake. Especially fried, barbecued and well-done meat is associated with colorectal cancer risk. Obesity and high caloric intake are the independent risk factors for colorectal cancer, excessive body mass gives a two fold increase in colorectal cancer, and this association in more expressed in men than women (12). Although studies carried out in humans and animal models have shown a positive correlation between the saturated fats/red meat consumption and the development of colorectal cancer, only a few of them are confirmed to be statistically significant. The total amount of fat in terms of daily caloric intake (>40%) and their form appears to have special significance. The conversion of dietary phospholipids to diacylglycerol by intestinal bacteria is assumed to be a potential mechanism of carcinogenesis. Diacylglycerol can enter the epithelial cell directly and by stimulating protein kinase C, it evokes intracellular signal transduction or mucosal proliferation. Another important mechanism is the formation of free radicals during fat metabolism and the mucosal damage induced by the secondary bile acids (lithocholic acid). Nitroso compounds, heat-generated heterocyclic amines and high protein intake (accelerated epithelial proliferation) have potential carcinogenic consequencies (12, 13).

Among the factors mentioned to reduce the risk are diet high in plant fibers and calcium, antioxidants (vitamin E, selenium etc.), menopausal hormone replacement therapy and administration of nonsteroidal anti-inflammatory drugs (12). In 1990 more than 13 studies showed significant reductions in colorectal cancer risk comparing the group with higher vs group with lower fibre intake. Some of the potential benefit mechanisms are: increased stool

Effects of Dietary Counseling on Patients with Colorectal Cancer 215

The symptoms of colorectal carcinoma depend on the anatomical location and size of the tumor. The tumor located in the cecoascendent portion will not necessarily produce obstruction since in this portion the stool has a liquid consistency, and the colonic lumen is wider than in the other parts. Patients complain of weakness, subfebrile temperature and blunt pain in the right lower hemiabdomen, and their laboratory tests show a high

Tumors of the transverse colon and on the left side of the half usually invade the colonic wall in a ring-shaped pattern mainly producing symptoms of the obstructive nature (cramping pain after meal, meteorism, change in stool form, occasional sudden ileus development and even bowel perforation). Symptoms of tumors confined to the rectosigmoid portion are most often false and/or painful urge to defecate (tenesmus),

The patient with suspicion of colorectal cancer should undergo a complete physical examination which must include digital rectal examination. In a large number of patients, the digital rectal examination already shows a hard lump inside the rectum, bleeding on touch. Colonoscopy is a procedure for visualizing colonic mucosa and obtaining samples for pathohistological analysis. Colonoscopy is the gold standard for detecting colorectal cancer. If for technical difficulties colonoscopy cannot be done, double-contrast irrigography may be considered although only 70-80% of lesions are detected by this method. Virtual colonoscopy and MR colonoscopy are also more and more often used. These radiology techniques use high-speed spiral CT and magnetic resonance imaging, and sophisticated software to process endoluminal images of the air-filled colon. Diagnostic techniques show a sensitivity of about 90% for tumors larger than 10 mm. Disadvantage is an inability to take biopsy samples and perform interventions available during colonoscopy. 'Colon capsule' for minutely detailed inspection of the colonic mucosa is also being gradually introduced, although this technique has the same disadvantage as the above mentioned techniques, and that is its inability to obtain biopsy samples (1, 13). Endoscopic ultrasound of the lower digestive tract is capable of providing assessment of tumor invasion into muscles and adjacent structures, as well as assessment of regional lymph node enlargement. The technique is employed to determine the extent of the spread of rectal tumors. Diagnosis of the spread of the disease involves imaging techniques (US, CT/MSCT/MRI of the abdomen and small pelvis, and CT/MSCT of the thorax). Serum CEA has limitations in sensitivity and specificity but was recommended for detection of recurrence. Molecular detection of tumor cells in circulation may prove to be more sensitive and specific than CEA (13, 18).

Treatment for colorectal cancer depends on the extent of cancer spread. Surgery is the method of choice for treatment of localized tumors. Colon resection surgery for colorectal cancer must be as radical as possible. Chemotherapy, immunotherapy and radiotherapy used may be adjuvant, neo-adjuvant, curative or palliative in nature. Adjuvant chemotherapy aims to destroy micrometastases following surgery, and neo-adjuvant chemotherapy is aimed at reducing the tumor mass to allow surgery for either the primary tumor or distant metastases (usually to the liver or lungs) (1, 13). Systemic therapy for disseminated disease has been gaining popularity over the past few years. Treatment options for colorectal cancer include a variety of chemotherapy and immunotherapy

**2.3 Clinical signs and diagnostic procedures** 

sedimentation rate and sideropenic anemia (1, 13).

narrow stool and hematochezia (13).

**2.4 Treatment** 

weight, dilution of potential carcinogens and increased colon transit rate. But other studies did not confirm such results, and today in this field we have inconclusive and controversial results (12, 13).

A complex interaction between inherited predispositions and external factors is responsible for the development of colorectal cancer (Table 1).


Table 1. Risk factors for colorectal cancer development

#### **2.2 Pathology**

Molecular basis of disease are genetic mutations of somatic cells and the inner innervation of the colon is important in carcinoma pathogenesis and spread. According to their macroscopic appearance, colorectal cancers are divided into exophytic, ulcerative and stenosing tumors. Exophytic tumors are most often located in the right half of the colon, while stenosing tumors are mostly found in its left half. The majority (up to 75%) of colorectal cancer occur within the descending colon, sigmoid colon and rectum, 15% of cases are located in the cecum and ascending colon, and only 10% in the transverse colon (13, 14, 15). Adenocarcinoma accounts for more than 95% of colorectal cancer cases. The prognosis of the disease is associated with the depth of tumor invasion through the colonic wall, peripheral lymph node involvement and absence or presence of distant metastases. The Dukes staging system (Table 2) as used in clinical practice divides this cancer into three stages, depending on the depth of cancer invasion into the colorectal wall (16, 17).


Table 2. The Dukes staging system for colorectal cancer

#### **2.3 Clinical signs and diagnostic procedures**

214 Colorectal Cancer – From Prevention to Patient Care

weight, dilution of potential carcinogens and increased colon transit rate. But other studies did not confirm such results, and today in this field we have inconclusive and controversial

A complex interaction between inherited predispositions and external factors is responsible

RISK FACTORS FOR COLORECTAL CANCER





Molecular basis of disease are genetic mutations of somatic cells and the inner innervation of the colon is important in carcinoma pathogenesis and spread. According to their macroscopic appearance, colorectal cancers are divided into exophytic, ulcerative and stenosing tumors. Exophytic tumors are most often located in the right half of the colon, while stenosing tumors are mostly found in its left half. The majority (up to 75%) of colorectal cancer occur within the descending colon, sigmoid colon and rectum, 15% of cases are located in the cecum and ascending colon, and only 10% in the transverse colon (13, 14, 15). Adenocarcinoma accounts for more than 95% of colorectal cancer cases. The prognosis of the disease is associated with the depth of tumor invasion through the colonic wall, peripheral lymph node involvement and absence or presence of distant metastases. The Dukes staging system (Table 2) as used in clinical practice divides this cancer into three

stages, depending on the depth of cancer invasion into the colorectal wall (16, 17).

DUKES C - metastases to lymph nodes

Table 2. The Dukes staging system for colorectal cancer

DUKES A - tumor confined within the colorectal wall DUKES B - tumor invaded through the colorectal wall

> B1-tumor limited to muscular mucosa B2-tumor protruding in/trough serosa






factors - age over 40 years

Table 1. Risk factors for colorectal cancer development


results (12, 13).

for the development of colorectal cancer (Table 1).

Genetic factors

Family factors

Pre-existing diseases

External factors

Other

**2.2 Pathology** 

The symptoms of colorectal carcinoma depend on the anatomical location and size of the tumor. The tumor located in the cecoascendent portion will not necessarily produce obstruction since in this portion the stool has a liquid consistency, and the colonic lumen is wider than in the other parts. Patients complain of weakness, subfebrile temperature and blunt pain in the right lower hemiabdomen, and their laboratory tests show a high sedimentation rate and sideropenic anemia (1, 13).

Tumors of the transverse colon and on the left side of the half usually invade the colonic wall in a ring-shaped pattern mainly producing symptoms of the obstructive nature (cramping pain after meal, meteorism, change in stool form, occasional sudden ileus development and even bowel perforation). Symptoms of tumors confined to the rectosigmoid portion are most often false and/or painful urge to defecate (tenesmus), narrow stool and hematochezia (13).

The patient with suspicion of colorectal cancer should undergo a complete physical examination which must include digital rectal examination. In a large number of patients, the digital rectal examination already shows a hard lump inside the rectum, bleeding on touch. Colonoscopy is a procedure for visualizing colonic mucosa and obtaining samples for pathohistological analysis. Colonoscopy is the gold standard for detecting colorectal cancer. If for technical difficulties colonoscopy cannot be done, double-contrast irrigography may be considered although only 70-80% of lesions are detected by this method. Virtual colonoscopy and MR colonoscopy are also more and more often used. These radiology techniques use high-speed spiral CT and magnetic resonance imaging, and sophisticated software to process endoluminal images of the air-filled colon. Diagnostic techniques show a sensitivity of about 90% for tumors larger than 10 mm. Disadvantage is an inability to take biopsy samples and perform interventions available during colonoscopy. 'Colon capsule' for minutely detailed inspection of the colonic mucosa is also being gradually introduced, although this technique has the same disadvantage as the above mentioned techniques, and that is its inability to obtain biopsy samples (1, 13). Endoscopic ultrasound of the lower digestive tract is capable of providing assessment of tumor invasion into muscles and adjacent structures, as well as assessment of regional lymph node enlargement. The technique is employed to determine the extent of the spread of rectal tumors. Diagnosis of the spread of the disease involves imaging techniques (US, CT/MSCT/MRI of the abdomen and small pelvis, and CT/MSCT of the thorax). Serum CEA has limitations in sensitivity and specificity but was recommended for detection of recurrence. Molecular detection of tumor cells in circulation may prove to be more sensitive and specific than CEA (13, 18).

#### **2.4 Treatment**

Treatment for colorectal cancer depends on the extent of cancer spread. Surgery is the method of choice for treatment of localized tumors. Colon resection surgery for colorectal cancer must be as radical as possible. Chemotherapy, immunotherapy and radiotherapy used may be adjuvant, neo-adjuvant, curative or palliative in nature. Adjuvant chemotherapy aims to destroy micrometastases following surgery, and neo-adjuvant chemotherapy is aimed at reducing the tumor mass to allow surgery for either the primary tumor or distant metastases (usually to the liver or lungs) (1, 13). Systemic therapy for disseminated disease has been gaining popularity over the past few years. Treatment options for colorectal cancer include a variety of chemotherapy and immunotherapy

Effects of Dietary Counseling on Patients with Colorectal Cancer 217

mucosa. Consequently, the absorption of nutrients is reduced. Some chemotherapeutics may affect the digestive system causing severe nausea, vomiting, abdominal pain, stomatitis and aversion to food. It should be noted that, in addition to the above mechanism for development of this syndrome, some chemotherapeutic agents also affect the taste buds of the tongue resulting in a changed and reduced sense of taste. It may also lead to reduced

Sometimes we cannot find a reason for anorexia and weight loss may be unrelated to nutritional intake. In this cases weight loss is reflection of elevated resting energy expenditures and over expression of pro-inflammatory cytokines (27). The most common factors to stimulate the production of proinflammatory cytokines include: TNF-alpha, interleukins, interferon gamma and leukemia inhibitory factor. It should be noted that, due to such complex mechanism, energy supplementation in cachectic patients does not result in

The pathophysiologic mechanism is correlated with the production of catabolic factors either by the tumor or via factors produced by the host. Cancer cachexia differ from starvation. It is an unquestionable fact today that cancer cachexia is pro-inflammatory condition. The pathophysiology of cachexia involves very complex pathways; cachexia is caused with numerous metabolic changes mediated with pro-inflammatory cytokines. The most known mediators are tumor necrosis factor α, interleukin-1 (IL-1), interleukin 6 (IL-6), interferon-Ƴ (from patients mononuclears) and molecules from tumor cells as lipid mobilisation factor (LMF) and proteolysis inducing factor (PIF). PIF is stimulating adenosine-triphosphate ubiquitin proteolitic pathway that is important in degradation of

The result of these changes is impairment of immune functions, quality of life, and performance status. The worst consequence is inability of patient to endure chemo, immunotherapy and radiotherapy. Cachexia decreases response to therapy due to frequent

Although increasing nutritional intake is insufficient to prevent the development of cachexia, nutritional support (taking into account the specific needs of the patient group), is required to reduce the consequences of nutritional decline and to improve quality of life and possibility to support the anticancer therapy (30). However, data from published studies are divided; some studies suggest that aggressive nutritional support can improve response to the antitumor treatment and decrease complications, but some deny any impact of nutritional support on tumor response, chemotherapy toxicity and survival (3, 30). Aggressive nutritional therapy does not significantly influence the outcome of patients with advanced cancer; "super"nutrition alone cannot reverse cachexia. But its use is still warranted because the patients QoL is significantly improved (3). The pharmacological treatments of cachexia antagonize the main symptoms (anorexia and chronic nausea) and improve the muscle metabolism. A significant number of studies (many uncontrolled) have suggested that anorexia and asthenia can be alleviated in cancer patient under corticosteroid

We must not underestimate advantage of nutritional treatment in improvement of asthenia and patients body image. Oral nutrition (after nutritional counseling) is ideal for cancer patients with a functional bowel. Enteral nutrition is useful in patients with advanced head and neck cancers or esophageal and gastric cancer and the use of parenteral nutrition (due

muscle mass and its stimulating synthesis of C-reactive protein (28, 29).

treatment; also, the feeling of well-being is observed (31).

toxicity and severe complications, what leads to shortened survival time (3, 20).

saliva production (26).

**3.2 Nutritional support** 

an increased body mass index (28).

regimens, with 5-fluorouracil/leucovorin, which may be added irinotecan and/or oxaliplatin, and bevacizumab, cetuximab and panitumumab as biological therapy, still remaining the mainstay for the management of patients with disseminated disease (3). The addition of this molecularly targeted therapy to standard chemotherapy improves treatment response, prolongs both the time to disease progression and eventually, median survival for disseminated or metastatic colorectal cancer, which currently is over 30 months (3, 19). In the future, prognostic and predictive factors will allow individual identification of patients who may benefit most from adjuvant chemotherapy, and which therapy should be used for the treatment of disseminated disease (personalized medicine). Therapy of rectal cancer includes adjuvant chemotherapy combined with radiation therapy (19).

Despite huge advances in diagnostic and surgery and despite global and national programs of prevention, about 50% of colorectal carcinomas are diagnosed in advanced stage (11). Advanced disease is largely refractory to conventional therapy and 5 years survival is still poor. Patients with advanced disease suffer from many stress symptoms (pain, vomiting, diarrhea, anorexia-cachexia syndrome, and etc.) and the therapeutic goal for them is maintenance of quality of life (QoL) (13). Many of those symptoms have implications for diagnostic and therapeutic procedures and can heavily disturb the process of chemoimmunotherapy and radiotherapy (3).

#### **3. Anorexia-cachexia syndrome**

#### **3.1 Pathophysiology of anorexia-cachexia syndrome**

Anorexia is defined as an unintentional reduction in food intake and anticipated cachexia. Cachexia develops as a result of progressive wasting of skeletal muscle mass and to a lesser extent adipose tissue (20). In cachexia, progressive wasting of skeletal muscle mass is replaced with adipose tissue and this occurs even before weight loss. Anorexia-cachexia syndrome is highly prevalent among patients with malignant diseases. Depending on primary tumor site anorexia-cachexia syndrome is present in 8-88% of cancer patients. Tumors of head and neck, stomach and pancreas have highest percentage of cachexia (21). At the time of diagnosis weight loss is present in about 50% of patients. Weight loss is independent predictive factor of survival (21). Cachexia-anorexia syndrome includes clinical features which are associated with growth of cancer. In addition, it has a large impact on morbidity, mortality and on patients' quality of life. Cancer cachexia develops in a majority of patients with advanced disease (22) (70 %) and directly causes death in 20% of cancer patients. Clinical signs of cancer cachexia are anorexia and weight loss. Abnormalities in carbohydrates, fat, protein and energy metabolism are clinically manifested as weakness, fatigue, malaise, loss of skeletal muscle and adipose tissue. In serum chemistry and haematology tests we can find anaemia, hypertrigliceridaemia, hypoproteinaemia with low albumines, hyperlacticacidaemia and glucose intolerance (insulin resistence) (23). Metabolic aberration in cancer cells and cells and microenvironment (inadequate energy intake, increased energy expenditure, mucositis, nausea, vomiting, change in taste or psychological problems as reaction to cancer disease) cause primary cancer cachexia. There are several conditions that can contribute decreased food intake (gastrointestinal obstruction, postchemotherapy nausea and vomiting, pain and etc) and cause secondary cancer cachexia (24, 25). Anorexia-cachexia syndrome often occurs or worsens after the administration of chemotherapy. Chemotherapeutic agents are toxic to malignant tissue, and also to the quickly proliferating cells. This group of cells also includes cells of the gastrointestinal

regimens, with 5-fluorouracil/leucovorin, which may be added irinotecan and/or oxaliplatin, and bevacizumab, cetuximab and panitumumab as biological therapy, still remaining the mainstay for the management of patients with disseminated disease (3). The addition of this molecularly targeted therapy to standard chemotherapy improves treatment response, prolongs both the time to disease progression and eventually, median survival for disseminated or metastatic colorectal cancer, which currently is over 30 months (3, 19). In the future, prognostic and predictive factors will allow individual identification of patients who may benefit most from adjuvant chemotherapy, and which therapy should be used for the treatment of disseminated disease (personalized medicine). Therapy of rectal cancer

Despite huge advances in diagnostic and surgery and despite global and national programs of prevention, about 50% of colorectal carcinomas are diagnosed in advanced stage (11). Advanced disease is largely refractory to conventional therapy and 5 years survival is still poor. Patients with advanced disease suffer from many stress symptoms (pain, vomiting, diarrhea, anorexia-cachexia syndrome, and etc.) and the therapeutic goal for them is maintenance of quality of life (QoL) (13). Many of those symptoms have implications for diagnostic and therapeutic procedures and can heavily disturb the process of chemo-

Anorexia is defined as an unintentional reduction in food intake and anticipated cachexia. Cachexia develops as a result of progressive wasting of skeletal muscle mass and to a lesser extent adipose tissue (20). In cachexia, progressive wasting of skeletal muscle mass is replaced with adipose tissue and this occurs even before weight loss. Anorexia-cachexia syndrome is highly prevalent among patients with malignant diseases. Depending on primary tumor site anorexia-cachexia syndrome is present in 8-88% of cancer patients. Tumors of head and neck, stomach and pancreas have highest percentage of cachexia (21). At the time of diagnosis weight loss is present in about 50% of patients. Weight loss is independent predictive factor of survival (21). Cachexia-anorexia syndrome includes clinical features which are associated with growth of cancer. In addition, it has a large impact on morbidity, mortality and on patients' quality of life. Cancer cachexia develops in a majority of patients with advanced disease (22) (70 %) and directly causes death in 20% of cancer patients. Clinical signs of cancer cachexia are anorexia and weight loss. Abnormalities in carbohydrates, fat, protein and energy metabolism are clinically manifested as weakness, fatigue, malaise, loss of skeletal muscle and adipose tissue. In serum chemistry and haematology tests we can find anaemia, hypertrigliceridaemia, hypoproteinaemia with low albumines, hyperlacticacidaemia and glucose intolerance (insulin resistence) (23). Metabolic aberration in cancer cells and cells and microenvironment (inadequate energy intake, increased energy expenditure, mucositis, nausea, vomiting, change in taste or psychological problems as reaction to cancer disease) cause primary cancer cachexia. There are several conditions that can contribute decreased food intake (gastrointestinal obstruction, postchemotherapy nausea and vomiting, pain and etc) and cause secondary cancer cachexia (24, 25). Anorexia-cachexia syndrome often occurs or worsens after the administration of chemotherapy. Chemotherapeutic agents are toxic to malignant tissue, and also to the quickly proliferating cells. This group of cells also includes cells of the gastrointestinal

includes adjuvant chemotherapy combined with radiation therapy (19).

immunotherapy and radiotherapy (3).

**3. Anorexia-cachexia syndrome** 

**3.1 Pathophysiology of anorexia-cachexia syndrome** 

mucosa. Consequently, the absorption of nutrients is reduced. Some chemotherapeutics may affect the digestive system causing severe nausea, vomiting, abdominal pain, stomatitis and aversion to food. It should be noted that, in addition to the above mechanism for development of this syndrome, some chemotherapeutic agents also affect the taste buds of the tongue resulting in a changed and reduced sense of taste. It may also lead to reduced saliva production (26).

Sometimes we cannot find a reason for anorexia and weight loss may be unrelated to nutritional intake. In this cases weight loss is reflection of elevated resting energy expenditures and over expression of pro-inflammatory cytokines (27). The most common factors to stimulate the production of proinflammatory cytokines include: TNF-alpha, interleukins, interferon gamma and leukemia inhibitory factor. It should be noted that, due to such complex mechanism, energy supplementation in cachectic patients does not result in an increased body mass index (28).

The pathophysiologic mechanism is correlated with the production of catabolic factors either by the tumor or via factors produced by the host. Cancer cachexia differ from starvation. It is an unquestionable fact today that cancer cachexia is pro-inflammatory condition. The pathophysiology of cachexia involves very complex pathways; cachexia is caused with numerous metabolic changes mediated with pro-inflammatory cytokines. The most known mediators are tumor necrosis factor α, interleukin-1 (IL-1), interleukin 6 (IL-6), interferon-Ƴ (from patients mononuclears) and molecules from tumor cells as lipid mobilisation factor (LMF) and proteolysis inducing factor (PIF). PIF is stimulating adenosine-triphosphate ubiquitin proteolitic pathway that is important in degradation of muscle mass and its stimulating synthesis of C-reactive protein (28, 29).

The result of these changes is impairment of immune functions, quality of life, and performance status. The worst consequence is inability of patient to endure chemo, immunotherapy and radiotherapy. Cachexia decreases response to therapy due to frequent toxicity and severe complications, what leads to shortened survival time (3, 20).

#### **3.2 Nutritional support**

Although increasing nutritional intake is insufficient to prevent the development of cachexia, nutritional support (taking into account the specific needs of the patient group), is required to reduce the consequences of nutritional decline and to improve quality of life and possibility to support the anticancer therapy (30). However, data from published studies are divided; some studies suggest that aggressive nutritional support can improve response to the antitumor treatment and decrease complications, but some deny any impact of nutritional support on tumor response, chemotherapy toxicity and survival (3, 30). Aggressive nutritional therapy does not significantly influence the outcome of patients with advanced cancer; "super"nutrition alone cannot reverse cachexia. But its use is still warranted because the patients QoL is significantly improved (3). The pharmacological treatments of cachexia antagonize the main symptoms (anorexia and chronic nausea) and improve the muscle metabolism. A significant number of studies (many uncontrolled) have suggested that anorexia and asthenia can be alleviated in cancer patient under corticosteroid treatment; also, the feeling of well-being is observed (31).

We must not underestimate advantage of nutritional treatment in improvement of asthenia and patients body image. Oral nutrition (after nutritional counseling) is ideal for cancer patients with a functional bowel. Enteral nutrition is useful in patients with advanced head and neck cancers or esophageal and gastric cancer and the use of parenteral nutrition (due

Effects of Dietary Counseling on Patients with Colorectal Cancer 219

6) reduction of skeletal and visceral muscle proteolysis; 7) prevention of bacterial translocation; 8) reduction in the frequency and severity of infectious complications and 9) shortening hospital stay (42). Fatty acids are thus formed providing a basic substrate for the colonic mucosa. Formulas containing omega-3 fatty acids from fish oil have been recently introduced. Omega-3 fatty acids reduce the synthesis of immunosuppressive and proinflammatory mediators. Meta-analyses of several studies have shown that immunomodulatory formulas do not significantly reduce mortality compared with standard enteral formulas. Their administration, however, achieves a lower rate of infection and septic complications, reduces dependency on assisted ventilation and shortens length of

Omega-3 polyunsaturated fatty acid, eicosapentaenoic acid (EPA) can down-regulate the production of pro-inflammatory cytokines such as IL-6, IL-1 and TNF in patients with cancer and in healthy individuals. EPA can also inhibit the effects of proteolysis inducing factor (PIF). EPA normalizes metabolic pathways changed due to malignant disease and stabilize weight gain through the competitive metabolism with arachidonic acid. EPA metabolites have lower inflammatory and immunosuppressive effect versus arachidonic acid metabolites. Especially interestingly is inhibitory effect of EPA on pancreatic and colorectal cancer cell line growth observed "in vitro" (44, 45). Wigmor and Bruera, like many other investigators, showed that EPA can stabilize body weight in cancer patients (46). We investigated if dietary counseling and oral nutrition supplement during chemotherapy affected nutritional status and symptom prevalence in our first study on 388 patients with colorectal cancer receiving chemotherapy for advanced disease (FOLFIRI/XELIRI/

Megestrol acetate is a type of medicine that comes in suspension form recommended in treatment guidelines for appetite and body weight loss in patients with malignant diseases. The drug belongs to a group of steroid hormones - progesterone. Its empirical formula is C24H32O4. Progestational derivate megestrol acetate has been evaluated in many studies; conclusion is that megestrol acetate significantly increases appetite, caloric intake and nutritional status with mild side effects as edema and hypercalcaemia. It is not completely clear through which mechanisms megestrol is acting. It is assumed that megestrol acetate changes the cytokines which are inhibiting TNF effects. Stimulation of appetite is due to stimulation neuropeptide Y in lateral hypothalamus. Megestrol acetate enhances appetite and increases food intake, enables the administration of specific treatments, and improves both patient treatment tolerance and their quality of life. Implementation of megestrol acetate in nutritional support plan is necessary; according to the highlights of the 2004 Cachexia Cancer Conference anorexia preceding to weight loss and orexigenics are necessary even when weight loss is absent. Furthermore, patients with cancer cachexia do not react on isolated over caloric food intake. Mild side effects (edema) are not enough pronounced over the social benefits caused by appetite stimulation; patients do not withdraw megestrol acetate therapy (49). Therefore, the International Association for Hospice and Palliative Care, NCCN Guidellines and European Palliative Care Research Collaborative Group recommend megestrol acetate as a mandatory drug for treatment cancer cachexia (50). The recommended starting dose is 400 mg (10 ml) once a day. The dose may be increased up to 800 mg (20 ml) / day. The most common side effects of megestrol acetate include edemas, insomnia, impaired libido, and very rarely thromboembolic

hospital treatment (43).

FOLFOX) (47, 48).

complications (48, 50).

to high costs and morbidity of 15%) with exception of high selected cases has no major role in care of cancer patients, especially in terminal disease (30, 31). In patients with colorectal cancer, enteral nutrition is usually provided by administration of food and/or commercial nutrient solutions and formulas (32). They either supplement daily diet or provide basic nutritional needs to patients who are unable to ingest sufficient amounts of food. The baseline requirements to administer such feeding include preserved swallowing function and the ability of the esophagus and stomach. There is a wide range of enteral nutrition formulas available for everyday use (33). Enteral nutrition formulas are classified into the following categories: monomeric (elementary) formulas, oligomeric formulas, polymeric formulas (32, 34, 35). The essential difference between them is in their size and/or the amount and type of molecules present. Accordingly, formulas containing a larger number of molecules that are also shorter at the same time, have a higher osmolality and can therefore cause side effects, such as diarrhea. The osmolality of an enteral formula depends on the type and amount of carbohydrates. Polysaccharides account for the vast majority of carbohydrate types present in the enteral feeding formulas. According to their solubility, fibers in the digestive system are divided into two categories: soluble and insoluble. Soluble fiber absorbs water in the intestinal lumen and increase the volume of the stool. They thus help regulate bowel motility. Soluble fibers are fermented by bowel bacteria using the aerobic pathway. Pectin slows down the emptying of the stomach and prolongs the passage of contents through the colon resulting in formation of the stool of satisfactory consistency even in tube fed patients (37). Normal metabolism requires daily protein intake of 0.8-1.0 g/kg body weight, and in the hypercatabolic state daily protein needs range from 1.2 to 1.6 g/kg body weight. According to the presence of nitrogen-containing compounds the diet may be divided into three groups: polymeric diet (includes natural proteins), oligomeric diet (includes small peptides), elementary diet (containing amino acids). Patients with the preserved gastrointestinal function require a diet in which complete proteins prevail (38). In case of compromised digestion, peptides should be the most represented. Among the amino acids, glutamine should be singled out. Glutamine helps maintain normal intestinal integrity by stimulating RNA, DNA and protein synthesis, resulting in an increase in the number and size of intestinal villi. Glutamine also prevents damage to intestinal permeability, preserves mucosal structure and prevents translocation of bacteria and toxins in the intestine. Glutamine is an important nutritional substrate for the intestinal cell line. In catabolic conditions including colorectal cancer, the intestinal system's requirements for glutamine are increased. The deficiency can be compensated for by addition of glutamine to the enteral nutrition formula (39). Arginine is another amino acid that plays a significant role in the immune events. It stimulates nitrogen oxide (NO) synthesis and the CD4/CD8 ratio, as well as the release of insulin, glucagon, prolactin and somatostatin (40, 41). The use of arginine requires caution as increased NO synthesis may accelerate the synthesis of proinflammatory cytokines and thereby cause a number of side effects. The main role of lipids in enteral formulas is to ensure large amounts of energy stored in relatively small volumes and sufficient amount of essential fatty acids which are a vital component of cell membranes and organelles. Corn oil and soybean oil used in enteral formulas provide longchain triglycerides (LCT), while coconut and palm oil provide medium-chain triglycerides (MCT). These products have a favorable effect on: 1) growth, differentiation and function of lymphocytes, macrophages and granulocytes; 2) release of trophic hormones or growth factors; 3) function of NK cells; 4) IL-2 synthesis; 5) improvement of mesenteric blood flow;

to high costs and morbidity of 15%) with exception of high selected cases has no major role in care of cancer patients, especially in terminal disease (30, 31). In patients with colorectal cancer, enteral nutrition is usually provided by administration of food and/or commercial nutrient solutions and formulas (32). They either supplement daily diet or provide basic nutritional needs to patients who are unable to ingest sufficient amounts of food. The baseline requirements to administer such feeding include preserved swallowing function and the ability of the esophagus and stomach. There is a wide range of enteral nutrition formulas available for everyday use (33). Enteral nutrition formulas are classified into the following categories: monomeric (elementary) formulas, oligomeric formulas, polymeric formulas (32, 34, 35). The essential difference between them is in their size and/or the amount and type of molecules present. Accordingly, formulas containing a larger number of molecules that are also shorter at the same time, have a higher osmolality and can therefore cause side effects, such as diarrhea. The osmolality of an enteral formula depends on the type and amount of carbohydrates. Polysaccharides account for the vast majority of carbohydrate types present in the enteral feeding formulas. According to their solubility, fibers in the digestive system are divided into two categories: soluble and insoluble. Soluble fiber absorbs water in the intestinal lumen and increase the volume of the stool. They thus help regulate bowel motility. Soluble fibers are fermented by bowel bacteria using the aerobic pathway. Pectin slows down the emptying of the stomach and prolongs the passage of contents through the colon resulting in formation of the stool of satisfactory consistency even in tube fed patients (37). Normal metabolism requires daily protein intake of 0.8-1.0 g/kg body weight, and in the hypercatabolic state daily protein needs range from 1.2 to 1.6 g/kg body weight. According to the presence of nitrogen-containing compounds the diet may be divided into three groups: polymeric diet (includes natural proteins), oligomeric diet (includes small peptides), elementary diet (containing amino acids). Patients with the preserved gastrointestinal function require a diet in which complete proteins prevail (38). In case of compromised digestion, peptides should be the most represented. Among the amino acids, glutamine should be singled out. Glutamine helps maintain normal intestinal integrity by stimulating RNA, DNA and protein synthesis, resulting in an increase in the number and size of intestinal villi. Glutamine also prevents damage to intestinal permeability, preserves mucosal structure and prevents translocation of bacteria and toxins in the intestine. Glutamine is an important nutritional substrate for the intestinal cell line. In catabolic conditions including colorectal cancer, the intestinal system's requirements for glutamine are increased. The deficiency can be compensated for by addition of glutamine to the enteral nutrition formula (39). Arginine is another amino acid that plays a significant role in the immune events. It stimulates nitrogen oxide (NO) synthesis and the CD4/CD8 ratio, as well as the release of insulin, glucagon, prolactin and somatostatin (40, 41). The use of arginine requires caution as increased NO synthesis may accelerate the synthesis of proinflammatory cytokines and thereby cause a number of side effects. The main role of lipids in enteral formulas is to ensure large amounts of energy stored in relatively small volumes and sufficient amount of essential fatty acids which are a vital component of cell membranes and organelles. Corn oil and soybean oil used in enteral formulas provide longchain triglycerides (LCT), while coconut and palm oil provide medium-chain triglycerides (MCT). These products have a favorable effect on: 1) growth, differentiation and function of lymphocytes, macrophages and granulocytes; 2) release of trophic hormones or growth factors; 3) function of NK cells; 4) IL-2 synthesis; 5) improvement of mesenteric blood flow; 6) reduction of skeletal and visceral muscle proteolysis; 7) prevention of bacterial translocation; 8) reduction in the frequency and severity of infectious complications and 9) shortening hospital stay (42). Fatty acids are thus formed providing a basic substrate for the colonic mucosa. Formulas containing omega-3 fatty acids from fish oil have been recently introduced. Omega-3 fatty acids reduce the synthesis of immunosuppressive and proinflammatory mediators. Meta-analyses of several studies have shown that immunomodulatory formulas do not significantly reduce mortality compared with standard enteral formulas. Their administration, however, achieves a lower rate of infection and septic complications, reduces dependency on assisted ventilation and shortens length of hospital treatment (43).

Omega-3 polyunsaturated fatty acid, eicosapentaenoic acid (EPA) can down-regulate the production of pro-inflammatory cytokines such as IL-6, IL-1 and TNF in patients with cancer and in healthy individuals. EPA can also inhibit the effects of proteolysis inducing factor (PIF). EPA normalizes metabolic pathways changed due to malignant disease and stabilize weight gain through the competitive metabolism with arachidonic acid. EPA metabolites have lower inflammatory and immunosuppressive effect versus arachidonic acid metabolites. Especially interestingly is inhibitory effect of EPA on pancreatic and colorectal cancer cell line growth observed "in vitro" (44, 45). Wigmor and Bruera, like many other investigators, showed that EPA can stabilize body weight in cancer patients (46). We investigated if dietary counseling and oral nutrition supplement during chemotherapy affected nutritional status and symptom prevalence in our first study on 388 patients with colorectal cancer receiving chemotherapy for advanced disease (FOLFIRI/XELIRI/ FOLFOX) (47, 48).

Megestrol acetate is a type of medicine that comes in suspension form recommended in treatment guidelines for appetite and body weight loss in patients with malignant diseases. The drug belongs to a group of steroid hormones - progesterone. Its empirical formula is C24H32O4. Progestational derivate megestrol acetate has been evaluated in many studies; conclusion is that megestrol acetate significantly increases appetite, caloric intake and nutritional status with mild side effects as edema and hypercalcaemia. It is not completely clear through which mechanisms megestrol is acting. It is assumed that megestrol acetate changes the cytokines which are inhibiting TNF effects. Stimulation of appetite is due to stimulation neuropeptide Y in lateral hypothalamus. Megestrol acetate enhances appetite and increases food intake, enables the administration of specific treatments, and improves both patient treatment tolerance and their quality of life. Implementation of megestrol acetate in nutritional support plan is necessary; according to the highlights of the 2004 Cachexia Cancer Conference anorexia preceding to weight loss and orexigenics are necessary even when weight loss is absent. Furthermore, patients with cancer cachexia do not react on isolated over caloric food intake. Mild side effects (edema) are not enough pronounced over the social benefits caused by appetite stimulation; patients do not withdraw megestrol acetate therapy (49). Therefore, the International Association for Hospice and Palliative Care, NCCN Guidellines and European Palliative Care Research Collaborative Group recommend megestrol acetate as a mandatory drug for treatment cancer cachexia (50). The recommended starting dose is 400 mg (10 ml) once a day. The dose may be increased up to 800 mg (20 ml) / day. The most common side effects of megestrol acetate include edemas, insomnia, impaired libido, and very rarely thromboembolic complications (48, 50).

Effects of Dietary Counseling on Patients with Colorectal Cancer 221

visit 0 and visit 12, Nottingham Screening Tool Questionnaire in groups I and II, at visit 0 and visit 12, Appetite Loss Scale in groups I and II, at visit 0 and visit 1, Weight loss in groups I and II , at visit 0 and visit 12, Karnofsky Performance Status in groups I and II, at visit 0 and visit 12, side effects of megestrol acetate. Evaluating the initial risk measurement according to BMI, decrease in weight gain and NST, we did not find any significant difference between the two groups. We performed 12 visits in follow-up according to chemotherapy schedule. Before initiation of chemotherapy, we re-evaluated nutritional status of our patients using evaluation tools. After chemotherapy was completed, in group I (consisted of 215 patients who were monitored prospectively and were given nutritional support) we observed weight gain of 1.5 kg (0.6-2.8 kg) and appetite improvement, the most commonly seen result after 4 weeks of therapy with megestrol acetate. The appetite also improved on Appetite Loss Scale from 3.1 (prechemotherapy) to 4.7 (post-chemotherapy). But KPS did not change significantly (74.2% before chemo versus 80.4% after chemo respectively) reflecting the impact of the disease itself. The most common side effects in patients receiving enteral nutrition were diarrhea (12% of patients) , abdominal pain (9%) and altered taste sensation(5%). The most frequently reported side effect in patients receiving megestrol acetate was the occurrence

This clinical study is ongoing and preliminary results from more than 600 patients are

**BMI Score** >20 0 18-20 1 <18 2

**during last 3 months? Score** 

Score: 0-2 Patient is not in nutritive risk and does not need nutritional support

resections, and also in patients who have chemo/radiotherapy treatment (3, 22).

≥ 5 Patient is in nutritive risk and needs nutritive support

No 0 A little, up to 3 kg 1 A lot, more than 3 kg 2

Anorexia-cachexia syndrome often occurs in patients with gastrointestinal cancers. Malnutrition has huge impact on outcome in patients who underwent major surgical

Although manifestations of chemotherapy injury on nutritional status is well-known, the potential role of nutritional supplementing is still not explored in detail. When treating cancer patients with chemotherapy we observed two problems and one of them is general failure in recognition of the weight loss early enough to perform nutritional support (30).

of edema (20% of patients).

**Has the patient unintentionally lost weight** 

Table 3. Nottingham Screening Tool Questionnaire

3-4 Patient need re-evaluation weekly

similar to this one.

**4. Discussion** 

The choice of enteral route depends on the underlying pathology, anticipated duration of enteral feeding and patient's preferences (30). In addition to the oral route of nutrition administration the transnasal route can also be considered. Indications for transnasal tube feeding include conditions or illnesses where normal feeding cannot be provided, and where the gastrointestinal tract maintains its function. For this purpose, nasogastric, nasoduodenal and nasojejunal types of tubes may be used. The tubes are usually placed 'blindly', however they may be placed by radiological and endoscopic means. The tubes are used when it is anticipated that tube feeding will be needed for up to 4 weeks. If enteral feeding is likely to be needed for more than 4 weeks, percutaneous endoscopic gastrostomy tubes or percutaneous endoscopic jejunostomies may be placed via an endoscopic access. The surgical placement of the gastrostomy or jejunostomy tube may also be taken into consideration. Two types of feeding can be used for patients requiring tube feeding: bolus (6 to 10 doses a day, each ranging from 50 to 200 ml, given over 5 to 30 minutes) or continuous feeding( 20 to 150 ml per hour during 16-18 hours). The method of 'bolus feeding' is more frequently reported to cause side effects than continuous feedings (30, 35).

In some clinical situations, enteral feeding may be unsafe or contraindicated. Reasons for postponing enteral nutrition administration are as follows: persistent nausea/vomiting, intensive postprandial pain, diarrhea, mechanical obstruction, diminished bowel motility, malabsorption, gastrointestinal bleeding. In mentioned situations, parenteral feeding is used. Parenteral feeding may be administered by peripheral or central vein access. Riskbenefit assessment of parenteral nutrition is necessary for each patient (30, 31, 35).

We can evaluate nutritional status of the cancer patient with quick screening methods (NRS-2002, NSTQ, ect) or more detailed examination (laboratory findings, anthropometric measurement, body composition measurement, BMI). Nottingham Screening Tool Questionnaire is simple, quick, and proper for re-evaluating. Another simple model (Fearon) is suggested for quick evaluation: if patient unintentionally decrease in weight gain more than 5% in 3 to 6 months, if caloric intake is less than 1500 kcal/day and C-reactive protein value is 10 and higher. Based on these data we can assume that cancer cachexia is developing (34, 35, 36, 43, 46, 50)

#### **3.3 Study results**

Our study was conducted at the Gastrointestinal Oncology Department, Clinic for Internal Medicine, University Hospital Center Rijeka, from January 2001 to December 2007. The aim of the study was to evaluate the effect of nutritional support in patients with colorectal cancer. The follow-up included 338 patients divided into two groups. Group I: patients receiving nutritional support (215 patients), and group II patients who did not receive nutritional support (173 patients); retrospectively collected data*.* Visit 0 took place one week before initiation of chemotherapy. The nutritional status was evaluated according to body weight change. The body mass index (BMI) was calculated for all patients and all patients were also evaluated through three questionnaires: Nottingham Screening Tool (Table 3), Appetite Loss Scale and Karnofsky Performance Status. The reassessments were done at control visits, each visit taking place before the next chemotherapy course. There were, in total, 12 visits performed. The aim of the study was to assess the effects of nutritional support in colorectal cancer patients on chemotherapy. For all patients, the following parameters were monitored and statistically evaluated: selection of nutritional support regimen in group I, BMI in groups I and II, at

The choice of enteral route depends on the underlying pathology, anticipated duration of enteral feeding and patient's preferences (30). In addition to the oral route of nutrition administration the transnasal route can also be considered. Indications for transnasal tube feeding include conditions or illnesses where normal feeding cannot be provided, and where the gastrointestinal tract maintains its function. For this purpose, nasogastric, nasoduodenal and nasojejunal types of tubes may be used. The tubes are usually placed 'blindly', however they may be placed by radiological and endoscopic means. The tubes are used when it is anticipated that tube feeding will be needed for up to 4 weeks. If enteral feeding is likely to be needed for more than 4 weeks, percutaneous endoscopic gastrostomy tubes or percutaneous endoscopic jejunostomies may be placed via an endoscopic access. The surgical placement of the gastrostomy or jejunostomy tube may also be taken into consideration. Two types of feeding can be used for patients requiring tube feeding: bolus (6 to 10 doses a day, each ranging from 50 to 200 ml, given over 5 to 30 minutes) or continuous feeding( 20 to 150 ml per hour during 16-18 hours). The method of 'bolus feeding' is more

In some clinical situations, enteral feeding may be unsafe or contraindicated. Reasons for postponing enteral nutrition administration are as follows: persistent nausea/vomiting, intensive postprandial pain, diarrhea, mechanical obstruction, diminished bowel motility, malabsorption, gastrointestinal bleeding. In mentioned situations, parenteral feeding is used. Parenteral feeding may be administered by peripheral or central vein access. Risk-

We can evaluate nutritional status of the cancer patient with quick screening methods (NRS-2002, NSTQ, ect) or more detailed examination (laboratory findings, anthropometric measurement, body composition measurement, BMI). Nottingham Screening Tool Questionnaire is simple, quick, and proper for re-evaluating. Another simple model (Fearon) is suggested for quick evaluation: if patient unintentionally decrease in weight gain more than 5% in 3 to 6 months, if caloric intake is less than 1500 kcal/day and C-reactive protein value is 10 and higher. Based on these data we can assume that cancer cachexia is

Our study was conducted at the Gastrointestinal Oncology Department, Clinic for Internal Medicine, University Hospital Center Rijeka, from January 2001 to December 2007. The aim of the study was to evaluate the effect of nutritional support in patients with colorectal cancer. The follow-up included 338 patients divided into two groups. Group I: patients receiving nutritional support (215 patients), and group II patients who did not receive nutritional support (173 patients); retrospectively collected data*.* Visit 0 took place one week before initiation of chemotherapy. The nutritional status was evaluated according to body weight change. The body mass index (BMI) was calculated for all patients and all patients were also evaluated through three questionnaires: Nottingham Screening Tool (Table 3), Appetite Loss Scale and Karnofsky Performance Status. The reassessments were done at control visits, each visit taking place before the next chemotherapy course. There were, in total, 12 visits performed. The aim of the study was to assess the effects of nutritional support in colorectal cancer patients on chemotherapy. For all patients, the following parameters were monitored and statistically evaluated: selection of nutritional support regimen in group I, BMI in groups I and II, at

frequently reported to cause side effects than continuous feedings (30, 35).

developing (34, 35, 36, 43, 46, 50)

**3.3 Study results** 

benefit assessment of parenteral nutrition is necessary for each patient (30, 31, 35).

visit 0 and visit 12, Nottingham Screening Tool Questionnaire in groups I and II, at visit 0 and visit 12, Appetite Loss Scale in groups I and II, at visit 0 and visit 1, Weight loss in groups I and II , at visit 0 and visit 12, Karnofsky Performance Status in groups I and II, at visit 0 and visit 12, side effects of megestrol acetate. Evaluating the initial risk measurement according to BMI, decrease in weight gain and NST, we did not find any significant difference between the two groups. We performed 12 visits in follow-up according to chemotherapy schedule. Before initiation of chemotherapy, we re-evaluated nutritional status of our patients using evaluation tools. After chemotherapy was completed, in group I (consisted of 215 patients who were monitored prospectively and were given nutritional support) we observed weight gain of 1.5 kg (0.6-2.8 kg) and appetite improvement, the most commonly seen result after 4 weeks of therapy with megestrol acetate. The appetite also improved on Appetite Loss Scale from 3.1 (prechemotherapy) to 4.7 (post-chemotherapy). But KPS did not change significantly (74.2% before chemo versus 80.4% after chemo respectively) reflecting the impact of the disease itself. The most common side effects in patients receiving enteral nutrition were diarrhea (12% of patients) , abdominal pain (9%) and altered taste sensation(5%). The most frequently reported side effect in patients receiving megestrol acetate was the occurrence of edema (20% of patients).

This clinical study is ongoing and preliminary results from more than 600 patients are similar to this one.



Table 3. Nottingham Screening Tool Questionnaire

Score: 0-2 Patient is not in nutritive risk and does not need nutritional support 3-4 Patient need re-evaluation weekly

≥ 5 Patient is in nutritive risk and needs nutritive support

#### **4. Discussion**

Anorexia-cachexia syndrome often occurs in patients with gastrointestinal cancers. Malnutrition has huge impact on outcome in patients who underwent major surgical resections, and also in patients who have chemo/radiotherapy treatment (3, 22).

Although manifestations of chemotherapy injury on nutritional status is well-known, the potential role of nutritional supplementing is still not explored in detail. When treating cancer patients with chemotherapy we observed two problems and one of them is general failure in recognition of the weight loss early enough to perform nutritional support (30).

Effects of Dietary Counseling on Patients with Colorectal Cancer 223

pathways or using alternative pathways in muscle cells. A number of molecules exhibiting anti cytokines activity have been tested without significant clinical data (20). Ghrelin is a hormone that stimulates the release of GH and increases appetite. In a phase II clinical

Despite cachexia is very common condition in cancers, there are still very few trials of drug therapies to reduce weight loss in cancer cachexia. Cachexia remains poorly studied and often undertreated condition that causes severe impairment of quality of life and increases

[2] Heavey PM, McKenna D, Rowland IR. Colorectal cancer and the relationship between

[3] Dobrila-Dintinjana R, Guina T, Krznarić Ž, Radić M, Dobrila M. Effects of Nutritional

[4] Markman B, Rodríguez-Freixinos V, Tabernero J. Biomarkers in colorectal cancer. Clin

[5] Sobczak A, Wawrzyn-Sobczak K, Sobaniec-Lotowska M. [The colorectal carcinoma risk

[8] Hans F.A. Vasen, Patrice Watson, Jukka–Pekka Mecklin, Henry T. Lynch and the ICG–

[9] Billioud V, Allen PB, Peyrin-Biroulet L. Update on Crohn's disease and ulcerative colitis.

[12] Sung MK, Bae YJ. Linking obesity to colorectal cancer: application of nutrigenomics.

[13] Wactawski-Wende J, Kotchen JM, Anderson GL. Calcium plus vitamin D

[14] Astin, M; Griffin, T, Neal, RD, Rose, P, Hamilton, W . The diagnostic value of symptoms

[15] Cervera P, Fléjou JF. Changing pathology with changing drugs: tumors of the

supplementation and the risk of colorectal cancer. N. Engl. J. Med*.* 2006; **354** (7):

for colorectal cancer in primary care: a systematic review. The British journal of general practice : the journal of the Royal College of General Practitioners 2011;**61**

[10] Watson AJ, Collins PD. Colon cancer: a civilization disorder. Dig Dis. 2011;29:222-8. [11] Kujundžić M, Banić M, Bokun T. Epidemiologija kolorektalnog karcinoma. Medix

HNPCC. New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative Group on

[6] Gala M, Chung DC. Hereditary colon cancer syndromes. Semin Oncol. 2011;38:490-9. [7] Migliore L, Migheli F, Spisni R, Coppedè F. Genetics, cytogenetics, and epigenetics of

Support in Patients with Colorectal Cancer during Chemotherapy. Collegium

study, ghrelin agonist anamorelin produced an improvement in total body mass (52).

[1] Brkić M, Grgić T. Kolorektalni karcinom. Medicus 15 (1): 89-97, 2006.

genes and the environment. Nutr Cancer, 48 (124), 2004.

colorectal cancer. J Biomed Biotechnol. 2011;2011:792362.

Antropologicum 32: 737–740, 2008.

factors]. Pol Merkuriusz Lek, 19:808, 2005.

HNPCC. Gastroenterology 1999;116:1453-1456.

Expert Rev Gastroenterol Hepatol. 2011;5:311-4.

gastrointestinal tract. Pathobiology. 2011;78:76-89.

Transl Oncol. 2010;12:261-70.

75/76: 70-76, 2008

684–96.

(586): 231–43.

Biotechnol J. 2010;5:930-41.

mortality.

**6. References** 

But if we know that patients will undergo to stress-full procedure which can have impact on his nutritional status (diagnostic procedures, colonoscopy for example, major gastrointestinal surgery) we have to give adequate nutritional support according to different clinical algorithms (3).

Adequate substitution with metabolites, increased caloric intake, inhibition of catabolic and inflammatory mediators leads to decrease of surgery, chemo and radiotherapy complications, but still has no significant impact on survival. Nutritional counseling, supplemental feeding and pharmacological support do temporarily stop weight loss and improve appetite, QoL and social life but this improvement has no implications on patient´s KPS and course of their disease. An improved knowledge of the pathophysiology of cancer induced cachexia will lead to development of more effective treatments (26).

In clinical practice, the role of nutrition therapy is often assumed to be less important than role of chemo, immunotherapy and radiotherapy as outcomes are less clear in literature (22, 26, 30). Our study showed that early nutritional intervention can decrease course of weight deterioration in the early course or locally advanced or metastatic colorectal cancer. Karnofsky Performance Status did not change significantly, what we expected.

Taking food is not only a physiologic necessity, but also cultural and a social event reflecting life and religious philosophy. Nutritive support can facilitate life of oncology patienta, their family support and caregivers understand (1). Therefore we have to recognize nutritionrelated issues and to implement strategies that will lead to a better outcome for patient and his caregivers. In the end of the life the wish of dying patient is most important factor regarding enteral/parenteral nutrition. The interaction between major syndromes in terminal disease (pain, cachexia, cognitive failure) should be better established because it seems that severity of them has impact on the others. If we improve pain and depression we can expect impact on cachexia syndrome (3, 21).

#### **5. Conclusion**

The role of nutritional therapy in oncology patients has been neglected. This mainly results from failure to recognize malnutrition and untimely introduction of nutritional support.

Our study shows that early introduction of nutritional support can decrease weight loss and in some cases even enable weight gain in patients with locally advanced and metastatic colorectal cancer.

To achieve better treatment results for patients with colorectal cancer, nutritional therapy should be considered as a highly important part of their treatment and more attention should be paid to timely recognition of malnutrition and introduction of nutritional support. Patients with anorexia-cachexia syndrome should undergo to individualized nutritional intervention where nutrition counseling is base for improvement of nutritional status, quality of life and social life. Anorectic patients have changes in taste and smell and do not support high-fat food and therefore frequent but small meals are highly recommended (20).

#### **Future perspectives:**

Cancer patients have increased level of growth hormone (GH), low serum concentrations of insulin growth factor-1 (IGF-1) and insulin resistence. Loss of lean mass and inflammatory processes are closely connected to the action of three signaling molecules: insulin, growth hormone and insulin growth factor-1 is essential (51). Basic stimuli of insulin, IGF-1 and GH does not provide response in muscle cells in cachexia, its reasonable to target post-receptor pathways or using alternative pathways in muscle cells. A number of molecules exhibiting anti cytokines activity have been tested without significant clinical data (20). Ghrelin is a hormone that stimulates the release of GH and increases appetite. In a phase II clinical study, ghrelin agonist anamorelin produced an improvement in total body mass (52).

Despite cachexia is very common condition in cancers, there are still very few trials of drug therapies to reduce weight loss in cancer cachexia. Cachexia remains poorly studied and often undertreated condition that causes severe impairment of quality of life and increases mortality.

#### **6. References**

222 Colorectal Cancer – From Prevention to Patient Care

But if we know that patients will undergo to stress-full procedure which can have impact on his nutritional status (diagnostic procedures, colonoscopy for example, major gastrointestinal surgery) we have to give adequate nutritional support according to different

Adequate substitution with metabolites, increased caloric intake, inhibition of catabolic and inflammatory mediators leads to decrease of surgery, chemo and radiotherapy complications, but still has no significant impact on survival. Nutritional counseling, supplemental feeding and pharmacological support do temporarily stop weight loss and improve appetite, QoL and social life but this improvement has no implications on patient´s KPS and course of their disease. An improved knowledge of the pathophysiology of cancer

In clinical practice, the role of nutrition therapy is often assumed to be less important than role of chemo, immunotherapy and radiotherapy as outcomes are less clear in literature (22, 26, 30). Our study showed that early nutritional intervention can decrease course of weight deterioration in the early course or locally advanced or metastatic colorectal cancer.

Taking food is not only a physiologic necessity, but also cultural and a social event reflecting life and religious philosophy. Nutritive support can facilitate life of oncology patienta, their family support and caregivers understand (1). Therefore we have to recognize nutritionrelated issues and to implement strategies that will lead to a better outcome for patient and his caregivers. In the end of the life the wish of dying patient is most important factor regarding enteral/parenteral nutrition. The interaction between major syndromes in terminal disease (pain, cachexia, cognitive failure) should be better established because it seems that severity of them has impact on the others. If we improve pain and depression we

The role of nutritional therapy in oncology patients has been neglected. This mainly results from failure to recognize malnutrition and untimely introduction of nutritional support. Our study shows that early introduction of nutritional support can decrease weight loss and in some cases even enable weight gain in patients with locally advanced and metastatic

To achieve better treatment results for patients with colorectal cancer, nutritional therapy should be considered as a highly important part of their treatment and more attention should be paid to timely recognition of malnutrition and introduction of nutritional support. Patients with anorexia-cachexia syndrome should undergo to individualized nutritional intervention where nutrition counseling is base for improvement of nutritional status, quality of life and social life. Anorectic patients have changes in taste and smell and do not support high-fat food and therefore frequent but small meals are highly recommended (20).

Cancer patients have increased level of growth hormone (GH), low serum concentrations of insulin growth factor-1 (IGF-1) and insulin resistence. Loss of lean mass and inflammatory processes are closely connected to the action of three signaling molecules: insulin, growth hormone and insulin growth factor-1 is essential (51). Basic stimuli of insulin, IGF-1 and GH does not provide response in muscle cells in cachexia, its reasonable to target post-receptor

induced cachexia will lead to development of more effective treatments (26).

Karnofsky Performance Status did not change significantly, what we expected.

can expect impact on cachexia syndrome (3, 21).

clinical algorithms (3).

**5. Conclusion** 

colorectal cancer.

**Future perspectives:** 


Effects of Dietary Counseling on Patients with Colorectal Cancer 225

[35] Coss CC, Bohl CE, Dalton JT. Curr Opin Clin Nutr Metab Care. Cancer cachexia

[36] Mattox TW, Treatment of unintentional weight loss in patients with cancer. Nutr Clin

[37] Klosterbuer A, Roughead ZF, Slavin J. Benefits of dietary fiber in clinical nutrition. Nutr

[38] Blum D, Strasser F. Cachexia assessment tools. Curr Opin Support Palliat Care. 2011,

[39] Evans MA, Shronts EP. Intestinal fuels: glutamine, short-chain fatty acids, and dietary

[40] Jobgen WS, Fried SK, Fu WJ, Meininger CJ, Wu G. J Nutr Biochem.Regulatory role for

[41] Parry RV, Ward SG. Protein arginine methylation: a new handle on T lymphocytes?

[42] Swails WS, Kenler AS, Driscoll DF, DeMichele SJ, Babineau TJ, Utsunamiya T, Chavali

[43] Mantovani G, Maccio A, Madeddu C, Gramignano G, Serpe R, Massa E, Dessi M, Tanca

[44] Calder PC. Immunomodulation by omega-3 fatty acids. Prostaglandins Leukot Essent

[45] La Guardia M, Giammanco S, Di Majo D, Tabacchi G, Tripoli E, Giammanco M. Omega

[46] Wigmore SJ, Ross JA, Falconer JS, Plester CE, Tisdale MJ, Carter DC, Fearon KC. The

[47] Dintinjana RD, Guina T, Krznarić Z, Radić M, Dintinjana M. Effects of nutritional

[48] Krznarić Ž, Juretić A, Šamija M, Dintinjana-Dobrila R, Vrdoljak E, Samaržija M, Kolaček

[49] Loprinzi CL, Kugler JW, Sloan JA, Mailliard JA, Krook JE, wilwerding MB, Rowland

Treatment of Cancer Anorexia/Cachexia . JCO 1999(17);10:3299-3306.

pancreatic cancer. Nutrition. 1996;12(1 Suppl):S27-30

the arginine-nitric oxide pathway in metabolism of energy substrates. 2006;17:571-

S, Forse RA, Bistrian BR. Effect of a fish oil structured lipid-based diet on prostaglandin release from mononuclear cells in cancer patients after surgery. JPEN

FM, Sanna E, Deiana L, Panzone F, Contu P, Floris C. Randomized phase III clinical trial of five different arms of treatment for patients with cancer cachexia: interim

3 fatty acids: biological activity and effects on human health. Panminerva Med.

effect of polyunsaturated fatty acids on the progress of cachexia in patients with

support in patients with colorectal cancer during chemotherapy. Coll Antropol.

S, Vrbanec D, Prgomet D, Ivkić M, Zelić M. [Croatian guidelines for use of eicosapentaenoic acid and megestrol acetate in cancer cachexia syndrome]. Lijec

KM, Camoriano JK, Novotny PJ, Christensen BJ. Randomised Comparison of Megestrole Acetate Versus Dexamethasone Versus Fluoxymestrone for the

Pharmacol Ther. 22: 583. 2005.

Trends Immunol. 2010;31:164-9.

results. Nutrition, 24:305, 2008.

Fatty Acids. 2007;77:327-35.

2005;47:245-57.

2008 ; 32:737-40.

Vjesn, 129:381, 2007.

J Parenter Enteral Nutr. 1997;21:266-74.

fiber. J Am Diet Assoc. 1992;92:1239-46,1249.

Pract, 20:400, 2005.

5(4):350-355.

88.

Clin Pract. 201.26:625-35.

therapy: a key weapon in the fight against cancer. 2011 May;14:268-73 Aliment


[16] Kyriakos M: The President cancer, the Dukes classification, and confusion, Arch Pathol

[17] Dukes CE. The classification of cancer of the rectum. Journal of Pathological

[18] Astin M, Griffin T, Neal RD, Rose P, Hamilton W. The diagnostic value of symptoms for

[19] Orbell J, West NJ. Improving detection of colorectal cancer. Practitioner. 2010;254:17-21,

[20] Laviano A, Meguid MM, Inui A, Muscaritoli M, Rossi-Fanelli F. Therapy insight: Cancer

[21] Penet MF, Winnard PT Jr, Jacobs MA, Bhujwalla Understanding cancer-induced cachexia: imaging the flame and its fuel. Curr Opin Support Palliat Care. 2011. [22] Van Cutsem E, Arends The causes and consequences of cancer-associated malnutrition.

[23] Bing C. Lipid Mobilization in cachexia: mechanisms and mediators. Curr Opin Support

[24] Inui A. Nippon Ronen Igakkai Zasshi, [Pathogenesis and treatment of cancer anorexia-

[25] Esper DH, Harb WA. The cancer cachexia syndrome: a review of metabolic and clinical

[26] Van Cutsem E, Arends J. The causes and consequences of cancer-associated

[27] Scheede-Bergdahl C, Watt HL, Trutschnigg B, Kilgour RD, Haggarty A, Lucar E, Vigano

[28] Ravasco P, Monteiro-Grillo I, Camilo M How relevant are cytokines in colorectal cancer

[29] Shibata M, Nezu T, Kanou H, Abe H, Takekawa M, Fukuzawa M. Decreased

[31] Bozzetti F. Guidelines on Artificial Nutrition versus Hydratation in terminal cancer

[32] Mirhosseini N, Fainsinger RL, Baracos V. Parenteral nutrition in advanced cancer: indications and clinical practice guidelines. J Palliat Med, 8:914. 2005. [33] Ries A, Trottenberg P, Elsner F, Stiel S, Haugen D, Kaasa S, Radbruch L. A systematic

[34] Ockenga J, Valentini L. Anorexia and cachexia in gastrointestinal cancer. Aliment

[30] Krznarić Ž: Klinička prehrana u gastroenterologiji. Medicus 15: 169-181, 2006.

EPCRC cachexia guidelines project. Palliat Med. 2011.

cachexia, with special emphasis on aged patients]. Nihon Ronen Igakkai Zasshi.

A Is IL-6 the best pro-inflammatory biomarker of clinical outcomes of cancer

production of interleukin-12 and type 2 immune responses are marked in cachectic patients with colorectal and gastric cancer. J Clin Gastroenterol. 2002;34:416-20.

review on the role of fish oil for the treatment of cachexia in advanced cancer: An

colorectal cancer in primary care: a systematic review. Br J Gen Pract. 2011;61:e231-

anorexia-cachexia syndrome--when all you can eat is yourself. Nat Clin Pract

Lab Med 109:1063, 1985.

Bacteriology 1932;35:323.

Oncol. 2:158, 2005.

20;4:460-7.

J. Eur J Oncol Nurs, 9:51, 2005.

Palliat Care. 2011; 5(4):356-360.

cachexia? Clin Nutr. 2011.

wasting? Cancer J. 2007;13:392-8.

patients.Nutrition 12:163-167, 1996.

Pharmacol Ther. 2005, 22(7):583-594.

manifestations. Nutr Clin Pract, 20:369.2005.

malnutrition. Eur J Oncol Nurs. 2005;9 Suppl 2:S51-63.

43.

2-3.


**Part 4** 

**Management and Treatment** 


## **Part 4**

**Management and Treatment** 

226 Colorectal Cancer – From Prevention to Patient Care

[50] Evans JW. Megestrol Acetate Use for Weight Gain should be Carefully Considered. The

[51] Trobec K, Haehling S, Anker SD, Lainscak M. Growth hormone, insulin-like growth

[52] Coats AJS, Surendran J, Vangipuram SRKG, Jain M, Shah S, Irhfan ABH, Fuang HG,

factor 1, and insulin signaling-a pharmacological target in body wasting and

Hassan MZM, Beadle J, Tilson J, Kirwan AB, Anker SD. The ACT-ONE trial, a multicentre, randomized, double blind, placebo-controlled, dose-finding study of the anabolic/catabolic transforming agent, MT-102 in subjects with cachexia related to stage III and IV non-small cell lung cancer and colorectal cancer: study design. J

Journal of Endocrinology and Metabolism 2007(92);2:420-421.

cachexia. J Cachexia Sarcopenia Muscle (2011) 2:191-200.

Cachexia Sarcopenia Muscle (2011) 2:201-207.

**12** 

*1USA 2Ireland* 

**Therapeutic Targets in Colorectal Cancer** 

Colon cancer is common worldwide: nearly a million people develop the disease every year and in the United States, colorectal cancer ranks third for frequency of occurrence and mortality in both men and women, with projected estimates for 2011 for occurrence and mortality put respectively at approximately 140,000 and 49,000 (American Cancer Society, 2011; Jemal et al, 2005). The projection for total deaths from all cancers in 2010 was 569,490

Significant progress in understanding colon cancer has produced a wealth of information that has aided improvements in aspects of diagnosis and disease management, contributing in the process to reduced mortality rates. The mechanisms that facilitate colorectal carcinogenesis and sustain progression and metastatic spread have been extensively investigated. The cause of colorectal cancer is multi-factorial. Notwithstanding the various contributing elements to the disease, the primary manifestation of colorectal carcinoma is the relentless and uncontrolled proliferation of cells and tissues in the intestinal mucosal epithelium. This pattern of abnormal proliferation is a disruption of the normal balance between new cell production by the epithelial cells in the mucosal crypts, and the release and loss of epithelial cells into the intestinal lumen i.e. cell-producing proliferation is normally finely and properly counter-balanced by regulated apoptotic and physical cell loss

Given the multistep, multifactor origins of colorectal cancer, the rationale for targeted therapies and the identification of therapeutic targets is that the disease can be (a) prevented prior to initiation (b) obstructed in its progression by blocking or inhibiting mechanisms that sustain progression and facilitate metastasis (c) reversed. The list of potential targets include microbes and bacteria that facilitate tumor initiation, molecular targets such as adenomatous polyposis coli (APC), and cancer stem cells (CSCs) where targeted destruction is thought to

As with all cancers, finding and delivering therapeutic targets in colorectal cancer is based on the premise that there is one originating cell type (van der Flier & Clevers, 2009). If this population of mutant originating cells is eliminated, the ability for new initiation, progression and distant seeding of tumor cells should be impaired and eventually abolished. Several therapeutic approaches have shown promising results in experimental

**1. Introduction** 

(Aliperti et al, 2011).

(Raz, 2002).

be central to preventing metastatic tumor spread.

Rajunor Ettarh1, Alvise Calamai2 and Anthony Cullen2

*2School of Medicine and Medical Sciences, University College, Dublin,* 

*1Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans,* 

### **Therapeutic Targets in Colorectal Cancer**

Rajunor Ettarh1, Alvise Calamai2 and Anthony Cullen2

*1Department of Structural and Cellular Biology, Tulane University School of Medicine, New Orleans, 2School of Medicine and Medical Sciences, University College, Dublin, 1USA 2Ireland* 

#### **1. Introduction**

Colon cancer is common worldwide: nearly a million people develop the disease every year and in the United States, colorectal cancer ranks third for frequency of occurrence and mortality in both men and women, with projected estimates for 2011 for occurrence and mortality put respectively at approximately 140,000 and 49,000 (American Cancer Society, 2011; Jemal et al, 2005). The projection for total deaths from all cancers in 2010 was 569,490 (Aliperti et al, 2011).

Significant progress in understanding colon cancer has produced a wealth of information that has aided improvements in aspects of diagnosis and disease management, contributing in the process to reduced mortality rates. The mechanisms that facilitate colorectal carcinogenesis and sustain progression and metastatic spread have been extensively investigated. The cause of colorectal cancer is multi-factorial. Notwithstanding the various contributing elements to the disease, the primary manifestation of colorectal carcinoma is the relentless and uncontrolled proliferation of cells and tissues in the intestinal mucosal epithelium. This pattern of abnormal proliferation is a disruption of the normal balance between new cell production by the epithelial cells in the mucosal crypts, and the release and loss of epithelial cells into the intestinal lumen i.e. cell-producing proliferation is normally finely and properly counter-balanced by regulated apoptotic and physical cell loss (Raz, 2002).

Given the multistep, multifactor origins of colorectal cancer, the rationale for targeted therapies and the identification of therapeutic targets is that the disease can be (a) prevented prior to initiation (b) obstructed in its progression by blocking or inhibiting mechanisms that sustain progression and facilitate metastasis (c) reversed. The list of potential targets include microbes and bacteria that facilitate tumor initiation, molecular targets such as adenomatous polyposis coli (APC), and cancer stem cells (CSCs) where targeted destruction is thought to be central to preventing metastatic tumor spread.

As with all cancers, finding and delivering therapeutic targets in colorectal cancer is based on the premise that there is one originating cell type (van der Flier & Clevers, 2009). If this population of mutant originating cells is eliminated, the ability for new initiation, progression and distant seeding of tumor cells should be impaired and eventually abolished. Several therapeutic approaches have shown promising results in experimental

Therapeutic Targets in Colorectal Cancer 231

The maintenance of functional and structural integrity and viability of the enteric mucosal epithelium depends on the preservation of the crypt cell renewing and emigration mechanisms for repopulating the continuously shedding epithelial cell cover (McGarvey et al, 2007a, 2007b). Several models for investigating the dynamics of colon cell regulation have been described (Boman et al, 2001; Hardy & Stark, 2002; Lander, 2009; Michor et al, 2004; Paulus et al, 1992; van Leeuwen et al, 2006; Wodarz, 2007). Many of these models have been employed in studies of the mechanisms that underlie normal colonic epithelial cell regulation and regeneration, as well as the dysregulated proliferation in colorectal

All of the new cells that are produced by proliferation of the cells in the stem cell compartment of the crypt, and numerically amplified in the semi-differentiated compartment, are distributed to the colonic mucosal epithelium to provide functionally important roles in absorption and secretion as well as providing a selectively permeable surface cover (Hall et al, 1994). The supply of new cells towards the upper crypt and surface epithelium is designed to satisfy the losses caused by cell injury, loss and programmed death (apoptosis). Surface cover cells are therefore removed or shed by processes that are as controlled and as balanced as the crypt-mediated cell renewal mechanism, and involves a cessation of proliferative processes in conjunction with the initiation of disposal and cell loss pathways (Leblond, 1964; Wright & Alison, 1984; Hall et al, 1994). Because the enteric epithelium is associated with underlying connective tissue fibroblasts, the accompanying fluxes in these cells are also correspondingly regulated in a controlled manner for proliferation and for cell loss (Marsh & Trier, 1974a, 1974b; Parker et al, 1974; Pascal et al, 1968a, 1968b). Together, the careful balance of cell production and cell loss maintains homeostasis in the colonic epithelium. Apoptosis does not occur randomly, rather it is seen towards the distal end of the cell migration route up the crypt (Hall et al, 1994). In colon cancer, proliferation is elevated and apoptosis is dysregulated, making the restoration of apoptosis an attractive proposition for therapeutic control of colon cancer growth (Evan &

A number of cyclooxygenase (COX) inhibitors induce apoptosis by activating mechanisms that are either upstream (via the lipid metabolite 13-S-hydroxyoctadecadienoic acid) or downstream (via 14-3-3ε proteins) of the nuclear hormone receptor PPAR∂ (Liou et al, 2007; Shureiqi et al, 2003), indicating that the pro-apoptotic effect of COX inhibitors on cancer cells is dependent on down-regulation of PPAR∂. In APC min mice, short-term treatment with nitric-oxide-donating aspirin (NO-ASA) induces apoptosis in differentiated intestinal epithelial cells while prolonged treatment with sulindac reverses the anti-apoptotic effect of APC (Mahmoud et al, 1998; Ouyang et al, 2006). In contrast, celecoxib administration

Other agents that have been shown to reduce colorectal cancer growth in vitro include CDDO-Me, an oleanane synthetic triterpenoid that achieves its apoptotic effect partly through the generation of reactive oxygen (ROS) and the activation of procaspases (Gao et al, 2011), green tea polyphenols that achieve their apoptotic effect through the induction of caspases (Oz & Ebersole, 2010), and tocotrienol, a member of the vitamin E family of compounds that induces morphological changes similar to apoptosis (paraptosis) and an accompanying reduction in Wnt signaling and its down-stream genes (Zhang et al, 2011).

cancer.

**3. Apoptosis** 

Vousden, 2001, Johnstone et al, 2002).

produces no effect on apoptosis (Williams et al, 2000).

studies. However, this chapter will focus largely on molecular targets in Wnt signaling, the nuclear receptor peroxisome proliferator-activated receptor (PPAR), and cancer stem cells (also known as cancer initiating cells).

#### **2. Colonic epithelial cell renewal**

The colon is the distal part of the intestinal tract and is lined internally by a simple layer of columnar epithelial cells (colonocytes) that send tube-like extensions called crypts into the mucosal layer of the intestinal wall. The crypts provide a conducive environment for the regulation and renewal of the epithelial covering of the colonic mucosa.

The epithelial cells in the crypt divide continuously and rapidly, achieving a turnover rate of epithelial renewal of between 5-6 days in mammals, with much shorter cell kinetic data reported for rodents (Di Garbo et al, 2010; Hall et al, 1994; Heath, 1996; Giles et al, 2003; Li et al, 1994; Loeffler et al, 1986; Potten & Loeffler, 1990; Okamoto & Watanabe, 2004; Wright & Alison, 1984). In the small intestine, between 8-9 cells are produced by each crypt epithelium every hour in mice; 2-3 dividing cells per crypt support cell production in the proximal intestine while up to 5 dividing cells are required to maintain cell production in the distal intestine (McGarvey et al, 2007a, 2007b). The renewal mechanism is sustained by a hierarchical arrangement of epithelial cells within the crypts, exemplified by the model described by Tomlinson and Bodmer (1995), with stem cells thought to reside in the lower part of the crypts, while differentiated cells populate the upper part of the crypt. By dividing and supplying transit (semi-differentiated) cells that migrate up the crypts, the stem cells are capable of and responsible for producing the various cell types that are found in the colonic epithelium. Differentiated cells at the top of the crypt and colonic mucosal surface eventually undergo spontaneous apoptosis and are released into the intestinal lumen (Fig 1).

Fig. 1. Schematic diagram of colonic crypt, illustrating the three zones and cell categories that constitute the kinetic framework for cell production and regeneration.

The maintenance of functional and structural integrity and viability of the enteric mucosal epithelium depends on the preservation of the crypt cell renewing and emigration mechanisms for repopulating the continuously shedding epithelial cell cover (McGarvey et al, 2007a, 2007b). Several models for investigating the dynamics of colon cell regulation have been described (Boman et al, 2001; Hardy & Stark, 2002; Lander, 2009; Michor et al, 2004; Paulus et al, 1992; van Leeuwen et al, 2006; Wodarz, 2007). Many of these models have been employed in studies of the mechanisms that underlie normal colonic epithelial cell regulation and regeneration, as well as the dysregulated proliferation in colorectal cancer.

#### **3. Apoptosis**

230 Colorectal Cancer – From Prevention to Patient Care

studies. However, this chapter will focus largely on molecular targets in Wnt signaling, the nuclear receptor peroxisome proliferator-activated receptor (PPAR), and cancer stem cells

The colon is the distal part of the intestinal tract and is lined internally by a simple layer of columnar epithelial cells (colonocytes) that send tube-like extensions called crypts into the mucosal layer of the intestinal wall. The crypts provide a conducive environment for the

The epithelial cells in the crypt divide continuously and rapidly, achieving a turnover rate of epithelial renewal of between 5-6 days in mammals, with much shorter cell kinetic data reported for rodents (Di Garbo et al, 2010; Hall et al, 1994; Heath, 1996; Giles et al, 2003; Li et al, 1994; Loeffler et al, 1986; Potten & Loeffler, 1990; Okamoto & Watanabe, 2004; Wright & Alison, 1984). In the small intestine, between 8-9 cells are produced by each crypt epithelium every hour in mice; 2-3 dividing cells per crypt support cell production in the proximal intestine while up to 5 dividing cells are required to maintain cell production in the distal intestine (McGarvey et al, 2007a, 2007b). The renewal mechanism is sustained by a hierarchical arrangement of epithelial cells within the crypts, exemplified by the model described by Tomlinson and Bodmer (1995), with stem cells thought to reside in the lower part of the crypts, while differentiated cells populate the upper part of the crypt. By dividing and supplying transit (semi-differentiated) cells that migrate up the crypts, the stem cells are capable of and responsible for producing the various cell types that are found in the colonic epithelium. Differentiated cells at the top of the crypt and colonic mucosal surface eventually undergo spontaneous apoptosis and are released into the intestinal lumen (Fig 1).

Fig. 1. Schematic diagram of colonic crypt, illustrating the three zones and cell categories

that constitute the kinetic framework for cell production and regeneration.

regulation and renewal of the epithelial covering of the colonic mucosa.

(also known as cancer initiating cells).

**2. Colonic epithelial cell renewal** 

All of the new cells that are produced by proliferation of the cells in the stem cell compartment of the crypt, and numerically amplified in the semi-differentiated compartment, are distributed to the colonic mucosal epithelium to provide functionally important roles in absorption and secretion as well as providing a selectively permeable surface cover (Hall et al, 1994). The supply of new cells towards the upper crypt and surface epithelium is designed to satisfy the losses caused by cell injury, loss and programmed death (apoptosis). Surface cover cells are therefore removed or shed by processes that are as controlled and as balanced as the crypt-mediated cell renewal mechanism, and involves a cessation of proliferative processes in conjunction with the initiation of disposal and cell loss pathways (Leblond, 1964; Wright & Alison, 1984; Hall et al, 1994). Because the enteric epithelium is associated with underlying connective tissue fibroblasts, the accompanying fluxes in these cells are also correspondingly regulated in a controlled manner for proliferation and for cell loss (Marsh & Trier, 1974a, 1974b; Parker et al, 1974; Pascal et al, 1968a, 1968b). Together, the careful balance of cell production and cell loss maintains homeostasis in the colonic epithelium. Apoptosis does not occur randomly, rather it is seen towards the distal end of the cell migration route up the crypt (Hall et al, 1994). In colon cancer, proliferation is elevated and apoptosis is dysregulated, making the restoration of apoptosis an attractive proposition for therapeutic control of colon cancer growth (Evan & Vousden, 2001, Johnstone et al, 2002).

A number of cyclooxygenase (COX) inhibitors induce apoptosis by activating mechanisms that are either upstream (via the lipid metabolite 13-S-hydroxyoctadecadienoic acid) or downstream (via 14-3-3ε proteins) of the nuclear hormone receptor PPAR∂ (Liou et al, 2007; Shureiqi et al, 2003), indicating that the pro-apoptotic effect of COX inhibitors on cancer cells is dependent on down-regulation of PPAR∂. In APC min mice, short-term treatment with nitric-oxide-donating aspirin (NO-ASA) induces apoptosis in differentiated intestinal epithelial cells while prolonged treatment with sulindac reverses the anti-apoptotic effect of APC (Mahmoud et al, 1998; Ouyang et al, 2006). In contrast, celecoxib administration produces no effect on apoptosis (Williams et al, 2000).

Other agents that have been shown to reduce colorectal cancer growth in vitro include CDDO-Me, an oleanane synthetic triterpenoid that achieves its apoptotic effect partly through the generation of reactive oxygen (ROS) and the activation of procaspases (Gao et al, 2011), green tea polyphenols that achieve their apoptotic effect through the induction of caspases (Oz & Ebersole, 2010), and tocotrienol, a member of the vitamin E family of compounds that induces morphological changes similar to apoptosis (paraptosis) and an accompanying reduction in Wnt signaling and its down-stream genes (Zhang et al, 2011).

Therapeutic Targets in Colorectal Cancer 233

Rao et al, 1995, 2009; Reddy et al, 1993). Prevention of tumorigenesis or tumor load reduction reflects either decreased cell proliferation or increased cell death but findings from animal studies are inconsistent (Table 1). For example, celecoxib treatment reduces tumor numbers and inhibits cell proliferation but data from studies using various sulindac preparations point to a variability that may be rodent species dependent (Jacoby et al, 2000;

**Model Inhibitor Dose & duration (wks) Inhibition effect Reference** 

mouse sulindac 160ppm 10 none *Shiff et al 1996*  mouse sulindac S2 20mg/kg 11 none *Swamy et al 2006*  mouse celecoxib 1500ppm 6 tumor number *Han et al 2008* 

mouse sulindac 5mg/kg 24 tumorigenesis *Shureiqi et al 2000*  mouse sulindac 5mg/kg 18 n/a *Kim et al 2009* 

mouse nimesulide 0.04%w/w 14 n/a *Shureiqi et al 2003*  rat celecoxib 300ppm 46 n/a *Guo et al 2009*  rat aspirin 200-400ppm 52 tumorigenesis *Piazza et al 1997* 

rat indomethacin 10ppm 1-30 tumorigenesis *Hanif et al 1996* 

Table 1. Effect of COX inhibitors on initiation and progression of experimental colon cancer in vivo. S2 = sulfide, NMNU = n-methyl-N-nitrosourea, AOM = azoxymethane, DMH = 1,2-

Some of the inconsistency in findings from animal studies is reflected in the results from clinical investigations in patients. Treatment with aspirin and celecoxib shows beneficial prevention of colorectal cancer in patients, and treatment with 150mg sulindac twice daily for nine months reduces number and size of colorectal adenomas. However, treatment with standard sulindac doses (25-150 mg twice daily) for 48 months did not prevent adenomas in patients (Giardiello et al, 1993, 2002; Giovannucci et al, 1994; Lanas & Fernandez, 2009; Thun

Peroxisome proliferator-activated receptors (PPAR) are part of the nuclear hormone receptor superfamily. While PPARα and PPARγ have been shown to be involved in various aspects of dietary lipid and glucose metabolism, PPAR∂ is implicated in the control of cell proliferation, differentiation and colorectal carcinogenesis (Desvergne & Wahil, 1999; Michalik et al, 2003; Wang & Dubois, 2010). Ligand activation of PPAR∂ is associated with suppressed induction of colon cancer (genetic and chemical treatment models) in mice via mechanisms that are linked to colonocyte differentiation and apoptosis (Harman et al, 2004;

dimethylhydrazine, APC = adenomatous polyposis coli, n/a = not measured

Mahmoud et al, 1998; Moorghen et al, 1988, 1998; Rao et al 1995, 2009).

**APC** 

 **DMH** 

 **AOM** 

**NMNU** 

et al, 1991).

**6. PPAR and COX inhibition** 

#### **4. Wnt and colorectal cancer**

One of the primary regulators of epithelial cell proliferation is Wnt signaling (Di Garbo et al, 2010). This signaling pathway involves the intermediate elements beta catenin, glycogen synthase kinase 3 beta (GSK3β), casein kinase I (CKI), axin, adenomatous polyposis coli (APC) and T-cell factor/lymphoid enhancer factor (TCF/LEF). Inappropriate activation or disruption of Wnt signaling upsets the careful regulatory balance in epithelial kinetics, leads to disorderly proliferation, and is an important contributor to the process of colorectal carcinogenesis. Wnt signaling helps to control the levels of cytoplasmic beta catenin, between pools bound to APC and to the cell adhesion molecule E-cadherin. The APC-bound pool of beta catenin is held in a stable complex of axin, GSK3β, CKI and APC that serves to regulate its cytoplasmic levels via targeted ubiquitin-mediated proteasomal degradation (Kikuchi et al, 2003; Pinto & Clevers, 2005). Wnt ligand signaling via membrane receptor proteins triggers a cascade that alters the relationship between the scaffold protein axin and GSK3β, interrupts regulated destruction of beta catenin, and leads to accumulation of nonphosphorylated beta catenin in the cytoplasm that then reaches the nucleus. Translocation of beta catenin into the nucleus after binding with TCF/LEF leads to the activation of target genes that regulate proliferation, differentiation and apoptosis (Araki et al, 2003; Coghlan et al, 2000; DiGarbo et al, 2010; Fagotto et al, 1998; He et al, 1998; Kishida et al, 1999; Shtutman et al, 1999; Tetsu & McCormick, 1999; van der Flier & Clevers, 2009; Yamamoto et al, 1999; Yanagawa et al, 1995; Yost et al, 1996). Direct binding of TCF to regulatory elements in downstream genes have aided identification of target genes and suggest that Wnt-activated gene expression shows a gradient-wise concentration of activity in intestinal crypts with the highest expression in the bottom of the crypt (Gregorieff et al, 2005). Most of these target genes are expressed in normal crypts and in adenomas (van der Flier et al, 2007; van der Wetering et al, 2002).

#### **5. Wnt and COX inhibition**

Colon cancer is associated with dysregulation and overexpression of COX, a key enzyme in the biosynthetic conversion of arachidonic acid to eicosanoids (Botting, 2006). Increased levels of expression of COX-2 are seen in up to 85% of colorectal adenomas and carcinomas (Eberhart et al, 1994; Fujita et al, 1998; Rigas et al, 1993; Sheng et al, 1997).

COX inhibitors demonstrate an ability to disrupt proliferation in several CRC cell lines. In HT29 colorectal adenoma cell lines, suppression of proliferation is evident as early as 48 hours after treatment with naproxen and piroxicam and at later timepoints with aspirin, indomethacin, aspirin and NS398 (Shiff et al, 1996; Shureiqi et al, 2000). But in some studies, naproxen and salicylic acid showed no effect on proliferation in the same cell lines pointing to differing potencies for inhibition of COX as well as effects on growth and apoptosis (Piazza et al, 1997). Although anti-proliferative effects have been reported in studies using HCA7, HT115 and SW620 cell lines which all express COX, the non-COX expressing cell line HT116 also shows reduced growth when treated with celecoxib for 72 hours (Shureiqi et al, 2003). Most of the evidence allows the conclusion that the anti-proliferative effects of COX inhibitors on colon cancer cell lines are not related to COX expression or activity.

When COX inhibitors are administered to APC min mice, initiation and progression of intestinal and colonic polyps is inhibited and polyp load is reduced (Jacoby et al, 1996, 2000; Kohno et al, 2005; Mahmoud et al, 1998, Moorghen et al, 1988, 1998; Narisawa et al, 1983;

One of the primary regulators of epithelial cell proliferation is Wnt signaling (Di Garbo et al, 2010). This signaling pathway involves the intermediate elements beta catenin, glycogen synthase kinase 3 beta (GSK3β), casein kinase I (CKI), axin, adenomatous polyposis coli (APC) and T-cell factor/lymphoid enhancer factor (TCF/LEF). Inappropriate activation or disruption of Wnt signaling upsets the careful regulatory balance in epithelial kinetics, leads to disorderly proliferation, and is an important contributor to the process of colorectal carcinogenesis. Wnt signaling helps to control the levels of cytoplasmic beta catenin, between pools bound to APC and to the cell adhesion molecule E-cadherin. The APC-bound pool of beta catenin is held in a stable complex of axin, GSK3β, CKI and APC that serves to regulate its cytoplasmic levels via targeted ubiquitin-mediated proteasomal degradation (Kikuchi et al, 2003; Pinto & Clevers, 2005). Wnt ligand signaling via membrane receptor proteins triggers a cascade that alters the relationship between the scaffold protein axin and GSK3β, interrupts regulated destruction of beta catenin, and leads to accumulation of nonphosphorylated beta catenin in the cytoplasm that then reaches the nucleus. Translocation of beta catenin into the nucleus after binding with TCF/LEF leads to the activation of target genes that regulate proliferation, differentiation and apoptosis (Araki et al, 2003; Coghlan et al, 2000; DiGarbo et al, 2010; Fagotto et al, 1998; He et al, 1998; Kishida et al, 1999; Shtutman et al, 1999; Tetsu & McCormick, 1999; van der Flier & Clevers, 2009; Yamamoto et al, 1999; Yanagawa et al, 1995; Yost et al, 1996). Direct binding of TCF to regulatory elements in downstream genes have aided identification of target genes and suggest that Wnt-activated gene expression shows a gradient-wise concentration of activity in intestinal crypts with the highest expression in the bottom of the crypt (Gregorieff et al, 2005). Most of these target genes are expressed in normal crypts and in adenomas (van der Flier et al, 2007; van der

Colon cancer is associated with dysregulation and overexpression of COX, a key enzyme in the biosynthetic conversion of arachidonic acid to eicosanoids (Botting, 2006). Increased levels of expression of COX-2 are seen in up to 85% of colorectal adenomas and carcinomas

COX inhibitors demonstrate an ability to disrupt proliferation in several CRC cell lines. In HT29 colorectal adenoma cell lines, suppression of proliferation is evident as early as 48 hours after treatment with naproxen and piroxicam and at later timepoints with aspirin, indomethacin, aspirin and NS398 (Shiff et al, 1996; Shureiqi et al, 2000). But in some studies, naproxen and salicylic acid showed no effect on proliferation in the same cell lines pointing to differing potencies for inhibition of COX as well as effects on growth and apoptosis (Piazza et al, 1997). Although anti-proliferative effects have been reported in studies using HCA7, HT115 and SW620 cell lines which all express COX, the non-COX expressing cell line HT116 also shows reduced growth when treated with celecoxib for 72 hours (Shureiqi et al, 2003). Most of the evidence allows the conclusion that the anti-proliferative effects of COX

When COX inhibitors are administered to APC min mice, initiation and progression of intestinal and colonic polyps is inhibited and polyp load is reduced (Jacoby et al, 1996, 2000; Kohno et al, 2005; Mahmoud et al, 1998, Moorghen et al, 1988, 1998; Narisawa et al, 1983;

(Eberhart et al, 1994; Fujita et al, 1998; Rigas et al, 1993; Sheng et al, 1997).

inhibitors on colon cancer cell lines are not related to COX expression or activity.

**4. Wnt and colorectal cancer** 

Wetering et al, 2002).

**5. Wnt and COX inhibition** 

Rao et al, 1995, 2009; Reddy et al, 1993). Prevention of tumorigenesis or tumor load reduction reflects either decreased cell proliferation or increased cell death but findings from animal studies are inconsistent (Table 1). For example, celecoxib treatment reduces tumor numbers and inhibits cell proliferation but data from studies using various sulindac preparations point to a variability that may be rodent species dependent (Jacoby et al, 2000; Mahmoud et al, 1998; Moorghen et al, 1988, 1998; Rao et al 1995, 2009).


Table 1. Effect of COX inhibitors on initiation and progression of experimental colon cancer in vivo. S2 = sulfide, NMNU = n-methyl-N-nitrosourea, AOM = azoxymethane, DMH = 1,2 dimethylhydrazine, APC = adenomatous polyposis coli, n/a = not measured

Some of the inconsistency in findings from animal studies is reflected in the results from clinical investigations in patients. Treatment with aspirin and celecoxib shows beneficial prevention of colorectal cancer in patients, and treatment with 150mg sulindac twice daily for nine months reduces number and size of colorectal adenomas. However, treatment with standard sulindac doses (25-150 mg twice daily) for 48 months did not prevent adenomas in patients (Giardiello et al, 1993, 2002; Giovannucci et al, 1994; Lanas & Fernandez, 2009; Thun et al, 1991).

#### **6. PPAR and COX inhibition**

Peroxisome proliferator-activated receptors (PPAR) are part of the nuclear hormone receptor superfamily. While PPARα and PPARγ have been shown to be involved in various aspects of dietary lipid and glucose metabolism, PPAR∂ is implicated in the control of cell proliferation, differentiation and colorectal carcinogenesis (Desvergne & Wahil, 1999; Michalik et al, 2003; Wang & Dubois, 2010). Ligand activation of PPAR∂ is associated with suppressed induction of colon cancer (genetic and chemical treatment models) in mice via mechanisms that are linked to colonocyte differentiation and apoptosis (Harman et al, 2004;

Therapeutic Targets in Colorectal Cancer 235

1981a, 1981b, 1999, 2006). On the basis of modelling studies, it is proposed that stem cells and crypts can suffer losses and be replaced (Cairnie & Millen, 1975; Nicolas et al, 2007;

Unequivocal stem cell identification has long remained elusive but, using genetic lineage tracing experiments, Barker et al (2007) showed that Lgr5, a G-protein-coupled receptor, is expressed in CBC cells. The study followed Lgr5-positive daughter cells up intestinal crypts and on to the intestinal villous epithelium, where all differentiated epithelial cell types could be demonstrated. The ability of Lgr5-positive stem cells in the crypt to give rise to cryptvillus units appear to be dependent on proximity to CD24+ cells at the bottom of the crypt (Sato et al, 2011). Stem cells have also been identified in mammalian epidermal hair follicles where they express Lgr6 (Snippert et al, 2010). Deletion of the APC gene in crypt stem cells in Lgr5 knock-in mice facilitates intestinal microadenoma growth; deletion of APC in transit-amplifying, semi-differentiated crypt cells in Lgr5 knock-in mice significantly reduces the growth of intestinal adenomas. Together this suggests that APC loss needs to be stem cell specific to propagate unrestrained tumor growth (Barker et al, 2009). The finding that single isolated Lgr5-positive stem cells can give rise to self-organizing crypt-villus units (Sato et al, 2009) raises the possibility that these cells may be useful in treatment strategies

There is experimental evidence for several proposed colon cancer stem cell markers including CD133, CD44, CD166, the extracellular matrix protein olfactomedin-4 (OLFM4), aldehyde dehydrogenase (ALDH1A1), Lgr5, and pleckstrin homology-like domain family A member 1 (PHLDA1). Some of these markers are associated with IL6-STAT3-JAK2 signaling (Becker et al, 2008; Dalerba et al, 2007; O'Brien et al, 2007; Ricci-Vitani et al, 2007; Sakthianandeswaren et al, 2011; Sanders & Majumdar, 2011; Shmelkov et al, 2008; Tsai et al,

In contrast to the idea that carcinogenic mutations can occur in any cell, the cancer stem cell model (first described in 1997 for hematologic malignancies) proposes that tumor transformation, progression and metastatic initiation is driven by the acquisition of oncogenic self-renewal properties by tissue stem cells, contributing to differentiation and the cellular heterogeneity of tumors (Chen et al, 2011; Sanders & Majumdar, 2011). This has led to the idea that conventional cancer therapies that target only proliferating cells in tumors may not necessarily be effective against cancer stem cells that mediate metastasis (Abdul Khalek et al, 2010, Sanders & Majumdar, 2011; Soltanian & Matin, 2011), and that these therapies may therefore be ineffective in producing long-term remissions. CSCs have greater DNA repair capacity and expression of ABC transporter genes, both of which contribute to relatively higher resistance to chemotherapy and radiation (Bao et al, 2006; Cho & Clarke, 2008; Hirschmann-Jax et al, 2004; Zhou et al, 2009). GO-Y030, a curcumin analogue has been shown to inhibit STAT3 phosphorylation signaling in colon cancer stem cells, offering the possibility of targeting STAT3 signaling in colon CSCs (Lin et al, 2011). The clonogenic and proliferative properties of CSCs are significantly interrupted by histone deacetylase (HDAC) inhibitors and this effect is associated with apoptotic cell death and modified Wnt signalling

1. When applied to colorectal cancer, the concept of hierarchical compartmentalization (as described in crypt kinetic models) offers target environments for stemness, proliferation

Yatabe et al, 2001).

that aim to repopulate enteric epithelia.

(Sikandar et al 2010).

**8. Conclusion** 

2011; Uchida et al, 2010; van der Flier et al, 2009).

Marin et al, 2006). Conversely, inactivation of PPAR∂ in APC-min mice enhances predisposition to multiple intestinal and colorectal polyps (Harman et al, 2004; Reed et al, 2004). Such evidence suggests that PPAR∂ attenuates colon cancer. However, Park and colleagues found a reduction in the ability of PPAR∂-/-(null) cells to form tumors in nude mice and they concluded that PPAR∂ might function to assist the tumor-suppressing function of adenomatous polyposis coli (APC) protein (Park et al, 2001).

Despite significant insights into the role of PPAR∂ in colorectal cancer, the physiological role of PPAR∂ in epithelia is still not completely understood. The unresolved nature of the available data has not prevented studies that have explored the possibility of targeting PPAR∂ therapeutically in colorectal cancer. Prostacyclin I2 can act as a natural ligand for PPAR∂ (Gupta et al, 2000), and because COX-2 inhibitors can suppress carcinogenesis and reduce intestinal polyposis (Hollingshead et al, 2008; Jacoby et al, 1996; Mahmoud et al, 1998), a number of studies examined the use of COX inhibtion to influence PPAR∂ activity. Sulindac and indomethacin inhibit colorectal carcinogenesis in vitro by rapidly downregulating transcriptional activity of PPAR∂ via disruption of DNA binding to PPAR∂-response elements (He et al, 1999). A similar effect on PPAR∂ is also observed following administration of sulindac and celecoxib but this is preceded by induction of the enzyme 15-lipoxygenase-1 (Shureiqi et al, 2003). Administration of nitric-oxide-donating aspirin reduces PPAR∂ expression and intestinal polyp numbers in mice but neither nimesulide nor GW0742 (a PPAR∂ ligand) has an effect on PPAR∂ mRNA levels, despite the fact that both agents reduce intestinal polyp numbers (Gupta et al, 2004; Hollingshead et al, 2008; Kohno et al, 2005).

COX-2 inhibitors and PPAR∂ ligands can separately attenuate cancer growth, however combinatorial protocols have so far failed to produce potentiated inhibition of colon cancer indicating that COX-inhibitory and PPAR∂ pathways are mechanistically separate (Hollingshead et al, 2008). In addition, concurrent expression of PPAR∂ and COX-2 in colorectal tumors has poor prognostic implications for patients (Yoshinaga et al, 2011).

Ligand activation of PPARγ is also anti-neoplastic in several tissues, but the data regarding its role in colorectal cancer is just as conflicting as the data for PPAR∂. PPARγ activation inhibits colon cancer cell growth in vitro whereas a mutation-dependent pro-tumorigenic effect has been reported in vivo (Girnun et al, 2002; Yoshizumi et al, 2004). The mechanistically interrelated and inter-dependent nature of colorectal cancer is illustrated by the finding that PPARγ agonists induce apoptosis by suppressing activation of NFκB and GSK3β (Ban et al, 2010). Other investigators have shown that PPARγ induces apoptosis via inactivation of survivin and activation of caspase-3 in colorectal cancer cell lines and were able to inhibit PPARγ-ligand induced apoptosis by activating PPAR∂ (Wang et al, 2011).

#### **7. Clones and stem cells**

The crypt structure of the colonic epithelium is maintained by the putative presence of pluripotent intestinal crypt stem cells (Schmidt et al, 1988). Initially crypts are polyclonal and subsequently become monoclonal. Two kinetic models of the stem-cell-sustained intestinal crypt have been described. In the classic model, intestinal stem cells are thought to reside in the 4th cell position from the bottom of the crypt (the +4 cell). These stem cells supply daughter cells to the proliferative, transit-amplifying zone of the crypt; stem cells can be replaced by these daughter cells if necessary (Marshman et al, 2002; Pottten, 1977; Potten et al, 1974, 2002). The zone model localizes stem cells to the bottom of the crypt; these cells are proposed to be the undifferentiated crypt base columnar (CBC) cells (Bjerknes & Cheng

Marin et al, 2006). Conversely, inactivation of PPAR∂ in APC-min mice enhances predisposition to multiple intestinal and colorectal polyps (Harman et al, 2004; Reed et al, 2004). Such evidence suggests that PPAR∂ attenuates colon cancer. However, Park and colleagues found a reduction in the ability of PPAR∂-/-(null) cells to form tumors in nude mice and they concluded that PPAR∂ might function to assist the tumor-suppressing

Despite significant insights into the role of PPAR∂ in colorectal cancer, the physiological role of PPAR∂ in epithelia is still not completely understood. The unresolved nature of the available data has not prevented studies that have explored the possibility of targeting PPAR∂ therapeutically in colorectal cancer. Prostacyclin I2 can act as a natural ligand for PPAR∂ (Gupta et al, 2000), and because COX-2 inhibitors can suppress carcinogenesis and reduce intestinal polyposis (Hollingshead et al, 2008; Jacoby et al, 1996; Mahmoud et al, 1998), a number of studies examined the use of COX inhibtion to influence PPAR∂ activity. Sulindac and indomethacin inhibit colorectal carcinogenesis in vitro by rapidly downregulating transcriptional activity of PPAR∂ via disruption of DNA binding to PPAR∂-response elements (He et al, 1999). A similar effect on PPAR∂ is also observed following administration of sulindac and celecoxib but this is preceded by induction of the enzyme 15-lipoxygenase-1 (Shureiqi et al, 2003). Administration of nitric-oxide-donating aspirin reduces PPAR∂ expression and intestinal polyp numbers in mice but neither nimesulide nor GW0742 (a PPAR∂ ligand) has an effect on PPAR∂ mRNA levels, despite the fact that both agents reduce intestinal polyp numbers (Gupta et al, 2004; Hollingshead et al, 2008; Kohno et al, 2005). COX-2 inhibitors and PPAR∂ ligands can separately attenuate cancer growth, however combinatorial protocols have so far failed to produce potentiated inhibition of colon cancer indicating that COX-inhibitory and PPAR∂ pathways are mechanistically separate (Hollingshead et al, 2008). In addition, concurrent expression of PPAR∂ and COX-2 in colorectal tumors has poor prognostic implications for patients (Yoshinaga et al, 2011). Ligand activation of PPARγ is also anti-neoplastic in several tissues, but the data regarding its role in colorectal cancer is just as conflicting as the data for PPAR∂. PPARγ activation inhibits colon cancer cell growth in vitro whereas a mutation-dependent pro-tumorigenic effect has been reported in vivo (Girnun et al, 2002; Yoshizumi et al, 2004). The mechanistically interrelated and inter-dependent nature of colorectal cancer is illustrated by the finding that PPARγ agonists induce apoptosis by suppressing activation of NFκB and GSK3β (Ban et al, 2010). Other investigators have shown that PPARγ induces apoptosis via inactivation of survivin and activation of caspase-3 in colorectal cancer cell lines and were able to inhibit PPARγ-ligand induced apoptosis by activating PPAR∂ (Wang et al, 2011).

The crypt structure of the colonic epithelium is maintained by the putative presence of pluripotent intestinal crypt stem cells (Schmidt et al, 1988). Initially crypts are polyclonal and subsequently become monoclonal. Two kinetic models of the stem-cell-sustained intestinal crypt have been described. In the classic model, intestinal stem cells are thought to reside in the 4th cell position from the bottom of the crypt (the +4 cell). These stem cells supply daughter cells to the proliferative, transit-amplifying zone of the crypt; stem cells can be replaced by these daughter cells if necessary (Marshman et al, 2002; Pottten, 1977; Potten et al, 1974, 2002). The zone model localizes stem cells to the bottom of the crypt; these cells are proposed to be the undifferentiated crypt base columnar (CBC) cells (Bjerknes & Cheng

function of adenomatous polyposis coli (APC) protein (Park et al, 2001).

**7. Clones and stem cells** 

1981a, 1981b, 1999, 2006). On the basis of modelling studies, it is proposed that stem cells and crypts can suffer losses and be replaced (Cairnie & Millen, 1975; Nicolas et al, 2007; Yatabe et al, 2001).

Unequivocal stem cell identification has long remained elusive but, using genetic lineage tracing experiments, Barker et al (2007) showed that Lgr5, a G-protein-coupled receptor, is expressed in CBC cells. The study followed Lgr5-positive daughter cells up intestinal crypts and on to the intestinal villous epithelium, where all differentiated epithelial cell types could be demonstrated. The ability of Lgr5-positive stem cells in the crypt to give rise to cryptvillus units appear to be dependent on proximity to CD24+ cells at the bottom of the crypt (Sato et al, 2011). Stem cells have also been identified in mammalian epidermal hair follicles where they express Lgr6 (Snippert et al, 2010). Deletion of the APC gene in crypt stem cells in Lgr5 knock-in mice facilitates intestinal microadenoma growth; deletion of APC in transit-amplifying, semi-differentiated crypt cells in Lgr5 knock-in mice significantly reduces the growth of intestinal adenomas. Together this suggests that APC loss needs to be stem cell specific to propagate unrestrained tumor growth (Barker et al, 2009). The finding that single isolated Lgr5-positive stem cells can give rise to self-organizing crypt-villus units (Sato et al, 2009) raises the possibility that these cells may be useful in treatment strategies that aim to repopulate enteric epithelia.

There is experimental evidence for several proposed colon cancer stem cell markers including CD133, CD44, CD166, the extracellular matrix protein olfactomedin-4 (OLFM4), aldehyde dehydrogenase (ALDH1A1), Lgr5, and pleckstrin homology-like domain family A member 1 (PHLDA1). Some of these markers are associated with IL6-STAT3-JAK2 signaling (Becker et al, 2008; Dalerba et al, 2007; O'Brien et al, 2007; Ricci-Vitani et al, 2007; Sakthianandeswaren et al, 2011; Sanders & Majumdar, 2011; Shmelkov et al, 2008; Tsai et al, 2011; Uchida et al, 2010; van der Flier et al, 2009).

In contrast to the idea that carcinogenic mutations can occur in any cell, the cancer stem cell model (first described in 1997 for hematologic malignancies) proposes that tumor transformation, progression and metastatic initiation is driven by the acquisition of oncogenic self-renewal properties by tissue stem cells, contributing to differentiation and the cellular heterogeneity of tumors (Chen et al, 2011; Sanders & Majumdar, 2011). This has led to the idea that conventional cancer therapies that target only proliferating cells in tumors may not necessarily be effective against cancer stem cells that mediate metastasis (Abdul Khalek et al, 2010, Sanders & Majumdar, 2011; Soltanian & Matin, 2011), and that these therapies may therefore be ineffective in producing long-term remissions. CSCs have greater DNA repair capacity and expression of ABC transporter genes, both of which contribute to relatively higher resistance to chemotherapy and radiation (Bao et al, 2006; Cho & Clarke, 2008; Hirschmann-Jax et al, 2004; Zhou et al, 2009). GO-Y030, a curcumin analogue has been shown to inhibit STAT3 phosphorylation signaling in colon cancer stem cells, offering the possibility of targeting STAT3 signaling in colon CSCs (Lin et al, 2011). The clonogenic and proliferative properties of CSCs are significantly interrupted by histone deacetylase (HDAC) inhibitors and this effect is associated with apoptotic cell death and modified Wnt signalling (Sikandar et al 2010).

#### **8. Conclusion**

1. When applied to colorectal cancer, the concept of hierarchical compartmentalization (as described in crypt kinetic models) offers target environments for stemness, proliferation

Therapeutic Targets in Colorectal Cancer 237

Bjerknes, M. & Cheng, H. (1981a). The stem-cell zone of the small intestinal epithelium. I. Evidence from Paneth cells in the adult mouse. *Am. J. Anat*, Vol.160, pp. 51–63 Bjerknes, M. & Cheng, H. (1981b). The stem-cell zone of the small intestinal epithelium. III.

Bjerknes, M. & Cheng, H. (1999). Clonal analysis of mouse intestinal epithelial progenitors.

Bjerknes, M. & Cheng, H. (2006). Intestinal epithelial stem cells and progenitors. *Methods* 

Boman, B. M.; Fields, J. Z.; Bonham-Carter, O. & Runquist, O. A. (2001). Computer modeling

Botting, R. M. (2006). Inhibitors of cyclooxygenases: mechanisms, selectivity and uses. *J* 

Cairnie, A. B. & Millen, B. H. (1975). Fission of crypts in the small intestine of the irradiated

Chen, S. Y.; Huang, Y. C.; Liu, S. P.; Tsai, F. J.; Shyu, W. C. & Lin, S. Z. (2011). An overview of concepts for cancer stem cells. *Cell Transplant,* Vol.20, No.1, pp. 113-20 Cho, R. W. & Clarke, M. F. (2008). Recent advances in cancer stem cells. *Curr Opin Genet Dev,*

Coghlan, M. P.; Culbert, A. A.; Cross, D. A.; Corcoran, S. L.; Yates, J. W.; Pearce, N. J.;

glycogen metabolism and gene transcription. *Chem Biol*, Vol.7, pp. 793–803. Dalerba, P.; Dylla, S. J.; Park, I. K.; Liu, R.; Wang, X.; Cho, R. W.; Hoey, T.; Gurney, A.;

Desvergne, B. & Wahli, W. (1999). Peroxisome proliferator-activated receptors: nuclear

Di Garbo, A.; Johnston, M. D.; Chapman, S. J. & Maini, P. K. (2010). Variable renewal rate

Eberhart, C. E.; Coffey, R. J.; Radhika, A.; Giardiello, F. M.; Ferrenbach, S. & DuBois, R. N.

Fagotto, F.; Gluck, U. & Gumbiner, B. M. (1998). Nuclear localization signal-independent

Fujita, T.; Matsui, M.; Takaku, K.; Uetake, H.; Ichikawa, W.; Taketo, M. M. & Sugihara, K.

adenomas and adenocarcinomas. *Gastroenterology,* Vol.107, pp. 1183–88. Evan, G. I. & Vousden, K. H. (2001). Proliferation, cell cycle and apoptosis in cancer. *Nature,*

Rausch, O. L, Murphy, G. J.; Carter, P. S.; Roxbee Cox, L.; Mills, D.; Brown, M. J.; Haigh, D.; Ward, R. W.; Smith, DG.; Murray, K. J.; Reith, A. D. & Holder, J. C. (2000). Selective small molecule inhibitors of glycogen synthase kinase-3 modulate

Huang, E. H.; Simeone, D. M.; Shelton, A. A.; Parmiani, G.; Castelli, C. & Clarke, M. F. (2007). Phenotypic characterization of human colorectal cancer stem cells. *Proc.* 

and growth properties of cell populations in colon crypts. *Physical Review,* Vol.81,

(1994). Up-regulation of cyclooxygenase-2 gene expression in human colorectal

and importin/karyopherin-independent nuclear import of β-catenin. *Curr Biol,*

(1998). Size- and invasion-dependent increase in cyclooxygenase-2 levels in human

*Am. J. Anat*, Vol.160, pp. 77–91

*Enzymol*, Vol.419, pp. 337– 83

Vol.18, No.1, pp. 48-53.

pp. 061909.1-12

Vol.411, pp. 342-48.

Vol.8, pp. 181–90.

pp. 8408–11

*Gastroenterology,* Vol.116, pp. 7–14

*Physiol Pharmacol*, Vol.57, Suppl 5, pp. 113-24.

mouse. *Cell Tissue Kinet*, Vol.8, pp. 189–96

*Natl. Acad. Sci. USA,* Vol.104, pp. 10158–63.

control of metabolism. *Endocr Rev,* Vol.20, pp. 649–88.

colorectal carcinomas. *Cancer Res,* Vol.58, pp. 4823–26.

Evidence from columnar, enteroendocrine, and mucous cells in the adult mouse.

implicates stem cell overproduction in colon cancer initiation. *Cancer Res,* Vol.61,

and differentiation. Potential targets in each compartment include dividing cells, apoptotic mechanisms and cancer stem cells.


#### **9. References**


2. Wnt signalling has been targeted for inhibition because of its relationship with proliferation. Activity in this pathway is highest in the stem zone which provides the

3. COX inhibitors have variable effects on proliferation that may be related to differing potencies, and the evidence suggests that these effects may not be due to any inhibitory action by the compounds on COX. Inconsistencies remain in trying to reproduce in patients the experimental outcomes on tumor loads seen following treatment with COX

4. A range of compounds, including nutritional and synthetic substances, induce apoptosis in colorectal cancer cell lines. Not all COX inhibitors induce apoptosis. 5. Some COX inhibitors down-regulate PPAR∂, other inhibitors do not. However, combination treatments do not produce the expected potentiation effect. The conflicting evidence of the roles of PPAR∂ and PPARγ in colorectal cancer remains unresolved. 6. Stem cells markers are increasingly being identified and involvement in signalling pathways such as IL6-STAT3 point to new targets that may be modulated using

Abdul Khalek, F. J.; Gallicano, G. I. & Mishra, L. (2010). Colon cancer stem cells. *Gastrointest* 

Aliperti, L. A.; Predina J. D.; Vachani, A. & Singhal, S. (2011). Local and systemic recurrence is the Achilles heel of cancer surgery. *Ann Surg Oncol,* Vol.18, No.3 pp. 603-7. American Cancer Society. (2011). *Colorectal Cancer Facts & Figures 2011-2013*, American

Araki, Y.; Okamura, S.; Hussain, S. P.; Nagashima, M.; He, P.; Shiseki, M.; Miura, K. &

Ban, J. O.; Kwak, D. H.; Oh, J. H.; Park, E. J.; Cho, M. C.; Song, H. S.; Song, M. J.; Han, S. B.;

Bao, S.; Wu, Q.; McLendon, R. E.; Hao, Y.; Shi, Q.; Hjelmeland, A. B.; Dewhirst, M. W.;

Barker, N.; Ridgway, R. A.; van Es, J. H.; van de Wetering, M.; Begthel, H.; van den Born, M.;

the cells-of-origin of intestinal cancer. *Nature,* Vol.457(7229), pp. 608-11 Barker, N.; van Es, J. H.; Kuipers, J.; Kujala, P.; van den Born, M.; Cozijnsen, M.; Haegebarth,

Harris, C. C. (2003). Regulation of cyclooxygenase-2 expression by the Wnt and ras

Moon, D. C.; Kang, K. W. & Hong, J. T. (2010). Suppression of NF-kappaB and GSK-3beta is involved in colon cancer cell growth inhibition by the PPAR agonist

Bigner, D. D. & Rich, J. N. (2006). Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. *Nature,* Vol.444, No.7120, pp.

Danenberg, E.; Clarke, A. R.; Sansom, O. J. & Clevers, H. (2009). Crypt stem cells as

A.; Korving, J.; Begthel, H.; Peters, P. J. & Clevers, H. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. *Nature,* Vol.449, pp. 1003–7 Becker, L.; Huang, Q. & Mashimo, H. (2008). Immunostaining of lgr5, an intestinal stem cell

marker, in normal and premalignant human gastrointestinal tissue. *Sci. World J,* 

apoptotic mechanisms and cancer stem cells.

therapeutic agents or genetic manipulations.

*Cancer Res* (Suppl 1), pp. S16-23

Cancer Society, Atlanta, United States

pathways. *Cancer Res*, Vol.63, pp. 728–34.

troglitazone. *Chem Biol Interact,* Vol.188, No.1, pp. 75-85

source of new cells.

inhibitors.

**9. References** 

756–60.

Vol.8, pp. 1168–76.

and differentiation. Potential targets in each compartment include dividing cells,


Therapeutic Targets in Colorectal Cancer 239

Harman FS, Nicol CJ, Marin HE, Ward JM, Gonzalez FJ, Peters JM. (2004). Peroxisome

He, T. C.; Chan, T. A.; Vogelstein, B. & Kinzler, K. W. (1999). PPARdelta is an APCregulated target of nonsteroidal anti-inflammatory drugs. *Cell,* Vol.99, pp. 335-45. He, T. C.; Sparks, A. B.; Rago, C.; Hermeking, H.; Zawel, L.; da Costa, L. T.; Morin, P. J.;

Heath, J. P. (1996). Epithelial cell migration in the intestine. *Cell Biol Int*, Vol.20, pp. 139–46. Hirschmann-Jax, C.; Foster, A. E.; Wulf, G. G.; Nuchtern, J. G.; Jax, T. W.; Gobel, U.; Goodell,

Hollingshead, H. E.; Borland, M. G.; Billin, A. N.; Willson, T. M.; Gonzalez, F. J. & Peters, J.

Jacoby, R. F.; Marshall, D. J.; Newton, M. A.; Novakovic, K.; Tutsch, K.; Cole, C. E.; Lubet, R.

Jacoby, R. F.; Seibert, K.; Cole, C. E.; Kelloff, G. & Lubet, R. A. (2000). The cyclooxygenase-2

Jemal, A.; Murray, T.; Ward, E.; Samuels, A.; Tiwari, R. C.; Ghafoor, A.; Feuer, E. J. & Thun, M. J. (2005). Cancer statistics, 2005. *CA Cancer J Clin*, Vol.55, No.1, pp. 10-30. Johnstone, R. W.; Ruefli, A. A. & Lowe, S. W. (2002). Apoptosis: a link between cancer

Kikuchi, A. (2003). Tumor formation by genetic mutations in the components of the Wnt

Kim, Y. H.; Kim, M. H.; Kim, B. J.; Kim, J. J.; Chang, D. K.; Son, H. J.; Rhee, P. L. & Rhee, J. C.

Kishida, S.; Yamamoto, H.; Hino, S.; Ikeda, S.; Kishida, M. & Kikuchi, A. (1999). DIX

Kohno, H.; Suzuki, R.; Sugie, S. & Tanaka, T. (2005). Suppression of colitis-related mouse

Lanas, A. & Ferrandez, A. (2009). NSAIDs and the colon. *Curr Opin Gastroenterol,* Vol.25, pp.

Lander, A. D. (2009). The 'stem cell' concept: is it holding us back? *J Biol*, Vol.8, No.8, pp. 70. Leblond, C. P. (1964). Classification of cell populations on the basis of their proliferative

Li, Y. Q.; Roberts, S. A.; Paulus, U.; Loeffler, M. & Potten, C. S. (1994). The crypt cycle in

(2009). Inhibition of cell proliferation and invasion in a human colon cancer cell line

domains of Dvl and axin are necessary for protein interactions and their ability to

colon carcinogenesis by a COX-2 inhibitor and PPAR ligands. *BMC Cancer,* Vol.5,

anti-inflammatory drug piroxicam *Cancer Res,* Vol.56, pp. 710–14

model of adenomatous polyposis *Cancer Res,* Vol.60, pp. 5040–44

genetics and chemotherapy. *Cell,* Vol.108, pp. 153-64.

by 5-aminosalicylic acid. *Dig Liver Dis*, Vol.41, pp. 328-37.

regulate beta-catenin stability. *Mol. Cell. Biol,* Vol.19, pp. 4414–22.

behaviour. *J. Nat. Cancer Inst. Monograph,* Vol.14, pp. 119-48.

mouse small intestinal epithelium. *J Cell Sci,* Vol.107, pp. 3271–79.

signaling pathway. *Cancer Sci,* Vol.94, pp. 225–29.

Vol.10, pp. 481-83.

14228–33

pp. 46

44–49

Vol.29, pp. 169-76.

APC pathway. *Science,* Vol.281, pp. 1509–12.

proliferator-activated receptor-delta attenuates colon carcinogenesis. *Nat. Med,*

Vogelstein, B. & Kinzler, K. W. (1998). Identification of c-MYC as a target of the

M. A. & Brenner, M. K. (2004). A distinct "side population" of cells with high drug efflux capacity in human tumor cells. *Proc Natl Acad Sci USA*, Vol.101, No.39, pp.

M. (2008). Ligand activation of peroxisome proliferator-activated receptorbeta/delta (PPARbeta/delta) and inhibition of cyclooxygenase 2 (COX2) attenuate colon carcinogenesis through independent signaling mechanisms. *Carcinogenesis,*

A.; Kelloff, G. J.; Verma, A.; Moser, A. R. & Dove, W. F. (1996). Chemoprevention of spontaneous intestinal adenomas in the Apc Min mouse model by the nonsteroidal

inhibitor celecoxib is a potent preventive and therapeutic agent in the min mouse


Gao, X.; Deeb, D.; Liu, P.; Liu, Y.; Arbab-Ali, S.; Dulchavsky, S. A. & Gautam, S. C. (2011).

Giardiello, F. M.; Yang, V. W.; Hylind, L. M.; Krush, A. J.; Petersen, G. M.; Trimbath, J. D.;

Giles, R. H.; van Es, J. H. & Clevers, H. (2003). Caught up in a Wnt storm: Wnt signaling in

Giovannucci, E.; Rimm, E. B.; Stampfer, M. J.; Colditz, G. A.; Ascherio, A. & Willett, W. C.

Girnun, G. D.; Smith, W. M.; Drori, S.; Sarraf, P.; Mueller, E.; Eng, C.; Nambiar, P.;

Gregorieff, A.; Pinto, D.; Begthel, H.; Destree, O.; Kileman, M. & Clevers, H. (2005).

Guo, Q.; Wu, M.; Lian, P.; Liao, M.; Xiao, Z.; Wang, X. & Shen, S. (2009). Synergistic effect of

Gupta, R. A.; Tan, J.; Krause, W. F.; Geraci, M. W.; Willson, T. M.; Dey, S. K. & DuBois, R. N.

Hall, P. A.; Coates, P. J.; Ansari, B. & Hopwood, D. (1994). Regulation of cell number in the

Han, A.; Song, Z.; Tong, C.; Hu, D.; Bi, X.; Augenlicht, L. H. & Yang, W. (2008). Sulindac

Hanif, R.; Pittas, A.; Feng, Y.; Koutsos, M. I.; Qiao, L.; Staiano-Coico, L.; Shiff, S. I. & Rigas, B.

Hardy, K. & Stark, J. (2002). Mathematical models of the balance between apoptosis and

accelerates intestinal adenoma growth. *Nat Med,* Vol.10, pp. 245-7.

pathway. *Biochemical pharmacology*, Vol.52, pp. 237-45.

proliferation. *Apoptosis,* Vol.7, pp. 373-81.

Vol.328, pp. 1313–16

sulindac. *N Engl J Med,* Vol.346, pp. 1054–59

cancer. *Biochim Biophys Acta,* Vol.1653, pp. 1–24.

professionals. *Ann Intern Med,* Vol.121, pp. 241–6

*Gastroenterology*, Vol.129, pp. 626–38

*Biochem*, Vol.330, pp. 71-81.

Vol.107(Pt. 12), pp. 3569–77

pp. 26-31.

PPARgamma. *Proc Natl Acad Sci USA*, Vol.99, pp. 13771–76.

Role of reactive oxygen species (ROS) in CDDO-Me-mediated growth inhibition and apoptosis in colorectal cancer cells. *J Exp Ther Oncol*, Vol.9, No.2, pp. 119-27. Giardiello, F. M.; Hamilton, S. R.; Krush, A. J.; Piantadosi, S.; Hylind, L. M.; Celano, P.;

Booker, S. V.; Robinson, C. R. & Offerhaus, G. J. (1993). Treatment of colonic and rectal adenomas with sulindac in familial adenomatous polyposis. *N Engl J Med,*

Piantadosi, S.; Garrett, E.; Geiman, D. E.; Hubbard, W.; Offerhaus, G. J. & Hamilton, S. R. (2002). Primary chemoprevention of familial adenomatous polyposis with

(1994). Aspirin use and the risk for colorectal cancer and adenoma in male health

Rosenberg, D. W.; Bronson, R. T.; Edelmann, W.; Kucherlapati, R.; Gonzalez, F. J. & Spiegelman, B. M. (2002). APC-dependent suppression of colon carcinogenesis by

Expression patterns of Wnt signaling components in the adult intestine.

indomethacin and NGX6 on proliferation and invasion by human colorectal cancer cells through modulation of the Wnt/beta-catenin signaling pathway. *Mol Cell* 

(2000). Prostacyclin-mediated activation of peroxisome proliferator-activated receptor delta in colorectal cancer. *Proc. Natl. Acad. Sci. USA,* Vol.97, pp. 13275-80. Gupta, R. A.; Wang, D.; Katkuri, S.; Wang, H.; Dey, SK. & DuBois, R. N. (2004). Activation of

nuclear hormone receptor peroxisome proliferator-activated receptor-delta

mammalian gastrointestinal tract: the importance of apoptosis. *J. Cell Sci*,

suppresses beta-catenin expression in human cancer cells. *Eur J Pharmacol*, Vol.583,

(1996). Effects of nonsteroidal anti-inflammatory drugs on proliferation and on induction of apoptosis in colon cancer cells by a prostaglandin-independent


Therapeutic Targets in Colorectal Cancer 241

O'Brien, C. A.; Pollett, A.; Gallinger, S. & Dick, J. E. (2007). A human colon cancer cell

Okamoto, R. & Watanabe, M. (2004). Molecular and clinical basis for the regeneration of

Ouyang, N.; Williams, J. L. & Rigas, B. (2006). NO-donating aspirin isomers downregulate

Oz, H. S. & Ebersole, J. L. (2010). Green tea polyphenols mediated apoptosis in intestinal

Park, B. H.; Vogelstein, B. & Kinzler, K. W. (2001). Genetic disruption of PPARdelta

Parker, F. G.; Barnes, E. N. & Kaye, G. I. (1974). The pericryptal fibroblast sheath. IV.

crypt and villus of the rabbit jejunum. *Gastroenterology*, Vol.67, pp. 607-21. Pascal, R. R.; Kaye, G. I. & Lane, N. (1968a). Colonic pericryptal fibroblast sheath:

Pascal, R. R.; Kaye, G. I. & Lane, N. (1968b). Colonic pericryptal fibroblast sheath:

Paulus, U.; Potten, C. S. & Loeffler, M. (1992). A model of the control of cellular regeneration

Piazza, G. A.; Rahm, A. K.; Finn, T. S.; Fryer, B. H.; Li, H.; Stoumen, A. L.; Pamukcu, R. &

Pinto, D. & Clevers, H. (2005). Wnt control of stem cells and differentiation in the intestinal

Potten, C. S. (1977). Extreme sensitivity of some intestinal crypt cells to X and gamma

Potten, C. S.; Kovacs, L. & Hamilton, E. (1974). Continuous labelling studies on mouse skin

Potten, C. S.; Owen, G. & Booth, D. (2002). Intestinal stem cells protect their genome by selective segregation of template DNA strands. *J. Cell Sci*, Vol.115, pp. 2381–88 Rao, C. V.; Rivenson, A.; Simi, B.; Zang, E.; Kelloff, G.; Steele, V. & Reddy, B. S. (1995).

Chemoprevention of colon carcinogenesis by sulindac a nonsteroidal anti-

human gastrointestinal epithelia. *J Gastroenterol*, Vol.39, pp. 1–6.

PPARdelta in carcinogenesis. *Carcinogenesis*, Vol.27, pp. 232-39.

106–10.

13.

Vol.98, pp. 2598-603.

Vol.54, pp. 835-51.

pp. 2452–59.

*Gastroenterology*, Vol.54, pp. 852-65.

epithelium. *Exp. Cell Res*, Vol.306, pp. 357–63.

Potten, C. S. & Loeffler, M. (1990). *Development,* Vol.110, pp. 1001–20.

inflammatory agent. *Cancer Res*, Vol.55, pp. 1464–72

and intestine. *Cell Tissue Kinet*, Vol.7, pp. 271–83

irradiation. *Nature*, Vol.269, pp. 518–21

*Cell Prolif,* Vol.25, pp. 559–78

capable of initiating tumour growth in immunodeficient mice. *Nature*, Vol.445, pp.

peroxisome proliferator-activated receptor (PPAR)delta expression in APC(min/+) mice proportionally to their tumor inhibitory effect: Implications for the role of

epithelial cells by a FADD-dependent pathway. *J Cancer Ther*, Vol.1, No.3, pp. 105-

decreases the tumorigenicity of human colon cancer cells. *Proc. Natl. Acad. Sci. USA*,

Replication, migration and differentiation of the subepithelial fibroblasts of the

replication, migration and cytodifferentiation of a mesenchymal cell system in adult tissue. I. Autoradiographic studies of normal rabbit colon. *Gastroenterology*,

replication, migration and cytodifferentiationof a mesenchymal cell system in adult tissue. II. Fine structural aspects of normal rabbit and human colon.

in the intestinal crypt after perturbation based solely on local stem cell regulation.

Ahnen, D. J. (1997). Apoptosis primarily accounts for the growth-inhibitory properties of sulindac metabolites and involves a mechanism that is independent of cyclooxygenase inhibition, cell cycle arrest, and p53 induction. *Cancer Res*, Vol.57,


Lin, L.; Liu, Y.; Li, H.; Li, P. K.; Fuchs, J.; Shibata, H.; Iwabuchi, Y. & Lin, J. (2011). Targeting

Liou, J. Y.; Ghelani, D.; Yeh, S. & Wu, K. K. (2007). Nonsteroidal anti-inflammatory drugs

Loeffler, M.; Stein, R.; Wichmann, H-E.; Potten, C. S.; Kaur, P. & Chwalinski, S. (1986). *Cell* 

Mahmoud, N. N.; Boolbol, S. K.; Dannenberg, A. J.; Mestre, J. R.; Bilinski, R. T.; Martucci, C.;

Marin, H. E.; Peraza, M. A.; Billin, A. N.; Willson, T. M.; Ward, J. M.; Kennett, M. J.;

Marsh, M. N. & Trier, J. S. (1974a). Morphology and cell proliferation of subepithelial

Marsh, M. N. & Trier, J. S. (1974b). Morphology and cell proliferation of subepithelial

Marshman, E.; Booth, C. & Potten C. S. (2002). The intestinal epithelial stem cell. *Bioessays*

McGarvey, M. A.; Bass, G. & Ettarh, R. R. (2007a). Nimesulide alters cell recruitment into

McGarvey, M. A.; O'Kelly, F. & Ettarh R. R. (2007b). Nimesulide inhibits crypt epithelial cell

Michalik, L.; Desvergne, B. & Wahli, W. (2003). Peroxisome proliferator-activated receptors

Michor, F.; Iwasa, Y.; Rajagopalan, H.; Lengauer, C. & Nowak, M. (2004). Linear model of

Moorghen, M.; Ince, P.; Finney, K. J.; Sunter, J. P.; Appleton, D. R. & Watson, A. J. (1988). A

Moorghen, M.; Orde, M.; Finney, K. J.; Appleton, D. R. & Watson, A. J. (1998). Sulindac

Narisawa, T.; Satoh, M.; Sano, M. & Takahashi, T. (1983). Inhibition of initiation and

familial adenomatous polyposis. *Carcinogenesis*, Vol.19, pp. 87–91

Vol.105, No.2, pp. 212-20

*Tissue Kinet,* Vol.19, pp. 627– 45.

Vol.67, pp. 3185-91.

401.

622-35.

2087-94.

Vol.67, pp. 636-45.

*Dig. Dis. Sci*, Vol.52, pp. 1471-78.

*Opin Lipidol*, Vol.14, pp. 129–35.

in mice. *J Pathol*, Vol.156, pp. 341–47

colon cancer initiation. *Cell Cycle*, Vol.3, pp. 358-62.

Vol.24, pp. 91–8

Vol.31, pp. 59–70

e28

colon cancer stem cells using a new curcumin analogue, GO-Y030. *Br J Cancer*,

induce colorectal cancer cell apoptosis by suppressing 14-3-3epsilon. *Cancer Res*,

Newmark, H. L.; Chadburn, A. & Bertagnolli, M. M. (1998). The sulfide metabolite of sulindac prevents tumors and restores enterocyte apoptosis in a murine model of

Gonzalez, F. J. & Peters, J. M. (2006). Ligand activation of peroxisome proliferatoractivated receptor beta inhibits colon carcinogenesis. *Cancer Res*, Vol.66, pp. 4394-

fibroblasts in adult mouse jejunum. I. Structural features. *Gastroenterol*, Vol.67, pp.

fibroblasts in adult mouse jejunum. II. Radioautographic studies. *Gastroenterology,*

mitosis in murine intestinal crypts without influencing the cell production rate.

proliferation at 6 hours in the small intestine in CD-1 mice. *Dig. Dis. Sci*, Vol.52, pp.

beta/delta: emerging roles for a previously neglected third family member. *Curr* 

protective effect of sulindac against chemically-induced primary colonic tumours

enhances cell proliferation in DMH-treated mouse colonic mucosa. *Cell Prolif*,

promotion by N-methylnitrosourea-induced colon carcinogenesis in rats by nonsteroid anti-inflammatory agent indomethacin. *Carcinogenesis*, Vol.4, pp. 1225–27 Nicolas, P.; Kim, K. M.; Shibata, D. & Tavare, S. (2007). The stem cell population of the

human colon crypt: analysis via methylation patterns. *PLoS Comput. Biol*, Vol.3, pp.


Therapeutic Targets in Colorectal Cancer 243

Shtutman, M.; Zhurinsky, J.; Simcha, I.; Albanese, C.; D'Amico, M.; Pestell, R. & Ben-Ze'ev,

Shureiqi, I.; Chen, D.; Lotan, R.; Yang, P.; Newman, R. A.; Fischer, S. M. & Lippman, S. M.

Shureiqi, I.; Jiang, W.; Zuo, X.; Wu, Y.; Stimmel, J. B.; Leesnitzer, L. M.; Morris, J. S.; Fan, H.

Snippert, H. J.; Haegebarth, A.; Kasper, M.; Jaks, V.; van Es, J. H.; Barker, N.; van de

Soltanian, S. & Matin, M. M. (2011). Cancer stem cells and cancer therapy. *Tumour Biol*,

Swamy, M. V.; Patlolla, J. M.; Steele, V. E.; Kopelovich, L.; Reddy, B. S. & Rao, C. V. (2006).

Tetsu, O. & McCormick, F. (1999). β-catenin regulates expression of cyclin D1 in colon

Thun, M. J.; Namboodiri, M. M.; Heath, C.W. Jr. (1991). Aspirin use and reduced risk of fatal

Tomlinson, I. P. & Bodmer, W. F. (1995). Failure of programmed cell death and

Tsai, K. S.; Yang, S. H.; Lei, Y. P.; Tsai, C. C.; Chen, H. W.; Hsu, C. Y.; Chen, L. L.; Wang, H.

Uchida, H.; Yamazaki, K.; Fukuma, M.; Yamada, T.; Hayashida, T.; Hasegawa, H.; Kitajima,

van der Flier, L. G. & Clevers, H. (2009). Stem Cells, Self-Renewal, and Differentiation in the

colorectal cancer cells. *Proc. Nat. Acad. Sci. USA*, Vol.100, pp. 9968–73. Sikandar, S.; Dizon, D.; Shen, X.; Li, Z.; Besterman, J. & Lipkin, S. M. (2010). The class I

lineages of the skin. *Science*, Vol.327(5971), pp. 1385-89.

Vol.118, pp. 2111–20.

Vol.60, pp. 6846–50

*Acad. Sci. USA*, Vol.96, pp. 5522–27.

*Oncotarget*, Vol.1, No.7, pp. 596-605.

Vol.32, No.3, pp. 425-40

*research*, Vol.66, pp. 7370-77.

carcinoma cells. *Nature*, Vol.398, pp. 422–26.

*Acad Sci USA*, Vol.92, No.24, pp. 11130-34.

doi:10.1053/j.gastro.2011.05.045

1731–37

colon cancer. *N Engl J Med,* Vol.325, pp. 1593–96

Intestinal Epithelium. *Annu Rev Physiol*, 71:241-60

D. & Rafii, S. (2008). Cd133 expression is not restricted to stem cells, and both cd133+ and cd133− metastatic colon cancer cells initiate tumors. *J. Clin. Invest*,

A. (1999). The cyclin D1 gene is a target of the β-catenin/LEF-1 pathway. *Proc. Nat.* 

(2000). 15-Lipoxygenase-1 mediates nonsteroidal anti-inflammatory drug-induced apoptosis independently of cyclooxygenase-2 in colon cancer cells. *Cancer Res*,

Z.; Fischer, S. M. & Lippman, S. M. (2003). The 15-lipoxygenase-1 product 13-Shydroxyoctadecadienoic acid down-regulates PPAR-δ to induce apoptosis in

HDAC inhibitor MGCD0103 induces cell cycle arrest and apoptosis in colon cancer initiating cells by upregulating Dickkopf-1 and non-canonical Wnt signaling.

Wetering, M.; van den Born, M.; Begthel, H.; Vries, R. G.; Stange, D. E.; Toftgård, R. & Clevers, H. (2010). Lgr6 marks stem cells in the hair follicle that generate all cell

Chemoprevention of familial adenomatous polyposis by low doses of atorvastatin and celecoxib given individually and in combination to APCmin mice. *Cancer* 

differentiation as causes of tumors: some simple mathematical models. *Proc Natl* 

W.; Miller, S. A.; Chiou, S. H.; Hung, M. C. & Hung, S. C. (2011). Mesenchymal Stem Cells Promote Formation of Colorectal Tumors in Mice. *Gastroenterology*,

M.; Kitagawa, Y. & Sakamoto, M. (2010). Overexpression of leucine-rich repeatcontaining g protein-coupled receptor 5 in colorectal cancer. *Cancer Sci*, Vol.101, pp.


Rao, C. V.; Steele, V. E.; Swamy, M. V.; Patlolla, J. M.; Guruswamy, S. & Kopelovich, L.

Raz, A. (2002). Is inhibition of cyclooxygenase required for the anti-tumorigenic effects of

Reddy, B. S.; Rao, C. V.; Rivenson, A. & Kelloff, G. (1993). Inhibitory effect of aspirin on

Reed, K. R.; Sansom, O. J.; Hayes, A. J.; Gescher, A. J.; Winton, D. J.; Peters, J. M. & Clarke, A.

Ricci-Vitiani, L.; Lombardi, D. G.; Pilozzi, E.; Biffoni, M.; Todaro, M.; Peschle, C. & De Maria,

Rigas, B.; Goldman, I. S. & Levine, L. (1993). Altered eicosanoid levels in human colon

Sakthianandeswaren, A.; Christie, M.; D'Andreti, C.; Tsui, C.; Jorissen, R. N.; Li, S.; Fleming,

Sanders, M. A. & Majumdar, A. P. (2011). Colon cancer stem cells: implications in

Sato, T.; van Es, J. H.; Snippert, H. J.; Stange, D. E.; Vries, R. G.; van den Born, M.; Barker, N.;

Schmidt, G. H.; Winton, D. J. & Ponder, B. A. (1988). Development of the pattern of cell

Sheng, H.; Shao, J.; Kirkland, S. C.; Isakson, P.; Coffey, R. J.; Morrow, J.; Beauchamp, R. D. &

Shiff, S. J.; Koutsos, M. I.; Qiao, L. & Rigas, B. (1996). Nonsteroidal antiinflammatory drugs

Shmelkov, S. V.; Butler, J. M.; Hooper, A. T.; Hormigo, A.; Kushner, J.; Milde, T.; St Clair, R.;

inhibition of cyclooxygenase-2. *J. Clin. Invest*, Vol.99, pp. 2254–59.

Shroyer, N. F.; van de Wetering, M. & Clevers, H. (2011). Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts. *Nature*, Vol.469(7330), pp. 415-8. Sato, T.; Vries, R. G.; Snippert, H. J.; van de Wetering, M.; Barker, N.; Stange, D. E.; van Es, J.

H.; Abo, A.; Kujala, P.; Peters, P. J. & Clevers, H. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. *Nature*, Vol.459(7244),

renewal in the crypt-villus unit of chimaeric mouse small intestine. *Development*,

DuBois, R. N. (1997). Inhibition of human colon cancer cell growth by selective

inhibit the proliferation of colon adenocarcinoma cells: effects on cell cycle and

Baljevic, M.; White, I.; Jin, D. K.; Chadburn, A.; Murphy, A. J.; Valenzuela, D. M.; Gale, N. W.; Thurston, G.; Yancopoulos, G. D.; D'Angelica, M.; Kemeny, N.; Lyden,

*Cancer Res*, Vol.69, pp. 8175–82

intestine. *Oncogene*, Vol.23, pp. 8992-96.

cancer. *J. Lab. Clin. Med*, Vol.122, pp. 518–23

carcinogenesis. *Front Biosci*, Vol.16, pp. 1651-62

apoptosis. *Exp Cell Res*, Vol.222, pp. 179-88.

*Nature*, Vol.445, pp. 111–15.

No.10, pp. 3709-19

pp. 262-65.

Vol.103, No.4, pp. 785-90.

Vol.63, pp. 343–47.

pp. 1493–97

(2009). Inhibition of azoxymethane-induced colorectal cancer by CP-31398 a TP53 modulator alone or in combination with low doses of celecoxib in male F344 rats.

nonsteroidal, anti-inflammatory drugs (NSAIDs)? In vitro versus in vivo results and the relevance for the prevention and treatment of cancer. *Biochem. Pharmacol*,

azoxymethane- induced colon carcinogenesis in F344 rats. *Carcinogenesis*, Vol.14,

R. (2004). PPARdelta status and Apc-mediated tumourigenesis in the mouse

R. (2007). Identification and expansion of human colon-cancer-initiating cells.

N. I.; Gibbs, P.; Lipton, L.; Malaterre, J.; Ramsay, R. G.; Phesse, T. J.; Ernst, M.; Jeffery, R. E.; Poulsom, R.; Leedham, S. J.; Segditsas, S.; Tomlinson, I. P.; Bernhard, O. K.; Simpson, R. J.; Walker, F.; Faux, M. C.; Church, N.; Catimel, B.; Flanagan, D. J.; Vincan, E. & Sieber, O. M. (2011). PHLDA1 expression marks the putative epithelial stem cells and contributes to intestinal tumorigenesis. *Cancer Res*, Vol.71, D. & Rafii, S. (2008). Cd133 expression is not restricted to stem cells, and both cd133+ and cd133− metastatic colon cancer cells initiate tumors. *J. Clin. Invest*, Vol.118, pp. 2111–20.


**13** 

*Denmark* 

**Anti-EGFR Treatment in Patients** 

The survival of patients with colorectal cancer (CRC) has increased constantly for many years due to superior surgical techniques, improved postoperative care, regular follow-up and an increased use of effective systemic therapy in the adjuvant and the palliative setting [1,2]. All of these advancements are important, but the establishment of multidisciplinary teams which facilitate optimal selection of therapy for individual patients may have been

In recent years a number of biologically active substances attacking specific signalling pathways in cancer cells (targeted therapy) have been developed and included in the treatment of patients with CRC. Three monoclonal antibodies (cetuximab, panitumumab, bevacizumab) have by now been approved for therapy in metastatic CRC (mCRC) [2,3]. Angiogenesis is necessary in tumour development and controlled in part by the vascular endothelial growth system which is inhibited by bevacizumab (Avastin ®) and many other

Cetuximab (Erbitux®) and panitumumab (Vectibix®) block the extracellular portion of the epidermal growth factor receptor (EGFR) and these two drugs will be discussed in detail in

EGFR is a trans-membrane glycoprotein that is involved in signaling pathways affecting cellular growth, differentiation, and proliferation and EGFR is expressed in many types of normal tissues. The EGFR is up regulated in a large number of cancers, in CRC in 60-80% of cases, and might be associated to a poor prognosis. Once a ligand binds to the extracellular domain of EGFR, receptor-dimerization occurs and down-stream signaling cascades are activated. Amongst the downstream effectors are the RAF/MEK/MAPK pathway and the

Two anti-EGFR monoclonal antibodies are approved by US Food and Drug Administration and European Medicines Agency for the treatment of CRC – cetuximab and panitumumab. Both are directed against the ligand-binding site of EGFR and competitively inhibiting ligand-induced activation, and thereby inhibiting EGFR induced cell growth, survival, and

**1. Introduction** 

anti-angiogenic drugs.

PI3K/PTEN/AKT pathway.

this chapter.

the most important concept on its own.

**2. Targeted therapy - Inhibition of EGFR** 

**with Colorectal Cancer** 

Camilla Qvortrup and Per Pfeiffer

*Odense University Hospital, Odense,* 

*Department of Oncology,* 


### **Anti-EGFR Treatment in Patients with Colorectal Cancer**

Camilla Qvortrup and Per Pfeiffer

*Department of Oncology, Odense University Hospital, Odense, Denmark* 

#### **1. Introduction**

244 Colorectal Cancer – From Prevention to Patient Care

van der Flier, L. G.; Haegebarth, A.; Stange, D. E.; van de Wetering, M. & Clevers, H. (2009).

van Leeuwen, I.; Byrne, H.; Jensen, O. & King, J. (2006). Crypt dynamics and colorectal

Wang, D. & DuBois, R. N. (2010). Therapeutic potential of peroxisome proliferator-activated

Wang, D.; Ning, W.; Xie, D.; Guo, L. & Dubois, R. N. (2011). Peroxisome proliferator-

Wright, N. A. & Alison, M. (1985). *The Biology of Epithelial Cell Populations*, Vol.2, ISBN 978-0-

Yamamoto, H.; Kishida, S.; Kishida, M.; Ikeda, S.; Takada, S. & Kikuchi, A. (1999).

Yanagawa, S.; van Leeuwen, F.; Wodarz, A.; Klingensmith, J. & Nusse, R. (1995). The

Yatabe, Y.; Tavare, S. & Shibata, D. (2001). Investigating stem cells in human colon by using methylation patterns. *Proc. Natl. Acad. Sci. USA*, Vol.98, pp. 10839–44 Yoshinaga, M.; Taki, K.; Somada, S.; Sakiyama, Y.; Kubo, N.; Kaku, T.; Tsuruta, S.;

Yoshizumi, T.; Ohta, T.; Ninomiya, I.; Terada, I.; Fushida, S.; Fujimura, T.; Nishimura, G.;

Yost, C.; Torres, M.; Miller, J. R.; Huang, E.; Kimelman, D. & Moon, R. T (1996). The axis-

Zhang, J. S.; Li, D. M.; He, N.; Liu, Y. H.; Wang, C. H.; Jiang, S. Q.; Chen, B. Q. & Liu, J. R.

Zhou, B. B.; Zhang, H.; Damelin, M.; Geles, K. G.; Grindley, J. C. & Dirks, P. B. (2009).

discovery. *Nat Rev Drug Discov*, Vol.8, No.10, pp. 806–23

between in vitro and in vivo models. *Cancer Res*, Vol.60, pp. 6045-51. Wodarz, D. (2007). Effect of stem cell turnover rates on protection against cancer and aging.

kinase-3β regulates its stability. *J. Biol. Chem*, Vol.274, pp. 10681–84.

cancer: advances in mathematical modelling. *Cell Prolif*, 39, 157-81.

colorectal cancer cells. *Gastroenterology*, 137:15–17

19-857615-0, Oxford University Press, United States

patients. *Dig Dis Sci*, Vol.56, No.4, pp. 1194-200.

pathway. *Toxicology*, Vol.285, pp. 8-17.

*Am*, Vol.39, No.3, pp. 697-707

*J. Theor. Biol*, Vol.245, pp. 449-58.

Vol.9, pp. 1087–97.

631–39

1443–54.

Olfm4 is a robust marker for stem cells in human intestine and marks a subset of

receptors in chronic inflammation and colorectal cancer. *Gastroenterol Clin North* 

activated receptor δ confers resistance to peroxisome proliferator-activated receptor γ-induced apoptosis in colorectal cancer cells. *Oncogene*, doi: 10.1038/onc.2011.299 Williams, C. S.; Watson, A. J.; Sheng, H.; Helou, R.; Shao, J. & DuBois, R. N. (2000). Celecoxib

prevents tumor growth in vivo without toxicity to normal gut: lack of correlation

Phosphorylation of axin, a Wnt signal negative regulator, by glycogen synthase

dishevelled protein is modified by wingless signaling in Drosophila. *Gene Develop*,

Kusumoto, T.; Sakai, H.; Nakamura, K.; Takayanagi, R. & Muto, Y. (2011). The expression of both peroxisome proliferator-activated receptor delta and cyclooxygenase-2 in tissues is associated with poor prognosis in colorectal cancer

Shimizu, K.; Yi, S. & Miwa, K. (2004). Thiazolidinedione, a peroxisome proliferatoractivated receptor-gamma ligand, inhibits growth and metastasis of HT-29 human colon cancer cells through differentiation-promoting effects. *Int J Oncol*, Vol.25, pp.

inducing activity, stability, and subcellular distribution of beta-catenin is regulated in Xenopus embryos by glycogen synthase kinase 3. *Gene. Develop*. Vol.10, pp.

(2011). A paraptosis-like cell death induced by δ-tocotrienol in human colon carcinoma SW620 cells is associated with the suppression of the Wnt signaling

Tumour-initiating cells: challenges and opportunities for anticancer drug

The survival of patients with colorectal cancer (CRC) has increased constantly for many years due to superior surgical techniques, improved postoperative care, regular follow-up and an increased use of effective systemic therapy in the adjuvant and the palliative setting [1,2]. All of these advancements are important, but the establishment of multidisciplinary teams which facilitate optimal selection of therapy for individual patients may have been the most important concept on its own.

In recent years a number of biologically active substances attacking specific signalling pathways in cancer cells (targeted therapy) have been developed and included in the treatment of patients with CRC. Three monoclonal antibodies (cetuximab, panitumumab, bevacizumab) have by now been approved for therapy in metastatic CRC (mCRC) [2,3].

Angiogenesis is necessary in tumour development and controlled in part by the vascular endothelial growth system which is inhibited by bevacizumab (Avastin ®) and many other anti-angiogenic drugs.

Cetuximab (Erbitux®) and panitumumab (Vectibix®) block the extracellular portion of the epidermal growth factor receptor (EGFR) and these two drugs will be discussed in detail in this chapter.

#### **2. Targeted therapy - Inhibition of EGFR**

EGFR is a trans-membrane glycoprotein that is involved in signaling pathways affecting cellular growth, differentiation, and proliferation and EGFR is expressed in many types of normal tissues. The EGFR is up regulated in a large number of cancers, in CRC in 60-80% of cases, and might be associated to a poor prognosis. Once a ligand binds to the extracellular domain of EGFR, receptor-dimerization occurs and down-stream signaling cascades are activated. Amongst the downstream effectors are the RAF/MEK/MAPK pathway and the PI3K/PTEN/AKT pathway.

Two anti-EGFR monoclonal antibodies are approved by US Food and Drug Administration and European Medicines Agency for the treatment of CRC – cetuximab and panitumumab. Both are directed against the ligand-binding site of EGFR and competitively inhibiting ligand-induced activation, and thereby inhibiting EGFR induced cell growth, survival, and

Anti-EGFR Treatment in Patients with Colorectal Cancer 247

Adjuvant fluoropyrimidine-based chemotherapy for 6 months is standard of care in patients with radical resected stage III colon cancer, whereas it is more controversial in patients with stage II colon cancer. Modest but definite benefit of 4-5 % in 5-years survival has been

Three large phase III trials have documented that addition of oxaliplatin to fluoropyrimidine (FOLFOX, FLOX, XELOX) are superior compared to single agent fluoropyrimidine in terms of disease-free survival and overall survival in stage III patients

Cetuximab has been tested in the adjuvant setting. The US Intergroup N0147 study assessed the potential benefit of cetuximab added to adjuvant FOLFOX after resection in patients with colon cancer stage III. The primary end point was 3 year disease-free survival. The initial concept was to treat patients regardless of KRAS status, but when the impact of KRAS status in the metastatic setting was established NO147 was amended to include only patients with KRAS wild type tumours. In 717 patients with KRAS mutations included before amendment both 3-year disease-free survival (FOLFOX: 75.8% versus FOLFOX + cetuximab: 72.3%) and 3-year survival (88.0% versus 80.4%) favoured FOLFOX alone [28]. Surprisingly, the addition of cetuximab to FOLFOX in the KRAS wild-type population did not add any benefit as well , with a 3-year disease-free survival of 75.8% in the FOLFOX arm

The FOxTROT trial is presently evaluating a neo-adjuvant strategy with oxaliplatin-based chemotherapy with or without panitumumab in patients with high-risk but resectable colon

Data from ongoing or completed adjuvant trials are awaited, but currently cetuximab,

Since the introduction of 5-fluorouracil in 1957, numerous well-conducted phase III studies have proven its efficacy and even nowadays fluoropyrimidine is the backbone of systemic therapy [30,31]. The era of modern combination therapy started when it was shown that irinotecan prolonged survival in patients with fluoropyrimidine-resistant disease. Since then irinotecan, oxaliplatin, and two oral formulations of 5-fluorouracil (capecitabine and uftoral)

Combination chemotherapy with fluoropyrimidine and irinotecan (e.g. FOLFIRI or XELIRI) or oxaliplatin (e.g. FOLFOX or XELOX) produces tumour regression in approximately half of patients with mCRC. PFS is prolonged from 6 to 9 months and the use of several sequential lines of chemotherapy has improved median survival from 6 months to more

When planning the treatment strategy for an individual patient in the daily clinic it is important to realize the goal of treatment – is there a possibility for cure or is the treatment of palliative character – which depends on the resectability of the metastases and on patient-

panitumumab and other targeted therapies should not be used outside clinical trials.

**4.1 Adjuvant chemotherapy after radical resection for colon cancer** 

**4.2 Adjuvant targeted therapy after radical resection for colon cancer** 

demonstrated in pooled analyses and in the Quasar study [21-24].

and probably also in high-risk stage II [25-27].

versus 72.3% in the FOLFOX-cetuximab arm [29].

**5. Systemic treatment of metastatic colorectal cancer** 

have been approved [32] and are used in the routine clinical practise.

related factors as performance status and co-morbidity.

cancer.

than 18 months.

**4. Adjuvant therapy** 

proliferation. In addition, cetuximab may act by inducing an antibody-dependent cellmediated cytotoxicity reaction as cetuximab is an IgG1 antibody.

Cetuximab is a human-murine chimeric monoclonal antibody with terminal half-life around 4-5 days (range 3-7 days) whereas panitumumab is a fully humanized IgG2 monoclonal antibody with terminal half-life around 7 days (range 4-11 days).

#### **3. Predicting efficacy of anti-EGFR therapy**

Unfortunately only a fraction of patients will benefit from the EGFR inhibition and therefore much research is ongoing to identify predictive markers in order to tailor therapy for the individual patient.

Several studies have shown a clear correlation between the severity of skin rash and outcome of therapy of the anti-EGFR antibodies [4-7].

In addition, development of hypomagnesaemia may be a surrogate marker for outcome of therapy [8,9].

Until recently, the development of skin rash during therapy was the most promising predictive factor, but focus has now changed towards assessment of tumour tissue.

Predicting efficacy of anti-EGFR treatment is naturally focused on the EGFR and effectors in the down-stream-signaling pathways in the tumour. Even though EGFR is the target of the anti-EGFR antibodies there is no difference in efficacy in patients with EGFR positive and EGFR negative tumours as assessed by immunohistochemistry [10] and therapy is therefore not restricted to tumours overexpressing EGFR.

KRAS is a member of one of the intra-cellular signal-transduction cascades. If KRAS harbors a mutation (KRASmut), then the growth signal is constitutively activated independently of ligand binding to the extracellular part of the receptor.

KRASmut is found in approximately 40% of mCRC patients [11].The KRAS mutation is an early event during the colorectal adenoma-carcinoma carcinogenic process [12] and thus there is a high concordance between KRAS mutations in the primary tumour and the metastasis [13]. Analyses of clinical trials with anti-EGFR antibodies in mCRC have demonstrated that the KRAS mutational status is central for the effect of anti-EGFR treatment. Patients with KRASmut do not or hardly ever respond to anti-EGFR therapy and progression-free survival (PFS) and survival is definitely shorter than in patients with KRAS wildtype tumours (KRASwt) [5,11,14-19]. However, recently it has been suggested that the different KRAS mutations might have different biological potentials, and that patients with codon 13 mutated tumours may be sensitive to therapy with cetuximab [20].

Normal expression and mutation status of other members of the down-stream signaling pathways (e.g. BRAF PTEN and PI3K), are also needed for normal function of the EGFR pathway. In a recent study including more than 600 patients with mCRC treated with cetuximab and irinotecan in the third line setting response rates were as high as 41 % in the population of patients with wildtype KRAS, BRAF, NRAS, and PIK3CA exon 20 compared to 24% in the unselected population [11].

Furthermore many other attempts have been made in order to identify other predictive marker including of expression of ligands to the EGFR, mutations in the EGFR resulting in structural changes in the receptor, expression of other members of the EGFR-family; however currently the K-RAS gene mutational status is the only established marker of sensitivity to panitumumab and cetuximab, and the use of anti-EGFR antibodies should be restricted to patients with KRASwt tumours.

### **4. Adjuvant therapy**

246 Colorectal Cancer – From Prevention to Patient Care

proliferation. In addition, cetuximab may act by inducing an antibody-dependent cell-

Cetuximab is a human-murine chimeric monoclonal antibody with terminal half-life around 4-5 days (range 3-7 days) whereas panitumumab is a fully humanized IgG2 monoclonal

Unfortunately only a fraction of patients will benefit from the EGFR inhibition and therefore much research is ongoing to identify predictive markers in order to tailor therapy for the

Several studies have shown a clear correlation between the severity of skin rash and

In addition, development of hypomagnesaemia may be a surrogate marker for outcome of

Until recently, the development of skin rash during therapy was the most promising

Predicting efficacy of anti-EGFR treatment is naturally focused on the EGFR and effectors in the down-stream-signaling pathways in the tumour. Even though EGFR is the target of the anti-EGFR antibodies there is no difference in efficacy in patients with EGFR positive and EGFR negative tumours as assessed by immunohistochemistry [10] and therapy is therefore

KRAS is a member of one of the intra-cellular signal-transduction cascades. If KRAS harbors a mutation (KRASmut), then the growth signal is constitutively activated independently of

KRASmut is found in approximately 40% of mCRC patients [11].The KRAS mutation is an early event during the colorectal adenoma-carcinoma carcinogenic process [12] and thus there is a high concordance between KRAS mutations in the primary tumour and the metastasis [13]. Analyses of clinical trials with anti-EGFR antibodies in mCRC have demonstrated that the KRAS mutational status is central for the effect of anti-EGFR treatment. Patients with KRASmut do not or hardly ever respond to anti-EGFR therapy and progression-free survival (PFS) and survival is definitely shorter than in patients with KRAS wildtype tumours (KRASwt) [5,11,14-19]. However, recently it has been suggested that the different KRAS mutations might have different biological potentials, and that patients with

Normal expression and mutation status of other members of the down-stream signaling pathways (e.g. BRAF PTEN and PI3K), are also needed for normal function of the EGFR pathway. In a recent study including more than 600 patients with mCRC treated with cetuximab and irinotecan in the third line setting response rates were as high as 41 % in the population of patients with wildtype KRAS, BRAF, NRAS, and PIK3CA exon 20 compared

Furthermore many other attempts have been made in order to identify other predictive marker including of expression of ligands to the EGFR, mutations in the EGFR resulting in structural changes in the receptor, expression of other members of the EGFR-family; however currently the K-RAS gene mutational status is the only established marker of sensitivity to panitumumab and cetuximab, and the use of anti-EGFR antibodies should be

codon 13 mutated tumours may be sensitive to therapy with cetuximab [20].

predictive factor, but focus has now changed towards assessment of tumour tissue.

mediated cytotoxicity reaction as cetuximab is an IgG1 antibody.

antibody with terminal half-life around 7 days (range 4-11 days).

**3. Predicting efficacy of anti-EGFR therapy** 

outcome of therapy of the anti-EGFR antibodies [4-7].

not restricted to tumours overexpressing EGFR.

to 24% in the unselected population [11].

restricted to patients with KRASwt tumours.

ligand binding to the extracellular part of the receptor.

individual patient.

therapy [8,9].

#### **4.1 Adjuvant chemotherapy after radical resection for colon cancer**

Adjuvant fluoropyrimidine-based chemotherapy for 6 months is standard of care in patients with radical resected stage III colon cancer, whereas it is more controversial in patients with stage II colon cancer. Modest but definite benefit of 4-5 % in 5-years survival has been demonstrated in pooled analyses and in the Quasar study [21-24].

Three large phase III trials have documented that addition of oxaliplatin to fluoropyrimidine (FOLFOX, FLOX, XELOX) are superior compared to single agent fluoropyrimidine in terms of disease-free survival and overall survival in stage III patients and probably also in high-risk stage II [25-27].

#### **4.2 Adjuvant targeted therapy after radical resection for colon cancer**

Cetuximab has been tested in the adjuvant setting. The US Intergroup N0147 study assessed the potential benefit of cetuximab added to adjuvant FOLFOX after resection in patients with colon cancer stage III. The primary end point was 3 year disease-free survival. The initial concept was to treat patients regardless of KRAS status, but when the impact of KRAS status in the metastatic setting was established NO147 was amended to include only patients with KRAS wild type tumours. In 717 patients with KRAS mutations included before amendment both 3-year disease-free survival (FOLFOX: 75.8% versus FOLFOX + cetuximab: 72.3%) and 3-year survival (88.0% versus 80.4%) favoured FOLFOX alone [28]. Surprisingly, the addition of cetuximab to FOLFOX in the KRAS wild-type population did not add any benefit as well , with a 3-year disease-free survival of 75.8% in the FOLFOX arm versus 72.3% in the FOLFOX-cetuximab arm [29].

The FOxTROT trial is presently evaluating a neo-adjuvant strategy with oxaliplatin-based chemotherapy with or without panitumumab in patients with high-risk but resectable colon cancer.

Data from ongoing or completed adjuvant trials are awaited, but currently cetuximab, panitumumab and other targeted therapies should not be used outside clinical trials.

### **5. Systemic treatment of metastatic colorectal cancer**

Since the introduction of 5-fluorouracil in 1957, numerous well-conducted phase III studies have proven its efficacy and even nowadays fluoropyrimidine is the backbone of systemic therapy [30,31]. The era of modern combination therapy started when it was shown that irinotecan prolonged survival in patients with fluoropyrimidine-resistant disease. Since then irinotecan, oxaliplatin, and two oral formulations of 5-fluorouracil (capecitabine and uftoral) have been approved [32] and are used in the routine clinical practise.

Combination chemotherapy with fluoropyrimidine and irinotecan (e.g. FOLFIRI or XELIRI) or oxaliplatin (e.g. FOLFOX or XELOX) produces tumour regression in approximately half of patients with mCRC. PFS is prolonged from 6 to 9 months and the use of several sequential lines of chemotherapy has improved median survival from 6 months to more than 18 months.

When planning the treatment strategy for an individual patient in the daily clinic it is important to realize the goal of treatment – is there a possibility for cure or is the treatment of palliative character – which depends on the resectability of the metastases and on patientrelated factors as performance status and co-morbidity.

Anti-EGFR Treatment in Patients with Colorectal Cancer 249

months and that OS is prolonged from around 5 months to 9 months, in patients treated

patients

RR (%)

BSC ? 285 0 1.8 4.6 Cet ? 287 7\* 1.9\* 6.1\*

BSC WT 113 0 1.9 4.8 Cet WT 117 13\* 3.8\* 9.5\*

BSC ? 232 0 1.7 6.5 Pan + BSC ? 231 10\* 1.8\* 6.5

BSC WT 119 0 1.7 7.6 Pan + BSC WT 124 17\* 2.8\* 8.1

Cet ? 111 11 1.5 6.9 Cet + Iri ? 218 23\* 4.1\* 8.5

Iri WT 43\* 5.5\* 13.2\*

+ Iri WT 23\* 5.5\* 12.1\*

Iri ? 650 4 2.6 10.0

Cet + Iri ? 648 16\* 4.0\* 10.7

FOLFIRI WT 294 10 3.9 12.5

Pan WT 303 35\* 5.9\* 14.5

Pan MUT 238 13 5.9 11.8

FOLFIRI MUT 248 14 4.9 11.1

281

165

Median PFS (months)

0 2.7 8.0

3 3.9 7.9

Median OS (months)

unaided by KRAS status.

**Third line therapy**

Jonker et al NEJM 2007

Karapetis et al NEJM 2008

Van Cutsem JCO 2007

Amado et al JCO 2008

Di Fiore et al ASCO 2008

Jensen et al ASCO 2010

*EPIC*  Sobrero et al JCO 2008

*181* 

Peeters et al JCO 2010

**Second line therapy**

Cunningham et al NEJM 2004

Author. year Regimen KRAS No of

Weekly Cet +

Weekly Cet +

Biweekly Cet

Biweekly Cet

FOLFIRI +

FOLFIRI +

panitumumab) in patients with chemo-resistent mCRC.

Table 1**.** Selected studies evaluating efficacy of EGFR-inhibition (cetuximab or

Iri MUT

+ Iri MUT

Patients with mCRC may be grouped according to the resectability of their metastases: resectable at diagnosis and initially unresectable. Patients with initially unresectable mCRC can be further subdivided into two groups: potential resectable mCRC which may become resectable after tumour shrinkage and non-resectable which is defined as unresectable despite major tumour regression [33]. This classification has to be done in a close collaboration between surgeons, oncologists, radiologist and pathologist – in a multidisciplinary team. For patients with non-resectable mCRC therapy is primarily of palliative character.

In patients with potential resectable or symptomatic mCRC, tumour shrinkage is absolutely mandatory and therefore the most effective combination should be used as initial therapy. However, in patients with unresectable mCRC AND no tumour-related symptoms a sequential approach (single agent immediately followed by combination therapy upon progression) seems to be a safe strategy.

Targeted therapy enhance efficacy of chemotherapy but should be restricted to selected patients.

#### **6. EGFR inhibition in patients with chemoresistent mCRC**

There are no established cytotoxic drugs or combination in the third-line settings after progression to irinotecan, oxaliplatin and fluoropyrimidine, but this changed dramatically when efficacy of EGFR inhibition was proven in patients with chemo-resistant mCRC [2,3]. Data are summarized in Table 1.

The promising activity observed in phase I and II studies was first confirmed in the pivotal BOND study [34] where 329 patients with irinotecan-resistant mCRC were randomised to receive either weekly single agent cetuximab alone or cetuximab in combination with irinotecan. This combination significantly increased response rate from 11% to 23% and prolonged PFS from 1.5 months to 4.1 months. Survival was not significantly prolonged, perhaps due to cross-over and use of combination therapy as salvage therapy. As a result of the BOND study, cetuximab was approved for patients with irinotecan-resistant disease in US and Europe in 2004.

One of the criticisms of the BOND study was the lack of a control group and therefore NCIC-CO.17 was planned and completed [35]. Patients pre-treated with irinotecan and oxaliplatin were randomised to receive best supportive care (BSC – no crossover upon progression) or cetuximab monotherapy. Compared to BSC, cetuximab prolonged OS from 4.6 months to 6.1 months (Table 1).

In a parallel study, a similar benefit in terms of response and PFS was established for panitumumab [6]. In contrast to NCIC-CO.17, OS was not significantly prolonged perhaps due to the possibility of cross-over to panitumumab after progression in patients randomized to BSC. Based on these data, panitumumab was approved for monotherapy of refractory mCRC by the US Food and Drug Administration in September 2006 and conditionally approved in patients with tumours harbouring wild-type KRAS by the European Medicines Agency in December 2007. Presently there are more data on the combination of irinotecan and cetuximab as salvage therapy but it may be expected that efficacy of irinotecan and panitumumab will be comparable. Indirectly these data suggested that irinotecan with cetuximab (and perhaps irinotecan with panitumumab) increase response rate to more than 20%, prolong PFS from less than 2 months to more than 4

Patients with mCRC may be grouped according to the resectability of their metastases: resectable at diagnosis and initially unresectable. Patients with initially unresectable mCRC can be further subdivided into two groups: potential resectable mCRC which may become resectable after tumour shrinkage and non-resectable which is defined as unresectable despite major tumour regression [33]. This classification has to be done in a close collaboration between surgeons, oncologists, radiologist and pathologist – in a multidisciplinary team. For patients with non-resectable mCRC therapy is primarily of

In patients with potential resectable or symptomatic mCRC, tumour shrinkage is absolutely mandatory and therefore the most effective combination should be used as initial therapy. However, in patients with unresectable mCRC AND no tumour-related symptoms a sequential approach (single agent immediately followed by combination therapy upon

Targeted therapy enhance efficacy of chemotherapy but should be restricted to selected

There are no established cytotoxic drugs or combination in the third-line settings after progression to irinotecan, oxaliplatin and fluoropyrimidine, but this changed dramatically when efficacy of EGFR inhibition was proven in patients with chemo-resistant mCRC [2,3].

The promising activity observed in phase I and II studies was first confirmed in the pivotal BOND study [34] where 329 patients with irinotecan-resistant mCRC were randomised to receive either weekly single agent cetuximab alone or cetuximab in combination with irinotecan. This combination significantly increased response rate from 11% to 23% and prolonged PFS from 1.5 months to 4.1 months. Survival was not significantly prolonged, perhaps due to cross-over and use of combination therapy as salvage therapy. As a result of the BOND study, cetuximab was approved for patients with irinotecan-resistant disease in

One of the criticisms of the BOND study was the lack of a control group and therefore NCIC-CO.17 was planned and completed [35]. Patients pre-treated with irinotecan and oxaliplatin were randomised to receive best supportive care (BSC – no crossover upon progression) or cetuximab monotherapy. Compared to BSC, cetuximab prolonged OS from

In a parallel study, a similar benefit in terms of response and PFS was established for panitumumab [6]. In contrast to NCIC-CO.17, OS was not significantly prolonged perhaps due to the possibility of cross-over to panitumumab after progression in patients randomized to BSC. Based on these data, panitumumab was approved for monotherapy of refractory mCRC by the US Food and Drug Administration in September 2006 and conditionally approved in patients with tumours harbouring wild-type KRAS by the European Medicines Agency in December 2007. Presently there are more data on the combination of irinotecan and cetuximab as salvage therapy but it may be expected that efficacy of irinotecan and panitumumab will be comparable. Indirectly these data suggested that irinotecan with cetuximab (and perhaps irinotecan with panitumumab) increase response rate to more than 20%, prolong PFS from less than 2 months to more than 4

**6. EGFR inhibition in patients with chemoresistent mCRC** 

palliative character.

patients.

progression) seems to be a safe strategy.

Data are summarized in Table 1.

US and Europe in 2004.

4.6 months to 6.1 months (Table 1).

months and that OS is prolonged from around 5 months to 9 months, in patients treated unaided by KRAS status.


Table 1**.** Selected studies evaluating efficacy of EGFR-inhibition (cetuximab or panitumumab) in patients with chemo-resistent mCRC.

Anti-EGFR Treatment in Patients with Colorectal Cancer 251

patients

RR (%)

FOLFIRI WT 350 40 8.4 20.0 FOLFIRI+Cet WT 316 57\* 9.9\* 23.5\* FOLFIRI MUT 183 36 7.7 16.7 FOLFIRI+Cet MUT 214 31 7.4 16.2

FOLFOX WT 331 48 8.0 19.7 FOLFOX+Pan WT 325 55 9.6\* 23.9 FOLFOX MUT 219 40 8.8\* 19.3\* FOLFOX+Pan MUT 221 40 7.3 15.5

FLOX WT 97 47 8.7 20.1 FLOX + Cet WT 97 46 7.9 22.0 FLOX MUT 58 40 7.8 20.4 FLOX + Cet MUT 72 49 9.2 21.1

"Ox" WT 367 50 8.6 17.9 "Ox"+Cet WT 362 59\* 8.6 17.0 "Ox" MUT 268 41 6.9 14.8 "Ox"+ Cet MUT 297 40 6.5 13.6

FOLFOX WT 97 34 7.2 18.5 FOLFOX + Cet WT 82 57\* 8.3\* 22.8 FOLFOX MUT 59 53\* 8.6\* 17.5 FOLFOX + Cet MUT 77 34 5.5 13.4

FOLFIRI WT 294 10 3.9 12.5 FOLFIRI+Pan WT 303 35\* 5.9\* 14.5 FOLFIRI MUT 248 14 4.9 11.1 FOLFIRI+Pan MUT 238 13 5.9 11.8

Table 2. Recent studies evaluating EGFR-inhibition (cetuximab or panitumumab) as first line

Median PFS (months)

Median OS (months)

Author. year Regimen KRAS No of

**First line therapy**

*CRYSTAL*  van Cutsem et al NEJM 2009 & JCO 2011

*PRIME*  Douillard et al JCO 2010

*NORDIC 7*  Tveit et al ASCO GI 2011

*COIN* 

*OPUS* 

*181* 

Peeters et al ECCO 2009

Maughan et al Lancet Oncol 2011

Bokemeyer et al Ann Oncol 2011

**Second line therapy** 

therapy according to KRAS-status.

In the second line setting, the EPIC and "181" studies (Tables 1) showed that irinotecan + cetuximab or FOLFIRI + panitumumab, respectively, significantly increased response rate. PFS was prolonged significantly in both studies but the higher response rate and longer PFS did not translate into an improvement in OS [36,37].

#### **7. EGFR inhibition in patients with chemo-naïve mCRC**

Excellent efficacy in patients with chemo-resistent mCRC started logically a number of phase II studies for chemotherapy-cetuximab or panitumumab combinations with response rates as high as 80%, high liver resection rates and long survival [38]. As a consequence of these promising data, phase III studies were planned and conducted (Table 2). All published randomized trials were initiated and started before the importance of KRAS was known and therefore these studies have included patients with both KRASwt and KRASmut. As described efficacy of EGFR monoclonal antibodies is restricted to patients with KRASwt however for comparison, data on patients with KRASmut are included in Table 2 but only data on KRASwt will be discussed.

Most trials combining anti-EGFR treatment with chemotherapy confirmed a much higher response rate (absolutely 10-20% difference) in the combination arm and most trials also showed that PFS was prolonged absolutely 1-2 months but this difference was not as long as anticipated or hoped. However, at this time only one phase III study could confirm that the benefit in response and PFS was translated to a significant and clinical meaningful improvement in survival [19].

In the CRYSTAL study more than 1200 patients with EGFR-expressing mCRC were randomised to FOLFIRI or FOLFIRI + cetuximab [19]. The investigators managed to collect tumour tissue and analyze KRAS status in an astonishing 89% of all patients. Response rate and resection rate was significantly higher and both median PFS (8.4 vs. 9.9 months) and median survival were significantly prolonged (20.0 vs. 23.5 months). A higher response rate and longer PFS were also observed in the OPUS [39] and PRIME [40] studies. In PRIME, median survival was non-significantly prolonged (19.7 vs. 23.9 months) at the same level as in the CRYSTAL trial. In the large COIN study only response rate was increased [41] and in the smaller NORDIC VII trial cetuximab did not improve efficacy of the Nordic bolus regimen [42].

Since addition of cetuximab or panitumumab improve response rate to combination chemotherapy, there has been a particular interest in the use of these agents in patients with potential resectable liver-only metastasis if it was anticipated that a major response could lead to potentially curative surgery. In the CELIM study, a randomized phase II trial with 111 patients with unresectable liver- metastasis, patients were randomized to FOLFOX + cetuximab or FOLFIRI + cetuximab [43]. In these selected patients the R0 resection rate was impressing 38% and 30%, respectively, which show the importance of selecting and evaluation patients at a multidisciplinary conference but also that the patients should receive the most effective systemic therapy to enhance the chance for curative surgery. In a retrospective analysis of response by KRAS status, a partial or complete response was noted in 70% of patients with KRASwt.

When cetuximab or panitumumab is chosen for patients with KRASwt, it must be concluded, that the chemotherapy combination should be carefully selected. A combination of fluoropyrimidine with oxaliplatin and cetuximab seem to have no or less additional

In the second line setting, the EPIC and "181" studies (Tables 1) showed that irinotecan + cetuximab or FOLFIRI + panitumumab, respectively, significantly increased response rate. PFS was prolonged significantly in both studies but the higher response rate and longer PFS

Excellent efficacy in patients with chemo-resistent mCRC started logically a number of phase II studies for chemotherapy-cetuximab or panitumumab combinations with response rates as high as 80%, high liver resection rates and long survival [38]. As a consequence of these promising data, phase III studies were planned and conducted (Table 2). All published randomized trials were initiated and started before the importance of KRAS was known and therefore these studies have included patients with both KRASwt and KRASmut. As described efficacy of EGFR monoclonal antibodies is restricted to patients with KRASwt however for comparison, data on patients with KRASmut are included in Table 2 but only

Most trials combining anti-EGFR treatment with chemotherapy confirmed a much higher response rate (absolutely 10-20% difference) in the combination arm and most trials also showed that PFS was prolonged absolutely 1-2 months but this difference was not as long as anticipated or hoped. However, at this time only one phase III study could confirm that the benefit in response and PFS was translated to a significant and clinical meaningful

In the CRYSTAL study more than 1200 patients with EGFR-expressing mCRC were randomised to FOLFIRI or FOLFIRI + cetuximab [19]. The investigators managed to collect tumour tissue and analyze KRAS status in an astonishing 89% of all patients. Response rate and resection rate was significantly higher and both median PFS (8.4 vs. 9.9 months) and median survival were significantly prolonged (20.0 vs. 23.5 months). A higher response rate and longer PFS were also observed in the OPUS [39] and PRIME [40] studies. In PRIME, median survival was non-significantly prolonged (19.7 vs. 23.9 months) at the same level as in the CRYSTAL trial. In the large COIN study only response rate was increased [41] and in the smaller NORDIC VII trial cetuximab did not improve efficacy of the Nordic bolus

Since addition of cetuximab or panitumumab improve response rate to combination chemotherapy, there has been a particular interest in the use of these agents in patients with potential resectable liver-only metastasis if it was anticipated that a major response could lead to potentially curative surgery. In the CELIM study, a randomized phase II trial with 111 patients with unresectable liver- metastasis, patients were randomized to FOLFOX + cetuximab or FOLFIRI + cetuximab [43]. In these selected patients the R0 resection rate was impressing 38% and 30%, respectively, which show the importance of selecting and evaluation patients at a multidisciplinary conference but also that the patients should receive the most effective systemic therapy to enhance the chance for curative surgery. In a retrospective analysis of response by KRAS status, a partial or complete response was noted

When cetuximab or panitumumab is chosen for patients with KRASwt, it must be concluded, that the chemotherapy combination should be carefully selected. A combination of fluoropyrimidine with oxaliplatin and cetuximab seem to have no or less additional

did not translate into an improvement in OS [36,37].

data on KRASwt will be discussed.

improvement in survival [19].

in 70% of patients with KRASwt.

regimen [42].

**7. EGFR inhibition in patients with chemo-naïve mCRC** 


Table 2. Recent studies evaluating EGFR-inhibition (cetuximab or panitumumab) as first line therapy according to KRAS-status.

Anti-EGFR Treatment in Patients with Colorectal Cancer 253

In the PACCE study more than 1000 patients were randomised to a combination of chemotherapy with bevacizumab (optional oxaliplatin-based regimen (n = 823) or irinotecan-based regimen (n = 230) with or without panitumumab [45]. The four-drug combination of oxaliplatin-based therapy with bevacizumab and panitumumab resulted in several serious adverse events and also a shorter PFS and survival, while there was no significant difference in efficacy data in the smaller group where therapy was based on irinotecan. Even in patients with KRAS wild-type there was evidence of a harmful effect of

In the CAIRO-2 study, 734 patients were randomised to XELOX + bevacizumab with or without cetuximab. Similar to PACCE study, PFS was significantly shorter in patients receiving double targeted therapy and subgroup analysis of patients with KRAS mutations

Double targeted therapy against angiogenesis and EGFR should not be used as first line

Cetuximab is approved as weekly administration with an initial loading dose of 400 mg/m2 followed by weekly administration of 250 mg/m2. However, as most cytotoxic regimens are administered in two-weeks (or longer) schedules it would be more convenient if cetuximab could be administered as a two-week schedule. Based on a study showing that there is no major differences in the pharmacokinetics and pharmacodynamics between the standard weekly cetuximab schedule and cetuximab 500 mg/m2 given every second week [47,48] a simplified biweekly administration schedule of cetuximab has been developed [47-49]. The biweekly regimen has efficacy and safety profile similar to the weekly schedule [49-52] and ongoing studies will prospectively evaluate the biweekly regimen (www.clinicaltrial.org NCT00660582). In many institutions the biweekly schedule is used in the daily clinical

Panitumumab is administered as an intravenous infusion at 6 mg/kg every 14 days or 9

Toxicity of the anti-EGFR therapy is related to the blockade of the EGFR in the normal tissue. The most often reported side-effect is a papulo-pustular rash primarily in the seborrheic areas seen in up to 90% of patients [6]. The onset is usually within the first three weeks after start of therapy and with spontaneous improvement within the next 4–5 weeks [53]. Most cases are mild to moderate but severe in 5% to 20% of patients [6,7,34,35,54]. Prophylactic treatment with systemic tetracyclines reduces the severity of skin reactions but not the incidence of rash [55-57].Other dermatological reactions are xerosis, fissures of palm and foot, paronychia and extensive growth of both eyelashes and eyebrows [53,54,58,59], but these side-effects are primarily seen after many months of exposure to anti-EGFR

Furthermore, cetuximab and panitumumab may induce severe hypomagnesaemia in as many as 25% of patients, but fortunately it is seldom symptomatic. Hypomagnesaemia results from inhibition of the EGFR in the kidneys - particularly in the ascending limb of the

showed that efficacy (response, PFS and survival) was significant worse [46].

double targeted therapy.

treatment outside of controlled studies.

**8. Weekly or biweekly cetuximab** 

setting based on the above-mentioned experiences.

mg/kg every 3 weeks. There is no loading dose.

**9. Toxicity of anti-EGFR therapy** 

therapy.

benefit over chemotherapy alone [41,42] and presently capecitabine or bolus 5-fluorouracil in combination with oxaliplatin can not be recommended outside clinical trials (Table 2). Until otherwise proven, cetuximab or panitumumab should be combined with FOLFIRI or FOLFOX.

#### **7.1 Combinations of targeted therapies**

In vitro studies have shown that simultaneous inhibition of angiogenesis and EGFR systems have additive and perhaps even synergistic effect, but surprisingly this benefit could not be confirmed in first line randomised studies (Table 3).

In a small randomised phase II study, a triple-combination of cetuximab, irinotecan and bevacizumab was more effective than cetuximab + bevacizumab alone [44]. Even more interesting, PFS and survival for the triple-combination were considerably longer than the historical double-combination in the BOND1 trial [34]. It was therefore expected that a similar combination would increase efficacy also as first line therapy.


Abbreviations in the tables:

RR = response rate, PFS = progression free survival, OS = overall survival, BSC = best supportive care, cet = cetuximab, pan = panitumumab, WT = wildtype, mut = mutant, iri = irinotecan, ox = oxaliplatin, bev = bevazicumab

Table 3. Recent studies evaluating double targeted therapy (inhibition of angiogenesis and EGFR) according to KRAS-status.

benefit over chemotherapy alone [41,42] and presently capecitabine or bolus 5-fluorouracil in combination with oxaliplatin can not be recommended outside clinical trials (Table 2). Until otherwise proven, cetuximab or panitumumab should be combined with FOLFIRI or

In vitro studies have shown that simultaneous inhibition of angiogenesis and EGFR systems have additive and perhaps even synergistic effect, but surprisingly this benefit could not be

In a small randomised phase II study, a triple-combination of cetuximab, irinotecan and bevacizumab was more effective than cetuximab + bevacizumab alone [44]. Even more interesting, PFS and survival for the triple-combination were considerably longer than the historical double-combination in the BOND1 trial [34]. It was therefore expected that a

patients

CapOx+Bev WT 156 50 10.6 22.4

CapOx+Bev MUT 108 59\* 12.5\* 24.9\*

"Ox"+Bev WT 203 56 11.5\* 24.5\*

"Ox"+Bev MUT 125 44 11.0 19.3

"Ir"+Bev WT 58 48 12.5 19.8

"Ir"+Bev MUT 39 38 11.9 20.5

CapOx+Bev+Cet WT 158 61 10.5 21.8

CapOx+Bev+Cet MUT 98 46 8.3 17.2

"Ox"+Bev+Pan WT 201 50 9.8 20.7

"Ox"+Bev+Pan MUT 135 47 10.5 19.3

"Ir"+Bev+Pan WT 57 54 10.0 NR

"Ir"+Bev+Pan MUT 47 30 8.3 17.8

RR = response rate, PFS = progression free survival, OS = overall survival, BSC = best supportive care, cet = cetuximab, pan = panitumumab, WT = wildtype, mut = mutant, iri = irinotecan, ox = oxaliplatin,

Table 3. Recent studies evaluating double targeted therapy (inhibition of angiogenesis and

RR (%)

Median PFS (months)

Median OS (months)

FOLFOX.

*CAIRO2*  Tol et al NEJM 2009

*PACCE*  Hecht et JCO 2009

Abbreviations in the tables:

EGFR) according to KRAS-status.

bev = bevazicumab

**7.1 Combinations of targeted therapies** 

confirmed in first line randomised studies (Table 3).

Author. year Regimen KRAS No of

similar combination would increase efficacy also as first line therapy.

In the PACCE study more than 1000 patients were randomised to a combination of chemotherapy with bevacizumab (optional oxaliplatin-based regimen (n = 823) or irinotecan-based regimen (n = 230) with or without panitumumab [45]. The four-drug combination of oxaliplatin-based therapy with bevacizumab and panitumumab resulted in several serious adverse events and also a shorter PFS and survival, while there was no significant difference in efficacy data in the smaller group where therapy was based on irinotecan. Even in patients with KRAS wild-type there was evidence of a harmful effect of double targeted therapy.

In the CAIRO-2 study, 734 patients were randomised to XELOX + bevacizumab with or without cetuximab. Similar to PACCE study, PFS was significantly shorter in patients receiving double targeted therapy and subgroup analysis of patients with KRAS mutations showed that efficacy (response, PFS and survival) was significant worse [46].

Double targeted therapy against angiogenesis and EGFR should not be used as first line treatment outside of controlled studies.

#### **8. Weekly or biweekly cetuximab**

Cetuximab is approved as weekly administration with an initial loading dose of 400 mg/m2 followed by weekly administration of 250 mg/m2. However, as most cytotoxic regimens are administered in two-weeks (or longer) schedules it would be more convenient if cetuximab could be administered as a two-week schedule. Based on a study showing that there is no major differences in the pharmacokinetics and pharmacodynamics between the standard weekly cetuximab schedule and cetuximab 500 mg/m2 given every second week [47,48] a simplified biweekly administration schedule of cetuximab has been developed [47-49]. The biweekly regimen has efficacy and safety profile similar to the weekly schedule [49-52] and ongoing studies will prospectively evaluate the biweekly regimen (www.clinicaltrial.org NCT00660582). In many institutions the biweekly schedule is used in the daily clinical setting based on the above-mentioned experiences.

Panitumumab is administered as an intravenous infusion at 6 mg/kg every 14 days or 9 mg/kg every 3 weeks. There is no loading dose.

#### **9. Toxicity of anti-EGFR therapy**

Toxicity of the anti-EGFR therapy is related to the blockade of the EGFR in the normal tissue. The most often reported side-effect is a papulo-pustular rash primarily in the seborrheic areas seen in up to 90% of patients [6]. The onset is usually within the first three weeks after start of therapy and with spontaneous improvement within the next 4–5 weeks [53]. Most cases are mild to moderate but severe in 5% to 20% of patients [6,7,34,35,54]. Prophylactic treatment with systemic tetracyclines reduces the severity of skin reactions but not the incidence of rash [55-57].Other dermatological reactions are xerosis, fissures of palm and foot, paronychia and extensive growth of both eyelashes and eyebrows [53,54,58,59], but these side-effects are primarily seen after many months of exposure to anti-EGFR therapy.

Furthermore, cetuximab and panitumumab may induce severe hypomagnesaemia in as many as 25% of patients, but fortunately it is seldom symptomatic. Hypomagnesaemia results from inhibition of the EGFR in the kidneys - particularly in the ascending limb of the

Anti-EGFR Treatment in Patients with Colorectal Cancer 255

[8] Vincenzi B, Galluzzo S, Santini D et al. Early magnesium modifications as a surrogate

[10] Chung KY, Shia J, Kemeny NE et al. Cetuximab Shows Activity in Colorectal Cancer

[11] De RW, Claes B, Bernasconi D et al. Effects of KRAS, BRAF, NRAS, and PIK3CA

[12] Vogelstein B, Fearon ER, Hamilton SR et al. Genetic Alterations during Colorectal-Tumor Development. New England Journal of Medicine 1988; 319: 525-532. [13] Knijn N, Mekenkamp LJM, Klomp M et al. KRAS mutation analysis: a comparison

[14] Amado RG, Wolf M, Peeters M et al. Wild-Type KRAS Is Required for Panitumumab

[15] De Roock W, Piessevaux H, De Schutter J et al. KRAS wild-type state predicts survival

[16] Di Fiore F, Blanchard F, Charbonnier F et al. Clinical relevance of KRAS mutation

[17] Karapetis CS, Khambata-Ford S, Jonker DJ et al. K-ras Mutations and Benefit from Cetuximab in Advanced Colorectal Cancer. N Engl J Med 2008; 359: 1757-1765. [18] Liévre A, Bachet JB, Boige Vr et al. KRAS Mutations As an Independent Prognostic

[19] Van Cutsem E, Kohne CH, Láng I et al. Cetuximab Plus Irinotecan, Fluorouracil, and

[20] De Roock W, Jonker DJ, Di Nicolantonio F et al. Association of KRAS p.G13D Mutation

[21] Figueredo A, Coombes ME, Mukherjee S. Adjuvant therapy for completely resected

[22] Gill S, Loprinzi CL, Sargent DJ et al. Pooled Analysis of Fluorouracil-Based Adjuvant

[23] Gray R, Barnwell J, McConkey C et al. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet 2007; 370: 2020-2029.

Cancer Treated With Cetuximab. JAMA 2010; 304: 1812-1820.

stage II colon cancer. Cochrane Database Syst Rev 2008; CD005390.

advanced colorectal cancer patients. Ann Oncol 2011; 22: 1141-1146. [9] Vincenzi B, Santini D, Tonini G. Biological interaction between anti-epidermal growth

by Immunohistochemistry. J Clin Oncol 2005; 23: 1803-1810.

9: 1267-1269.

1634.

Oncol 2010; 11: 753-762.

patients. Br J Cancer 2011; 104: 1020-1026.

Br J Cancer 2007; 96: 1166-1169.

Clin Oncol 2008; 26: 374-379.

Oncol 2004; 22: 1797-1806.

Status. J Clin Oncol 2011; 29: 2011-2019.

treated with cetuximab. Ann Oncol 2008; 19: 508-515.

marker of efficacy of cetuximab-based anticancer treatment in KRAS wild-type

factor receptor agent cetuximab and magnesium. Expert Opin Pharmacother 2008;

Patients With Tumors That Do Not Express the Epidermal Growth Factor Receptor

mutations on the efficacy of cetuximab plus chemotherapy in chemotherapyrefractory metastatic colorectal cancer: a retrospective consortium analysis. Lancet

between primary tumours and matched liver metastases in 305 colorectal cancer

Efficacy in Patients With Metastatic Colorectal Cancer. J Clin Oncol 2008; 26: 1626-

and is associated to early radiological response in metastatic colorectal cancer

detection in metastatic colorectal cancer treated by Cetuximab plus chemotherapy.

Factor in Patients With Advanced Colorectal Cancer Treated With Cetuximab. J

Leucovorin As First-Line Treatment for Metastatic Colorectal Cancer: Updated Analysis of Overall Survival According to Tumor KRAS and BRAF Mutation

With Outcome in Patients With Chemotherapy-Refractory Metastatic Colorectal

Therapy for Stage II and III Colon Cancer: Who Benefits and by How Much? J Clin

loop of Henle. The hypomagnesaemia may be corrected by oral or IV supplements [60,61]. Anti-EGFR antibody therapy may as well cause nausea and diarrhea due to affection of the EFGR in the gastro-intestinal tract [34].

In addition, administration of chimeric antibodies also may give rise to severe allergic reactions in 1.4-4.5% [3]. The incidence of infusion reactions is reduced by the prophylactic use of antihistamines and corticosteroids as premedication [62]. No study has compared side effects of cetuximab and panitumumab, but cross-trial comparison shows that the spectrum of side effects is similar. However, as panitumumab is a human antibody anaphylactic reactions are rarely seen with panitumumab, and treatment with panitumumab does not require premedication [63]. A switch to panitumumab may be used after severe hypersensitivity reaction to cetuximab [64-66].

#### **10. Conclusion**

Optimal therapy of patients with CRC has increased in complexity with the introduction of targeted therapies, but unfortunately our expectations for these new drugs have not quite been settled. The largest benefits have been achieved with modern chemotherapy, which remains the backbone of treatment of patients with mCRC. However, targeted therapy has clinically significant effect, but we must learn to identify the correct regimes for the right patients. KRAS status is currently the most important predictive marker for efficacy of anti-EGFR therapy. To ensure the optimal treatment strategy, every patient with mCRC must be assessed by a multidisciplinary team.

#### **11. References**


loop of Henle. The hypomagnesaemia may be corrected by oral or IV supplements [60,61]. Anti-EGFR antibody therapy may as well cause nausea and diarrhea due to affection of the

In addition, administration of chimeric antibodies also may give rise to severe allergic reactions in 1.4-4.5% [3]. The incidence of infusion reactions is reduced by the prophylactic use of antihistamines and corticosteroids as premedication [62]. No study has compared side effects of cetuximab and panitumumab, but cross-trial comparison shows that the spectrum of side effects is similar. However, as panitumumab is a human antibody anaphylactic reactions are rarely seen with panitumumab, and treatment with panitumumab does not require premedication [63]. A switch to panitumumab may be used after

Optimal therapy of patients with CRC has increased in complexity with the introduction of targeted therapies, but unfortunately our expectations for these new drugs have not quite been settled. The largest benefits have been achieved with modern chemotherapy, which remains the backbone of treatment of patients with mCRC. However, targeted therapy has clinically significant effect, but we must learn to identify the correct regimes for the right patients. KRAS status is currently the most important predictive marker for efficacy of anti-EGFR therapy. To ensure the optimal treatment strategy, every patient with mCRC must be

[1] Meyerhardt JA, Mayer RJ. Systemic Therapy for Colorectal Cancer. N Engl J Med 2005;

[2] Pfeiffer P, Qvortrup C, Eriksen JG. Current role of antibody therapy in patients with

[3] Pfeiffer P, Qvortrup C, Bjerregaard JK. Current status of treatment of metastatic

[4] Lenz HJ, Van Cutsem E, Khambata-Ford S et al. Multicenter Phase II and Translational

patients receiving panitumumab monotherapy. Cancer 2009; 115: 1544-1554. [6] Van Cutsem E, Siena S, Humblet Y et al. An open-label, single-arm study assessing

[7] Van Cutsem E, Peeters M, Siena S et al. Open-Label Phase III Trial of Panitumumab Plus

Oxaliplatin, and Fluoropyrimidines. J Clin Oncol 2006; 24: 4914-4921. [5] Peeters M, Siena S, Van Cutsem E et al. Association of progression-free survival, overall

refractory to standard chemotherapy. Ann Oncol 2008; 19: 92-98.

colorectal cancer with special reference to cetuximab and elderly patients. Onco

Study of Cetuximab in Metastatic Colorectal Carcinoma Refractory to Irinotecan,

survival, and patient-reported outcomes by skin toxicity and KRAS status in

safety and efficacy of panitumumab in patients with metastatic colorectal cancer

Best Supportive Care Compared With Best Supportive Care Alone in Patients With Chemotherapy-Refractory Metastatic Colorectal Cancer. J Clin Oncol 2007; 25: 1658-

metastatic colorectal cancer. Oncogene 2007; 26: 3661-3678.

EFGR in the gastro-intestinal tract [34].

assessed by a multidisciplinary team.

Targets Ther 2009; 2: 17-27.

352: 476-487.

1664.

**10. Conclusion** 

**11. References** 

severe hypersensitivity reaction to cetuximab [64-66].


Anti-EGFR Treatment in Patients with Colorectal Cancer 257

[40] Douillard JY, Siena S, Cassidy J et al. Randomized, Phase III Trial of Panitumumab With

Colorectal Cancer: The PRIME Study. J Clin Oncol 2010; 28: 4697-4705. [41] Maughan TS, Adams RA, Smith CG et al. Addition of cetuximab to oxaliplatin-based

ASCO GI 2011. Abstr 365

Cancer 2008; 99: 455-458.

with platinum. Gastric Cancer 2011.

Academy of Dermatology 2006; 55: 657-670.

25: 4557-4561.

Infusional Fluorouracil, Leucovorin, and Oxaliplatin (FOLFOX4) Versus FOLFOX4 Alone As First-Line Treatment in Patients With Previously Untreated Metastatic

first-line combination chemotherapy for treatment of advanced colorectal cancer: results of the randomised phase 3 MRC COIN trial. Lancet 2011; 377: 2103-2114. [42] Tveit,K, Guren,T, Glimelius,B et al Randomized phase III study of 5-fluorouracil/

folinate/oxaliplatin given continuously or intermittently with or without cetuximab, as first-line treatment of metastatic colorectal cancer: The NORDIC VII study (NCT00145314), by the Nordic Colorectal Cancer Biomodulation Group.

resectability of colorectal liver metastases following neoadjuvant chemotherapy with cetuximab: the CELIM randomised phase 2 trial. Lancet Oncol 2010; 11: 38-47.

Bevacizumab, and Irinotecan Compared With Cetuximab and Bevacizumab Alone in Irinotecan-Refractory Colorectal Cancer: The BOND-2 Study. J Clin Oncol 2007;

Bevacizumab, and Panitumumab Compared With Chemotherapy and Bevacizumab

week to patients with metastatic colorectal cancer: a two-part pharmacokinetic/ pharmacodynamic phase I dose-escalation study. Ann Oncol 2010; 21: 1537-1545. [48] Tabernero J, Pfeiffer P, Cervantes A. Administration of Cetuximab Every 2 Weeks in the

Treatment of Metastatic Colorectal Cancer: An Effective, More Convenient

line therapy in patients with advanced colorectal cancer after failure to irinotecan,

in patients with metastatic colorectal cancer refractory to 5-FU, oxaliplatin, and

irinotecan in advanced colorectal cancer patients progressing after at least one previous line of chemotherapy: results of a phase II single institution trial. Br J

as second-line therapy in patients with gastro-esophageal cancer previously treated

with the epidermal growth factor receptor inhibitors. Journal of the American

[43] Folprecht G, Gruenberger T, Bechstein WO et al. Tumour response and secondary

[44] Saltz LB, Lenz HJ, Kindler HL et al. Randomized Phase II Trial of Cetuximab,

[45] Hecht JR, Mitchell E, Chidiac T et al. A Randomized Phase IIIB Trial of Chemotherapy,

[47] Tabernero J, Ciardiello F, Rivera F et al. Cetuximab administered once every second

[50] Jensen,BV, Schou,JV, Johannesen,HH et al Cetuximab every second week with irinotecan

[52] Schoennemann KR, Bjerregaard JK, Hansen TP et al. Biweekly cetuximab and irinotecan

[53] Agero AL, Dusza SW, Benvenuto-Andrade C et al. Dermatologic side effects associated

Alone for Metastatic Colorectal Cancer. J Clin Oncol 2009; 27: 672-680. [46] Tol J, Koopman M, Cats A et al. Chemotherapy, Bevacizumab, and Cetuximab in

Alternative to Weekly Administration? Oncologist 2008; 13: 113-119. [49] Pfeiffer P, Nielsen D, Bjerregaard J et al. Biweekly cetuximab and irinotecan as third-

irinotecan: KRAS mutation status and efficacy. ASCO 2010. abstr 3573 [51] Martin-Martorell P, Rosello S, Rodriguez-Braun E et al. Biweekly cetuximab and

oxaliplatin and 5-fluorouracil. Ann Oncol 2008; 19: 1141-1145.

Metastatic Colorectal Cancer. N Engl J Med 2009; 360: 563-572.


[24] Sargent D, Sobrero A, Grothey A et al. Evidence for Cure by Adjuvant Therapy in

[25] Andre T, Boni C, Navarro M et al. Improved Overall Survival With Oxaliplatin,

[26] Haller DG, Tabernero J, Maroun J et al. Capecitabine Plus Oxaliplatin Compared With

[27] Kuebler JP, Wieand HS, O'Connell MJ et al. Oxaliplatin Combined With Weekly Bolus

[30] Ragnhammar P, Hafstrom L, Nygren P et al. A Systematic Overview of Chemotherapy

[31] Sobrero AF, Aschele C, Bertino JR. Fluorouracil in colorectal cancer--a tale of two drugs: implications for biochemical modulation. J Clin Oncol 1997; 15: 368-381. [32] Pfeiffer P, Rasmussen F. A Multidisciplinary Approach to the Treatment of Colorectal

[33] Poston GJ, Figueras J, Giuliante F et al. Urgent Need for a New Staging System in

[34] Cunningham D, Humblet Y, Siena S et al. Cetuximab Monotherapy and Cetuximab plus

[35] Jonker DJ, O'Callaghan CJ, Karapetis CS et al. Cetuximab for the Treatment of

[36] Peeters M, Price TJ, Cervantes As et al. Randomized Phase III Study of Panitumumab

[37] Sobrero AF, Maurel J, Fehrenbacher L et al. EPIC: Phase III Trial of Cetuximab Plus

[38] Macarulla T, Ramos FJ, Elez E et al. Update on novel strategies to optimize cetuximab

[39] Bokemeyer C, Bondarenko I, Hartmann JT et al. Efficacy according to biomarker status

Irinotecan in Irinotecan-Refractory Metastatic Colorectal Cancer. N Engl J Med

With Fluorouracil, Leucovorin, and Irinotecan (FOLFIRI) Compared With FOLFIRI Alone As Second-Line Treatment in Patients With Metastatic Colorectal Cancer. J

Irinotecan After Fluoropyrimidine and Oxaliplatin Failure in Patients With

therapy in patients with metastatic colorectal cancer. Clin Colorectal Cancer 2008; 7:

of cetuximab plus FOLFOX-4 as first-line treatment for metastatic colorectal cancer:

Liver Metastases Is Mandatory. Acta Radiologica 2009; 50: 707-708.

Advanced Colorectal Cancer. J Clin Oncol 2008; 26: 4828-4833.

Metastatic Colorectal Cancer. J Clin Oncol 2008; 26: 2311-2319.

Colorectal Cancer. N Engl J Med 2007; 357: 2040-2048.

the OPUS study. Ann Oncol 2011; 22: 1535-1546.

Effects in Colorectal Cancer. Acta Oncol 2001; 40: 282-308.

on 18 Randomized Trials. J Clin Oncol 2009; 27: 872-877.

Clin Oncol 2011; 29: 1465-1471.

2011. abstr 3607

2004; 351: 337-345.

300-308.

Clin Oncol 2010; 28: 4706-4713.

Cancer in the MOSAIC Trial. J Clin Oncol 2009; 27: 3109-3116.

Colon Cancer: Observations Based on Individual Patient Data From 20,898 Patients

Fluorouracil, and Leucovorin As Adjuvant Treatment in Stage II or III Colon

Fluorouracil and Folinic Acid As Adjuvant Therapy for Stage III Colon Cancer. J

Fluorouracil and Leucovorin As Surgical Adjuvant Chemotherapy for Stage II and III Colon Cancer: Results From NSABP C-07. J Clin Oncol 2007; 25: 2198-2204. [28] Goldberg,R, Sargent,D, Thibodeau,SN et al Adjuvant mFOLFOX6 plus or minus

cetuximab (Cmab) in patients (pts) with KRAS mutant (m) resected stage III colon cancer (CC): NCCTG Intergroup Phase III Trial N0147. ASCO 2010. abstr 3508 [29] Alberts,S, Thibodeau,SN, Sargent,D et al Influence of KRAS and BRAF mutational

status and rash on disease-free survival (DFS) in patients with resected stage III colon cancer receiving cetuximab (Cmab): Results from NCCTG N0147. ASCO


**14** 

*France* 

**Pharmacogenetics and Pharmacogenomics** 

Colorectal cancer (CRC) remains one of the most deadly diseases in the western world, and starts to become a concern in developed countries (Labianca et al., 2010). However, significant steps have been made recently in CRC therapy. Until the 80's, 5-fluorouracil (5- FU) was the only available drug to treat patients, with limited efficacy. Today, 4 cytotoxic agents (5-FU associated with folinic acid, capecitabine, oxaliplatin, irinotecan) and three monoclonal antibodies (cetuximab, panitumumab, bevacizumab) are available, mostly as part of combinations (Koutras et al., 2011). In particular, the rise of targeted therapies in digestive oncology has fueled a new hope by significantly stretching the therapeutic options available so far. Despite these improvements, treatment of metastatic CRC (mCRC) remains a challenging task, and it is acknowledged now that although improving response rates, the introduction of the latest targeted therapies only marginally impacts on either progression free survival (PFS) or overall survival (OS) of mCRC patients. Because of the cost of these new therapies, identifying biomarkers likely to sort patients on their ability to benefit or not, from these new drugs is paradigmatic of the current trend to move towards a more personalized medicine in oncology. Because genetic variability is one of the main factor regulating efficacy and toxicity of most anticancer agents, addressing the issues of pharmacogenomics and pharmacogenetics (PGx) in CRC patient becomes critical, far beyond the only use of costly targeted therapies. Although often used interchangeably, the term "pharmacogenetics" refers historically to inherited changes in genes coding for drug metabolizing enzymes or membrane transporters, thus impacting on the pharmacokinetic (PK) profile and exposure levels eventually, whereas "pharmacogenomics" is a broader definition encompassing genetic changes at the tumor level potentially affecting drug response (Amstutz et al., 2011). Whether they are somatic or found in the germline, all these mutations can potentially have deleterious impacts on the clinical outcome of patients with CRC cancer. At the tumor level, genetic changes affecting the expression of pharmacological

**1. Introduction** 

**of Colorectal Cancer: Moving Towards** 

**Personalized Medicine** 

Joseph Ciccolini1,2,3, Fréderic Fina2,3,

*1APHM, University-Hospital Timone,* 

*2APHM, University-Hospital Nord,* 

L'Houcine Ouafik2,3 and Bruno Lacarelle1,3

*3Aix-Marseille Univ, CRO2 UMR 911, Marseille,* 

*Laboratoire de Pharmacocinétique et Toxicocinétique, Marseille,* 

*Laboratoire de Transfert d'Oncologie Biologique,, Marseille,* 


### **Pharmacogenetics and Pharmacogenomics of Colorectal Cancer: Moving Towards Personalized Medicine**

Joseph Ciccolini1,2,3, Fréderic Fina2,3, L'Houcine Ouafik2,3 and Bruno Lacarelle1,3 *1APHM, University-Hospital Timone, Laboratoire de Pharmacocinétique et Toxicocinétique, Marseille, 2APHM, University-Hospital Nord, Laboratoire de Transfert d'Oncologie Biologique,, Marseille, 3Aix-Marseille Univ, CRO2 UMR 911, Marseille, France* 

#### **1. Introduction**

258 Colorectal Cancer – From Prevention to Patient Care

[54] Saltz LB, Meropol NJ, Loehrer PJ et al. Phase II Trial of Cetuximab in Patients With

[55] Scope A, Agero AL, Dusza SW et al. Randomized Double-Blind Trial of Prophylactic

[56] Jatoi A, Rowland K, Sloan JA et al. Tetracycline to prevent epidermal growth factor

[57] Lacouture ME, Mitchell EP, Piperdi B et al. Skin Toxicity Evaluation Protocol With

[58] Bouché O, Brixi-Benmansour H, Bertin A et al. Trichomegaly of the eyelashes following

[59] Robert C, Soria JC, Spatz A et al. Cutaneous side-effects of kinase inhibitors and

[60] Fakih MG, Wilding G, Lombardo J. Cetuximab-induced hypomagnesemia in patients

[61] Schrag D, Chung KY, Flombaum C et al. Cetuximab Therapy and Symptomatic Hypomagnesemia. Journal of the National Cancer Institute 2005; 97: 1221-1224. [62] Siena S, Glynne-Jones R, Adenis A et al. Reduced incidence of infusion-related reactions

[63] Peeters M, Balfour J, Arnold D. Review article: panitumumab--a fully human anti-EGFR

[64] Cartwright TH, Genther R. Successful administration of panitumumab alone after

[65] Nielsen DL, Pfeiffer P, Jensen BV. Six cases of treatment with panitumumab in patients with severe hypersensitivity reactions to cetuximab. Ann Oncol 2009; 20: 798. [66] Nielsen DL, Pfeiffer P, Jensen BV. Re-treatment with cetuximab in patients with severe

in metastatic colorectal cancer during treatment with cetuximab plus irinotecan with combined corticosteroid and antihistamine premedication. Cancer 2010; 116:

monoclonal antibody for treatment of metastatic colorectal cancer. Aliment

severe infusion reaction to cetuximab in a patient with metastatic colorectal cancer.

hypersensitivity reactions to cetuximab. Two case reports. Acta Oncol 2006; 45:

receptor inhibitor-induced skin rashes. Cancer 2008; 113: 847-853.

treatment with cetuximab. Ann Oncol 2005; 16: 1711-1712.

blocking antibodies. The Lancet Oncology 2005; 6: 491-500.

with colorectal cancer. Clin Colorectal Cancer 2006; 6: 152-156.

Receptor. J Clin Oncol 2004; 22: 1201-1208.

Eruption. J Clin Oncol 2007; 25: 5390-5396.

1827-1837.

1137-1138.

Pharmacol Ther 2008; 28: 269-281.

Clin Colorectal Cancer 2008; 7: 202-203.

Refractory Colorectal Cancer That Expresses the Epidermal Growth Factor

Oral Minocycline and Topical Tazarotene for Cetuximab-Associated Acne-Like

Panitumumab (STEPP), a Phase II, Open-Label, Randomized Trial Evaluating the Impact of a Pre-Emptive Skin Treatment Regimen on Skin Toxicities and Quality of Life in Patients With Metastatic Colorectal Cancer. J Clin Oncol 2010; 28: 1351-1357.

> Colorectal cancer (CRC) remains one of the most deadly diseases in the western world, and starts to become a concern in developed countries (Labianca et al., 2010). However, significant steps have been made recently in CRC therapy. Until the 80's, 5-fluorouracil (5- FU) was the only available drug to treat patients, with limited efficacy. Today, 4 cytotoxic agents (5-FU associated with folinic acid, capecitabine, oxaliplatin, irinotecan) and three monoclonal antibodies (cetuximab, panitumumab, bevacizumab) are available, mostly as part of combinations (Koutras et al., 2011). In particular, the rise of targeted therapies in digestive oncology has fueled a new hope by significantly stretching the therapeutic options available so far. Despite these improvements, treatment of metastatic CRC (mCRC) remains a challenging task, and it is acknowledged now that although improving response rates, the introduction of the latest targeted therapies only marginally impacts on either progression free survival (PFS) or overall survival (OS) of mCRC patients. Because of the cost of these new therapies, identifying biomarkers likely to sort patients on their ability to benefit or not, from these new drugs is paradigmatic of the current trend to move towards a more personalized medicine in oncology. Because genetic variability is one of the main factor regulating efficacy and toxicity of most anticancer agents, addressing the issues of pharmacogenomics and pharmacogenetics (PGx) in CRC patient becomes critical, far beyond the only use of costly targeted therapies. Although often used interchangeably, the term "pharmacogenetics" refers historically to inherited changes in genes coding for drug metabolizing enzymes or membrane transporters, thus impacting on the pharmacokinetic (PK) profile and exposure levels eventually, whereas "pharmacogenomics" is a broader definition encompassing genetic changes at the tumor level potentially affecting drug response (Amstutz et al., 2011). Whether they are somatic or found in the germline, all these mutations can potentially have deleterious impacts on the clinical outcome of patients with CRC cancer. At the tumor level, genetic changes affecting the expression of pharmacological

Pharmacogenetics and Pharmacogenomics

of Colorectal Cancer: Moving Towards Personalized Medicine 261

as for TS expression level in tumors, the actual clinical relevance of these polymorphisms is far from being consensual. Lower response rates have been reported in colorectal cancer patients with the TS 5'-UTR 3R genotypes (ie, TSER\*3), as compared to individuals harboring the homozygous TS 5'-UTR 2R/2R genotype (Salgado et al., 2007). Of note, other groups (Stoehlmacher et al., 2004, Kostopoulos et al., 2009) failed in observing any significant difference in the clinical outcome according to the TS 5'-UTR genotypes, whereas conversely, other authors (Jakobsen et al., 2005; Dotor et al., 2006) found longer survival in carriers of TS 5'- UTR 3R genotypes as compared with those carrying the TS 5'-UTR 2R/2R genotypes. Such conflicting results for predicting outcome from *TYMS* genomic status is not surprising. Several factors such as genetic and epigenetic regulations may interfere with the genotype-tophenotype relationships (Pullmann et al., 2006). For instance, the loss of heterozygosity in tumours at the TS locus may cause the heterozygous TS 5'-UTR 2R/3R risk genotype to acquire either the 2R/loss or the 3R/loss genotype. Consequently, individuals theoretically at risk of treatment failure on the basis of their TS 5'-UTR 2R/3R genomic status may harbor actually the favorable 2R/loss genotype in cancer cells and exhibit higher response eventually when treated with 5-FU (Ruzzo et al., 2007). In addition to target TS, other non-synonymous SNPs (677C>T: MTHFR\*4 and 1298A>C:MTHFR\*6 allelic variants) affecting methylene tetrahydrofolate reductase (MTHFR), one of the key-enzyme involved in the synthesis of reduced folate cofactor, could lead to lack of efficacy when down-regulated (Etienne-Grimaldi et al., 2007, Zintsaras et al., 2009, Braun et al., 2009). However, as for *TYMS*, the actual impact of *MTHFR* genetic polymorphisms on the clinical outcome with 5-FU or 5-FU-derivatives remains to be fully elucidated because inconsistent data have been generated so far (Sharma et al., 2008, Ruzzo et al., 2007). All these contradictory findings with *TYMS* and the associated *MTHFR* genomic status are better understood when one keeps in mind that TS is not the main locus of action of 5-FU. Incorporation into RNA and DNA can be alternative mechanisms of actions for the cytotoxic effects of 5-FU, depending on the way the drug will be metabolized within tumor cells (Ciccolini et al., 2000a). In this respect, the expression levels of activating/deactivating enzymes at the tumor level (eg, orotate phosphoribosyl transferase, thymidine kinase, thymidine phosphorylase, dihydropyrimidine dehydrogenase) have been associated with clinical outcome in patients treated with 5-FU-containing regimen, although once again the data collected so far proved to be rather conflicting (Ciccolini et al., 2004; Soong et al., 2008). For instance, thymidine kinase is implicated both in the activation of 5-FU to active metabolite FdUMP with subsequent theoretical better TS inhibition if highly expressed, and in the *de novo* salvage pathway likely to help cancer cells to survive to 5-FU-induced thymineless stress (Fanciullino et al., 2007). Similarly, thymidine phosphorylase (TP) is involved in the tumoral activation of both 5-FU and capecitabine, but could promote neoangiogenesis too, thus rendering the clinical impact of TP levels in tumors hardly predictable (Ciccolini et al., 2004). Furthermore, deregulation of downstream proteins involved in the transmission of apoptosis in cells exposed to thymineless stress can affect 5-FU or capecitabine antiproliferative efficacy, despite proper inhibition of target TS. Because 5-FU exerts its cytotoxic effects partly through a p53/Fas-dependent apoptotic pathway involving Bax translocation and mitochondrial permeabilization, deregulations affecting each of these steps can interfere with the actual upstream *TYMS* status or the extent of TS inhibition (Borralho et al., 2007). For instance, down-expression of Apo-1 Fas CD95 receptor has been associated with resistance to 5-FU or capecitabine in non-clinical colorectal models, including after that a near-total inhibition of TS activity was achieved (Ciccolini et al., 2000b; 2001).

targets or downstream signaling pathways can lead to treatment failure, as highlighted by the now canonical KRAS mutational status in patients undergoing anti-EGFR therapies. Constitutive mutations are mostly associated with increased toxic risk, as largely publicized by the dihydropyrimidine dehydrogenase (DPD) deficiency syndrome, a condition that puts 5-FU patients at risk of life-threatening toxicities. Of note, when not directly life-threatening, inherited genetic mutations affecting drug disposition in the body and pharmacokinetics can ultimately lead to treatment failure too, because the induced-toxicities often require discontinuation of the treatments until the patient recovers. For all these reasons, developing pharmacogenetic and pharmacogenomic testing in routine clinical practice is now seen as a major issue in oncology.

#### **2. Pharmacogenomics: A matter of life & death at the tumor level**

#### **2.1 Cytotoxics: Why should we not forget that they are targeted therapies too**

Standard care of colorectal cancer includes the use of a variety of cytotoxic agents, used either alone or more frequently as part of combinations (e.g., the canonical Folfiri and Folfox4 regimen). Each of these drugs have their own specific target (e.g., thymidylate synthase for 5-FU, DNA for oxaliplatin, topoisomerase I for irinotecan) and in this respect, numerous studies have focused on the deregulations affecting these targets, either at the genetic or the molecular level, as an attempt to predict treatment efficacy. Indeed, variations in the expression level of the targeted protein, polymorphisms inducing conformation changes, or increase in the repair systems/salvage pathways have been identified as major causes for treatment failure in mCRC patients.

#### **2.1.1 5-FU & Oral 5-FU: The older, the better**

5-FU remains the pivotal drug for treating CRC. Initially used alone, it soon turned to be systematically associated with folinic acid so as to enhance its effect as an antimetabolite, before being introduced as the backbone of several polychemotherapies including irinotecan (a.k.a. Folfiri regimen) or oxaliplatin (a.k.a. Folfox regimen). 5-FU's main target is thymidylate synthase (TS), an enzyme essential to the DNA synthesis and cell replication. Several genetic polymorphisms can affect both *TYMS*, the gene coding for TS, and the folate cycle necessary for the synthesis of methylene tetrahydrofolate, the cofactor required for a complete inhibition of the target through the formation of a stable ternary complex between the enzyme, the cofactor, and fluorodeoxyuridine monophosphate (FdUMP). TS overexpression in tumors is generally associated with resistance to 5-FU treatment, both in vitro and at the bedside (Popat et al., 2004, Lenz et al., 2004). Conversely, another pivotal study has demonstrated that higher TS expression was predictive of higher response with adjuvant fluoropyrimidine (Edler et al., 2002). However, other clinical reports failed in demonstrating such relationship (Locker et al., 2006, Lurje et al., 2009), thus preventing substantial step to be undertaken for implementing screening for TS expression in tumors in routine clinical practice. Variations in TS expression are, at least in part, related to mutations affecting the *TYMS* gene promoter. For instance, the TSER\*3 genotype has been associated with increased mRNA production, thus potentially leading to lower response rates in mCRC patients treated with 5-FU (Uetake et al., 1999). Beside the issue of over-expressing TS tumors likely to resist to 5-FU, constitutive polymorphisms in the 5' and 3'UTRs of the *TYMS* gene responsible for downregulation of TS, have been associated with increased toxicities in patients treated with 5-FU or oral capecitabine (Larguiller et al., 2006). However,

targets or downstream signaling pathways can lead to treatment failure, as highlighted by the now canonical KRAS mutational status in patients undergoing anti-EGFR therapies. Constitutive mutations are mostly associated with increased toxic risk, as largely publicized by the dihydropyrimidine dehydrogenase (DPD) deficiency syndrome, a condition that puts 5-FU patients at risk of life-threatening toxicities. Of note, when not directly life-threatening, inherited genetic mutations affecting drug disposition in the body and pharmacokinetics can ultimately lead to treatment failure too, because the induced-toxicities often require discontinuation of the treatments until the patient recovers. For all these reasons, developing pharmacogenetic and pharmacogenomic testing in routine clinical practice is

**2. Pharmacogenomics: A matter of life & death at the tumor level** 

**2.1 Cytotoxics: Why should we not forget that they are targeted therapies too** 

Standard care of colorectal cancer includes the use of a variety of cytotoxic agents, used either alone or more frequently as part of combinations (e.g., the canonical Folfiri and Folfox4 regimen). Each of these drugs have their own specific target (e.g., thymidylate synthase for 5-FU, DNA for oxaliplatin, topoisomerase I for irinotecan) and in this respect, numerous studies have focused on the deregulations affecting these targets, either at the genetic or the molecular level, as an attempt to predict treatment efficacy. Indeed, variations in the expression level of the targeted protein, polymorphisms inducing conformation changes, or increase in the repair systems/salvage pathways have been identified as major

5-FU remains the pivotal drug for treating CRC. Initially used alone, it soon turned to be systematically associated with folinic acid so as to enhance its effect as an antimetabolite, before being introduced as the backbone of several polychemotherapies including irinotecan (a.k.a. Folfiri regimen) or oxaliplatin (a.k.a. Folfox regimen). 5-FU's main target is thymidylate synthase (TS), an enzyme essential to the DNA synthesis and cell replication. Several genetic polymorphisms can affect both *TYMS*, the gene coding for TS, and the folate cycle necessary for the synthesis of methylene tetrahydrofolate, the cofactor required for a complete inhibition of the target through the formation of a stable ternary complex between the enzyme, the cofactor, and fluorodeoxyuridine monophosphate (FdUMP). TS overexpression in tumors is generally associated with resistance to 5-FU treatment, both in vitro and at the bedside (Popat et al., 2004, Lenz et al., 2004). Conversely, another pivotal study has demonstrated that higher TS expression was predictive of higher response with adjuvant fluoropyrimidine (Edler et al., 2002). However, other clinical reports failed in demonstrating such relationship (Locker et al., 2006, Lurje et al., 2009), thus preventing substantial step to be undertaken for implementing screening for TS expression in tumors in routine clinical practice. Variations in TS expression are, at least in part, related to mutations affecting the *TYMS* gene promoter. For instance, the TSER\*3 genotype has been associated with increased mRNA production, thus potentially leading to lower response rates in mCRC patients treated with 5-FU (Uetake et al., 1999). Beside the issue of over-expressing TS tumors likely to resist to 5-FU, constitutive polymorphisms in the 5' and 3'UTRs of the *TYMS* gene responsible for downregulation of TS, have been associated with increased toxicities in patients treated with 5-FU or oral capecitabine (Larguiller et al., 2006). However,

now seen as a major issue in oncology.

causes for treatment failure in mCRC patients.

**2.1.1 5-FU & Oral 5-FU: The older, the better** 

as for TS expression level in tumors, the actual clinical relevance of these polymorphisms is far from being consensual. Lower response rates have been reported in colorectal cancer patients with the TS 5'-UTR 3R genotypes (ie, TSER\*3), as compared to individuals harboring the homozygous TS 5'-UTR 2R/2R genotype (Salgado et al., 2007). Of note, other groups (Stoehlmacher et al., 2004, Kostopoulos et al., 2009) failed in observing any significant difference in the clinical outcome according to the TS 5'-UTR genotypes, whereas conversely, other authors (Jakobsen et al., 2005; Dotor et al., 2006) found longer survival in carriers of TS 5'- UTR 3R genotypes as compared with those carrying the TS 5'-UTR 2R/2R genotypes. Such conflicting results for predicting outcome from *TYMS* genomic status is not surprising. Several factors such as genetic and epigenetic regulations may interfere with the genotype-tophenotype relationships (Pullmann et al., 2006). For instance, the loss of heterozygosity in tumours at the TS locus may cause the heterozygous TS 5'-UTR 2R/3R risk genotype to acquire either the 2R/loss or the 3R/loss genotype. Consequently, individuals theoretically at risk of treatment failure on the basis of their TS 5'-UTR 2R/3R genomic status may harbor actually the favorable 2R/loss genotype in cancer cells and exhibit higher response eventually when treated with 5-FU (Ruzzo et al., 2007). In addition to target TS, other non-synonymous SNPs (677C>T: MTHFR\*4 and 1298A>C:MTHFR\*6 allelic variants) affecting methylene tetrahydrofolate reductase (MTHFR), one of the key-enzyme involved in the synthesis of reduced folate cofactor, could lead to lack of efficacy when down-regulated (Etienne-Grimaldi et al., 2007, Zintsaras et al., 2009, Braun et al., 2009). However, as for *TYMS*, the actual impact of *MTHFR* genetic polymorphisms on the clinical outcome with 5-FU or 5-FU-derivatives remains to be fully elucidated because inconsistent data have been generated so far (Sharma et al., 2008, Ruzzo et al., 2007). All these contradictory findings with *TYMS* and the associated *MTHFR* genomic status are better understood when one keeps in mind that TS is not the main locus of action of 5-FU. Incorporation into RNA and DNA can be alternative mechanisms of actions for the cytotoxic effects of 5-FU, depending on the way the drug will be metabolized within tumor cells (Ciccolini et al., 2000a). In this respect, the expression levels of activating/deactivating enzymes at the tumor level (eg, orotate phosphoribosyl transferase, thymidine kinase, thymidine phosphorylase, dihydropyrimidine dehydrogenase) have been associated with clinical outcome in patients treated with 5-FU-containing regimen, although once again the data collected so far proved to be rather conflicting (Ciccolini et al., 2004; Soong et al., 2008). For instance, thymidine kinase is implicated both in the activation of 5-FU to active metabolite FdUMP with subsequent theoretical better TS inhibition if highly expressed, and in the *de novo* salvage pathway likely to help cancer cells to survive to 5-FU-induced thymineless stress (Fanciullino et al., 2007). Similarly, thymidine phosphorylase (TP) is involved in the tumoral activation of both 5-FU and capecitabine, but could promote neoangiogenesis too, thus rendering the clinical impact of TP levels in tumors hardly predictable (Ciccolini et al., 2004). Furthermore, deregulation of downstream proteins involved in the transmission of apoptosis in cells exposed to thymineless stress can affect 5-FU or capecitabine antiproliferative efficacy, despite proper inhibition of target TS. Because 5-FU exerts its cytotoxic effects partly through a p53/Fas-dependent apoptotic pathway involving Bax translocation and mitochondrial permeabilization, deregulations affecting each of these steps can interfere with the actual upstream *TYMS* status or the extent of TS inhibition (Borralho et al., 2007). For instance, down-expression of Apo-1 Fas CD95 receptor has been associated with resistance to 5-FU or capecitabine in non-clinical colorectal models, including after that a near-total inhibition of TS activity was achieved (Ciccolini et al., 2000b; 2001).

Pharmacogenetics and Pharmacogenomics

this observation have been found.

**2.2 Biotherapies: Where are my keys?** 

of Colorectal Cancer: Moving Towards Personalized Medicine 263

complex, thus causing double-strand DNA breakage and cell death eventually. Expression levels of target topo-I has been associated with clinical outcome in multivariate analysis performed from large studies including several hundreds of patients undergoing irinotecanbased therapy (Braun et al., 2008; Kostopoulos et al., 2009). However, the lack of randomized, prospective trial prevents, for the time being, the evaluation of Topo-1 level in tumours to be proposed in routine clinical setting as a predictive biomarker for irinotecan efficacy, and little is known about the genetic or epigenetic events affecting the Topo-1 gene likely to modify expression levels of the target protein. However, in the Focus trial, Topo-1 expression level was found to be related to efficacy, although it remains unclear whether the expression level is to be considered as a predictive or a prognostic marker (Braun et al., 2008). In addition, as for oxaliplatin, deregulations affecting DNA-repairing enzymes like *XRCC1*, ERCC1 and *GSTP1* have been found to be predictive of the clinical outcome in irinotecan-treated patients. Polymorphism affecting the *XRCC1* gene (eg, the GGCC-G haplotype) was associated with improved response rates in patients given irinotecan, much probably in relation with loss of ability to repair DNA damage (Hoskins et al., 2008). Conversely, better response and, in some cases, improved PFS was observed in patients undergoing FolFiri regimen with tumors overexpressing GSTP1 and ERCC1 (Vallbohmer et al., 2006). This finding may be confusing because higher expression in DNA-repair enzymes is normally associated with resistance to DNA-targeting agents. Here, high ERCC1 levels could be indicative of a higher DNA damage, thus making the tumor cells more sensitive to Topo-I inhibition by irinotecan. In the same study, EGFR expression was found to be associated too with better response, although to date, no molecular mechanisms underlying

Treatment of colorectal cancer has taken benefit from the rise of the biotherapies in clinical oncology, because both anti-VEGF and anti-EGFR monoclonal antibodies can be used now in association with cytotoxics agents. However, the efficacy of most targeted therapies is

Cetuximab and panitumumab are two anti-Her1 monoclonal antibodies indicated for treating metastatic colorectal cancer. Initially proposed alone, both drugs showed better efficacy and improved survival when combined with standard Folfox4 or Folfiri regimen. Although cetuximab is a chimeric IgG1 and panitumumab a 100% human IgG2, these both antibodies target the extracellular domain of EGFR-1, thus blocking the downstream signaling pathway normally leading to cell proliferation and differenciation, neoangiogenesis and invasion patterns associated with colorectal cancer. Cetuximab and panitumumab prescription is contingent upon the completion of pharmacogenomics testing. Expression level of target EGFR is the first condition, although in clinical practice, the relevance of this test is more and more debated and controversial at the bedside. However, several studies have demonstrated how patients with elevated *EGFR* gene copy number are more likely to respond to cetuximab or panitumumab therapy (Moroni et al., 2005; Sartore-Bianchi et al., 2007, Heinemann et al., 2009). More interestingly and consensual, determination of the mutational status of KRAS soon turned to be the paradigm of implementing pharmacogenomic testing prior to initiating treatment with a targeted

generally contingent upon a number of biomarkers at the tumor level to be checked.

**2.2.1 Anti-EGFR monoclonal antibodies: Why hitting the target is not enough** 

However, subsequent clinical studies failed in demonstrating the role Fas expression could play as a predictive marker in patients with colorectal cancer (Backus et al., 2001; Bezulier et al., 2003).

#### **2.1.2 Oxaliplatin: A metal precious to the patients**

In clinical practice, oxaliplatin is given in mCRC patients in association with 5-FU/folinic acid, a combination known as the Folfox regimen. It can be further combined now with the latest monoclonal antibodies targeting VEGF or EGFR-1. Oxaliplatin is a third-generation platinum derivative that targets complementary DNA strands, thus inducing cell death eventually. However, the nucleotide excision repair (NER) system is designed to remove the oxaliplatin-induced DNA-adducts, and several factors (XPD (a.k.a. ERCC2), XPC and XPA) are implicated in the repair process of DNA helixes once adducts have been formed. In addition, XPG and ERCC1 are implicated in the cleavage of the damaged DNA strand and participate to the repairing pattern of cells exposed to oxaliplatin. Any changes in those repair mechanisms can lead to increase of sensitivity or loss of efficacy in patients. Several genotypes at the tumor level have been associated with clinical outcome in oxaliplatinregimen. In particular, it has been demonstrated that polymorphisms affecting *ERCC1* and *XPD* genes are related to patient survival. For instance, ERCC1-118 T/T, or XPD-751 A/C and C/C genotypes have been associated with reduced disease-free survival in patients treated with oxaliplatin (Ruzzo et al., 2007). In another study, the Lys751Gln polymorphism of the *XPD* gene has been identified as a predictive marker in mCRC patients undergoing FolFox treatment (Le Morvan et al., 2007). Beside the NER, basepair excision repair is also involved in the chemosentivity to oxaliplatin. *XRCC1* gene is affected by several polymorphisms, and expression of the wild-type allele has been associated with better clinical outcome in patients with mCRC (Suh et al., 2006, Stoehlmacher et al., 2001), although subsequent studies failed in confirming the relevance of establishing *XRCC1* genotype as a predictive biomarker with oxaliplatin (Ruzzo et al., 2007). Along with the issue of efficacy, mutations affecting Glutathione-S Transferase (GST), the enzyme responsible for the cell detoxification of oxaliplatin, could have an impact on the clinical outcome with oxaliplatin. Overexpression of tumoral GSTP1 has been found in CRC patients, thus leading to lack of efficacy (Glasgow et al., 2005). However, the exact role the genetic status *GSTP1* plays in patients treated with oxaliplatin remains controversial. For instance, the GSTP1 ile105val genotype has been associated with improved survival in patients treated with Folfox regimen (Stoehlmacher et al., 2002), although the same genotype was predictive of reduced survival in another study (Sun et al., 2005) . In addition, the GSTP1-105 G allele, could explain higher incidence of severe neurotoxicities, the most common side-effect of oxaliplatin, observed in some patients (Ruzzo et al., 2007). Another polymorphism affecting the *AGXT* gene coding for the enzyme responsible for the metabolism of oxalate, which peaks during oxaliplatin infusion, could explain higher risk of neurotoxicity in patients (Gamelin et al., 2007).

#### **2.1.3 Irinotecan: Twist again 'till double-strand DNA breakage**

Irinotecan (CPT-11) is a topoisomerase-I (Topo-1) inhibitor usually combined with 5- FU/folinic acid regimen, an association known as the FolFiri regimen. Topo-1 relieves torsional strain in DNA, thus allowing DNA replication, recombination, and repair. Irinotecan prevents religation of the DNA strand by binding to topoisomerase I-DNA

However, subsequent clinical studies failed in demonstrating the role Fas expression could play as a predictive marker in patients with colorectal cancer (Backus et al., 2001; Bezulier et

In clinical practice, oxaliplatin is given in mCRC patients in association with 5-FU/folinic acid, a combination known as the Folfox regimen. It can be further combined now with the latest monoclonal antibodies targeting VEGF or EGFR-1. Oxaliplatin is a third-generation platinum derivative that targets complementary DNA strands, thus inducing cell death eventually. However, the nucleotide excision repair (NER) system is designed to remove the oxaliplatin-induced DNA-adducts, and several factors (XPD (a.k.a. ERCC2), XPC and XPA) are implicated in the repair process of DNA helixes once adducts have been formed. In addition, XPG and ERCC1 are implicated in the cleavage of the damaged DNA strand and participate to the repairing pattern of cells exposed to oxaliplatin. Any changes in those repair mechanisms can lead to increase of sensitivity or loss of efficacy in patients. Several genotypes at the tumor level have been associated with clinical outcome in oxaliplatinregimen. In particular, it has been demonstrated that polymorphisms affecting *ERCC1* and *XPD* genes are related to patient survival. For instance, ERCC1-118 T/T, or XPD-751 A/C and C/C genotypes have been associated with reduced disease-free survival in patients treated with oxaliplatin (Ruzzo et al., 2007). In another study, the Lys751Gln polymorphism of the *XPD* gene has been identified as a predictive marker in mCRC patients undergoing FolFox treatment (Le Morvan et al., 2007). Beside the NER, basepair excision repair is also involved in the chemosentivity to oxaliplatin. *XRCC1* gene is affected by several polymorphisms, and expression of the wild-type allele has been associated with better clinical outcome in patients with mCRC (Suh et al., 2006, Stoehlmacher et al., 2001), although subsequent studies failed in confirming the relevance of establishing *XRCC1* genotype as a predictive biomarker with oxaliplatin (Ruzzo et al., 2007). Along with the issue of efficacy, mutations affecting Glutathione-S Transferase (GST), the enzyme responsible for the cell detoxification of oxaliplatin, could have an impact on the clinical outcome with oxaliplatin. Overexpression of tumoral GSTP1 has been found in CRC patients, thus leading to lack of efficacy (Glasgow et al., 2005). However, the exact role the genetic status *GSTP1* plays in patients treated with oxaliplatin remains controversial. For instance, the GSTP1 ile105val genotype has been associated with improved survival in patients treated with Folfox regimen (Stoehlmacher et al., 2002), although the same genotype was predictive of reduced survival in another study (Sun et al., 2005) . In addition, the GSTP1-105 G allele, could explain higher incidence of severe neurotoxicities, the most common side-effect of oxaliplatin, observed in some patients (Ruzzo et al., 2007). Another polymorphism affecting the *AGXT* gene coding for the enzyme responsible for the metabolism of oxalate, which peaks during oxaliplatin infusion, could explain higher risk of

**2.1.2 Oxaliplatin: A metal precious to the patients** 

neurotoxicity in patients (Gamelin et al., 2007).

**2.1.3 Irinotecan: Twist again 'till double-strand DNA breakage** 

Irinotecan (CPT-11) is a topoisomerase-I (Topo-1) inhibitor usually combined with 5- FU/folinic acid regimen, an association known as the FolFiri regimen. Topo-1 relieves torsional strain in DNA, thus allowing DNA replication, recombination, and repair. Irinotecan prevents religation of the DNA strand by binding to topoisomerase I-DNA

al., 2003).

complex, thus causing double-strand DNA breakage and cell death eventually. Expression levels of target topo-I has been associated with clinical outcome in multivariate analysis performed from large studies including several hundreds of patients undergoing irinotecanbased therapy (Braun et al., 2008; Kostopoulos et al., 2009). However, the lack of randomized, prospective trial prevents, for the time being, the evaluation of Topo-1 level in tumours to be proposed in routine clinical setting as a predictive biomarker for irinotecan efficacy, and little is known about the genetic or epigenetic events affecting the Topo-1 gene likely to modify expression levels of the target protein. However, in the Focus trial, Topo-1 expression level was found to be related to efficacy, although it remains unclear whether the expression level is to be considered as a predictive or a prognostic marker (Braun et al., 2008). In addition, as for oxaliplatin, deregulations affecting DNA-repairing enzymes like *XRCC1*, ERCC1 and *GSTP1* have been found to be predictive of the clinical outcome in irinotecan-treated patients. Polymorphism affecting the *XRCC1* gene (eg, the GGCC-G haplotype) was associated with improved response rates in patients given irinotecan, much probably in relation with loss of ability to repair DNA damage (Hoskins et al., 2008). Conversely, better response and, in some cases, improved PFS was observed in patients undergoing FolFiri regimen with tumors overexpressing GSTP1 and ERCC1 (Vallbohmer et al., 2006). This finding may be confusing because higher expression in DNA-repair enzymes is normally associated with resistance to DNA-targeting agents. Here, high ERCC1 levels could be indicative of a higher DNA damage, thus making the tumor cells more sensitive to Topo-I inhibition by irinotecan. In the same study, EGFR expression was found to be associated too with better response, although to date, no molecular mechanisms underlying this observation have been found.

#### **2.2 Biotherapies: Where are my keys?**

Treatment of colorectal cancer has taken benefit from the rise of the biotherapies in clinical oncology, because both anti-VEGF and anti-EGFR monoclonal antibodies can be used now in association with cytotoxics agents. However, the efficacy of most targeted therapies is generally contingent upon a number of biomarkers at the tumor level to be checked.

#### **2.2.1 Anti-EGFR monoclonal antibodies: Why hitting the target is not enough**

Cetuximab and panitumumab are two anti-Her1 monoclonal antibodies indicated for treating metastatic colorectal cancer. Initially proposed alone, both drugs showed better efficacy and improved survival when combined with standard Folfox4 or Folfiri regimen. Although cetuximab is a chimeric IgG1 and panitumumab a 100% human IgG2, these both antibodies target the extracellular domain of EGFR-1, thus blocking the downstream signaling pathway normally leading to cell proliferation and differenciation, neoangiogenesis and invasion patterns associated with colorectal cancer. Cetuximab and panitumumab prescription is contingent upon the completion of pharmacogenomics testing. Expression level of target EGFR is the first condition, although in clinical practice, the relevance of this test is more and more debated and controversial at the bedside. However, several studies have demonstrated how patients with elevated *EGFR* gene copy number are more likely to respond to cetuximab or panitumumab therapy (Moroni et al., 2005; Sartore-Bianchi et al., 2007, Heinemann et al., 2009). More interestingly and consensual, determination of the mutational status of KRAS soon turned to be the paradigm of implementing pharmacogenomic testing prior to initiating treatment with a targeted

Pharmacogenetics and Pharmacogenomics

numerous parameters affecting each drug.

**3.1 Cytotoxics: Improving the efficacy/toxicity balance** 

loss of chance and poor clinical outcome eventually.

**3.1.1 5-FU & Oral 5-FU** 

of Colorectal Cancer: Moving Towards Personalized Medicine 265

with PFS was found (Loupakis et al., 2011). This latter study illustrates the difficulty in identifying relevant biomarkers for response in heavily treated mCRC patients receiving several drugs in a row, the observed efficacy being the resulting combination of the

Beside those affecting tumors, several constitutive genetic mutations can impact on the disposition of anticancer drugs, especially when they concern genes coding for detoxifying enzymes in the liver. Although for years, such polymorphisms were mostly associated with increased risk of developing severe and sometimes deadly toxicities upon drug intake, they may impact as well on treatment efficacy eventually. Indeed, when they are not directly lifethreatening, drug-induced toxicities and their management often require treatment discontinuation, delays in subsequent radiotherapy courses if scheduled, with a subsequent

Fluoropyrimidines pharmacokinetics is primarily dependent upon an intense liver first pass effect mediated by dihydropyrimidine dehydrogenase (DPD), the enzyme that converts uracil into dihydrouracil. It is generally estimated that about 90-95% of an administered 5- FU dose will be metabolized in the liver before being distributed throughout the body. DPD exhibits a similar pivotal role in the disposition of oral fluoropyrimidines like capecitabine or UFT, all generating 5-FU eventually. *DPYD* gene is highly polymorphic because several dozen of mutations have been described thus far (Van Kuilenburg, 2004). Mutational inactivation of the *DPYD* gene has been characterized as an autosomal recessive disease in Caucasians' population, with probably a higher impact in black American (Mercier C et al., 2006). Genetic and epigenetic regulations, such as promoter hypermethylation or variations in transcriptional factor expression, could play as well a critical role in *DPYD* dysregulations (Etienne MC et al., 1994, Zhang et al., 2006), although this issue remains debated today. Admittedly, three relevant mutations (canonical IV14+1G>A (*DPYD*\*2A), plus 2846A>T, and 1679T>G) should be screened at bedside to anticipate 5-FU-related side effects (Morel et al., 2006). Numerous clinical reports have demonstrated the deleterious effect of DPD genetic polymorphism in patients undergoing 5-FU based regimen. Regardless of the upstream genetic events leading to the loss of enzymatic activity, impaired DPD has been systematically associated with increased risk of developing severe/lethal toxicities upon 5- FU exposure. In a proof-of-concept study, DPD deficiency was retrospectively identified as the culprit for 70% of the severe toxicities and 80% of the toxic-death cases monitored over a two-year observation period, and when performed, drug monitoring confirmed strong overexposure to 5-FU in DPD-deficient individuals (Ciccolini et al., 2006). However, some reports failed in providing data for this pivotal role *DPYD* genetic polymorphism could play in the incidence of severe toxicities with 5-FU. In a gene-candidate study, Schwab et al. have investigated the role several polymorphisms, including the *DPYD*\*2A allelic variant, could play in the tolerance to 5-FU. Surprisingly, this genotype was found to be only marginally associated with toxicities, but it has to be underlined that in this study, no complementary functional investigations were undertaken to evaluate globally the DPD status in those

**3. Pharmacogenetics: When genetics help finding the right exposure** 

therapy. The EGFR/KRAS/Raf pathway is implicated in signal transduction from receptors to the nucleus, thus promoting cell proliferation and differentiation. KRAS transmits signal after binding to guanosine triphosphate (GTP), and becomes inactive when GTP is converted to GDP. Mutations affecting *KRAS* will maintain the protein continuously activated in a switch-on position, even if the upstream receptor is inhibited by a monoclonal antibody. It was demonstrated in the mid-2000's that specific *KRAS* mutations (eg, codons 12/13) was associated with lack of response in cetuximab-treated patients (Lievre et al., 2006). Subsequent studies all confirmed the predictive value of wild-type (WT) *KRAS* for the response with anti-EGFR biotherapies, either cetuximab or panitumumab, regardless of their use as monotherapy or combined with cytotoxics (Heinemann et al., 2009, Asghar et al., 2010). However, WT *KRAS* is a mandatory but no sufficient condition to guarantee an optimal efficacy with anti-EGFR therapies. Mutations affecting *BRaf*, an effector of KRAS, has been associated with treatment failure, although it remains unclear whether *BRaf* mutational status should be used as a prognostic or a predictive marker (Di Nicolantonio et al., 2008). Similarly, correlation was found in cetuximab-treated patients between *EGFR* gene amplification, WT *KRAS* status, PTEN expression, and response. Of note, loss of PTEN expression was systematically associated with treatment failure, thus suggesting that PTEN could be a novel predictive biomarker for anti-EGFR therapies (Frattini et al., 2007). Along with PTEN, several other parameters like epiregulin and amphiregulin expression have been recently identified as putative biomarkers (Jacobs et al., 2009; Laurent-Puig et al., 2009; Di Fiore et al., 2010), although larger prospective studies will be necessary to validate their clinical relevance to predict clinical outcome with EGFR-inhibitors.

#### **2.2.2 Anti-VEGF therapy: Desesperatly seeking biomarkers**

Bevacizumab is the only *stricto-sensu* antiangiogenic therapy approved for treating mCRC patients in association with cytotoxics. This humanized monoclonal antibody targets circulating VEGF-A. To date, no predictive biomarkers have been identified with bevacizumab. Overexpression of VEGF is usually associated with poor survival in mCRC patients, but VEGF level is generally considered as a prognostic, rather than a predictive, biomarker. Even in a prognostic setting, the actual role *VEGF* polymorphism plays remains unclear. For instance, in some studies, the -460CC genotype was found to have a favorable impact on OS in gastric cancer patients (Kim et al., 2007), but deleterious in breast cancer patients (Lu et al., 2005). Beside, some studies in breast cancer patients have found a relationship between *VEGF* polymorphisms (eg, -2578A/A and -1154A/A genotypes) and better survival in patients treated with the paclitaxel + bevacizumab regimen (Schneider et al., 2008). Similar relationship between *VEGF-A* polymorphism and both toxicity and DFS has been evidenced more recently (Etienne-Grimaldi et al., 2010). A similar trend has been found with digestive cancers (Formica et al., 2010). Additionally, circulating PDGF could be implicated in resistance to anti-angiogenic drugs (Crawford et al., 2009), as well as SDF1 and FGF2 factors (Batchelor et al., 2007). Finally, plasma cytokines and vascular factors could be associated with clinical outcome in patients undergoing bevacizumab-based therapy (Kopetz et al., 2010). However in a recent study, Loupakis et al. have investigated the molecular and genetic markers likely to predict efficacy in mCRC patients treated with the Folfoxiri plus bevacizumab quadruple combination. Among the various bevacizumabrelated biomarkers they monitored in plasma (ie VEGF, PIGF, sVEGFR2, TSP-1 plasma level) and the screening of several polymorphims affecting *VEGF* (eg., -2578C/A, -1498C/T, -1154G/A, 936C/T) and *VEGFR-2* (-604A/G, 1192C/T, 1719T/A), little relevant association with PFS was found (Loupakis et al., 2011). This latter study illustrates the difficulty in identifying relevant biomarkers for response in heavily treated mCRC patients receiving several drugs in a row, the observed efficacy being the resulting combination of the numerous parameters affecting each drug.

#### **3. Pharmacogenetics: When genetics help finding the right exposure**

#### **3.1 Cytotoxics: Improving the efficacy/toxicity balance**

Beside those affecting tumors, several constitutive genetic mutations can impact on the disposition of anticancer drugs, especially when they concern genes coding for detoxifying enzymes in the liver. Although for years, such polymorphisms were mostly associated with increased risk of developing severe and sometimes deadly toxicities upon drug intake, they may impact as well on treatment efficacy eventually. Indeed, when they are not directly lifethreatening, drug-induced toxicities and their management often require treatment discontinuation, delays in subsequent radiotherapy courses if scheduled, with a subsequent loss of chance and poor clinical outcome eventually.

#### **3.1.1 5-FU & Oral 5-FU**

264 Colorectal Cancer – From Prevention to Patient Care

therapy. The EGFR/KRAS/Raf pathway is implicated in signal transduction from receptors to the nucleus, thus promoting cell proliferation and differentiation. KRAS transmits signal after binding to guanosine triphosphate (GTP), and becomes inactive when GTP is converted to GDP. Mutations affecting *KRAS* will maintain the protein continuously activated in a switch-on position, even if the upstream receptor is inhibited by a monoclonal antibody. It was demonstrated in the mid-2000's that specific *KRAS* mutations (eg, codons 12/13) was associated with lack of response in cetuximab-treated patients (Lievre et al., 2006). Subsequent studies all confirmed the predictive value of wild-type (WT) *KRAS* for the response with anti-EGFR biotherapies, either cetuximab or panitumumab, regardless of their use as monotherapy or combined with cytotoxics (Heinemann et al., 2009, Asghar et al., 2010). However, WT *KRAS* is a mandatory but no sufficient condition to guarantee an optimal efficacy with anti-EGFR therapies. Mutations affecting *BRaf*, an effector of KRAS, has been associated with treatment failure, although it remains unclear whether *BRaf* mutational status should be used as a prognostic or a predictive marker (Di Nicolantonio et al., 2008). Similarly, correlation was found in cetuximab-treated patients between *EGFR* gene amplification, WT *KRAS* status, PTEN expression, and response. Of note, loss of PTEN expression was systematically associated with treatment failure, thus suggesting that PTEN could be a novel predictive biomarker for anti-EGFR therapies (Frattini et al., 2007). Along with PTEN, several other parameters like epiregulin and amphiregulin expression have been recently identified as putative biomarkers (Jacobs et al., 2009; Laurent-Puig et al., 2009; Di Fiore et al., 2010), although larger prospective studies will be necessary to validate their

Bevacizumab is the only *stricto-sensu* antiangiogenic therapy approved for treating mCRC patients in association with cytotoxics. This humanized monoclonal antibody targets circulating VEGF-A. To date, no predictive biomarkers have been identified with bevacizumab. Overexpression of VEGF is usually associated with poor survival in mCRC patients, but VEGF level is generally considered as a prognostic, rather than a predictive, biomarker. Even in a prognostic setting, the actual role *VEGF* polymorphism plays remains unclear. For instance, in some studies, the -460CC genotype was found to have a favorable impact on OS in gastric cancer patients (Kim et al., 2007), but deleterious in breast cancer patients (Lu et al., 2005). Beside, some studies in breast cancer patients have found a relationship between *VEGF* polymorphisms (eg, -2578A/A and -1154A/A genotypes) and better survival in patients treated with the paclitaxel + bevacizumab regimen (Schneider et al., 2008). Similar relationship between *VEGF-A* polymorphism and both toxicity and DFS has been evidenced more recently (Etienne-Grimaldi et al., 2010). A similar trend has been found with digestive cancers (Formica et al., 2010). Additionally, circulating PDGF could be implicated in resistance to anti-angiogenic drugs (Crawford et al., 2009), as well as SDF1 and FGF2 factors (Batchelor et al., 2007). Finally, plasma cytokines and vascular factors could be associated with clinical outcome in patients undergoing bevacizumab-based therapy (Kopetz et al., 2010). However in a recent study, Loupakis et al. have investigated the molecular and genetic markers likely to predict efficacy in mCRC patients treated with the Folfoxiri plus bevacizumab quadruple combination. Among the various bevacizumabrelated biomarkers they monitored in plasma (ie VEGF, PIGF, sVEGFR2, TSP-1 plasma level) and the screening of several polymorphims affecting *VEGF* (eg., -2578C/A, -1498C/T, -1154G/A, 936C/T) and *VEGFR-2* (-604A/G, 1192C/T, 1719T/A), little relevant association

clinical relevance to predict clinical outcome with EGFR-inhibitors.

**2.2.2 Anti-VEGF therapy: Desesperatly seeking biomarkers** 

Fluoropyrimidines pharmacokinetics is primarily dependent upon an intense liver first pass effect mediated by dihydropyrimidine dehydrogenase (DPD), the enzyme that converts uracil into dihydrouracil. It is generally estimated that about 90-95% of an administered 5- FU dose will be metabolized in the liver before being distributed throughout the body. DPD exhibits a similar pivotal role in the disposition of oral fluoropyrimidines like capecitabine or UFT, all generating 5-FU eventually. *DPYD* gene is highly polymorphic because several dozen of mutations have been described thus far (Van Kuilenburg, 2004). Mutational inactivation of the *DPYD* gene has been characterized as an autosomal recessive disease in Caucasians' population, with probably a higher impact in black American (Mercier C et al., 2006). Genetic and epigenetic regulations, such as promoter hypermethylation or variations in transcriptional factor expression, could play as well a critical role in *DPYD* dysregulations (Etienne MC et al., 1994, Zhang et al., 2006), although this issue remains debated today. Admittedly, three relevant mutations (canonical IV14+1G>A (*DPYD*\*2A), plus 2846A>T, and 1679T>G) should be screened at bedside to anticipate 5-FU-related side effects (Morel et al., 2006). Numerous clinical reports have demonstrated the deleterious effect of DPD genetic polymorphism in patients undergoing 5-FU based regimen. Regardless of the upstream genetic events leading to the loss of enzymatic activity, impaired DPD has been systematically associated with increased risk of developing severe/lethal toxicities upon 5- FU exposure. In a proof-of-concept study, DPD deficiency was retrospectively identified as the culprit for 70% of the severe toxicities and 80% of the toxic-death cases monitored over a two-year observation period, and when performed, drug monitoring confirmed strong overexposure to 5-FU in DPD-deficient individuals (Ciccolini et al., 2006). However, some reports failed in providing data for this pivotal role *DPYD* genetic polymorphism could play in the incidence of severe toxicities with 5-FU. In a gene-candidate study, Schwab et al. have investigated the role several polymorphisms, including the *DPYD*\*2A allelic variant, could play in the tolerance to 5-FU. Surprisingly, this genotype was found to be only marginally associated with toxicities, but it has to be underlined that in this study, no complementary functional investigations were undertaken to evaluate globally the DPD status in those

Pharmacogenetics and Pharmacogenomics

2009).

of Colorectal Cancer: Moving Towards Personalized Medicine 267

exposure levels. For instance, patients harboring the 34A>G SNP on the *ABCG2* gene could be more at risk of treatment failure, as compared with WT patients (Mc Leod et al., 2008). Conversely, other SNPs like the 421C>A polymorphism seems to have limited impact on irinotecan pharmacokinetics and clinical outcome (de jong et al., 2004) whereas some mutations were associated with higher incidence of drug-induced toxicities (Cha et al.,

For years, the importance of pharmacokinetic issues such as residual plasma levels or drug concentrations at the tumor site has been largely underestimated with targeted therapies. For instance, it took 5 years since its first approval in Chronic Myeloid Leukemia to acknowledge the fact that the residual concentrations of imatinib were predictive for the major molecular response in patients, thus highlighting the utility to perform drug monitoring and subsequently developing dose-tailoring strategies to ensure a better efficacy (Egorin et al., 2009). Although similar strategies are now developed with other small molecules such as pazopanib (Suttle et al., 2010), no such trend is currently proposed with monoclonal antibodies, despite the fact that dose/exposure/efficacy and dose/exposure/toxicities relationships have been described (Lu et al., 2009, Keiser et al., 2010). Both non-clinical and clinical studies suggest that 90% of target inhibition should be continuously achieved to ensure a maximum efficacy, thus stressing the usefulness to monitor residual concentrations of monoclonal antibodies such as panitumumab, as for other target therapies (Yang et al., 2010). For instance, plasma residual concentrations of 10- 30 ug/ml are considered necessary with bevacizumab for an optimal efficacy (Data on File Genentech Inc). However, little is known about the pharmacokinetics of monoclonal antibodies and there is a clear lack for markers of inter-patient variability. Proteolytic degradation along with target-mediated drug disposition are the main patterns implicated in the clearance of monoclonal antibodies, and several factors such as antibodies antitherapeutic antibodies, target expression, number of metastatic sites or inflammatory syndromes are likely to modify drug levels in plasma. Of note, genetic polymorphism affecting immunoglobulin G fragment receptor Fc--R has been identified as a putative

**3.2 Pharmacokinetics of targeted therapies: The hidden biomarker?** 

marker for rituximab clearance, but the clinical importance of *Fc-*

the anti-EGFR therapy (Graziano et al., 2008).

**finally do it?** 

more related to the Antibody-Dependent Cell Cytotoxicity (ADCC) of rituximab that involves Fc--R, rather than a pharmacokinetics issue (Cartron et al., 2002). In digestive oncology, Fc--R genotype has been identified in mCRC patients treated with cetuximab as predictive for PFS, but as for rituximab, this could be related to changes in the ADCC described sometimes with cetuximab rather than changes in pharmacokinetics (Zhang et al., 2007), and other studies failed in confirming the impact this polymorphism could have with

**4. Conclusions: One patient, one disease, one drug, one dosage…. Can we** 

Developing strategies to implement personalized medicine in digestive oncology is now an irreversible trend (Ciccolini et al., 2011). However, identifying predictive biomarkers associated with either treatment efficacy or tolerance remains an uneasy task, because CRC patients are usually treated with up to 6 different drugs in combination over several lines.

*-R* genotype could be

patients (Schwab et al., 2008). In addition to 5-FU, several reports have suggested that *DPYD* genetic polymorphism could be an issue with capecitabine too. The very first toxic-death case has been first observed in the late-2000' in a patient treated with capecitabine who was found to be profoundly DPD deficient after post-mortem investigations (Mercier et al., 2007a). Several other clinical reports have demonstrated how *DPYD* genetic polymorphism could put deficient patients at risk of experiencing severe toxicities if given capecitabine (Mercier et al., 2007b). Lastly, another genetic polymorphism could be a rising concern with capecitabine. Deregulations affecting cytidine deaminase (CDA), one of the three enzymes responsible for the conversion of prodrug capecitabine to 5-FU, could lead to severe toxicities. As for DPD, the gene coding for CDA is highly polymorphic with either loss (poor metabolizer) or gain (ultra-metabolizer, UM) of enzymatic activity. The first life-threatening toxicity in a patient displaying the UM phenotype was reported in the late 2000's (Mercier C et al., 2009). The role CDA could play in severe toxicities with capecitabine has been next confirmed in another larger study showing that deletion in the promoter region of the CDA gene with increased transcription was a predictive marker for hand-foot syndrome (Caronia et al., 2011). Lastly, the first toxic-death case in a capecitabine-treated patient harboring several polymorphisms on the CDA gene, including the Caronia deletion, has been published recently (Dahan et al., 2011), thus highlighting the fact that beside *DPYD*, other genetic polymorphisms should be screened to ensure a better safety when handling oral fluoropyrimidines.

#### **3.1.2 Irinotecan**

Irinotecan is a prodrug that can be either metabolized by the Cyp3A sub-family to form the inactive APC derivative, or be converted by carboxylesterase into SN38, a highly cytotoxic metabolite responsible for both the efficacy and the toxicity of irinotecan. SN38 is next mainly detoxified after conjugation by the UGT1A1 to yield inactive SN-38G that will be excreted by the kidneys and the bile eventually. Numerous polymorphisms have been described for the gene coding for UGT1A1, and variations in the promoter region consisting in 7 instead of 6 TA-repeats (UGT1A1\*28) is admittedly associated with increased risk of severe toxicities in mCRC patients administered with high dose (e.g., above 250 mg/m2) irinotecan (Kweekel et al., 2010). A strong influence of ethnicity has been observed with this allelic variant because its population frequency is as high as 43% heterozygotes in the Caucasians but much lower in the Asians (Innocenti et al., 2005; deJong et al., 2006). Several independent studies have demonstrated how individuals with the UGT1A1\*28 genotype were up to 7-time more at risk to experience haematological or gastrointestinal severe toxicities when treated with irinotecan (Ando et al., 2000; Marcuello et al., 2004). Of note, some authors have reported an association between the UGT1A1\*28 genotype and irinotecan efficacy (Toffoli et al., 2006), although other studies have failed in providing evidence for such a relationship (Kweekel et al., 2008). Along with the UGT1A1\*28 genotype, other variations such has the UGT1A1\*6 most frequently found in Asian populations has been associated with increased severe neutropenia after irinotecan intake (Han et al., 2006), although other studies failed in confirming such relationship (Ando et al., 2000). Additionally, polymorphisms affecting transmembrane pumps involved in the excretion of toxic metabolites could be related to drug resistance. Pharmacogenetics of the ATP-binding cassette proteins has been associated with changes in the pharmacokinetics of irinotecan, because they impact of the renal clearance of the drug and ultimatelly on

patients (Schwab et al., 2008). In addition to 5-FU, several reports have suggested that *DPYD* genetic polymorphism could be an issue with capecitabine too. The very first toxic-death case has been first observed in the late-2000' in a patient treated with capecitabine who was found to be profoundly DPD deficient after post-mortem investigations (Mercier et al., 2007a). Several other clinical reports have demonstrated how *DPYD* genetic polymorphism could put deficient patients at risk of experiencing severe toxicities if given capecitabine (Mercier et al., 2007b). Lastly, another genetic polymorphism could be a rising concern with capecitabine. Deregulations affecting cytidine deaminase (CDA), one of the three enzymes responsible for the conversion of prodrug capecitabine to 5-FU, could lead to severe toxicities. As for DPD, the gene coding for CDA is highly polymorphic with either loss (poor metabolizer) or gain (ultra-metabolizer, UM) of enzymatic activity. The first life-threatening toxicity in a patient displaying the UM phenotype was reported in the late 2000's (Mercier C et al., 2009). The role CDA could play in severe toxicities with capecitabine has been next confirmed in another larger study showing that deletion in the promoter region of the CDA gene with increased transcription was a predictive marker for hand-foot syndrome (Caronia et al., 2011). Lastly, the first toxic-death case in a capecitabine-treated patient harboring several polymorphisms on the CDA gene, including the Caronia deletion, has been published recently (Dahan et al., 2011), thus highlighting the fact that beside *DPYD*, other genetic polymorphisms should be screened to ensure a better safety when handling oral

Irinotecan is a prodrug that can be either metabolized by the Cyp3A sub-family to form the inactive APC derivative, or be converted by carboxylesterase into SN38, a highly cytotoxic metabolite responsible for both the efficacy and the toxicity of irinotecan. SN38 is next mainly detoxified after conjugation by the UGT1A1 to yield inactive SN-38G that will be excreted by the kidneys and the bile eventually. Numerous polymorphisms have been described for the gene coding for UGT1A1, and variations in the promoter region consisting in 7 instead of 6 TA-repeats (UGT1A1\*28) is admittedly associated with increased risk of severe toxicities in mCRC patients administered with high dose (e.g., above 250 mg/m2) irinotecan (Kweekel et al., 2010). A strong influence of ethnicity has been observed with this allelic variant because its population frequency is as high as 43% heterozygotes in the Caucasians but much lower in the Asians (Innocenti et al., 2005; deJong et al., 2006). Several independent studies have demonstrated how individuals with the UGT1A1\*28 genotype were up to 7-time more at risk to experience haematological or gastrointestinal severe toxicities when treated with irinotecan (Ando et al., 2000; Marcuello et al., 2004). Of note, some authors have reported an association between the UGT1A1\*28 genotype and irinotecan efficacy (Toffoli et al., 2006), although other studies have failed in providing evidence for such a relationship (Kweekel et al., 2008). Along with the UGT1A1\*28 genotype, other variations such has the UGT1A1\*6 most frequently found in Asian populations has been associated with increased severe neutropenia after irinotecan intake (Han et al., 2006), although other studies failed in confirming such relationship (Ando et al., 2000). Additionally, polymorphisms affecting transmembrane pumps involved in the excretion of toxic metabolites could be related to drug resistance. Pharmacogenetics of the ATP-binding cassette proteins has been associated with changes in the pharmacokinetics of irinotecan, because they impact of the renal clearance of the drug and ultimatelly on

fluoropyrimidines.

**3.1.2 Irinotecan** 

exposure levels. For instance, patients harboring the 34A>G SNP on the *ABCG2* gene could be more at risk of treatment failure, as compared with WT patients (Mc Leod et al., 2008). Conversely, other SNPs like the 421C>A polymorphism seems to have limited impact on irinotecan pharmacokinetics and clinical outcome (de jong et al., 2004) whereas some mutations were associated with higher incidence of drug-induced toxicities (Cha et al., 2009).

#### **3.2 Pharmacokinetics of targeted therapies: The hidden biomarker?**

For years, the importance of pharmacokinetic issues such as residual plasma levels or drug concentrations at the tumor site has been largely underestimated with targeted therapies. For instance, it took 5 years since its first approval in Chronic Myeloid Leukemia to acknowledge the fact that the residual concentrations of imatinib were predictive for the major molecular response in patients, thus highlighting the utility to perform drug monitoring and subsequently developing dose-tailoring strategies to ensure a better efficacy (Egorin et al., 2009). Although similar strategies are now developed with other small molecules such as pazopanib (Suttle et al., 2010), no such trend is currently proposed with monoclonal antibodies, despite the fact that dose/exposure/efficacy and dose/exposure/toxicities relationships have been described (Lu et al., 2009, Keiser et al., 2010). Both non-clinical and clinical studies suggest that 90% of target inhibition should be continuously achieved to ensure a maximum efficacy, thus stressing the usefulness to monitor residual concentrations of monoclonal antibodies such as panitumumab, as for other target therapies (Yang et al., 2010). For instance, plasma residual concentrations of 10- 30 ug/ml are considered necessary with bevacizumab for an optimal efficacy (Data on File Genentech Inc). However, little is known about the pharmacokinetics of monoclonal antibodies and there is a clear lack for markers of inter-patient variability. Proteolytic degradation along with target-mediated drug disposition are the main patterns implicated in the clearance of monoclonal antibodies, and several factors such as antibodies antitherapeutic antibodies, target expression, number of metastatic sites or inflammatory syndromes are likely to modify drug levels in plasma. Of note, genetic polymorphism affecting immunoglobulin G fragment receptor Fc--R has been identified as a putative marker for rituximab clearance, but the clinical importance of *Fc--R* genotype could be more related to the Antibody-Dependent Cell Cytotoxicity (ADCC) of rituximab that involves Fc--R, rather than a pharmacokinetics issue (Cartron et al., 2002). In digestive oncology, Fc--R genotype has been identified in mCRC patients treated with cetuximab as predictive for PFS, but as for rituximab, this could be related to changes in the ADCC described sometimes with cetuximab rather than changes in pharmacokinetics (Zhang et al., 2007), and other studies failed in confirming the impact this polymorphism could have with the anti-EGFR therapy (Graziano et al., 2008).

#### **4. Conclusions: One patient, one disease, one drug, one dosage…. Can we finally do it?**

Developing strategies to implement personalized medicine in digestive oncology is now an irreversible trend (Ciccolini et al., 2011). However, identifying predictive biomarkers associated with either treatment efficacy or tolerance remains an uneasy task, because CRC patients are usually treated with up to 6 different drugs in combination over several lines.

Pharmacogenetics and Pharmacogenomics

**5. References** 

pp.274-81.

of Colorectal Cancer: Moving Towards Personalized Medicine 269

PK/PD/PGx modeling support should help to develop easy-to-implement tools designed to

Amstutz U, & Carleton BC. (2011). Pharmacogenetic testing: time for clinical practice

Ando Y, Saka H, Ando M, Sawa T, Muro K, Ueoka H, Yokoyama A, Saitoh S, Shimokata K, &

Backus HH, Dukers DF, van Groeningen CJ, Vos W, Bloemena E, Wouters D, van Riel JM,

Batchelor T, et al. (2007). AZD 2171, a pan-VEGF receptor tyrosine kinase inhibitor,

Bezulier K, Fina F, Roussel M, Bun SS, Ciccolini J, Martin PM, Milano G, Aubert C, & Barra

Borralho PM, Moreira da Silva IB, Aranha MM, Albuquerque C, Nobre Leitão C, Steer CJ, &

Braun MS, Richman SD, Thompson L, Daly CL, Meade AM, Adlard JW, Allan JM, Parmar

Caronia D, Martin M, Sastre J, de la Torre J, García-Sáenz JA, Alonso MR, Moreno LT, Pita

Cartron G, Dacheux L, Salles G, Solal-Celigny P, Bardos P, Colombat P, & Watier H. (2002).

Cha PC, Mushiroda T, Zembutsu H, Harada H, Shinoda N, Kawamoto S, Shimoyama R,

Ciccolini J, Peillard L, Aubert C, Formento P, Milano G, & Catalin J. (2000a). Monitoring of

fluoropyrimidines? *J Clin Pharm Ther*, Vol. 28, N°5, pp.403-8.

cancer: the FOCUS trial*. J Clin Oncol,* Vol. 27, N°33, pp.5519-28.

syndrome. *Clin Cancer Res,* Vol. 17, N°7, pp.2006-13

myelosuppression. *J Hum Genet,* Vol. 54, N°10, pp.572-80.

study. *Fundam Clin Pharmacol,* Vol. 14, N°2, pp.147-54.

Hasegawa Y. (2000). Polymorphisms of UDP-glucuronosyltransferase gene and irinotecan toxicity: a pharmacogenetic analysis. *Cancer Res*, Vol. 60, N°24, pp.6921-6. Asghar U, Hawkes E, & Cunningham D. (2010) Predictive and prognostic biomarkers for

targeted therapy in metastatic colorectal cancer. *Clin Colorectal Cancer*, Vol. 9, N°5,

Smid K, Giaccone G, Pinedo HM, & Peters GJ. (2011) 5-Fluorouracil induced Fas upregulation associated with apoptosis in liver metastases of colorectal cancer

normalizes tumor vasculature and alleviates edema in glioblastoma patients.

Y. (2003). Fas/FasL expression in tumor biopsies: a prognostic response factor to

Rodrigues CM. (2007). Inhibition of Fas expression by RNAi modulates 5 fluorouracil-induced apoptosis in HCT116 cells expressing wild-type p53. *Biochim* 

MK, Quirke P, & Seymour MT. (2009). Association of molecular markers with toxicity outcomes in a randomized trial of chemotherapy for advanced colorectal

G, Díaz-Rubio E, Benítez J, & González-Neira A. (2006). A polymorphism in the cytidine deaminase promoter predicts severe capecitabine-induced hand-foot

Therapeutic activity of humanized anti-CD20 monoclonal antibody and polymorphism in IgG Fc receptor FcgammaRIIIa gene. *Blood,* Vol. 99, N°3, pp.754-8.

Nishidate T, Furuhata T, Sasaki K, Hirata K, & Nakamura Y. (2009). Single nucleotide polymorphism in ABCG2 is associated with irinotecan-induced severe

the intracellular activation of 5-fluorouracil to deoxyribonucleotides in HT29 human colon cell line: application to modulation of metabolism and cytotoxicity

individualize dosing based upon the patients genotypes or phenotypes.

guidelines. *Clin Pharmacol Ther*, Vol. 89, N°6, pp.924-7.

patients. *Ann Oncol*, Vol. 12, N°2, pp.209-16.

*Cancer Cell,* Vol. 11, N°6, pp.83-95.

*Biophys Acta*, Vol. 1772, N°1, pp.40-7.

Consequently, and despite the abundant literature published, the heterogeneity in the clinical settings can hinder the relevance of some markers, thus preventing standardized guidelines to be issued. However, oncogenetic, pharmacogenetic and pharmacogenomic tools are now developed as a new mean to help oncologists to choose the optimal strategy for each patient, regarding the staging of the disease, the status of the various response markers, and eventually information about specificities in pharmacokinetics and detoxification patterns, once the right drugs have been chosen. However, this later and critical step remains today the forgotten item in routine clinical setting. Although implementing KRAS pharmacogenomic testing is now a systematic practice prior to administrate cetuximab or panitumumab to mCRC patients, little is done to further develop pharmacogenetics-based dose tailoring strategies to reach next the right exposure likely to ensure an optimal efficacy/toxicity balance. Screening for *DPYD* or *UGT1A1* genetic polymorphisms, despite countless clinical reports demonstrating their role in lifethreatening toxicities, and therefore the usefulness of preliminary testing in patients undergoing 5-FU, capecitabine or irinotecan-based regimen to anticipate treatment-related toxicities, is far from being a common practice. However, when performed, pharmacoeconomic studies suggest that implementation of such screening is cost-effective, thus suggesting that routine pharmacogenetics should benefit both to the patients and to the institute ultimately, by dramatically cutting the costs dedicated to managing the treatmentrelated toxicities (Mercier et al., 2009). Of note, no regulatory official step has been undertaken to date to prompt oncologists to require such tests when prescribing cytotoxics to mCRC patients. Changes in the drug label informing physicians about the toxic risks with irinotecan related to the *UGT1A1* genetic polymorphism has been done by the F.D.A in the mid-2000's and official warning issued as a as a level-2 priority, but with little impact in clinical practice, partly because the UGT1A1 test is not reimbursed in the U.S. by most insurance companies (Ikediobi et al., 2009, Meckley et al. 2010), and partly because of the lack of tools to customize the irinotecan dosage once the *UGT1A1* status has been obtained. As of today, it is acknowledged that the UGT1A1\*28 genotype is a concern in patients scheduled for irinotecan dosage above 200 mg/m² only, with little further guidelines made available about adaptive dosing strategies to treat patients harboring this polymorphism (Hoskins et al., 2007) because. Usually, an empirical 25-50% reduction in irinotecan starting dose is recommended in patients with the homozygous variant. Similarly, screening for *DPYD* genetic polymorphism is an exceptional, rather than a routine test in most institutes. DPD testing can be required at best once the severe toxicities have already shown in a patient treated with a 5-FU-containing regimen to keep or discard the fluoropyrimidine in the forthcoming course. As for UGT1A1, little tools are available to tailor dosage based upon the DPD status of the patient. However, a case-control study has demonstrated the immediate advantages patients could benefit from prospective DPD testing associated with adaptive-dosing, with a sharp reduction in the incidence of 5-FU-related toxicities in patients screened for DPD deficiency with tailored dosage as compared with patients treated with standard regimen (Yang et al., 2009). Of note, efficacy remained the same in this study despite markedly lower doses in patients with DPD deficiency, thus illustrating how pharmacogenetics-based adaptive dosing could improve indeed the efficacy/toxicity balance of canonical 5-FU. In this respect, prospective clinical trials investigating pharmacogenetics of drugs given in mCRC patients with strong PK, PK/PD and PK/PD/PGx modeling support should help to develop easy-to-implement tools designed to individualize dosing based upon the patients genotypes or phenotypes.

#### **5. References**

268 Colorectal Cancer – From Prevention to Patient Care

Consequently, and despite the abundant literature published, the heterogeneity in the clinical settings can hinder the relevance of some markers, thus preventing standardized guidelines to be issued. However, oncogenetic, pharmacogenetic and pharmacogenomic tools are now developed as a new mean to help oncologists to choose the optimal strategy for each patient, regarding the staging of the disease, the status of the various response markers, and eventually information about specificities in pharmacokinetics and detoxification patterns, once the right drugs have been chosen. However, this later and critical step remains today the forgotten item in routine clinical setting. Although implementing KRAS pharmacogenomic testing is now a systematic practice prior to administrate cetuximab or panitumumab to mCRC patients, little is done to further develop pharmacogenetics-based dose tailoring strategies to reach next the right exposure likely to ensure an optimal efficacy/toxicity balance. Screening for *DPYD* or *UGT1A1* genetic polymorphisms, despite countless clinical reports demonstrating their role in lifethreatening toxicities, and therefore the usefulness of preliminary testing in patients undergoing 5-FU, capecitabine or irinotecan-based regimen to anticipate treatment-related toxicities, is far from being a common practice. However, when performed, pharmacoeconomic studies suggest that implementation of such screening is cost-effective, thus suggesting that routine pharmacogenetics should benefit both to the patients and to the institute ultimately, by dramatically cutting the costs dedicated to managing the treatmentrelated toxicities (Mercier et al., 2009). Of note, no regulatory official step has been undertaken to date to prompt oncologists to require such tests when prescribing cytotoxics to mCRC patients. Changes in the drug label informing physicians about the toxic risks with irinotecan related to the *UGT1A1* genetic polymorphism has been done by the F.D.A in the mid-2000's and official warning issued as a as a level-2 priority, but with little impact in clinical practice, partly because the UGT1A1 test is not reimbursed in the U.S. by most insurance companies (Ikediobi et al., 2009, Meckley et al. 2010), and partly because of the lack of tools to customize the irinotecan dosage once the *UGT1A1* status has been obtained. As of today, it is acknowledged that the UGT1A1\*28 genotype is a concern in patients scheduled for irinotecan dosage above 200 mg/m² only, with little further guidelines made available about adaptive dosing strategies to treat patients harboring this polymorphism (Hoskins et al., 2007) because. Usually, an empirical 25-50% reduction in irinotecan starting dose is recommended in patients with the homozygous variant. Similarly, screening for *DPYD* genetic polymorphism is an exceptional, rather than a routine test in most institutes. DPD testing can be required at best once the severe toxicities have already shown in a patient treated with a 5-FU-containing regimen to keep or discard the fluoropyrimidine in the forthcoming course. As for UGT1A1, little tools are available to tailor dosage based upon the DPD status of the patient. However, a case-control study has demonstrated the immediate advantages patients could benefit from prospective DPD testing associated with adaptive-dosing, with a sharp reduction in the incidence of 5-FU-related toxicities in patients screened for DPD deficiency with tailored dosage as compared with patients treated with standard regimen (Yang et al., 2009). Of note, efficacy remained the same in this study despite markedly lower doses in patients with DPD deficiency, thus illustrating how pharmacogenetics-based adaptive dosing could improve indeed the efficacy/toxicity balance of canonical 5-FU. In this respect, prospective clinical trials investigating pharmacogenetics of drugs given in mCRC patients with strong PK, PK/PD and


Pharmacogenetics and Pharmacogenomics

*Oncol,* Vol. 12, N°11, pp.2248-53.

N°4, pp.463-70

pp.1139-45.

pp.6359-68.

Vol. 26, N°2, pp.143-51.

*Br J Cancer,* Vol. 92, N°2, pp.259-64.

of Colorectal Cancer: Moving Towards Personalized Medicine 271

Edler D, Glimelius B, Hallström M, Jakobsen A, Johnston PG, Magnusson I, Ragnhammar P,

Egorin MJ, Mauro MJ, & Trent JC.(2009). Drug plasma monitoring in CML and GIST: A case-based discussion. *Clin Adv Hematol Oncol,* Vol. 7, N°11-S1, pp. S3-11. Etienne MC, Lagrange JL, Dassonville O, Fleming R, Thyss A, Renee N, & Milano G. (1994).

Etienne-Grimaldi MC, Formento P, Degeorges A, Pierga JY, Delva R, Pivot X, Dalenc F,

Etienne-Grimaldi MC, Francoual M, Formento JL, & Milano G. (2007).

treatments in colorectal cancer. *Pharmacogenomics,* Vol. 8, N°11, pp.1561-6. Fanciullino R, Evrard A, Cuq P, Giacometti S, Peillard L, Mercier C, Aubert C, Milano G, &

Formica V, Palmirotta R, Del Monte G, Savonarola A, Ludovici G, De Marchis ML, Grenga I,

Frattini M, Saletti P, Romagnani E, Martin V, Molinari F, Ghisletta M, Camponovo A,

Gamelin L, Capitain O, Morel A, Dumont A, Traore S, Anne le B, Gilles S, Boisdron-Celle M,

Glasgow SC, Yu J, Carvalho LP, Shannon WD, Fleshman JW, & McLeod HL. (2005).

Graziano F, Ruzzo A, Loupakis F, Canestrari E, Santini D, Catalano V, Bisonni R, Torresi U,

Han JY, Lim HS, Shin ES, Yoo YK, Park YH, Lee JE, Jang IJ, Lee DH, & Lee JS. (2006).

advanced colorectal cancer. *J Clin Oncol,* Vol. 26, N°9, pp.1427-34.

chemotherapy. *J Clin Oncol,* Vol. 20, N°7, pp.1721-8.

patients. *Br J Clin Pharmacol,* Vol. 71, N°6, pp.921-8

& Blomgren H. (2002). Thymidylate synthase expression in colorectal cancer: a prognostic and predictive marker of benefit from adjuvant fluorouracil-based

Population study of dihydropyrimidine dehydrogenase in cancer patients. *J Clin* 

Espié M, Veyret C, Formento JL, Francoual M, Piutti M, de Crémoux P, & Milano G. (2011). Prospective analysis of the impact of VEGF-A gene polymorphisms on pharmacodynamics of bevacizumab-based therapy in metastatic breast cancer

Methylenetetrahydrofolate reductase (MTHFR) variants and fluorouracil-based

Ciccolini J. (2006). Genetic and biochemical modulation of 5-fluorouracil through the overexpression of thymidine kinase: an in-vitro study. *Anticancer Drugs,* Vol. 17,

Schirru M, Guadagni F, & Roselli M. Predictive value of VEGF gene polymorphisms for metastatic colorectal cancer patients receiving first-line treatment including fluorouracil, irinotecan, and bevacizumab. *Int J Colorectal Dis,*

Etienne LL, Cavalli F, & Mazzucchelli L. (2007). PTEN loss of expression predicts cetuximab efficacy in metastatic colorectal cancer patients. *Br J Cancer,* Vol. 97, N°8,

& Gamelin E. (2007). Predictive factors of oxaliplatin neurotoxicity: the involvement of the oxalate outcome pathway. *Clin Cancer Res,* Vol. 13, N°21,

Unfavourable expression of pharmacologic markers in mucinous colorectal cancer.

Floriani I, Schiavon G, Andreoni F, Maltese P, Rulli E, Humar B, Falcone A, Giustini L, Tonini G, Fontana A, Masi G, & Magnani M. (2008). Pharmacogenetic profiling for cetuximab plus irinotecan therapy in patients with refractory

Comprehensive analysis of UGT1A polymorphisms predictive for pharmacokinetics and treatment outcome in patients with non-small-cell lung cancer treated with irinotecan and cisplatin. *J Clin Oncol,* Vol. 24, N°11, pp.2237-44.


Ciccolini J, Peillard L, Evrard A, Cuq P, Aubert C, Pelegrin A, Formento P, Milano G, &

Ciccolini J, Cuq P, Evrard A, Giacometti S, Pelegrin A, Aubert C, Cano JP, & Iliadis A.

Ciccolini J, Evrard A, & Cuq P. (2004). Thymidine phosphorylase and fluoropyrimidines

Ciccolini J, Mercier C, Evrard A, Dahan L, Boyer JC, Duffaud F, Lacarelle B, & Seitz JF.

Ciccolini J, Mercier C, Dahan L, & André N. (2011). Integrating pharmacogenetics into gemcitabine dosing-Time for a change? *Nat Rev Clin Oncol,* Vol. 8, N°7, pp.439-44. Crawford Y, Kasman I, Yu L, Zhong C, Wu X, Modrusan Z, Kaminker J, & Ferrara N. (2009).

Dahan L, Ciccolini J, Evrard A, Mbatchi L, Tibbits J, Ries P, Norguet E, Mercier C, Iliadis A,

de Jong FA, Marsh S, Mathijssen RH, King C, Verweij J, Sparreboom A, & McLeod HL.

de Jong FA, Kehrer DF, Mathijssen RH, Creemers GJ, de Bruijn P, van Schaik RH, Planting

Di Fiore F, Sesboüé R, Michel P, Sabourin JC, & Frebourg T. (2010). Molecular determinants

Di Nicolantonio F, Martini M, Molinari F, Sartore-Bianchi A, Arena S, Saletti P, De Dosso S,

Dotor E, Cuatrecases M, Martínez-Iniesta M, Navarro M, Vilardell F, Guinó E, Pareja L,

implications. *J Clin Oncol*, Vol. 29, E-Pub ahead of Print.

placebo-controlled study. *Oncologist,* Vol. 11, N°8, pp.944-54.

adjuvant treatment. *J Clin Oncol,* Vol. 249, N°10, pp.1603-11.

xenografts. *Clin Cancer Res,* Vol. 6, N°4, pp.1529-35.

*Mol Cancer Ther,* Vol. 1, N°2, pp.133-9.

pp.71-81.

pp.678-85.

pp.21-34.

pp.5889-94.

Vol. 103, N°12, pp.1765-72

*J Clin Oncol,* Vol. 26, N°35, pp.5705-12.

Catalin J. (2000b). Enhanced antitumor activity of 5-fluorouracil in combination with 2'-deoxyinosine in human colorectal cell lines and human colon tumor

(2001). Combination of thymidine phosphorylase gene transfer and deoxyinosine treatment greatly enhances 5-fluorouracil antitumor activity in vitro and in vivo.

efficacy: a Jekyll and Hyde story. *Curr Med Chem Anticancer Agents,* Vol. 4, N°2,

(2006). A rapid and inexpensive method for anticipating severe toxicity to fluorouracil and fluorouracil-based chemotherapy. *Ther Drug Monit,* Vol. 28,

PDGF-C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractory to anti-VEGF treatment*. Cancer Cell,* Vol. 15, N°1,

Ouafik LH, Lacarelle B, & Seitz JF. (2011) Sudden toxic-death in a patient upon Xeliri (capecitabine + irinotecan) plus bevacizumab intake: pharmacogenetics

(2004). ABCG2 pharmacogenetics: ethnic differences in allele frequency and assessment of influence on irinotecan disposition. *Clin Cancer Res,* Vol. 10, N°17,

AS, van der Gaast A, Eskens FA, Janssen JT, Ruit JB, Verweij J, Sparreboom A, & de Jonge MJ. (2006). Prophylaxis of irinotecan-induced diarrhea with neomycin and potential role for UGT1A1\*28 genotype screening: a double-blind, randomized,

of anti-EGFR sensitivity and resistance in metastatic colorectal cancer. *Br J Cancer,*

Mazzucchelli L, Frattini M, Siena S, & Bardelli A. (2008). Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer.

Figueras A, Molleví DG, Serrano T, de Oca J, Peinado MA, Moreno V, Germà JR, Capellá G, & Villanueva A. (2006). Tumor thymidylate synthase 1494del6 genotype as a prognostic factor in colorectal cancer patients receiving fluorouracil-based


Pharmacogenetics and Pharmacogenomics

of Colorectal Cancer: Moving Towards Personalized Medicine 273

Kweekel DM, Gelderblom H, Van der Straaten T, Antonini NF, Punt CJ, & Guchelaar HJ;

Labianca R, & Merelli B. (2010). Screening and diagnosis for colorectal cancer: present and

Largillier R, Etienne-Grimaldi MC, Formento JL, Ciccolini J, Nebbia JF, Ginot A, Francoual

Laurent-Puig P, Cayre A, Manceau G, Buc E, Bachet JB, Lecomte T, Rougier P, Lievre A,

Laurent-Puig P, Lievre A,& Blons H. (2009). Mutations and response to epidermal growth

Le Morvan V, Smith D, Laurand A, Brouste V, Bellott R, Soubeyran I, Mathoulin-Pelissier S,

Lenz HJ. (2004). Pharmacogenomics and colorectal cancer. *Ann Oncol,* Vol. 15, N°S4:iv,

Lièvre A, Bachet JB, Le Corre D, Boige V, Landi B, Emile JF, Côté JF, Tomasic G, Penna C,

Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, Somerfield MR,

Lu JF, Bruno R, Eppler S, Novotny W, Lum B, & Gaudreault J. (2008). Clinical

Lurje G, Manegold PC, Ning Y, Pohl A, Zhang W, & Lenz HJ. (2009). Thymidylate synthase

Marcuello E, Altés A, Menoyo A, Del Rio E, Gómez-Pardo M, & Baiget M. (2004). UGT1A1

Meckley LM, Neumann PJ. (2010) Personalized medicine: factors influencing

factor receptor inhibitors. *Clin Cancer Res,* Vol. 15, N°4, pp.1133-9.

Group study. *Br J Cancer,* Vol. 99, N°2, pp.275-82

future. *Tumori,* Vol. 96, N°6, pp.889-901.

*Res,* Vol. 12, N°18, pp.5496-502.

*Oncol,* Vol. 27, N°35, pp.5924-30.

*Res,* Vol. 66, N°8, pp.3992-5.

pp.173-7

pp.1000-7.

*Pharmacogenomics,* Vol. 8, N°12, pp.1693-703.

survival. *Cancer Res,* Vol. 65, N°12, pp.5015-9.

cancer. *Br J Cancer,* Vol. 91, N°4, pp. 678-82.

reimbursement. *Health Policy*, Vol.94, N°2, pp.91-100.

*Pharmacol,* Vol. 62, N°5, pp.779-86

Dutch Colorectal Cancer Group study.(2008). UGT1A1\*28 genotype and irinotecan dosage in patients with metastatic colorectal cancer: a Dutch Colorectal Cancer

M, Renée N, Ferrero JM, Foa C, Namer M, Lacarelle B, & Milano G. (2006). Pharmacogenetics of capecitabine in advanced breast cancer patients. *Clin Cancer* 

Landi B, Boige V, Ducreux M, Ychou M, Bibeau F, Bouché O, Reid J, Stone S, & Penault-Llorca F. (2009). Analysis of PTEN, BRAF, and EGFR status in determining benefit from cetuximab therapy in wild-type KRAS metastatic colon cancer. *J Clin* 

& Robert J. (2007). Determination of ERCC2 Lys751Gln and GSTP1 Ile105Val gene polymorphisms in colorectal cancer patients: relationships with treatment outcome.

Ducreux M, Rougier P, Penault-Llorca F, & Laurent-Puig P. (2006). KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. *Cancer* 

Hayes DF, & Bast RC. (2006). ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. *J Clin Oncol,* Vol. 24, N°33, pp.5313-27. Lu H, Shu XO, Cui Y, Kataoka N, Wen W, Cai Q, Ruan ZX, Gao YT, & Zheng W. (2005).

Association of genetic polymorphisms in the VEGF gene with breast cancer

pharmacokinetics of bevacizumab in patients with solid tumors. *Cancer Chemother* 

gene variations: predictive and prognostic markers. *Mol Cancer Ther,* Vol. 8, N°5,

gene variations and irinotecan treatment in patients with metastatic colorectal


Heinemann V, Stintzing S, Kirchner T, Boeck S, & Jung A. (2009). Clinical relevance of

Hoskins JM, Marcuello E, Altes A, Marsh S, Maxwell T, Van Booven DJ, Paré L,

Innocenti F, Liu W, Chen P, Desai AA, Das S, & Ratain MJ. (2005). Haplotypes of variants in

Jacobs B, De Roock W, Piessevaux H, Van Oirbeek R, Biesmans B, De Schutter J, Fieuws S,

predictors of fluorouracil sensitivity. *J Clin Oncol,* Vol. 23, N°7, pp.1365-9. Keizer RJ, Huitema AD, Schellens JH, & Beijnen JH. (2010). Clinical pharmacokinetics of therapeutic monoclonal antibodies. *Clin Pharmacokinet,* Vol. 49, N°8, pp.493-507. Kim JG, Sohn SK, Chae YS, Cho YY, Bae HI, Yan G, Park JY, Lee MH, Chung HY, & Yu W.

therapeutic resistance. *J Clin Oncol,* Vol. 28, N°3, pp.453-9.

chemotherapy. *BMC Cancer,* 2009 Sep 24; 9:339.

challenges. *Curr Med Chem,* Vol. 18, N°11, pp.1599-612.

*Clin Pharmacol Ther,* Vol. 86, N°1, pp.28-31.

pp.1290-5

15, N°5, pp.295-301.

pp.90-105.

EGFR- and KRAS-status in colorectal cancer patients treated with monoclonal antibodies directed against the EGFR. *Cancer Treat Rev,* Vol. 35, N°3, pp.262-71 Hoskins JM, Goldberg RM, Qu P, Ibrahim JG, & McLeod HL. (2007). UGT1A1\*28 genotype

and irinotecan-induced neutropenia: dose matters. *J Natl Cancer Inst,* Vol. 99, N°7,

Culverhouse R, McLeod HL, & Baiget M. (2008). Irinotecan pharmacogenetics: influence of pharmacodynamic genes. *Clin Cancer Res,* Vol. 14, N°6, pp.1788-96. Ikediobi ON, Shin J, Nussbaum RL, Phillips KA; UCSF Center for Translational and Policy

Research on Personalized Medicine, Walsh JM, Ladabaum U, & Marshall D. (2009). Addressing the challenges of the clinical application of pharmacogenetic testing.

the UDP-glucuronosyltransferase1A9 and 1A1 genes. *Pharmacogenet Genomics,* Vol.

Vandesompele J, Peeters M, Van Laethem JL, Humblet Y, Pénault-Llorca F, De Hertogh G, Laurent-Puig P, Van Cutsem E, & Tejpar S. (2009). Amphiregulin and epiregulin mRNA expression in primary tumors predicts outcome in metastatic colorectal cancer treated with cetuximab. *J Clin Oncol,* Vol. 27, N°13, pp.5068-74 Jakobsen A, Nielsen JN, Gyldenkerne N, & Lindeberg J. (2005). Thymidylate synthase and

methylenetetrahydrofolate reductase gene polymorphism in normal tissue as

(2007). Vascular endothelial growth factor gene polymorphisms associated with prognosis for patients with gastric cancer. *Ann Oncol,* Vol. 18, N°6, pp.1030-6. Kopetz S, Hoff PM, Morris JS, Wolff RA, Eng C, Glover KY, Adinin R, Overman MJ, Valero

V, Wen S, Lieu C, Yan S, Tran HT, Ellis LM, Abbruzzese JL, & Heymach JV. (2010). Phase II trial of infusional fluorouracil, irinotecan, and bevacizumab for metastatic colorectal cancer: efficacy and circulating angiogenic biomarkers associated with

Papakostas P, Vrettou E, & Fountzilas G. (2009). Topoisomerase I but not thymidylate synthase is associated with improved outcome in patients with resected colorectal cancer treated with irinotecan containing adjuvant

HP. (2011). Targeted therapy in colorectal cancer: current status and future

use of pharmacogenomics to individualize therapy. *Cancer Treat Rev,* Vol. 31, N°2,

Kostopoulos I, Karavasilis V, Karina M, Bobos M, Xiros N, Pentheroudakis G, Kafiri G,

Koutras AK, Starakis I, Kyriakopoulou U, Katsaounis P, Nikolakopoulos A, & Kalofonos

Kweekel DM, Gelderblom H, & Guchelaar HJ. (2005). Pharmacology of oxaliplatin and the


Pharmacogenetics and Pharmacogenomics

Vol. 26, N°13, pp. 2131-8.

*Clin Cancer Res,* Vol. 14, N°3, pp. 817-25.

*Cancer Inst,* Vol. 94, N°12, pp. 936-42.

*Cancer,* Vol. 91, N°2, pp. 344-54.

N°11, pp. 1379-85.

of Colorectal Cancer: Moving Towards Personalized Medicine 275

advanced breast cancer: ECOG 2100. *J Clin Oncol,* Vol. 26, N°28, pp. 4672-8. Schwab M, Zanger UM, Marx C, Schaeffeler E, Klein K, Dippon J et al. (2008). Role of genetic

Sharma R, Hoskins JM, Rivory LP, Zucknick M, London R, Liddle C, & Clarke SJ. (2008).

Soong R, Shah N, Salto-Tellez M, Tai BC, Soo RA, Han HC, Ng SS, Tan WL, Zeps N, Joseph

fluorouracil-based chemotherapy. *Ann Oncol,* Vol. 19, N°5, pp. 915-9 Stoehlmacher J, Ghaderi V, Iobal S, Groshen S, Tsao-Wei D, Park D, & Lenz HJ. (2001). A

Miller KD; & ECOG 2100. (2008). Association of vascular endothelial growth factor and vascular endothelial growth factor receptor-2 genetic polymorphisms with outcome in a trial of paclitaxel compared with paclitaxel plus bevacizumab in

and nongenetic factors for fluorouracil treatment-related severe toxicity: a prospective clinical trial by the German 5-FU Toxicity Study Group. *J Clin Oncol,*

Thymidylate synthase and methylenetetrahydrofolate reductase gene polymorphisms and toxicity to capecitabine in advanced colorectal cancer patients.

D, Diasio RB, & Lacopetta B. (2008). Prognostic significance of thymidylate synthase, dihydropyrimidine dehydrogenase and thymidine phosphorylase protein expression in colorectal cancer patients treated with or without 5-

polymorphism of the XRCC1 gene predicts for response to platinum based treatment in advanced colorectal cancer. *Anticancer Res,* Vol. 21, N°4B, pp. 3075-9. Stoehlmacher J, Park DJ, Zhang W, Groshen S, Tsao-Wei DD, Yu MC, & Lenz HJ. (2002).

Association between glutathione S-transferase P1, T1, and M1 genetic polymorphism and survival of patients with metastatic colorectal cancer. *J Natl* 

multivariate analysis of genomic polymorphisms: prediction of clinical outcome to 5-FU/oxaliplatin combination chemotherapy in refractory colorectal cancer. *Br J* 

predictor of response during FOLFOX chemotherapy for the treatment of metastatic colorectal cancer, the MTHFR or XRCC1 gene? *Ann Surg Oncol,* Vol. 13,

sulfotransferase 1A1 and glutathione S-transferase P1 genes in relation to colorectal cancer risk and patients' survival. *World J Gastroenterol,* Vol. 11, N°43, pp. 6875-9 Suttle B, Ball HA, Molimard M, Rajagopalan D, Swann S, Amado R, & Pandite L. (2010).

Relationship between exposure to pazopanib and efficacy in patients with advanced renal cell carcinoma. *J Clin Oncol* Vol. 28, N°15s (suppl; abstr 3048). Toffoli G, Cecchin E, Corona G, Russo A, Buonadonna A, D'Andrea M, Pasetto LM, Pessa S,

Errante D, De Pangher V, Giusto M, Medici M, Gaion F, Sandri P, Galligioni E, Bonura S, Boccalon M, Biason P, & Frustaci S. (2006). The role of UGT1A1\*28 polymorphism in the pharmacodynamics and pharmacokinetics of irinotecan in patients with metastatic colorectal cancer. *J Clin Oncol,* Vol. 24, N°19, pp. 3061-8. Uetake H, Ichikawa W, Takechi T, Fukushima M, Nihei Z, & Sugihara K. (1999). Relationship

between intratumoral dihydropyrimidine dehydrogenase activity and gene expression in human colorectal cancer. *Clin Cancer Res,* Vol. 5, N°10, pp. 2836-9.

Stoehlmacher J, Park DJ, Zhang W, Yang D, Groshen S, Zahedy S, & Lenz HJ. (2004). A

Suh KW, Kim JH, Kim do Y, Kim YB, Lee C, & Choi S. (2006). Which gene is a dominant

Sun XF, Ahmadi A, Arbman G, Wallin A, Asklid D, & Zhang H. (2005). Polymorphisms in


Mercier C, & Ciccolini J. (2006). Profiling dihydropyrimidine dehydrogenase deficiency in

Mercier C, Raynal C, Dahan L, Ortiz A, Evrard A, Dupuis C, Blesius A, Duluc M,

Mercier C, & Ciccolini J. (2007b). Severe or lethal toxicities upon capecitabine intake: is

Mercier C, Brunet C, Yang C, Dupuis C, Bagarry-Liegey D, Duflo S, Giovanni A, Zanaret M,

Morel A, Boisdron-Celle M, Fey L, Soulie P, Craipeau MC, Traore S & Gamelin E. (2006).

Moroni M, Veronese S, Benvenuti S, Marrapese G, Sartore-Bianchi A, Di Nicolantonio F,

Pullmann R Jr, Abdelmohsen K, Lal A, Martindale JL, Ladner RD, & Gorospe M. (2006).

Salgado J, Zabalegui N, Gil C, Monreal I, Rodríguez J, & García-Foncillas J. (2007).

Sartore-Bianchi A, Moroni M, Veronese S, Carnaghi C, Bajetta E, Luppi G, Sobrero A, Barone

Schneider BP, Wang M, Radovich M, Sledge GW, Badve S, Thor A, Flockhart DA, Hancock

colorectal cancer: a cohort study. *Lancet Oncol,* Vol. 6, N°5, pp. 279-86. Popat S, Matakidou A, & Houlston RS. (2004). Thymidylate synthase expression and

variants regulated by AUF1. *J Biol Chem,* Vol. 281, N°33, pp. 23456-63. Ruzzo A, Graziano F, Loupakis F, Rulli E, Canestrari E, Santini D, Catalano V, Ficarelli R,

chemotherapy. *J Clin Oncol,* Vol. 25, N°10, pp. 1247-54.

colorectal cancer. *Oncol Rep,* Vol. 17, N°2, pp. 325-8.

*Colorectal Cancer,* Vol. 6, N°4, pp. 288-96.

17, N°10, pp.841-4.

N°15s, (suppl; abstr 6515).

N°11, pp. 2895-904.

Vol. 22, N°3, pp. 529-36.

pp. 3238-45.

pp. 597-8.

patients with cancer undergoing 5-fluorouracil/capecitabine therapy. *Clin* 

Franceschini F, Giacometti S, Salas S, Milano G, Favre R, Seitz JF, & Ciccolini J. (2007a). Toxic death case in a patient undergoing gemcitabine-based chemotherapy in relation with cytidine deaminase downregulation. *Pharmacogenet Genomics,* Vol.

*DPYD* genetic polymorphism the ideal culprit? *Trends Pharmacol.Sci,* Vol. 28, N°12,

Lacarelle B, Duffaud F, & Ciccolini J. (2009). Pharmacoeconomic study in head and neck cancer patients: Impact of prospective DPD deficiency screening with 5 fluorouracil (5-FU) dose tailoring on toxicities-related costs. *J Clin Oncol,* Vol. 27,

Clinical relevance of different dihydropyrimidine dehydrogenase gene single nucleotide polymorphisms on 5-fluorouracil tolerance. *Mol Cancer Ther,* Vol. 5,

Gambacorta M, Siena S, & Bardelli A. (2005). Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in

prognosis in colorectal cancer: a systematic review and meta-analysis. *J Clin Oncol,*

Differential stability of thymidylate synthase 3'-untranslated region polymorphic

Maltese P, Bisonni R, Masi G, Schiavon G, Giordani P, Giustini L, Falcone A, Tonini G, Silva R, Mattioli R, Floriani I, & Magnani M. (2007). Pharmacogenetic profiling in patients with advanced colorectal cancer treated with first-line FOLFOX-4

Polymorphisms in the thymidylate synthase and dihydropyrimidine dehydrogenase genes predict response and toxicity to capecitabine-raltitrexed in

C, Cascinu S, Colucci G, Cortesi E, Nichelatti M, Gambacorta M, & Siena S. (2007). Epidermal growth factor receptor gene copy number and clinical outcome of metastatic colorectal cancer treated with panitumumab. *J Clin Oncol,* Vol. 25, N°22,

B, Davidson N, Gralow J, Dickler M, Perez EA, Cobleigh M, Shenkier T, Edgerton S,

Miller KD; & ECOG 2100. (2008). Association of vascular endothelial growth factor and vascular endothelial growth factor receptor-2 genetic polymorphisms with outcome in a trial of paclitaxel compared with paclitaxel plus bevacizumab in advanced breast cancer: ECOG 2100. *J Clin Oncol,* Vol. 26, N°28, pp. 4672-8.


**15** 

*USA* 

**Animal Models of Colorectal** 

**Therapeutics Development** 

**Cancer in Chemoprevention and** 

*Department of Biochemistry and Molecular & Cellular Biology,* 

Shubhankar Suman, Albert J. Fornace Jr. and Kamal Datta\*

*Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC,* 

Considering the complexity of genetic and epigenetic events that occurs during colorectal carcinogenesis and the uncertainty in relying solely on extrapolation from cell culture models, it is essential to use animal models relevant to the molecular characteristics of colorectal cancer (CRC). Requirements for a mouse model of CRC are that the model has relevance to the molecular pathways involved in human CRC, that there are correlates with factors that affect the frequency of the disease in human populations, and that the chemopreventive/therapeutic agent-induced signal is sufficient to carry out the project with an affordable and statistically significant number of mice. In this chapter we will discuss how animal models have not only advanced our understanding of CRC initiation and progression but have also greatly facilitated the development of newer chemopreventive

Use of animal models could significantly expedite not only the delineation of molecular pathogenesis of colorectal carcinogenesis but could also aid in the development of newer preventive and therapeutic strategies. Animal tumor models can be classified as spontaneous and artificially transplanted systems. Spontaneous tumor models are now being widely considered for studying the biology of carcinogenesis and development of

Initial animal models of CRC involved use of chemical carcinogens in mouse, rat, as well as in guinea pig. In the last two decades genetically modified mice such as APCMin/+ (Min: multiple intestinal neoplasia) with germline APC mutations at different sites have been extensively usedfor the investigation of therapeutic, chemopreventive and dietary factors for management of colorectal cancer (Hu et al., 2006; Gerner, 2007). Furthermore, studies to identify genetic modifiers of CRC are undertaken by generating mouse models representing molecular events involved in colorectal carcinogenesis like mismatch repair deficiency (MSH2-/-) and crossing

and therapeutic strategies to reduce mortality and incidence.

**2. Animal models of colorectal cancer** 

chemopreventive or chemosuppressive drugs.

**1. Introduction** 

 \*

Corresponding Author


### **Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development**

Shubhankar Suman, Albert J. Fornace Jr. and Kamal Datta\* *Department of Biochemistry and Molecular & Cellular Biology, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC, USA* 

#### **1. Introduction**

276 Colorectal Cancer – From Prevention to Patient Care

Vallböhmer D, Iqbal S, Yang DY, Rhodes KE, Zhang W, Gordon M, Fazzone W, Schultheis

Van Kuilenburg AB. (2004). Dihydropyrimidine dehydrogenase and the efficacy and toxicity

Yang BB, Lum P, Chen A, Arends R, Roskos L, Smith B, & Pérez Ruixo JJ. (2010).

development of panitumumab. *Clin Pharmacokinet,* Vol. 49, N°11, pp.729-40 Yang CG, Ciccolini J, Blesius A, Dahan L, Bagarry-Liegey D, Brunet C, Varoquaux A,

Zhang W, Gordon M, Schultheis AM, Yang DY, Nagashima F, Azuma M, Chang HM,

Zhang X, Li L, Fourie J, Davie JR, Guarcello V, & Diasio RB. (2006). The role of Sp1 and Sp3

Zintzaras E, Ziogas DC, Kitsios GD, Papathanasiou AA, Lau J, & Raman G. (2009). MTHFR

single-agent cetuximab. *J Clin Oncol,* Vol. 25, N°24, pp. 3712-8.

analysis. *Pharmacogenomics,* Vol. 10, N°8, pp. 1285-94.

247-56.

irinotecan efficacy. *Int J Cancer,* Vol. 119, N°10, pp. 2435-42.

of 5-fluorouracil. *Eur J Cancer,* Vol. 40, N°7, pp. 939-50.

AM, Sherrod AE, Danenberg KD, & Lenz HJ. (2006). Molecular determinants of

Pharmacokinetic and pharmacodynamic perspectives on the clinical drug

Frances N, Marouani H, Giovanni A, Ferri-Dessens RM, Chefrour M, Favre R, Duffaud F, Seitz JF, Zanaret M, Lacarelle B, & Mercier C. (2011). DPD-based adaptive dosing of 5-FU in patients with head and neck cancer: impact on treatment efficacy and toxicity. *Cancer Chemother Pharmacol,* Vol. 67, N°1, pp. 49-56.

Borucka E, Lurje G, Sherrod AE, Iqbal S, Groshen S, & Lenz HJ. (2007). FCGR2A and FCGR3A polymorphisms associated with clinical outcome of epidermal growth factor receptor expressing metastatic colorectal cancer patients treated with

in the constitutive *DPYD* gene expression. *Biochim Biophys Acta,* Vol. 1759, N°5, pp.

gene polymorphisms and response to chemotherapy in colorectal cancer: a meta-

Considering the complexity of genetic and epigenetic events that occurs during colorectal carcinogenesis and the uncertainty in relying solely on extrapolation from cell culture models, it is essential to use animal models relevant to the molecular characteristics of colorectal cancer (CRC). Requirements for a mouse model of CRC are that the model has relevance to the molecular pathways involved in human CRC, that there are correlates with factors that affect the frequency of the disease in human populations, and that the chemopreventive/therapeutic agent-induced signal is sufficient to carry out the project with an affordable and statistically significant number of mice. In this chapter we will discuss how animal models have not only advanced our understanding of CRC initiation and progression but have also greatly facilitated the development of newer chemopreventive and therapeutic strategies to reduce mortality and incidence.

#### **2. Animal models of colorectal cancer**

Use of animal models could significantly expedite not only the delineation of molecular pathogenesis of colorectal carcinogenesis but could also aid in the development of newer preventive and therapeutic strategies. Animal tumor models can be classified as spontaneous and artificially transplanted systems. Spontaneous tumor models are now being widely considered for studying the biology of carcinogenesis and development of chemopreventive or chemosuppressive drugs.

Initial animal models of CRC involved use of chemical carcinogens in mouse, rat, as well as in guinea pig. In the last two decades genetically modified mice such as APCMin/+ (Min: multiple intestinal neoplasia) with germline APC mutations at different sites have been extensively usedfor the investigation of therapeutic, chemopreventive and dietary factors for management of colorectal cancer (Hu et al., 2006; Gerner, 2007). Furthermore, studies to identify genetic modifiers of CRC are undertaken by generating mouse models representing molecular events involved in colorectal carcinogenesis like mismatch repair deficiency (MSH2-/-) and crossing

 \* Corresponding Author

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 279

Symbiotic association of Bifidobacterium lactis and carbohydrate 'resistant starch

High amylose maize starch and butyrylated

methoxyserratan-21 beta-ol), a triterpenoid

MNU model TAC-101 (a retinobenzoic acid derivative) (Nakayama et al.,

irinotecan (CPT-11) as postoperative

drug (SAAND) exisulind (EXS)

Combinatorial therapy using HMG-CoA reductase inhibitor (HRI) lovastatin (LOV) and the selective apoptotic antineoplastic

Protective effect of Fullerenol on heart and liver toxicity induced by doxorubicin

high amylose maize starch

adjuvant chemotherapy

DMBDD model PJJ-34 (13 alpha, 14alpha-epoxy-3beta-

Orthotropic model Combined use of bevacizumab and

Table 1. Rat models used in chemoprevention & chemotherapy.

Etoricoxib (selective COX-2 inhibitor) (Kaur Saini &Nath

Diclofenac (preferential COX-2 inhibitor) (Kaur Saini &Nath

Adlay bran ethanol extract (ABE-Ea) (Chung et al., 2010) Soy isoflavones (Min et al., 2010) Arabinoxylan-oligosaccharide (Femia et al., 2010) Probiotic soy products (Silva et al., 2009) Physical exercise (Silva et al., 2009) Astaxanthine (Prabhu et al., 2009) Soy isoflavones (Raju et al., 2009) R-Flurbiprofen (Martin et al., 2010) Copper-indomethacin (Bonin et al., 2010) Naproxen (Steele et al., 2009) Nitric Oxide-Naproxen (Steele et al., 2009) CP-31398 (a p53 modulator) (Rao et al., 2009) Celecoxib (Rao et al., 2009)

Melatonin (Tanaka et al., 2003) Prebiotic germinated barley foodstuff (GBF) (Kanauchi et al.,

Ursodeoxycholic acid (UDCA) (Hess et al., 2004)

Sanyal, 2010)

Sanyal, 2010)

(Le Leu et al., 2010)

(Clarke et al., 2008)

(Doi et al., 2010)

(Mizobe et al., 2008)

(Kim et al., 2004)

(Injac et al., 2009)

2008)

2009)

Rat model Name of the compound Reference

Azoxymethane (AOM) /Dimethylhydrazine

model

AOM model or Dimethylhydrazine

model

them with APCMin/+(Kwong & Dove, 2009). Other models for CRC targeting relevant pathway are also being developed to elucidate chemopreventive and chemotherapeutic response to newer molecules (Kwong & Dove, 2009). Because there is no ideal animal tumor model, which can mimic all the complexities associated with human CRC, the selection of an appropriate experimental model is crucial to study specific biologic end points towards the understanding of mechanistic, preventive and therapeutic aspects of the cancer.

#### **2.1 Rat models**

In the last few decades many murine models have been established that are useful for the investigation of initiation, expansion and progression of gastrointestinal (GI) cancers. In most of these models, Wnt signaling, mismatch repair, and TGF pathways are targeted not only to understand initiation and progression of CRC but also to evaluate various pharmaceutical and biological agents for prevention and treatment of CRC.

Normally CRC is not observed in rats (<0.1%) (Goodman et al., 1980) except the Wistar-Furth/Osaka strain that spontaneously develops adenocarcinomas in 30-40% of the animal (Miyamoto & Takizawa, 1975). The rat models of colon cancer are developed using common carcinogens like AOM (azoxymethane), DMH (dimethylhydrazine), or PhIP (2-amino-1 methyl-6-phenylimidazo [4,5-*b*]pyridine)(Corpet & Pierre, 2003, 2005). Most of the carcinogen-treated rat models develop tumors in the colon and often progress to adenocarcinomas. However, long latency period of tumor development is a distinct disadvantage. Efforts have been made to generate target-selected mutations, including nonsense alleles by several laboratories resulting into development of a rat strain carrying a nonsense allele in codon 1137 of APC (Corpet & Pierre, 2005). Interestingly, multiple intestinal neoplasms mostly in the colon were observed in F344 rats, commonly known as PIRC (polyposis in the rat colon) model, heterozygous for this allele and these animals survive for about one year (Zan et al., 2003; Smits et al., 2006; Amos-Landgraf et al., 2007). The rat models due to their size allows investigators to perform procedures like endoscopy, microCT (Computerized Tomography), and microPET (Positron Emission Tomography) imaging to evaluate chemoprevention or therapeutic interventions without sacrificing the animal.

#### **2.1.1 Rat models in chemoprevention and therapeutics development**

Although, dimethylhydrazine and its metabolites azoxymethane (AOM) and methylazoxymethanol are commonly used in the induction of colonic tumors in rat models, other carcinogens, like nitrosomethyl urea (MNU), specific nitrosamines and heterocyclic amines are also in frequent use. Many potential chemopreventive agents of colorectal cancer have been assessed in rat models. The effects of chemopreventive and therapeutic agents on initiation and progression of carcinogen-induced colonic tumors can be studied by varying the time of intervention. In rat models over 160 compounds have been screened for chemopreventive properties (Corpet et al., 2008) and the compounds found to be of chemopreventive and therapeutic importance are summarized in Table 1. Complete inhibition of cancer induction has been detected in ursodeoxycholic, polyethylene glycol (PEG), methylmethanethiosulfonate (MMTS) treated rats and in rats given exercise. Also, celecoxib, acetoxychavicol, selenium, p53 vaccination, piroxicam with difluoromethylornithine (DFMO), cellulose, aspirin, S-allylcysteine, obacunone, sulindac sulfone and hesperidin (flavanone glycoside) reduced the incidence of adenocarcinoma more than 78% (Corpet et al., 2008). Moreover, a DMBDD (7-hydroxy-7'-methoxy-4,4'-

them with APCMin/+(Kwong & Dove, 2009). Other models for CRC targeting relevant pathway are also being developed to elucidate chemopreventive and chemotherapeutic response to newer molecules (Kwong & Dove, 2009). Because there is no ideal animal tumor model, which can mimic all the complexities associated with human CRC, the selection of an appropriate experimental model is crucial to study specific biologic end points towards the understanding

In the last few decades many murine models have been established that are useful for the investigation of initiation, expansion and progression of gastrointestinal (GI) cancers. In most of these models, Wnt signaling, mismatch repair, and TGF pathways are targeted not only to understand initiation and progression of CRC but also to evaluate various

Normally CRC is not observed in rats (<0.1%) (Goodman et al., 1980) except the Wistar-Furth/Osaka strain that spontaneously develops adenocarcinomas in 30-40% of the animal (Miyamoto & Takizawa, 1975). The rat models of colon cancer are developed using common carcinogens like AOM (azoxymethane), DMH (dimethylhydrazine), or PhIP (2-amino-1 methyl-6-phenylimidazo [4,5-*b*]pyridine)(Corpet & Pierre, 2003, 2005). Most of the carcinogen-treated rat models develop tumors in the colon and often progress to adenocarcinomas. However, long latency period of tumor development is a distinct disadvantage. Efforts have been made to generate target-selected mutations, including nonsense alleles by several laboratories resulting into development of a rat strain carrying a nonsense allele in codon 1137 of APC (Corpet & Pierre, 2005). Interestingly, multiple intestinal neoplasms mostly in the colon were observed in F344 rats, commonly known as PIRC (polyposis in the rat colon) model, heterozygous for this allele and these animals survive for about one year (Zan et al., 2003; Smits et al., 2006; Amos-Landgraf et al., 2007). The rat models due to their size allows investigators to perform procedures like endoscopy, microCT (Computerized Tomography), and microPET (Positron Emission Tomography) imaging to evaluate chemoprevention or therapeutic interventions without sacrificing the

of mechanistic, preventive and therapeutic aspects of the cancer.

pharmaceutical and biological agents for prevention and treatment of CRC.

**2.1.1 Rat models in chemoprevention and therapeutics development** 

Although, dimethylhydrazine and its metabolites azoxymethane (AOM) and methylazoxymethanol are commonly used in the induction of colonic tumors in rat models, other carcinogens, like nitrosomethyl urea (MNU), specific nitrosamines and heterocyclic amines are also in frequent use. Many potential chemopreventive agents of colorectal cancer have been assessed in rat models. The effects of chemopreventive and therapeutic agents on initiation and progression of carcinogen-induced colonic tumors can be studied by varying the time of intervention. In rat models over 160 compounds have been screened for chemopreventive properties (Corpet et al., 2008) and the compounds found to be of chemopreventive and therapeutic importance are summarized in Table 1. Complete inhibition of cancer induction has been detected in ursodeoxycholic, polyethylene glycol (PEG), methylmethanethiosulfonate (MMTS) treated rats and in rats given exercise. Also, celecoxib, acetoxychavicol, selenium, p53 vaccination, piroxicam with difluoromethylornithine (DFMO), cellulose, aspirin, S-allylcysteine, obacunone, sulindac sulfone and hesperidin (flavanone glycoside) reduced the incidence of adenocarcinoma more than 78% (Corpet et al., 2008). Moreover, a DMBDD (7-hydroxy-7'-methoxy-4,4'-

**2.1 Rat models** 

animal.


Table 1. Rat models used in chemoprevention & chemotherapy.

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 281

**Heterozygous phenotype** 

Multiple adenoma in GI-tract (mainly small intestine)

Multiple adenoma in GI-tract (mainly small intestine)

Multiple adenoma in GI-tract

**Average no. of spontaneo us tumors**  **References** 

Su et al., 1992)

et al., 2007)

et al., 1995)

et al., 1997)

et al., 1994)

et al., 1999)

et al., 2000)

et al., 2004)

et al., 2005)

(Shibata et al., 1997)

50-70 (Moser et al., 1990;

Up to 600 (Kwong

Up to 300 (Oshima

7-10 (Shibata

3-4 (Fodde

120 (Sasai

40-50 (Colnot

N/R (Gounari

1.1 (Li et al., 2005)

0.3 (Li et al., 2005)

up to 1year of age

**Homozyg ous phenotype** 

lethality

lethality

lethality

lethality

lethality

lethality

lethality

C57BL/6J N/R Small

C57BL/6J N/R Small

Table 2. Adematous polyposis coli (APC) based intestinal tumor models

\*Conditional expression; N/A=Not applicable; N/R=Not reported

C57BL/6J Embryonic

14/580\* N/A C57BL/6J N/A Multiple

468/468 N/R N/R Embryonic

N/A Mixed N/A N/A No tumor

Multiple adenomas at 4 week after conditional deletion

Adenoma and adenocarcinomas in GI-tract, desmoids tumors

C57BL/6J Viable Normal 0 (Smits

Develops tumor mainly in small intestine, but also in colon and stomach

1309/1309 N/R C57BL/6J N/R 25-40 (Niho et al., 2003)

age

adenomas

Polyps starting after 2 months of

microadenoma

microadenoma

C57BL/6J Embryonic

C57BL/6J Embryonic

**Allele/trunc ation location** 

Min/850 4-6

716/716 5-7

1638T/1638 Up to 2

474/474 <6

Ex13NeoR/ Full-length

Ex13NeoF/ Full-length

Years

months

>15 months

>15 months

580S/580\* (based on Cre-LoxP recombinatio n system)

**Average lifespan** 

months

months

Min/850 3-5 BTBR/Pas Embryonic

580D/580\* N/A Mixed Embryonic

1638N/1638 >1Year C57BL/6J Embryonic

**Genetic background** 

bis(1,3-benzodioxole)-5,5'-dicarboxylic acid dimethyl ester) rat multiorgan carcinogenicity model has also been developed for carcinogen testing (Takahashi et al., 1992; Imaida et al., 2003). The DMBDD model can be used for prediction of intestinal carcinogenesis risk assessment as well as for chemoprevention studies.

#### **2.2 Mouse models**

#### **2.2.1 APC related models**

The first heritable mouse model of colon cancer, APCMin/+, was reported in 1990 as a result of ethylnitrosourea (ENU)-induced germline truncating mutation at the codon 850 of APC (Moser et al., 1990, 1992; Su et al., 1992). In the C57BL/6J mouse background APCMin/+ mice develop about 30 small intestinal polyps with occasional adenocarcinoma and essentially no tumor in the colon (McCart et al., 2008). Although this contrasts human FAP (familial adenomatous polyposis) where most of the adenomas are in the colon and these adenomas certainly progresses to invasive adenocarcinoma, the APCMin/+ models due to their phenotypic and histopathological similarities to human intestinal neoplasm are used not only to test therapeutic and chemopreventive interventions but also to understand the role of APC gene in CRC (Fodde & Smits, 2001). The APCMin/+ models also proved to be important for the study of genetic modifiers-of-Min (*Mom*) locus. When APCMin/+ C57BL/6J mice were crossed to AKR and MA mice, only 6 to 7 intestinal adenomas were observed and backcrossing F1 hybrids to the C57BL/6J helped identify a number of loci modifying number and distribution of adenomas (Moser et al., 1992; Gould et al., 1996; Fodde & Smits, 2001). The adenomas in the small intestine of the APCMin/+ mouse have dysplastic and hyperplastic crypts and villi but the colonic tumors are spherical and peduncular with dysplastic cells. Importantly, these adenomas display higher mitotic index than surrounding normal crypts (Kwong & Dove, 2009). The Min mice due to their many advantages have been extensively used by scientists to study molecular pathogenesis of CRC – i] Min mice contains a single genetic change that produces a organ-specific, consistent, and discrete tumor phenotype, ii] Adenomas in Min mice develop rapidly, with lesions visible as early as 60 days, iii] high tumor multiplicities (>100 /intestinal tract) providing strong statistical power, and iv] multiple pathways impacting tumorigenesis enable many entry points for basic or applied study (Kwong & Dove, 2009). Importantly, other mouse models with targeted genetic manipulations at different locations on APC have also been generated (summarized in Table 2). When heterozygous, the Δ474, Δ14, Δ716, lacZ, and Δ1309 mouse models show phenotypes similar to that of Min (Sasai et al., 2000; Oshima et al., 2001; Niho et al., 2003; Colnot et al., 2004). In contrast, heterozygosity for the 1638N allele results in 0-2 tumors (none in the colon) while the 1638T model is tumor-free and unlike any other truncating allele, 1638T homozygous is viable. The 1638N has only approximately 1-2% of the truncated protein and is referred to as leaky allele (Fodde & Smits, 2001). In contrast, 1638T has the full expression level of the truncated protein and is known as truncated allele. Furthermore, Li Q et al., (2005) inserted a neomycin cassette in either orientation (reverse (neoR) or forward (neoF)) into the 13th intron of APC to generate full-length hypomorphic (expression reduced to 10-20%) alleles and showed that these heterozygous mice developed fewer than two adenomas per mouse (Li et al., 2005). The Cre/*loxP* conditional gene targeting system is developed to generate additional APC models to induce tumors specifically in the colon (Shibata et al., 1997; Colnot et al., 2004; Gounari et al., 2005; Hinoi et al., 2007).

bis(1,3-benzodioxole)-5,5'-dicarboxylic acid dimethyl ester) rat multiorgan carcinogenicity model has also been developed for carcinogen testing (Takahashi et al., 1992; Imaida et al., 2003). The DMBDD model can be used for prediction of intestinal carcinogenesis risk

The first heritable mouse model of colon cancer, APCMin/+, was reported in 1990 as a result of ethylnitrosourea (ENU)-induced germline truncating mutation at the codon 850 of APC (Moser et al., 1990, 1992; Su et al., 1992). In the C57BL/6J mouse background APCMin/+ mice develop about 30 small intestinal polyps with occasional adenocarcinoma and essentially no tumor in the colon (McCart et al., 2008). Although this contrasts human FAP (familial adenomatous polyposis) where most of the adenomas are in the colon and these adenomas certainly progresses to invasive adenocarcinoma, the APCMin/+ models due to their phenotypic and histopathological similarities to human intestinal neoplasm are used not only to test therapeutic and chemopreventive interventions but also to understand the role of APC gene in CRC (Fodde & Smits, 2001). The APCMin/+ models also proved to be important for the study of genetic modifiers-of-Min (*Mom*) locus. When APCMin/+ C57BL/6J mice were crossed to AKR and MA mice, only 6 to 7 intestinal adenomas were observed and backcrossing F1 hybrids to the C57BL/6J helped identify a number of loci modifying number and distribution of adenomas (Moser et al., 1992; Gould et al., 1996; Fodde & Smits, 2001). The adenomas in the small intestine of the APCMin/+ mouse have dysplastic and hyperplastic crypts and villi but the colonic tumors are spherical and peduncular with dysplastic cells. Importantly, these adenomas display higher mitotic index than surrounding normal crypts (Kwong & Dove, 2009). The Min mice due to their many advantages have been extensively used by scientists to study molecular pathogenesis of CRC – i] Min mice contains a single genetic change that produces a organ-specific, consistent, and discrete tumor phenotype, ii] Adenomas in Min mice develop rapidly, with lesions visible as early as 60 days, iii] high tumor multiplicities (>100 /intestinal tract) providing strong statistical power, and iv] multiple pathways impacting tumorigenesis enable many entry points for basic or applied study (Kwong & Dove, 2009). Importantly, other mouse models with targeted genetic manipulations at different locations on APC have also been generated (summarized in Table 2). When heterozygous, the Δ474, Δ14, Δ716, lacZ, and Δ1309 mouse models show phenotypes similar to that of Min (Sasai et al., 2000; Oshima et al., 2001; Niho et al., 2003; Colnot et al., 2004). In contrast, heterozygosity for the 1638N allele results in 0-2 tumors (none in the colon) while the 1638T model is tumor-free and unlike any other truncating allele, 1638T homozygous is viable. The 1638N has only approximately 1-2% of the truncated protein and is referred to as leaky allele (Fodde & Smits, 2001). In contrast, 1638T has the full expression level of the truncated protein and is known as truncated allele. Furthermore, Li Q et al., (2005) inserted a neomycin cassette in either orientation (reverse (neoR) or forward (neoF)) into the 13th intron of APC to generate full-length hypomorphic (expression reduced to 10-20%) alleles and showed that these heterozygous mice developed fewer than two adenomas per mouse (Li et al., 2005). The Cre/*loxP* conditional gene targeting system is developed to generate additional APC models to induce tumors specifically in the colon (Shibata et al., 1997; Colnot et al., 2004; Gounari et al., 2005; Hinoi et

assessment as well as for chemoprevention studies.

**2.2 Mouse models** 

al., 2007).

**2.2.1 APC related models**


Table 2. Adematous polyposis coli (APC) based intestinal tumor models

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 283

Fig. 1. Integrated molecular pathway implicated in development of colorectal cancer (Courtesy: KEGG pathway, www.genome.jp/kegg). Alterations in Wnt signaling including Wnt, APC, Axin and TCFs are associated increased -catenin level and increased cell proliferation. The MAPK signaling is associated with the oncogenic activation of RAS and ERK signaling leading to increase cell proliferation). The TGF- pathway is mainly a growth inhibitory pathway and any perturbations leads to suppressed apoptosis and increased cell proliferation. The mismatch repair (MMR) pathway maintains DNA homeostasis by facilitating post-replication repair and dysfunction results in accumulation of potential mutations and genetic instability implicated in the development of CRC. Important

adenoma

Carcinoma

Early

adenoma

candidate proteins altered in CRC are highlighted in red color.

Normal Dysplas c ACF

epithelium

**Figure**

**1**

Because APCMin/+ mice rarely develop invasive cancer, efforts were made to develop compound mouse models by introducing alterations of genes known to be involved in CRC signaling pathways and produce invasive tumors mimicking human disease. The oncogene Kras, which is mutated in 40-50% of human CRC, is not altered in APCMin/+ polyps and the APCMin/+ mice with the introduction of Kras develop aggressive adenocarcinoma (Fodde & Smits, 2001). Loss of EphB receptor is an important event in the progression of CRC and APCMin/+ carrying a dominant negative EphB2 transgene showed 10 fold more tumor formation with greater number of invasive adenocarcinomas (Fodde & Smits, 2001; Batlle et al., 2005). The APCMin/+ mouse model has been quite useful as an experimental system for studying colorectal tumorigenesis and CRC chemoprevention strategies (Moser et al., 1990) because Min mice have two distinct advantages i] numerous adenomas with the same inherited APC mutation are available for analysis and ii] these adenomas develop in animals of uniform genetic background (Luongo et al., 1994).

#### **2.2.2 Mismatch repair (MMR) deficient models**

The HNPCC (hereditary nonpolyposis colorectal cancer) is an inherited condition with inactivated DNA mismatch repair (MMR) genes, like *MLH1, MSH2, MSH6*, and *PMS2*  (Fishel et al., 1993; Lynch & de la Chapelle, 2003) and leads to the development of a variety of cancers including that of the colon (Lynch & Smyrk, 1996). Mouse models with loss of function of MMR genes have been generated and mice lacking *Mlh1*, *Msh2* and *Msh6*  develop tumors in stomach, small intestine, and colon. However, these mice also develop cancers of the lymphatic system, skin and lung (Reitmair et al., 1996; Edelmann et al., 1997, 1999, 2000). Enhanced development of adenomas was observed in the APCMin/+ mice lacking *Msh2* with increase in colonic adenoma numbers. Interestingly, these mice show normal growth and can reproduce but have reduced life span (Reitmair et al., 1996). Although loss of *Msh3* is not associated with increased tumors, loss of both the *Msh3* and *Msh6* leads to an increase in GI tumors at a younger age, similar to *Mlh1* or *Msh2-deficient mice* (Edelmann et al., 2000). Mice bearing mutations in the *Msh6* gene have a life span of 18 months and develop GI tumors within one year. The MMR mouse models carrying one functional copy of APC showed increasing mutation of APC and an enhanced frequency of intestinal neoplasia (Reitmair et al., 1996; Kuraguchi et al., 2001). Furthermore, mice lacking *Mlh1* in APC1638N model have a 40-fold increase in adenomas compared to APC1638N mice alone (Edelmann et al., 1999). Interestingly, the PMS2-/- mice are vulnerable to lymphomas but they do not develop GI tumors. However, the PMS2-/- mice in APCMin background showed an increased number of adenomas in the GI-tract compared to Min alone (Prolla et al., 1998; Prolla, 1998; Baker et al., 1995). In contrast, the Mlh1/APC1638N mice showed a greater percentage of tumors progressed to invasive carcinomas (Edelmann et al., 1999). The MMR models are useful for screening of agents known to interfere with DNA mismatch repair processes for their therapeutic or carcinogenic effects.

#### **2.2.3 TGF- models**

The TGF- (transforming growth factor-) signaling pathway regulates a number of cellular processes including cellular differentiation, growth suppression, deposition of extracellular matrix and apoptosis (Figure 1). The TGF ligands through a heteromeric receptor mediate their effects on cells and dysregulation of the TGF- receptor 2 (TGF-R2) is the most common occurrence in the CRC (Bellam & Pasche, 2010; Grady et al., 1999). Although TGFR2 has been suggested to have a tumor suppressor function in CRC, recent reports indicate that it could act as a tumor suppressor as well as a tumor promoter (Tang et al., 1998;

Because APCMin/+ mice rarely develop invasive cancer, efforts were made to develop compound mouse models by introducing alterations of genes known to be involved in CRC signaling pathways and produce invasive tumors mimicking human disease. The oncogene Kras, which is mutated in 40-50% of human CRC, is not altered in APCMin/+ polyps and the APCMin/+ mice with the introduction of Kras develop aggressive adenocarcinoma (Fodde & Smits, 2001). Loss of EphB receptor is an important event in the progression of CRC and APCMin/+ carrying a dominant negative EphB2 transgene showed 10 fold more tumor formation with greater number of invasive adenocarcinomas (Fodde & Smits, 2001; Batlle et al., 2005). The APCMin/+ mouse model has been quite useful as an experimental system for studying colorectal tumorigenesis and CRC chemoprevention strategies (Moser et al., 1990) because Min mice have two distinct advantages i] numerous adenomas with the same inherited APC mutation are available for analysis and ii] these adenomas develop in

The HNPCC (hereditary nonpolyposis colorectal cancer) is an inherited condition with inactivated DNA mismatch repair (MMR) genes, like *MLH1, MSH2, MSH6*, and *PMS2*  (Fishel et al., 1993; Lynch & de la Chapelle, 2003) and leads to the development of a variety of cancers including that of the colon (Lynch & Smyrk, 1996). Mouse models with loss of function of MMR genes have been generated and mice lacking *Mlh1*, *Msh2* and *Msh6*  develop tumors in stomach, small intestine, and colon. However, these mice also develop cancers of the lymphatic system, skin and lung (Reitmair et al., 1996; Edelmann et al., 1997, 1999, 2000). Enhanced development of adenomas was observed in the APCMin/+ mice lacking *Msh2* with increase in colonic adenoma numbers. Interestingly, these mice show normal growth and can reproduce but have reduced life span (Reitmair et al., 1996). Although loss of *Msh3* is not associated with increased tumors, loss of both the *Msh3* and *Msh6* leads to an increase in GI tumors at a younger age, similar to *Mlh1* or *Msh2-deficient mice* (Edelmann et al., 2000). Mice bearing mutations in the *Msh6* gene have a life span of 18 months and develop GI tumors within one year. The MMR mouse models carrying one functional copy of APC showed increasing mutation of APC and an enhanced frequency of intestinal neoplasia (Reitmair et al., 1996; Kuraguchi et al., 2001). Furthermore, mice lacking *Mlh1* in APC1638N model have a 40-fold increase in adenomas compared to APC1638N mice alone (Edelmann et al., 1999). Interestingly, the PMS2-/- mice are vulnerable to lymphomas but they do not develop GI tumors. However, the PMS2-/- mice in APCMin background showed an increased number of adenomas in the GI-tract compared to Min alone (Prolla et al., 1998; Prolla, 1998; Baker et al., 1995). In contrast, the Mlh1/APC1638N mice showed a greater percentage of tumors progressed to invasive carcinomas (Edelmann et al., 1999). The MMR models are useful for screening of agents known to interfere with

DNA mismatch repair processes for their therapeutic or carcinogenic effects.

The TGF- (transforming growth factor-) signaling pathway regulates a number of cellular processes including cellular differentiation, growth suppression, deposition of extracellular matrix and apoptosis (Figure 1). The TGF ligands through a heteromeric receptor mediate their effects on cells and dysregulation of the TGF- receptor 2 (TGF-R2) is the most common occurrence in the CRC (Bellam & Pasche, 2010; Grady et al., 1999). Although TGFR2 has been suggested to have a tumor suppressor function in CRC, recent reports indicate that it could act as a tumor suppressor as well as a tumor promoter (Tang et al., 1998;

animals of uniform genetic background (Luongo et al., 1994).

**2.2.2 Mismatch repair (MMR) deficient models** 

**2.2.3 TGF- models** 

**Figure**

**1**

Fig. 1. Integrated molecular pathway implicated in development of colorectal cancer (Courtesy: KEGG pathway, www.genome.jp/kegg). Alterations in Wnt signaling including Wnt, APC, Axin and TCFs are associated increased -catenin level and increased cell proliferation. The MAPK signaling is associated with the oncogenic activation of RAS and ERK signaling leading to increase cell proliferation). The TGF- pathway is mainly a growth inhibitory pathway and any perturbations leads to suppressed apoptosis and increased cell proliferation. The mismatch repair (MMR) pathway maintains DNA homeostasis by facilitating post-replication repair and dysfunction results in accumulation of potential mutations and genetic instability implicated in the development of CRC. Important candidate proteins altered in CRC are highlighted in red color.

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 285

Models have been developed to investigate the role of immune system in colorectal carcinogenesis. Immune cells are involved in inflammatory response and inflammation is intimately related to CRC. Therefore, the IL-2 (interleukin-2), IL-10, and TCR knockout models were developed and used in studying role of diet and inflammation in colorectal cancer initiation and progression. (Rudolph et al., 1995; Kullberg et al., 1998; Mizoguchi et

Although important information on familial and sporadic colorectal carcinogenesis was obtained from genetic animal models, it is the carcinogen-induced animal models that were instrumental in delineating molecular events specifically in the sporadic CRC. It is worth mentioning that the colon-specific carcinogen dimethylhyrdrazine (DMH) along with AOM has been useful in developing our current understanding of the molecular mechanisms underlying sporadic colorectal carcinogenesis (Druckrey et al., 1966). To its advantage the carcinogen-induced mouse model develops tumor which show much similarities to the pathophysiology of the human CRC (Kaiser et al., 2007; Uronis et al., 2007). The carcinogenmediated tumors show alterations in the WNT/-catenin pathway (Takahashi et al., 2000). Interestingly, the AOM-induced colorectal carcinogenesis, unlike in the APCMin/+, is mainly due to the mutations in the *Ctnnb1 gene*, which encodes -catenin protein. Mutations in the *Ctnnb1* results in ubiquitination-resistant stabilization of the -catenin leading to growth of colorectal adenomas associated with upregulation of proliferation markers like cyclin D1 and c*Myc* (Wang et al., 1998; Kaiser et al., 2007). The carcinogen-induced model was also a key player in the identification of the modifier loci like the *Ptprj* (a receptor-type protein tyrosine phosphatase), which has been shown to modify susceptibility to DMH and has shown frequent loss of heterozygosity in human colon cancer (Ruivenkamp et al., 2002). Additionally, the carcinogen-induced models led to the recognition of *Pref1* as a modifier of CRC and the promoter of the *Pref1* is suggested to contain a -catenin/TCF response

element (Dong et al., 2004; Ruivenkamp et al., 2002; Uronis et al., 2007).

more useful than APCMin/+ mice (Trani et al., 2010).

**2.2.7 Mouse models in chemoprevention and therapeutics development** 

The genetic mouse models as well as the carcinogen-induced models of CRC (APC, TGF and mismatch repair based models) are used to evaluate the effect of diets and chemopreventive agents. The chemopreventive and dietary interventions are usually started between 3 and 6 weeks of age and invariably the principal biological endpoint is the number and grade of the tumors. It is important to note that in contrast to human CRC where no small intestinal tumors are observed, the tumors in the mutant models are mostly in the small intestine. However, the concept that the non-steroidal anti-inflammatory drugs (NSAID) have a chemopreventive role in the CRC was first established in these models showing reduction in tumor number in the small intestine. The chemopreventive properties of NSAIDs like celecoxib and piroxicam were also validated in the carcinogen-induced models showing decrease in polyp number and size. Due to commercial availability, the APCMin/+ is the model of choice in many studies and a list of agents screened in APCMin/+ mouse model is shown in Table 3. Most of the other CRC mouse models are used for specific biological end points. But a few of them such as APC1638N model because of its longer life span and good signal to noise ratio is used for carcinogen testing and is suggested to be

**2.2.5 Immune system related models**

**2.2.6 Carcinogen-induced model***s*

al., 2000; Seril et al., 2005).

Yang et al., 2002a, b; Blobe et al., 2000). Mouse models related to the TGF- pathway have been used to delineate the multifaceted role of this pathway during colorectal carcinogenesis. It has been reported that the *TGF-1* deficient mice due to widespread inflammation die around three weeks of age (Shull et al., 1992; Kulkarni & Karlsson, 1993). Importantly, in the absence of *Rag2* the *TGF-1-*deficient mice survive until adulthood (Diebold et al., 1995). Due to the significant incidence of carcinoma of the cecum and colon in these mice, the *Rag2*/*TGF-1* deficient mice remains a useful model to study the role of inflammation in CRC in relation to the TGF1 signaling (Engle et al., 1999; Maggio-Price et al., 2006). Mouse lacking TGF-2 in the colonic epithelium showed increased adenoma and adenocarcinoma formation after carcinogen treatment. Apart from TGF-, mouse models representing alterations in the other factors in this important pathway like SMAD2, SMAD4 and SMAD3 have been reported. (Eppert et al., 1996). Although mice lacking SMAD2 and SMAD4 are embryonically lethal, the SMAD3 deficient mice are viable and are a useful model of CRC (Zhu et al., 1998). Importance of the TGF- related models lies in the fact that the APC remains intact in the adenomas of these mice and these models could serve as a valuable tool to investigate non-WNT/APC/ catenin-mediated colorectal carcinogenesis (Kaiser et al., 2007). Interestingly, the TGF- related models in the APCMin/+ background showed increased incidence of invasive carcinoma specifically in the distal colon (Takaku et al., 1998; Sodir et al., 2006).

#### **2.2.4 Inflammation mediated model***s*

Inflammatory bowel diseases (IBD) like ulcerative colitis (UC) and Crohn's Disease are predisposing conditions of CRC (Itzkowitz & Harpaz, 2004; Itzkowitz & Yio, 2004). Prolonged administration of dextransulfate sodium (DSS) in mice resulting in chronic colitis and formation of high-grade dysplasia confirmed the involvement of chronic inflammation in colorectal carcinogenesis (Okayasu et al., 1990). Interestingly intestinal tumorigenesis was augmented by combined administration of AOM and DSS. (Tanaka et al., 2003). To demonstrate that deficiency of *Sigirr* (single immunoglobulin and tollinterleukin 1 receptor domain) along with bacteria-induced inflammation increases susceptibility to CRC investigators effectively utilized the combined AOM/DSS model (Wald et al., 2003; Xiao et al., 2007; Uronis & Threadgill, 2009). Furthermore, the combined AOM/DSS model was also an important instrument in defining the role of the JAK/STAT (Janus kinase/signal transducers and activators of transcription) and NFkB (nuclear factor of kappa light chain gene enhancer in B-cells) pathways in inflammation-induced CRC (Wirtz & Neurath, 2007). The role of the JAK/STAT pathway in colorectal carcinogenesis was further confirmed by the fact that dysfunctional *Socs1* and *Socs3* (suppressors of cytokine signaling) leads to enhanced activation of STAT1, STAT3 and NFkB and subsequent growth of colorectal tumors (Hanada et al., 2006). Although lack of only *Socs3*  in the intestinal epithelial cells is not associated with chronic inflammation or tumor formation, these mice when treated with AOM/DSS showed distinct inflammatory response followed by colonic tumors (Hanada et al., 2006; Rigby et al., 2007). In contrast to increased nuclear -catenin in the *Socs3* deficient mice, the colorectal tumors in *Socs1*  deficient mice display enhanced expression of *Myc* (Sutherland et al., 2006). In addition, the Muc2-/- is an important animal model to study the role of inflammation in the colorectal carcinogenesis. This model specifically targets mucin-forming Muc2 and unlike other models tumor formation is also observed in the rectum (Mack & Hollingsworth, 1994; Yang et al., 2005; Femia et al., 2009).

#### **2.2.5 Immune system related models**

284 Colorectal Cancer – From Prevention to Patient Care

Yang et al., 2002a, b; Blobe et al., 2000). Mouse models related to the TGF- pathway have been used to delineate the multifaceted role of this pathway during colorectal carcinogenesis. It has

three weeks of age (Shull et al., 1992; Kulkarni & Karlsson, 1993). Importantly, in the absence

deficient mice remains a useful model to study the role of inflammation in CRC in relation to the TGF1 signaling (Engle et al., 1999; Maggio-Price et al., 2006). Mouse lacking TGF-2 in the colonic epithelium showed increased adenoma and adenocarcinoma formation after carcinogen treatment. Apart from TGF-, mouse models representing alterations in the other factors in this important pathway like SMAD2, SMAD4 and SMAD3 have been reported. (Eppert et al., 1996). Although mice lacking SMAD2 and SMAD4 are embryonically lethal, the SMAD3 deficient mice are viable and are a useful model of CRC (Zhu et al., 1998). Importance of the TGF- related models lies in the fact that the APC remains intact in the adenomas of these mice and these models could serve as a valuable tool to investigate non-WNT/APC/ catenin-mediated colorectal carcinogenesis (Kaiser et al., 2007). Interestingly, the TGF- related models in the APCMin/+ background showed increased incidence of invasive carcinoma

Inflammatory bowel diseases (IBD) like ulcerative colitis (UC) and Crohn's Disease are predisposing conditions of CRC (Itzkowitz & Harpaz, 2004; Itzkowitz & Yio, 2004). Prolonged administration of dextransulfate sodium (DSS) in mice resulting in chronic colitis and formation of high-grade dysplasia confirmed the involvement of chronic inflammation in colorectal carcinogenesis (Okayasu et al., 1990). Interestingly intestinal tumorigenesis was augmented by combined administration of AOM and DSS. (Tanaka et al., 2003). To demonstrate that deficiency of *Sigirr* (single immunoglobulin and tollinterleukin 1 receptor domain) along with bacteria-induced inflammation increases susceptibility to CRC investigators effectively utilized the combined AOM/DSS model (Wald et al., 2003; Xiao et al., 2007; Uronis & Threadgill, 2009). Furthermore, the combined AOM/DSS model was also an important instrument in defining the role of the JAK/STAT (Janus kinase/signal transducers and activators of transcription) and NFkB (nuclear factor of kappa light chain gene enhancer in B-cells) pathways in inflammation-induced CRC (Wirtz & Neurath, 2007). The role of the JAK/STAT pathway in colorectal carcinogenesis was further confirmed by the fact that dysfunctional *Socs1* and *Socs3* (suppressors of cytokine signaling) leads to enhanced activation of STAT1, STAT3 and NFkB and subsequent growth of colorectal tumors (Hanada et al., 2006). Although lack of only *Socs3*  in the intestinal epithelial cells is not associated with chronic inflammation or tumor formation, these mice when treated with AOM/DSS showed distinct inflammatory response followed by colonic tumors (Hanada et al., 2006; Rigby et al., 2007). In contrast to increased nuclear -catenin in the *Socs3* deficient mice, the colorectal tumors in *Socs1*  deficient mice display enhanced expression of *Myc* (Sutherland et al., 2006). In addition, the Muc2-/- is an important animal model to study the role of inflammation in the colorectal carcinogenesis. This model specifically targets mucin-forming Muc2 and unlike other models tumor formation is also observed in the rectum (Mack & Hollingsworth,

significant incidence of carcinoma of the cecum and colon in these mice, the *Rag2*/*TGF-*

*1* deficient mice due to widespread inflammation die around

*1-*

*1-*deficient mice survive until adulthood (Diebold et al., 1995). Due to the

been reported that the *TGF-*

**2.2.4 Inflammation mediated model***s*

1994; Yang et al., 2005; Femia et al., 2009).

of *Rag2* the *TGF-*

specifically in the distal colon (Takaku et al., 1998; Sodir et al., 2006).

Models have been developed to investigate the role of immune system in colorectal carcinogenesis. Immune cells are involved in inflammatory response and inflammation is intimately related to CRC. Therefore, the IL-2 (interleukin-2), IL-10, and TCR knockout models were developed and used in studying role of diet and inflammation in colorectal cancer initiation and progression. (Rudolph et al., 1995; Kullberg et al., 1998; Mizoguchi et al., 2000; Seril et al., 2005).

#### **2.2.6 Carcinogen-induced model***s*

Although important information on familial and sporadic colorectal carcinogenesis was obtained from genetic animal models, it is the carcinogen-induced animal models that were instrumental in delineating molecular events specifically in the sporadic CRC. It is worth mentioning that the colon-specific carcinogen dimethylhyrdrazine (DMH) along with AOM has been useful in developing our current understanding of the molecular mechanisms underlying sporadic colorectal carcinogenesis (Druckrey et al., 1966). To its advantage the carcinogen-induced mouse model develops tumor which show much similarities to the pathophysiology of the human CRC (Kaiser et al., 2007; Uronis et al., 2007). The carcinogenmediated tumors show alterations in the WNT/-catenin pathway (Takahashi et al., 2000). Interestingly, the AOM-induced colorectal carcinogenesis, unlike in the APCMin/+, is mainly due to the mutations in the *Ctnnb1 gene*, which encodes -catenin protein. Mutations in the *Ctnnb1* results in ubiquitination-resistant stabilization of the -catenin leading to growth of colorectal adenomas associated with upregulation of proliferation markers like cyclin D1 and c*Myc* (Wang et al., 1998; Kaiser et al., 2007). The carcinogen-induced model was also a key player in the identification of the modifier loci like the *Ptprj* (a receptor-type protein tyrosine phosphatase), which has been shown to modify susceptibility to DMH and has shown frequent loss of heterozygosity in human colon cancer (Ruivenkamp et al., 2002). Additionally, the carcinogen-induced models led to the recognition of *Pref1* as a modifier of CRC and the promoter of the *Pref1* is suggested to contain a -catenin/TCF response element (Dong et al., 2004; Ruivenkamp et al., 2002; Uronis et al., 2007).

#### **2.2.7 Mouse models in chemoprevention and therapeutics development**

The genetic mouse models as well as the carcinogen-induced models of CRC (APC, TGF and mismatch repair based models) are used to evaluate the effect of diets and chemopreventive agents. The chemopreventive and dietary interventions are usually started between 3 and 6 weeks of age and invariably the principal biological endpoint is the number and grade of the tumors. It is important to note that in contrast to human CRC where no small intestinal tumors are observed, the tumors in the mutant models are mostly in the small intestine. However, the concept that the non-steroidal anti-inflammatory drugs (NSAID) have a chemopreventive role in the CRC was first established in these models showing reduction in tumor number in the small intestine. The chemopreventive properties of NSAIDs like celecoxib and piroxicam were also validated in the carcinogen-induced models showing decrease in polyp number and size. Due to commercial availability, the APCMin/+ is the model of choice in many studies and a list of agents screened in APCMin/+ mouse model is shown in Table 3. Most of the other CRC mouse models are used for specific biological end points. But a few of them such as APC1638N model because of its longer life span and good signal to noise ratio is used for carcinogen testing and is suggested to be more useful than APCMin/+ mice (Trani et al., 2010).

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 287

Colon cancer xenograft models are created by implantation of cells subcutaneously, intrasplenically, or into the renal capsule. It is important to implant the xenograft into the immunocompromised mice and commonly the T-cell deficient ''nude'' mice or NOD-SCID (non-obese diabetic/severe combined immunodeficiency) mice are used (Rygaard & Povlsen, 1969). The xenograft models of CRC are commonly used to assess newer therapeutics and understand the pathogenesis of human disease. Indeed, subcutaneous xenografts have found an important place in CRC research due to the fact that anesthetics are not required and the tumors are accessible for external measurement. Some of the disadvantages of the subcutaneous model are i] lack of tumor microenvironment representative of the CRC, and ii] in contrast to the >50% hepatic metastatic incidence of CRC, no metastasis is observed in the subcutaneous xenograft models. However, xenograft models involving intrasplenic or intra-renal-capsule, although have shown metastasis similar to human CRC, does not represent tumor microenvironment of CRC and signaling pathway could be different than the human disease (Furukawa et al., 1993; Fidler, 1991a, b, c). Consequently, implantation of CRC xenografts into mouse colon, the orthotopic model, is much preferred by the investigators due to their similar characteristics of the human ailment.

An orthotopic mouse model involves placing of colorectal cancer cell or tumor tissue into the intestinal sub-mucosa (Tseng et al., 2007). The orthotopic model, unlike the subcutaneous model, is associated with all of the components of the tumor microenvironment as well as all of the angiogenic and growth factors, and cytokines. In addition to mimicking the human CRC in terms of metastasis and microenvironments, the orthotopic model also allows assessment of the alterations in the microenvironment on tumor initiation and progression. From a technical point of view, generation of orthotopic models demands specific expertise and more time than subcutaneous models. Because of technical difficulties in the physical measurement of the tumors, the orthotopic model also requires that an appropriate reporter like luciferase be in place for measuring tumor growth to determine the efficacy of a drug treatment. As with any animal model of human diseases there are inherent shortcoming and the orthotopic animal model is no exception. Because the tumors are in the colon, the orthotopic model requires sacrifice of the animals at a predetermined time for quantitative and qualitative analysis of the tumor. However, the orthotopic model has the advantage of mimicking human CRC including tumor

Signaling pathways involved in colorectal carcinogenesis are conserved across species and zebrafish, a well-characterized simple model system for human disease, are widely used to understand the molecular basis of cancer including CRC. Water borne carcinogens induce a wide variety of benign and malignant tumors in many organs of zebrafish. Zebrafish due to its easy maintainence and breeding along with conservation of human cancer-relevant oncogenes, and tumor suppressor and cell cycle genes makes it a useful model to study carcinogenesis. Interestingly, the zebrafish mutants display phenotypes similar to many human disorders, including cancer, cardiovascular disease, and neurodegeneration. Zebrafish carrying a mutation in the region representing most of the observed human APC

**3. Zebrafish – A non-murine model of colorectal cancer** 

**2.3 Xenograft models of colon cancer**

**2.4 Orthotopic mouse model** 

microenvironment.


Table 3. Use of APCMin/+ mouse model in chemoprevention development.

#### **2.3 Xenograft models of colon cancer**

286 Colorectal Cancer – From Prevention to Patient Care

(Sale et al., 2006)

(Floyd et al., 2010)

(Raufman et al., 2011)

Name of compound References

DMU-135 (3-4 methylenedioxy-3'4'5'-trimethoxy chalcone), a potent tyrosine kinase inhibitor prodrug

Alpha-phenyl-tert-butyl-nitrone (PBN) and 4

Scopolamine butylbromide (muscarinic receptor

hydroxy-PBN

antagonist)

Aspirin (Barnes&Lee, 1998) Piroxicam (Ritland&Gendler, 1999) Phenanthridinone derivative (PJ 34) (Mabley et al., 2004) Non-steroidal anti-inflammatory drugs (NSAIDs) (Gescher, 2004)

Atorvastatin (Swamy et al., 2006) Celecoxib (Swamy et al., 2006) Sulforaphane (Isothiocynate compound) (Khor et al., 2006) Anthocyanin (Bobe et al., 2006) Difluoro methylornithine (DFMO) (Telang&Katdare, 2007) Epigallocatechin 3- gallate (EGCG) (Telang&Katdare, 2007) Dibenzoylmethane (Tammariello&Milner, 2010) PPAR ligand MCC-555 (Yamaguchi et al., 2008) Metformin (Tomimoto et al., 2008)

Chafuroside (Tammariello&Milner, 2010)

Sodium Taurocholate (Smith et al., 2010) COX-2 inhibitors (Nakanishi et al., 2011)

Curcumin (Murphy et al., 2011) Silibinin (Rajamanickam et al., 2010) Ellagic acid (Mutanen et al., 2008) Epigallocatechin gallate (EGCG) (Telang&Katdare, 2007) Dietary sphingolipids (Symolon et al., 2004) Dietary Folate (Song et al., 2000) Dietary isoflavones (Sorensen et al., 1998) Apple polyphenol extract (APE) (Fini et al., 2011)

Grape Seed extract (GSE) (Velmurugan et al., 2010) Berries (bilberry, lingonberry, cloudberry) (Mutanen et al., 2008) Green tea (Issa et al., 2007) Orange peel extract (OPE) (Fan et al., 2007) Anthocyanin rich tart cherry extract (Bobe et al., 2006)

Fermented brown rice and rice bran (Phutthaphadoong et al., 2010)

Dietary /caloric restriction (Tammariello&Milner, 2010) Exercise/physical activity (Baltgalvis et al., 2008) Table 3. Use of APCMin/+ mouse model in chemoprevention development.

Fish oil (Bose et al., 2007)

Colon cancer xenograft models are created by implantation of cells subcutaneously, intrasplenically, or into the renal capsule. It is important to implant the xenograft into the immunocompromised mice and commonly the T-cell deficient ''nude'' mice or NOD-SCID (non-obese diabetic/severe combined immunodeficiency) mice are used (Rygaard & Povlsen, 1969). The xenograft models of CRC are commonly used to assess newer therapeutics and understand the pathogenesis of human disease. Indeed, subcutaneous xenografts have found an important place in CRC research due to the fact that anesthetics are not required and the tumors are accessible for external measurement. Some of the disadvantages of the subcutaneous model are i] lack of tumor microenvironment representative of the CRC, and ii] in contrast to the >50% hepatic metastatic incidence of CRC, no metastasis is observed in the subcutaneous xenograft models. However, xenograft models involving intrasplenic or intra-renal-capsule, although have shown metastasis similar to human CRC, does not represent tumor microenvironment of CRC and signaling pathway could be different than the human disease (Furukawa et al., 1993; Fidler, 1991a, b, c). Consequently, implantation of CRC xenografts into mouse colon, the orthotopic model, is much preferred by the investigators due to their similar characteristics of the human ailment.

#### **2.4 Orthotopic mouse model**

An orthotopic mouse model involves placing of colorectal cancer cell or tumor tissue into the intestinal sub-mucosa (Tseng et al., 2007). The orthotopic model, unlike the subcutaneous model, is associated with all of the components of the tumor microenvironment as well as all of the angiogenic and growth factors, and cytokines. In addition to mimicking the human CRC in terms of metastasis and microenvironments, the orthotopic model also allows assessment of the alterations in the microenvironment on tumor initiation and progression. From a technical point of view, generation of orthotopic models demands specific expertise and more time than subcutaneous models. Because of technical difficulties in the physical measurement of the tumors, the orthotopic model also requires that an appropriate reporter like luciferase be in place for measuring tumor growth to determine the efficacy of a drug treatment. As with any animal model of human diseases there are inherent shortcoming and the orthotopic animal model is no exception. Because the tumors are in the colon, the orthotopic model requires sacrifice of the animals at a predetermined time for quantitative and qualitative analysis of the tumor. However, the orthotopic model has the advantage of mimicking human CRC including tumor microenvironment.

#### **3. Zebrafish – A non-murine model of colorectal cancer**

Signaling pathways involved in colorectal carcinogenesis are conserved across species and zebrafish, a well-characterized simple model system for human disease, are widely used to understand the molecular basis of cancer including CRC. Water borne carcinogens induce a wide variety of benign and malignant tumors in many organs of zebrafish. Zebrafish due to its easy maintainence and breeding along with conservation of human cancer-relevant oncogenes, and tumor suppressor and cell cycle genes makes it a useful model to study carcinogenesis. Interestingly, the zebrafish mutants display phenotypes similar to many human disorders, including cancer, cardiovascular disease, and neurodegeneration. Zebrafish carrying a mutation in the region representing most of the observed human APC

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 289

in gut motility, hormones and immune surveillance, and differences in genetic events in somatic cells during adenoma to carcinoma transition in mouse compared to human

With few exceptions, a significant correlation was observed between animal and human studies. Both the AOM rat model and the mutant mouse models supported the chemopreventive effects of the NSAID. This is supportive of the epidemiological studies proposing that, NSAIDs might reduce the colorectal cancer incidence by at least 45% in humans. Also supportive is the effects of celecoxib and sulindac shown to decrease the number of polyps in FAP patient trials. Similar to human data, rats and mice fed a high-fat diet showed increased adenomas than those fed a low-fat diet thus establishing the relationship between the colon cancer incidence and the intake of fat. Fatty diets with high linoleic acid content and n-6-polyunsaturated fatty acids seem to increase the number of tumors in rodents. Moreover, caloric reduction is a strategy that seems very efficient in animals. A reasonable agreement is observed between the results of these animal studies

and the more limited clinical studies with few differences (Corpet & Pierre, 2003).

Currently a number of animal models are available to dissect various facets of CRC and to undertake risk estimation studies. Mutant mouse models provide a unique opportunity in studying numerous adenomas under defined experimental conditions and uniform genetic background. However, use of animal models in studying human disease has its own limitations. For example, in carcinogen-induced models of CRC, the tumor incidence and latency period could be modulated by amount of carcinogen used – higher amount of carcinogen leading to higher incidence of tumors. However, high ethanol consumption reduced carcinogen (DMH)-induced tumorigenesis suggesting that DMH model is not useful in determining the role of alcohol in CRC. This discrepancy was resolved using the APCMin/+ mice model where ethanol consumption was observed as a risk factor for CRC (Roy et al., 2002). Careful consideration is essential for the selection of animal model to study a particular agent and requires validation is two or more models for the unequivocal

Future advances in animal model development will require combinations of dietary and genetic manipulation of rodents or other inexpensive animals to more accurately mimic the various factors that contribute to colorectal neoplasia in humans. As epidemiologic and molecular studies demonstrate the heterogeneity of colorectal tumor development in diverse populations (e.g. the microsatellite instability or CIMP pathways), it is expected that any one model will not answer all the questions about the CRC chemoprevention or therapeutic

This work was supported by funding from NASA (Grant#NNX09AU95G).

**5. Comparison of human data with animal model data** 

colorectal neoplasia.

**6. Limitations of animal models** 

intervention strategies under investigation.

demonstration.

**7. Conclusions** 

**8. Acknowledgment** 

mutations were identified recently and like murine models the heterozygous fish develop intestinal adenomas (12%), which resembles its murine counterpart (Goessling et al., 2007). The APC-heterozygous fish when exposed to dimethylbenzanthracene, showed significant increase in the tumor number with 44% developing liver tumors and 35% developing intestinal tumors and could serve as an important model system to screen carcinogens (Goessling et al., 2007).

#### **3.1 Zebrafish model in chemoprevention and therapy**

Zebrafish model is in use for the target selection, bioactive compound screening as well as in the drug toxicity and efficacy studies (Lieschke & Currie, 2007). Angiogenesis supports cancer progression including CRC and anti-angiogenic therapy inhibits cancer growth. On the contrary anti-angiogenic therapy has been implicated in inflammation a known risk factor of colorectal carcinogenesis. Our understanding of the correlation between tumor angiogenesis, inflammation, and metastasis was much enhanced by studies in the zebrafish model (Moshal et al., 2010). Furthermore, the zebrafish model has been used to study potential therapeutic agents like SKLB610 (inhibitor of angiogenesis related tyrosine kinase), which was reported to inhibit angiogenesis in zebrafish sub-intestinal veins (Cao et al., 2011). The zebrafish model was useful in determining the role of DNA demthylase in maintaining intestinal epithelial cells lacking APC in an undifferentiated state (Rai et al., 2010). Additionally, it has been reported that the zebrafish expressing a truncated form of APC with either retinoic acid or a selective COX-2 inhibitor decreased -catenin in the cell. Curcumin-loaded biodegradable polymeric micelles (Cur/MPEG-PCL) has been shown to efficiently block angiogenesis in transgenic zebrafish model (Gou et al., 2011). Similarly, the incorporation of doxorubicin in MPEG-PCL micelles enhanced the anticancer activity and decreased the systemic toxicity of doxorubicin in Zebrafish and has implications for CRC treatment (Lee et al., 2006).

#### **4. Animal models in chemoprevention and chemotherapeutics development**

Rodent models have been used for CRC research providing insight into the complex oncogenic events contributing to the loss of cell growth and differentiation control. These models also offer prospects to identify and study both therapeutic and chemopreventive agents. In general, almost all popular human colorectal cancer prevention strategies, from dietary manipulations (such as folate or calcium supplementation), to drug testing (such as (NSAIDs) have been evaluated in both carcinogen-induced and genetically modified animal models. In most cases, suppression of polyps has demonstrated the preventive effects of these strategies, and in some cases, investigators have been able to dissect pathways where these agents block the development of aberrant crypt foci (ACF). Moreover, approaches that are effective in preventing the early stages of colorectal tumorigenesis have been shown to actually promote tumor growth in later stages of the adenoma-carcinoma sequence. This type of observation in mice is important in polyp prevention studies in humans where folate supplementation may actually be harmful in subjects already predisposed to colorectal neoplasia. Although it is important to note that caution has to be exercised in extrapolating animal model data to human, at least studies with NSAID have shown similar protective effects both in human and animal. However, some differential response of chemopreventive drugs has been observed between animal studies and human response. This could be because of higher genetic homogeneity of mice compared to humans, physiologic differences in gut motility, hormones and immune surveillance, and differences in genetic events in somatic cells during adenoma to carcinoma transition in mouse compared to human colorectal neoplasia.

#### **5. Comparison of human data with animal model data**

With few exceptions, a significant correlation was observed between animal and human studies. Both the AOM rat model and the mutant mouse models supported the chemopreventive effects of the NSAID. This is supportive of the epidemiological studies proposing that, NSAIDs might reduce the colorectal cancer incidence by at least 45% in humans. Also supportive is the effects of celecoxib and sulindac shown to decrease the number of polyps in FAP patient trials. Similar to human data, rats and mice fed a high-fat diet showed increased adenomas than those fed a low-fat diet thus establishing the relationship between the colon cancer incidence and the intake of fat. Fatty diets with high linoleic acid content and n-6-polyunsaturated fatty acids seem to increase the number of tumors in rodents. Moreover, caloric reduction is a strategy that seems very efficient in animals. A reasonable agreement is observed between the results of these animal studies and the more limited clinical studies with few differences (Corpet & Pierre, 2003).

#### **6. Limitations of animal models**

Currently a number of animal models are available to dissect various facets of CRC and to undertake risk estimation studies. Mutant mouse models provide a unique opportunity in studying numerous adenomas under defined experimental conditions and uniform genetic background. However, use of animal models in studying human disease has its own limitations. For example, in carcinogen-induced models of CRC, the tumor incidence and latency period could be modulated by amount of carcinogen used – higher amount of carcinogen leading to higher incidence of tumors. However, high ethanol consumption reduced carcinogen (DMH)-induced tumorigenesis suggesting that DMH model is not useful in determining the role of alcohol in CRC. This discrepancy was resolved using the APCMin/+ mice model where ethanol consumption was observed as a risk factor for CRC (Roy et al., 2002). Careful consideration is essential for the selection of animal model to study a particular agent and requires validation is two or more models for the unequivocal demonstration.

#### **7. Conclusions**

288 Colorectal Cancer – From Prevention to Patient Care

mutations were identified recently and like murine models the heterozygous fish develop intestinal adenomas (12%), which resembles its murine counterpart (Goessling et al., 2007). The APC-heterozygous fish when exposed to dimethylbenzanthracene, showed significant increase in the tumor number with 44% developing liver tumors and 35% developing intestinal tumors and could serve as an important model system to screen carcinogens

Zebrafish model is in use for the target selection, bioactive compound screening as well as in the drug toxicity and efficacy studies (Lieschke & Currie, 2007). Angiogenesis supports cancer progression including CRC and anti-angiogenic therapy inhibits cancer growth. On the contrary anti-angiogenic therapy has been implicated in inflammation a known risk factor of colorectal carcinogenesis. Our understanding of the correlation between tumor angiogenesis, inflammation, and metastasis was much enhanced by studies in the zebrafish model (Moshal et al., 2010). Furthermore, the zebrafish model has been used to study potential therapeutic agents like SKLB610 (inhibitor of angiogenesis related tyrosine kinase), which was reported to inhibit angiogenesis in zebrafish sub-intestinal veins (Cao et al., 2011). The zebrafish model was useful in determining the role of DNA demthylase in maintaining intestinal epithelial cells lacking APC in an undifferentiated state (Rai et al., 2010). Additionally, it has been reported that the zebrafish expressing a truncated form of APC with either retinoic acid or a selective COX-2 inhibitor decreased -catenin in the cell. Curcumin-loaded biodegradable polymeric micelles (Cur/MPEG-PCL) has been shown to efficiently block angiogenesis in transgenic zebrafish model (Gou et al., 2011). Similarly, the incorporation of doxorubicin in MPEG-PCL micelles enhanced the anticancer activity and decreased the systemic toxicity of doxorubicin in Zebrafish and has implications for CRC

**4. Animal models in chemoprevention and chemotherapeutics development**  Rodent models have been used for CRC research providing insight into the complex oncogenic events contributing to the loss of cell growth and differentiation control. These models also offer prospects to identify and study both therapeutic and chemopreventive agents. In general, almost all popular human colorectal cancer prevention strategies, from dietary manipulations (such as folate or calcium supplementation), to drug testing (such as (NSAIDs) have been evaluated in both carcinogen-induced and genetically modified animal models. In most cases, suppression of polyps has demonstrated the preventive effects of these strategies, and in some cases, investigators have been able to dissect pathways where these agents block the development of aberrant crypt foci (ACF). Moreover, approaches that are effective in preventing the early stages of colorectal tumorigenesis have been shown to actually promote tumor growth in later stages of the adenoma-carcinoma sequence. This type of observation in mice is important in polyp prevention studies in humans where folate supplementation may actually be harmful in subjects already predisposed to colorectal neoplasia. Although it is important to note that caution has to be exercised in extrapolating animal model data to human, at least studies with NSAID have shown similar protective effects both in human and animal. However, some differential response of chemopreventive drugs has been observed between animal studies and human response. This could be because of higher genetic homogeneity of mice compared to humans, physiologic differences

(Goessling et al., 2007).

treatment (Lee et al., 2006).

**3.1 Zebrafish model in chemoprevention and therapy** 

Future advances in animal model development will require combinations of dietary and genetic manipulation of rodents or other inexpensive animals to more accurately mimic the various factors that contribute to colorectal neoplasia in humans. As epidemiologic and molecular studies demonstrate the heterogeneity of colorectal tumor development in diverse populations (e.g. the microsatellite instability or CIMP pathways), it is expected that any one model will not answer all the questions about the CRC chemoprevention or therapeutic intervention strategies under investigation.

#### **8. Acknowledgment**

This work was supported by funding from NASA (Grant#NNX09AU95G).

Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 291

Corpet, DE. & Pierre, F. (2003). Point: From animal models to prevention of colon cancer.

Corpet, DE. & Pierre, F. (2005). How good are rodent models of carcinogenesis in predicting

Diebold, RJ.; Eis, MJ.; Yin, M.; Ormsby, I.; Boivin, GP.; Darrow, BJ.; Saffitz, JE. &

Doi, K.; Sakai, K.; Tanaka, R.; Toma, K.; Yamaguchi, T.; Wei, M.; Fukushima, S. &

Dong, M.; Guda, K.; Nambiar, PR.; Nakanishi, M.; Lichtler, AC.; Nishikawa, M.; Giardina, C.

Druckrey, H.; Ivankovic, S. & Preussmann, R. (1966). Teratogenic and carcinogenic effects in

Edelmann, W.; Cohen, PE.; Kneitz, B.; Winand, N.; Lia, M.; Heyer, J.; Kolodner, R.; Pollard,

Edelmann, W.; Umar, A.; Yang, K.; Heyer, J.; Kucherlapati, M.; Lia, M.; Kneitz, B.;

Edelmann, W.; Yang, K.; Kuraguchi, M.; Heyer, J.; Lia, M.; Kneitz, B.; Fan, K.; Brown, AM.;

Edelmann, W.; Yang, K.; Umar, A.; Heyer, J.; Lau, K.; Fan, K.; Liedtke, W.; Cohen, PE.; Kane,

Engle, SJ.; Hoying, JB.; Boivin, GP.; Ormsby, I.; Gartside, PS. & Doetschman, T. (1999).

cooperate in intestinal tumor suppression. *Cancer Res*, 60, 803-807.

chemoprevention in rats, mice and men. *Eur J Cancer*, 41, 1911-1922. Corpet, DE.; Tache, S.; Archer, MC. & Bruce, WR. (2008). Dehydroalanine and lysinoalanine

environmental modifiers. *Lab Invest*, 84, 1619-1630.

system. *Cancer Epidemiol Biomarkers Prev*, 12, 391-400.

del Junco, GW. (2002). Pathology for Colon and Rectal Surgeons: web2.airmail.net/uthman/pdf\_documents/colorect.pdf.

organ carcinogenesis bioassay. *Cancer Lett*, 289, 161-169.

chromosome pairing in meiosis. *Nat Genet*, 21, 123-127.

mutant mice. *Cancer Res*, 59, 1301-1307.

causes cancer susceptibility. *Cell*, 91, 467-477.

early stage of tumorigenesis. *Cancer Res*, 59, 3379-3386.

3037-3042.

92, 12215-12219.

2239-2246.

1378-1379.

a new mouse model of familial adenomatous polyposis: influence of genetic and

Systematic review of chemoprevention in min mice and choice of the model

efficacy in humans? A systematic review and meta-analysis of colon

in thermolyzed casein do not promote colon cancer in the rat. *Food Chem Toxicol*, 46,

Doetschman, T. (1995). Early-onset multifocal inflammation in the transforming growth factor beta 1-null mouse is lymphocyte mediated. *Proc Natl Acad Sci U S A*,

Wanibuchi, H. (2010). Chemopreventive effects of 13alpha,14alpha-epoxy-3betamethoxyserratan-21beta-ol (PJJ-34), a serratane-type triterpenoid, in a rat multi-

& Rosenberg, DW. (2004). Azoxymethane-induced pre-adipocyte factor 1 (Pref-1) functions as a differentiation inhibitor in colonic epithelial cells. *Carcinogenesis*, 25,

the offspring after single injection of ethylnitrosourea to pregnant rats. *Nature*, 210,

JW. & Kucherlapati, R. (1999). Mammalian MutS homologue 5 is required for

Avdievich, E.; Fan, K.; Wong, E.; Crouse, G.; Kunkel, T.; Lipkin, M.; Kolodner, RD. & Kucherlapati, R. (2000). The DNA mismatch repair genes Msh3 and Msh6

Lipkin, M. & Kucherlapati, R. (1999). Tumorigenesis in Mlh1 and Mlh1/Apc1638N

MF.; Lipford, JR.; Yu, N.; Crouse, GF.; Pollard, JW.; Kunkel, T.; Lipkin, M.; Kolodner, R. & Kucherlapati, R. (1997). Mutation in the mismatch repair gene Msh6

Transforming growth factor beta1 suppresses nonmetastatic colon cancer at an

#### **9. References**


Amos-Landgraf, JM.; Kwong, LN.; Kendziorski, CM.; Reichelderfer, M.; Torrealba, J.;

Baker, SM.; Bronner, CE.; Zhang, L.; Plug, AW.; Robatzek, M.; Warren, G.; Elliott, EA.; Yu, J.;

Baltgalvis, KA.; Berger, FG.; Pena, MM.; Davis, JM. & Carson, JA. (2008). Effect of exercise

Barnes, CJ. & Lee, M. (1998). Chemoprevention of spontaneous intestinal adenomas in the

Batlle, E.; Bacani, J.; Begthel, H.; Jonkheer, S.; Gregorieff, A.; van de Born, M.; Malats, N.;

Blobe, GC.; Schiemann, WP. & Lodish, HF. (2000). Role of transforming growth factor beta

Bobe, G.; Wang, B.; Seeram, NP.; Nair, MG. & Bourquin, LD. (2006). Dietary anthocyanin-

Bonin, AM.; Yanez, JA.; Fukuda, C.; Teng, XW.; Dillon, CT.; Hambley, TW.; Lay, PA. &

Bose, M.; Hao, X.; Ju, J.; Husain, A.; Park, S.; Lambert, JD. & Yang, CS. (2007). Inhibition of

Cao, ZX.; Zheng, RL.; Lin, HJ.; Luo, SD.; Zhou, Y.; Xu, YZ.; Zeng, XX.; Wang, Z.; Zhou, LN.;

Chung, CP.; Hsu, HY.; Huang, DW.; Hsu, HH.; Lin, JT.; Shih, CK. & Chiang, W. (2010). Ethyl

Clarke, JM.; Topping, DL.; Bird, AR.; Young, GP. & Cobiac, L. (2008). Effects of high-

an anti-inflammatory pathway. *J Agric Food Chem*, 58, 7616-7623.

suboptimal levels of sulindac. *J Agric Food Chem*, 54, 9322-9328.

colon cancer. *Proc Natl Acad Sci U S A*, 104, 4036-4041.

in human disease. *N Engl J Med*, 342, 1350-1358.

and fish oil. *J Agric Food Chem*, 55, 7695-7700.

*Chemother Pharmacol*, 66, 755-764.

meiosis. *Cell*, 82, 309-319.

1137-1143.

873-877.

565-574.

*Res*, 155, 85-103.

Weichert, J.; Haag, JD.; Chen, KS.; Waller, JL.; Gould, MN. & Dove, WF. (2007). A target-selected Apc-mutant rat kindred enhances the modeling of familial human

Ashley, T.; Arnheim, N.; Flavell, RA. & Liskay, RM. (1995). Male mice defective in the DNA mismatch repair gene PMS2 exhibit abnormal chromosome synapsis in

on biological pathways in ApcMin/+ mouse intestinal polyps. *J Appl Physiol*, 104,

adenomatous polyposis coli Min mouse model with aspirin. *Gastroenterology*, 114,

Sancho, E.; Boon, E.; Pawson, T.; Gallinger, S.; Pals, S. & Clevers, H. (2005). EphB receptor activity suppresses colorectal cancer progression. *Nature*, 435, 1126-1130. Bellam, N. & Pasche, B. (2010). Tgf-beta signaling alterations and colon cancer. *Cancer Treat* 

rich tart cherry extract inhibits intestinal tumorigenesis in APC(Min) mice fed

Davies, NM. (2010). Inhibition of experimental colorectal cancer and reduction in renal and gastrointestinal toxicities by copper-indomethacin in rats. *Cancer* 

tumorigenesis in ApcMin/+ mice by a combination of (-)-epigallocatechin-3-gallate

Mao, YQ.; Yang, L.; Wei, YQ.; Yu, LT.; Yang, SY. & Zhao, YL. (2011). SKLB610: A Novel Potential Inhibitor of Vascular Endothelial Growth Factor Receptor Tyrosine Kinases Inhibits Angiogenesis and Tumor Growth in Vivo. *Cell Physiol Biochem*, 27,

acetate fraction of adlay bran ethanolic extract inhibits oncogene expression and suppresses DMH-induced preneoplastic lesions of the colon in F344 rats through

amylose maize starch and butyrylated high-amylose maize starch on azoxymethane-induced intestinal cancer in rats. *Carcinogenesis*, 29, 2190-2194. Colnot, S.; Niwa-Kawakita, M.; Hamard, G.; Godard, C.; Le Plenier, S.; Houbron, C.;

Romagnolo, B.; Berrebi, D.; Giovannini, M. & Perret, C. (2004). Colorectal cancers in

**9. References** 

a new mouse model of familial adenomatous polyposis: influence of genetic and environmental modifiers. *Lab Invest*, 84, 1619-1630.


Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 293

Gescher, A. (2004). Polyphenolic phytochemicals versus non-steroidal anti-inflammatory drugs: which are better cancer chemopreventive agents? *J Chemother*, 16 Suppl 4, 3-6. Goessling, W.; North, TE. & Zon, LI. (2007). New waves of discovery: modeling cancer in

Goodman, DG.; Ward, JM.; Squire, RA.; Paxton, MB.; Reichardt, WD.; Chu, KC. & Linhart,

Gou, M.; Men, K.; Shi, H.; Xiang, M.; Zhang, J.; Song, J.; Long, J.; Wan, Y.; Luo, F.; Zhao, X. &

Gould, KA.; Dietrich, WF.; Borenstein, N.; Lander, ES. & Dove, WF. (1996). Mom1 is a semi-

Gounari, F.; Chang, R.; Cowan, J.; Guo, Z.; Dose, M.; Gounaris, E. & Khazaie, K. (2005). Loss

Grady, WM.; Myeroff, LL.; Swinler, SE.; Rajput, A.; Thiagalingam, S.; Lutterbaugh, JD.;

Hanada, T.; Kobayashi, T.; Chinen, T.; Saeki, K.; Takaki, H.; Koga, K.; Minoda, Y.; Sanada,

Hess, LM.; Krutzsch, MF.; Guillen, J.; Chow, HH.; Einspahr, J.; Batta, AK.; Salen, G.; Reid,

Hu, R.; Khor, TO.; Shen, G.; Jeong, WS.; Hebbar, V.; Chen, C.; Xu, C.; Reddy, B.; Chada, K. &

Imaida, K.; Tamano, S.; Hagiwara, A.; Fukushima, S.; Shirai, T. & Ito, N. (2003). Application

Injac, R.; Perse, M.; Cerne, M.; Potocnik, N.; Radic, N.; Govedarica, B.; Djordjevic, A.; Cerar,

Issa, AY.; Volate, SR.; Muga, SJ.; Nitcheva, D.; Smith, T. & Wargovich, MJ. (2007). Green tea

of endocrine active substances. *Pure Appl. Chem*, 75, 2491-2495.

on somatic Apc inactivation. *Cancer Res*, 67, 9721-9730.

cancer therapy in vitro and in vivo. *Nanoscale*, 3, 1558-1567.

MS. (1980). Neoplastic and nonneoplastic lesions in aging Osborne-Mendel rats.

Qian, Z. (2011). Curcumin-loaded biodegradable polymeric micelles for colon

dominant modifier of intestinal adenoma size and multiplicity in Min/+ mice.

of adenomatous polyposis coli gene function disrupts thymic development. *Nat* 

Neumann, A.; Brattain, MG.; Chang, J.; Kim, SJ.; Kinzler, KW.; Vogelstein, B.; Willson, JK. & Markowitz, S. (1999). Mutational inactivation of transforming growth factor beta receptor type II in microsatellite stable colon cancers. *Cancer Res*,

T.; Yoshioka, T.; Mimata, H.; Kato, S. & Yoshimura, A. (2006). IFNgammadependent, spontaneous development of colorectal carcinomas in SOCS1-deficient

ME.; Earnest, DL. & Alberts, DS. (2004). Results of a phase I multiple-dose clinical study of ursodeoxycholic Acid. *Cancer Epidemiol Biomarkers Prev*, 13, 861-867. Hinoi, T.; Akyol, A.; Theisen, BK.; Ferguson, DO.; Greenson, JK.; Williams, BO.; Cho, KR. &

Fearon, ER. (2007). Mouse model of colonic adenoma-carcinoma progression based

Kong, AN. (2006). Cancer chemoprevention of intestinal polyposis in ApcMin/+ mice by sulforaphane, a natural product derived from cruciferous vegetable.

of rat medium-term bioassays for detecting carcinogenic and modifying potentials

A. & Strukelj, B. (2009). Protective effects of fullerenol C60(OH)24 against doxorubicin-induced cardiotoxicity and hepatotoxicity in rats with colorectal

selectively targets initial stages of intestinal carcinogenesis in the AOM-ApcMin

zebrafish. *J Clin Oncol*, 25, 2473-2479.

*Toxicol Appl Pharmacol*, 55, 433-447.

*Genetics*, 144, 1769-1776.

mice. *J Exp Med*, 203, 1391-1397.

*Carcinogenesis*, 27, 2038-2046.

cancer. *Biomaterials*, 30, 1184-1196.

mouse model. *Carcinogenesis*, 28, 1978-1984.

*Immunol*, 6, 800-809.

59, 320-324.


Eppert, K.; Scherer, SW.; Ozcelik, H.; Pirone, R.; Hoodless, P.; Kim, H.; Tsui, LC.; Bapat, B.;

Fan, K.; Kurihara, N.; Abe, S.; Ho, CT.; Ghai, G. & Yang, K. (2007). Chemopreventive effects

Femia, AP.; Dolara, P.; Luceri, C.; Salvadori, M. & Caderni, G. (2009). Mucin-depleted foci

Femia, AP.; Salvadori, M.; Broekaert, WF.; Francois, IE.; Delcour, JA.; Courtin, CM. &

Fidler, IJ. (1991a). New developments in in vivo models of neoplasia. *Cancer Metastasis Rev*,

Fidler, IJ. (1991b). Orthotopic implantation of human colon carcinomas into nude mice

Fidler, IJ. (1991c). The biology of cancer metastasis or, 'you cannot fix it if you do not know

Fini, L.; Piazzi, G.; Daoud, Y.; Selgrad, M.; Maegawa, S.; Garcia, M.; Fogliano, V.; Romano,

Fishel, R.; Lescoe, MK.; Rao, MR.; Copeland, NG.; Jenkins, NA.; Garber, J.; Kane, M. &

Floyd, RA.; Towner, RA.; Wu, D.; Abbott, A.; Cranford, R.; Branch, D.; Guo, WX.; Foster, SB.;

Fodde, R. & Smits, R. (2001). Disease model: familial adenomatous polyposis. *Trends Mol* 

Furukawa, T.; Kubota, T.; Watanabe, M.; Kuo, TH.; Nishibori, H.; Kase, S.; Saikawa, Y.;

Gerner, EW. (2007). Impact of dietary amino acids and polyamines on intestinal carcinogenesis and chemoprevention in mouse models. *Biochem Soc Trans*, 35, 322-325.

is functionally mutated in colorectal carcinoma. *Cell*, 86, 543-552.

mice. *J Med Food*, 10, 11-17.

*Metastasis Rev*, 10, 229-243.

*Acad Sci U S A*, 91, 8969-8973.

*Med*, 7, 369-373.

*Surg Today*, 23, 420-423.

how it works'. *Bioessays*, 13, 551-554.

Polyphenol Extract. *Cancer Prev Res (Phila)*, 4, 907-915.

with hereditary nonpolyposis colon cancer. *Cell*, 75, 1027-1038.

49, 127-132.

10, 191-192.

sulfate. *Int J Cancer*, 125, 541-547.

Gallinger, S.; Andrulis, IL.; Thomsen, GH.; Wrana, JL. & Attisano, L. (1996). MADR2 maps to 18q21 and encodes a TGFbeta-regulated MAD-related protein that

of orange peel extract (OPE). I: OPE inhibits intestinal tumor growth in ApcMin/+

show strong activation of inflammatory markers in 1,2-dimethylhydrazine-induced carcinogenesis and are promoted by the inflammatory agent sodium dextran

Caderni, G. (2010). Arabinoxylan-oligosaccharides (AXOS) reduce preneoplastic lesions in the colon of rats treated with 1,2-dimethylhydrazine (DMH). *Eur J Nutr*,

provides a valuable model for the biology and therapy of metastasis. *Cancer* 

M.; Graziani, G.; Vitaglione, P.; Carmack, SW.; Gasbarrini, A.; Genta, RM.; Issa, JP.; Boland, CR. & Ricciardiello, L. (2011). Chemoprevention of Intestinal Polyps in ApcMin/+ Mice Fed with Western or Balanced Diets by Drinking Annurca Apple

Kolodner, R. (1993). The human mutator gene homolog MSH2 and its association

Jones, I.; Alam, R.; Moore, D.; Allen, T. & Huycke, M. (2010). Anti-cancer activity of nitrones in the Apc(Min/+) model of colorectal cancer. *Free Radic Res*, 44, 108-117. Fodde, R.; Edelmann, W.; Yang, K.; van Leeuwen, C.; Carlson, C.; Renault, B.; Breukel, C.;

Alt, E.; Lipkin, M.; Khan, PM. & et, a. (1994). A targeted chain-termination mutation in the mouse Apc gene results in multiple intestinal tumors. *Proc Natl* 

Tanino, H.; Teramoto, T.; Ishibiki, K. & et, a. (1993). A metastatic model of human colon cancer constructed using cecal implantation of cancer tissue in nude mice.


Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 295

Lee, CC.; Gillies, ER.; Fox, ME.; Guillaudeu, SJ.; Frechet, JM.; Dy, EE. & Szoka, FC. (2006). A

Li, Q.; Ishikawa, TO.; Oshima, M. & Taketo, MM. (2005). The threshold level of

Lieschke, GJ. & Currie, PD. (2007). Animal models of human disease: zebrafish swim into

Luongo, C.; Moser, AR.; Gledhill, S. & Dove, WF. (1994). Loss of Apc+ in intestinal

Lynch, HT. & de la Chapelle, A. (2003). Hereditary colorectal cancer. *N Engl J Med*, 348, 919-

Lynch, HT. & Smyrk, T. (1996). Hereditary nonpolyposis colorectal cancer (Lynch

Mabley, JG.; Pacher, P.; Bai, P.; Wallace, R.; Goonesekera, S.; Virag, L.; Southan, GJ. & Szabo,

Maggio-Price, L.; Treuting, P.; Zeng, W.; Tsang, M.; Bielefeldt-Ohmann, H. & Iritani, BM.

Martin, JE.; Le Leu, RK.; Hu, Y. & Young, GP. (2010). R-flurbiprofen suppresses distal

McCart, AE.; Vickaryous, NK. & Silver, A. (2008). Apc mice: models, modifiers and mutants.

Min, WK.; Sung, HY. & Choi, YS. (2010). Suppression of colonic aberrant crypt foci by soy

Miyamoto, M. & Takizawa, S. (1975). Colon carcinoma of highly inbred rats. *J Natl Cancer* 

Mizobe, T.; Ogata, Y.; Murakami, H.; Akagi, Y.; Ishibashi, N.; Mori, S.; Sasatomi, T. &

Mizoguchi, E.; Mizoguchi, A.; Preffer, FI. & Bhan, AK. (2000). Regulatory role of mature B cells in a murine model of inflammatory bowel disease. *Int Immunol*, 12, 597-605. Moser, AR.; Dove, WF.; Roth, KA. & Gordon, JI. (1992). The Min (multiple intestinal

Moser, AR.; Pitot, HC. & Dove, WF. (1990). A dominant mutation that predisposes to

Moshal, KS.; Ferri-Lagneau, KF. & Leung, T. (2010). Zebrafish model: worth considering in

multiple intestinal neoplasia in the mouse. *Science*, 247, 322-324.

defining tumor angiogenesis. *Trends Cardiovasc Med*, 20, 114-119.

nitric oxide or poly(ADP-ribose) pathways. *Mutat Res*, 548, 107-116. Mack, DR. & Hollingsworth, MA. (1994). Alteration in expression of MUC2 and MUC3

C. (2004). Suppression of intestinal polyposis in Apcmin/+ mice by targeting the

mRNA levels in HT29 colonic carcinoma cells. *Biochem Biophys Res Commun*, 199,

(2006). Helicobacter infection is required for inflammation and colon cancer in

nonmucin-producing colorectal tumors in azoxymethane-treated rats, without suppressing eicosanoid production. *Am J Physiol Gastrointest Liver Physiol*, 298,

isoflavones is dose-independent in dimethylhydrazine-treated rats. *J Med Food*, 13,

Shirouzu, K. (2008). Efficacy of the combined use of bevacizumab and irinotecan as a postoperative adjuvant chemotherapy in colon carcinoma. *Oncol Rep*, 20, 517-523.

neoplasia) mutation: its effect on gut epithelial cell differentiation and interaction

26 colon carcinomas. *Proc Natl Acad Sci U S A*, 103, 16649-16654.

adenomas from Min mice. *Cancer Res*, 54, 5947-5952.

syndrome). An updated review. *Cancer*, 78, 1149-1167.

SMAD3-deficient mice. *Cancer Res*, 66, 828-838.

with a modifier system. *J Cell Biol*, 116, 1517-1526.

65, 8622-8627.

932.

1012-1018.

G860-4.

495-502.

*Inst*, 55, 1471-1472.

*Pathol Res Pract*, 204, 479-490.

view. *Nat Rev Genet*, 8, 353-367.

single dose of doxorubicin-functionalized bow-tie dendrimer cures mice bearing C-

adenomatous polyposis coli protein for mouse intestinal tumorigenesis. *Cancer Res*,


Itzkowitz, SH. & Harpaz, N. (2004). Diagnosis and management of dysplasia in patients with inflammatory bowel diseases. *Gastroenterology*, 126, 1634-1648. Itzkowitz, SH. & Yio, X. (2004). Inflammation and cancer IV. Colorectal cancer in

Kaiser, S.; Park, YK.; Franklin, JL.; Halberg, RB.; Yu, M.; Jessen, WJ.; Freudenberg, J.; Chen,

Kanauchi, O.; Mitsuyama, K.; Andoh, A. & Iwanaga, T. (2008). Modulation of intestinal

Kaur Saini, M. & Nath Sanyal, S. (2010). Evaluation of chemopreventive response of two

Khor, TO.; Hu, R.; Shen, G.; Jeong, WS.; Hebbar, V.; Chen, C.; Xu, C.; Nair, S.; Reddy, B.;

Kim, KP.; Whitehead, C.; Piazza, G. & Wargovich, MJ. (2004). Combinatorial

Kulkarni, AB. & Karlsson, S. (1993). Transforming growth factor-beta 1 knockout mice. A

Kullberg, MC.; Ward, JM.; Gorelick, PL.; Caspar, P.; Hieny, S.; Cheever, A.; Jankovic, D. &

Kuraguchi, M.; Yang, K.; Wong, E.; Avdievich, E.; Fan, K.; Kolodner, RD.; Lipkin, M.;

Kwong, LN. & Dove, WF. (2009). APC and its modifiers in colon cancer. *Adv Exp Med Biol*,

Kwong, LN.; Shedlovsky, A.; Biehl, BS.; Clipson, L.; Pasch, CA. & Dove, WF. (2007).

Le Leu, RK.; Hu, Y.; Brown, IL.; Woodman, RJ. & Young, GP. (2010). Synbiotic intervention

*Liver Physiol*, 287, G7-17.

and human colon cancer. *Genome Biol*, 8, R131.

*Biopharm Drug Dispos*, 27, 407-420.

*Cancer Res*, 61, 7934-7942.

*Genetics*, 176, 1237-1244.

656, 85-106.

143, 3-9.

colonic carcinogenesis in rats. *Oncol Rep*, 20, 793-801.

and NMR spectroscopic techniques. *Nutr Hosp*, 25, 577-585.

progression of aberrant crypt foci. *Anticancer Res*, 24, 1805-1811.

dependent mechanism. *Infect Immun*, 66, 5157-5166.

development in rats. *Carcinogenesis*, 31, 246-251.

inflammatory bowel disease: the role of inflammation. *Am J Physiol Gastrointest* 

X.; Haigis, K.; Jegga, AG.; Kong, S.; Sakthivel, B.; Xu, H.; Reichling, T.; Azhar, M.; Boivin, GP.; Roberts, RB.; Bissahoyo, AC.; Gonzales, F.; Bloom, GC.; Eschrich, S.; Carter, SL.; Aronow, JE.; Kleimeyer, J.; Kleimeyer, M.; Ramaswamy, V.; Settle, SH.; Boone, B.; Levy, S.; Graff, JM.; Doetschman, T.; Groden, J.; Dove, WF.; Threadgill, DW.; Yeatman, TJ.; Coffey, RJJ. & Aronow, BJ. (2007). Transcriptional recapitulation and subversion of embryonic colon development by mouse colon tumor models

environment by prebiotic germinated barley foodstuff prevents chemo-induced

cycloxygenase-2 inhibitors, etoricoxib and diclofenac in rat colon cancer using FTIR

Chada, K. & Kong, AN. (2006). Pharmacogenomics of cancer chemopreventive isothiocyanate compound sulforaphane in the intestinal polyps of ApcMin/+ mice.

chemoprevention: efficacy of lovostatin and exisulind on the formation and

mutation in one cytokine gene causes a dramatic inflammatory disease. *Am J Pathol*,

Sher, A. (1998). Helicobacter hepaticus triggers colitis in specific-pathogen-free interleukin-10 (IL-10)-deficient mice through an IL-12- and gamma interferon-

Brown, AM.; Kucherlapati, R. & Edelmann, W. (2001). The distinct spectra of tumor-associated Apc mutations in mismatch repair-deficient Apc1638N mice define the roles of MSH3 and MSH6 in DNA repair and intestinal tumorigenesis.

Identification of Mom7, a novel modifier of Apc(Min/+) on mouse chromosome 18.

of Bifidobacterium lactis and resistant starch protects against colorectal cancer


Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 297

Rajamanickam, S.; Velmurugan, B.; Kaur, M.; Singh, RP. & Agarwal, R. (2010).

Raju, J.; Bielecki, A.; Caldwell, D.; Lok, E.; Taylor, M.; Kapal, K.; Curran, I.; Cooke, GM.;

Rao, CV.; Steele, VE.; Swamy, MV.; Patlolla, JM.; Guruswamy, S. & Kopelovich, L. (2009).

Raufman, JP.; Shant, J.; Xie, G.; Cheng, K.; Gao, XM.; Shiu, B.; Shah, N.; Drachenberg, C.;

Reitmair, AH.; Cai, JC.; Bjerknes, M.; Redston, M.; Cheng, H.; Pind, MT.; Hay, K.; Mitri, A.;

Roy, HK.; Gulizia, JM.; Karolski, WJ.; Ratashak, A.; Sorrell, MF. & Tuma, D. (2002). Ethanol

Rudolph, U.; Finegold, MJ.; Rich, SS.; Harriman, GR.; Srinivasan, Y.; Brabet, P.; Bradley, A.

Ruivenkamp, CA.; van Wezel, T.; Zanon, C.; Stassen, AP.; Vlcek, C.; Csikos, T.; Klous, AM.;

Rygaard, J. & Povlsen, CO. (1969). Heterotransplantation of a human malignant tumour to

Sale, S.; Tunstall, RG.; Ruparelia, KC.; Butler, PC.; Potter, GA.; Steward, WP. & Gescher, AJ.

development in the ApcMin+ mouse. *Invest New Drugs*, 24, 459-464.

*Cancer Res*, 70, 2368-2378.

*Cancer Res*, 69, 8175-8182.

*Carcinogenesis*, 20, 51-58.

295-300.

receptor-beta. *J Nutr*, 139, 474-481.

Neoplasia in Apcmin/+ Mice. *Carcinogenesis*,

*Cancer Epidemiol Biomarkers Prev*, 11, 1499-1502.

bowel disease. *J Clin Immunol*, 15, 101S-105S.

"Nude" mice. *Acta Pathol Microbiol Scand*, 77, 758-760.

Burt, R.; Cairns, BR. & Jones, DA. (2010). DNA demethylase activity maintains intestinal cells in an undifferentiated state following loss of APC. *Cell*, 142, 930-942.

Chemoprevention of intestinal tumorigenesis in APCmin/+ mice by silibinin.

Bird, RP. & Mehta, R. (2009). Soy isoflavones modulate azoxymethane-induced rat colon carcinogenesis exposed pre- and postnatally and inhibit growth of DLD-1 human colon adenocarcinoma cells by increasing the expression of estrogen

Inhibition of azoxymethane-induced colorectal cancer by CP-31398, a TP53 modulator, alone or in combination with low doses of celecoxib in male F344 rats.

Heath, J.; Wess, J. & Khurana, S. (2011). Muscarinic Receptor Subtype-3 Gene Ablation and Scopolamine ButylbromideTreatment Attenuate Small Intestinal

Bapat, BV.; Mak, TW. & Gallinger, S. (1996). MSH2 deficiency contributes to accelerated APC-mediated intestinal tumorigenesis. *Cancer Res*, 56, 2922-2926. Rigby, RJ.; Simmons, JG.; Greenhalgh, CJ.; Alexander, WS. & Lund, PK. (2007). Suppressor

of cytokine signaling 3 (SOCS3) limits damage-induced crypt hyper-proliferation and inflammation-associated tumorigenesis in the colon. *Oncogene*, 26, 4833-4841. Ritland, SR. & Gendler, SJ. (1999). Chemoprevention of intestinal adenomas in the ApcMin

mouse by piroxicam: kinetics, strain effects and resistance to chemosuppression.

promotes intestinal tumorigenesis in the MIN mouse. Multiple intestinal neoplasia.

& Birnbaumer, L. (1995). Gi2 alpha protein deficiency: a model of inflammatory

Tripodis, N.; Perrakis, A.; Boerrigter, L.; Groot, PC.; Lindeman, J.; Mooi, WJ.; Meijjer, GA.; Scholten, G.; Dauwerse, H.; Paces, V.; van Zandwijk, N.; van Ommen, GJ. & Demant, P. (2002). Ptprj is a candidate for the mouse colon-cancer susceptibility locus Scc1 and is frequently deleted in human cancers. *Nat Genet*, 31,

(2006). Effects of the potential chemopreventive agent DMU-135 on adenoma


Murphy, EA.; Davis, JM.; McClellan, JL.; Gordon, BT. & Carmichael, MD. (2011). Curcumin's

Mutanen, M.; Pajari, AM.; Paivarinta, E.; Misikangas, M.; Rajakangas, J.; Marttinen, M. &

Nakanishi, Y.; Nakatsuji, M.; Seno, H.; Ishizu, S.; Akitake-Kawano, R.; Kanda, K.; Ueo, T.;

Nakayama, Y.; Inoue, Y.; Minagawa, N.; Onitsuka, K.; Nagata, J.; Shibao, K.; Hirata, K.;

Niho, N.; Takahashi, M.; Shoji, Y.; Takeuchi, Y.; Matsubara, S.; Sugimura, T. & Wakabayashi,

Okayasu, I.; Hatakeyama, S.; Yamada, M.; Ohkusa, T.; Inagaki, Y. & Nakaya, R. (1990). A

Oshima, M.; Murai, N.; Kargman, S.; Arguello, M.; Luk, P.; Kwong, E.; Taketo, MM. &

Oshima, M.; Takahashi, M.; Oshima, H.; Tsutsumi, M.; Yazawa, K.; Sugimura, T.;

Papageorgis, P.; Cheng, K.; Ozturk, S.; Gong, Y.; Lambert, AW.; Abdolmaleky, HM.; Zhou,

Phutthaphadoong, S.; Yamada, Y.; Hirata, A.; Tomita, H.; Hara, A.; Limtrakul, P.; Iwasaki,

Prabhu, PN.; Ashokkumar, P. & Sudhandiran, G. (2009). Antioxidative and antiproliferative

Prolla, TA.; Baker, SM.; Harris, AC.; Tsao, JL.; Yao, X.; Bronner, CE.; Zheng, B.; Gordon, M.;

Rai, K.; Sarkar, S.; Broadbent, TJ.; Voas, M.; Grossmann, KF.; Nadauld, LD.; Dehghanizadeh,

Prolla, TA. (1998). DNA mismatch repair and cancer. *Curr Opin Cell Biol*, 10, 311-316.

carcinogenesis in a rat model. *Anticancer Res*, 29, 2059-2065.

colitis in mice. *Gastroenterology*, 98, 694-702.

resistance of colon cancer. *Cancer Res*, 71, 998-1008.

*Interferon Cytokine Res*, 31, 219-226.

mouse polyps. *Carcinogenesis*,

*Carcinogenesis*, 16, 2605-2607.

ApcMin/+ mice. *Oncol Rep*, 23, 53-59.

DNA mismatch repair. *Nat Genet*, 18, 276-279.

123-125.

964.

effect on intestinal inflammation and tumorigenesis in the ApcMin/+ mouse. *J* 

Oikarinen, S. (2008). Berries as chemopreventive dietary constituents--a mechanistic approach with the ApcMin/+ mouse. *Asia Pac J Clin Nutr*, 17 Suppl 1,

Komekado, H.; Kawada, M.; Minami, M. & Chiba, T. (2011). COX-2 inhibition alters the phenotype of tumor-associated macrophages from M2 to M1 in ApcMin/+

Sako, T.; Nagata, N. & Yamaguchi, K. (2009). Chemopreventive effect of 4-[3,5- Bis(trimethylsilyl) benzamido] benzoic acid (TAC-101) on MNU-induced colon

K. (2003). Dose-dependent suppression of hyperlipidemia and intestinal polyp formation in Min mice by pioglitazone, a PPAR gamma ligand. *Cancer Sci*, 94, 960-

novel method in the induction of reliable experimental acute and chronic ulcerative

Evans, JF. (2001). Chemoprevention of intestinal polyposis in the Apcdelta716 mouse by rofecoxib, a specific cyclooxygenase-2 inhibitor. *Cancer Res*, 61, 1733-1740.

Nishimura, S.; Wakabayashi, K. & Taketo, MM. (1995). Effects of docosahexaenoic acid (DHA) on intestinal polyp development in Apc delta 716 knockout mice.

JR. & Thiagalingam, S. (2011). Smad4 inactivation promotes malignancy and drug

T.; Kobayashi, H. & Mori, H. (2010). Chemopreventive effect of fermented brown rice and rice bran (FBRA) on the inflammation-related colorectal carcinogenesis in

effects of astaxanthin during the initiation stages of 1,2-dimethyl hydrazineinduced experimental colon carcinogenesis. *Fundam Clin Pharmacol*, 23, 225-234.

Reneker, J.; Arnheim, N.; Shibata, D.; Bradley, A. & Liskay, RM. (1998). Tumour susceptibility and spontaneous mutation in mice deficient in Mlh1, Pms1 and Pms2

S.; Hagos, FT.; Li, Y.; Toth, RK.; Chidester, S.; Bahr, TM.; Johnson, WE.; Sklow, B.;

Burt, R.; Cairns, BR. & Jones, DA. (2010). DNA demethylase activity maintains intestinal cells in an undifferentiated state following loss of APC. *Cell*, 142, 930-942.


Animal Models of Colorectal Cancer in Chemoprevention and Therapeutics Development 299

Swamy, MV.; Patlolla, JM.; Steele, VE.; Kopelovich, L.; Reddy, BS. & Rao, CV. (2006).

Symolon, H.; Schmelz, EM.; Dillehay, DL. & Merrill, AHJ. (2004). Dietary soy sphingolipids

Takahashi, M.; Nakatsugi, S.; Sugimura, T. & Wakabayashi, K. (2000). Frequent mutations of

Takahashi, S.; Hasegawa, R.; Masui, T.; Mizoguchi, M.; Fukushima, S. & Ito, N. (1992).

Takaku, K.; Oshima, M.; Miyoshi, H.; Matsui, M.; Seldin, MF. & Taketo, MM. (1998).

Tammariello, AE. & Milner, JA. (2010). Mouse models for unraveling the importance of diet

Tanaka, T.; Kohno, H.; Suzuki, R.; Yamada, Y.; Sugie, S. & Mori, H. (2003). A novel

Tang, B.; Bottinger, EP.; Jakowlew, SB.; Bagnall, KM.; Mariano, J.; Anver, MR.; Letterio, JJ. &

Telang, N. & Katdare, M. (2007). Combinatorial prevention of carcinogenic risk in a model

Tomimoto, A.; Endo, H.; Sugiyama, M.; Fujisawa, T.; Hosono, K.; Takahashi, H.; Nakajima,

Tseng, W.; Leong, X. & Engleman, E. (2007). Orthotopic mouse model of colorectal cancer. *J* 

Uronis, JM.; Herfarth, HH.; Rubinas, TC.; Bissahoyo, AC.; Hanlon, K. & Threadgill, DW.

Velmurugan, B.; Singh, RP.; Agarwal, R. & Agarwal, C. (2010). Dietary-feeding of grape

without going through a polypoid stage. *Cancer Res*, 67, 11594-11600. Uronis, JM. & Threadgill, DW. (2009). Murine models of colorectal cancer. *Mamm Genome*,

suppressor with true haploid insufficiency. *Nat Med*, 4, 802-807.

radiation risk estimates. *Radiat Environ Biophys*, 49, 389-396.

combination of five carcinogens. *J. Toxicol. Pathol*, 5, 151-156.

in colon cancer prevention. *J Nutr Biochem*, 21, 77-88.

and dextran sodium sulfate. *Cancer Sci*, 94, 965-973.

for familial colon cancer. *Oncol Rep*, 17, 909-914.

in fischer 344 rats. *Mol Carcinog*, 49, 641-652.

mammary glands lacking Socs3. *EMBO J*, 25, 5805-5815.

mice and ApcMin/+ mice. *J Nutr*, 134, 1157-1161.

66, 7370-7377.

*Carcinogenesis*, 21, 1117-1120.

genes. *Cell*, 92, 645-656.

*Vis Exp*, 484.

20, 261-268.

Visvader, JE. (2006). c-myc as a mediator of accelerated apoptosis and involution in

Chemoprevention of familial adenomatous polyposis by low doses of atorvastatin and celecoxib given individually and in combination to APCMin mice. *Cancer Res*,

suppress tumorigenesis and gene expression in 1,2-dimethylhydrazine-treated CF1

the beta-catenin gene in mouse colon tumors induced by azoxymethane.

Establishment of multiorgan carcinogenesis bioassay using rats treated with a

Intestinal tumorigenesis in compound mutant mice of both Dpc4 (Smad4) and Apc

inflammation-related mouse colon carcinogenesis model induced by azoxymethane

Wakefield, LM. (1998). Transforming growth factor-beta1 is a new form of tumor

N.; Nagashima, Y.; Wada, K.; Nakagama, H. & Nakajima, A. (2008). Metformin suppresses intestinal polyp growth in ApcMin/+ mice. *Cancer Sci*, 99, 2136-2141. Trani, D.; Datta, K.; Doiron, K.; Kallakury, B. & Fornace, AJJ. (2010). Enhanced intestinal

tumor multiplicity and grade in vivo after HZE exposure: mouse models for space

(2007). Flat colorectal cancers are genetically determined and progress to invasion

seed extract prevents azoxymethane-induced colonic aberrant crypt foci formation


Sasai, H.; Masaki, M. & Wakitani, K. (2000). Suppression of polypogenesis in a new mouse

Seril, DN.; Liao, J.; Yang, CS. & Yang, GY. (2005). Systemic iron supplementation replenishes

Shibata, H.; Toyama, K.; Shioya, H.; Ito, M.; Hirota, M.; Hasegawa, S.; Matsumoto, H.;

Silva, MF.; Sivieri, K. & Rossi, EA. (2009). Effects of a probiotic soy product and physical

Smith, DL.; Keshavan, P.; Avissar, U.; Ahmed, K. & Zucker, SD. (2010). Sodium taurocholate

Smits, BM.; Mudde, JB.; van de Belt, J.; Verheul, M.; Olivier, J.; Homberg, J.; Guryev, V.;

Smits, R.; Kielman, MF.; Breukel, C.; Zurcher, C.; Neufeld, K.; Jagmohan-Changur, S.;

Sodir, NM.; Chen, X.; Park, R.; Nickel, AE.; Conti, PS.; Moats, R.; Bading, JR.; Shibata, D. &

Song, J.; Medline, A.; Mason, JB.; Gallinger, S. & Kim, YI. (2000). Effects of dietary folate on intestinal tumorigenesis in the apcMin mouse. *Cancer Res*, 60, 5434-5440. Sorensen, IK.; Kristiansen, E.; Mortensen, A.; Nicolaisen, GM.; Wijnands, JA.; van Kranen,

Steele, VE.; Rao, CV.; Zhang, Y.; Patlolla, J.; Boring, D.; Kopelovich, L.; Juliana, MM.;

Su, LK.; Kinzler, KW.; Vogelstein, B.; Preisinger, AC.; Moser, AR.; Luongo, C.; Gould, KA. &

Sutherland, KD.; Vaillant, F.; Alexander, WS.; Wintermantel, TM.; Forrest, NC.; Holroyd,

intestinal neoplasia in ApcMin mouse. *Cancer Lett*, 130, 217-225.

activation of the farnesoid X receptor. *Carcinogenesis*, 31, 1100-1109.

initiated by conditional targeting of the Apc gene. *Science*, 278, 120-123. Shull, MM.; Ormsby, I.; Kier, AB.; Pawlowski, S.; Diebold, RJ.; Yin, M.; Allen, R.; Sidman, C.;

colitis: comparison with iron-enriched diet. *Dig Dis Sci*, 50, 696-707.

*Carcinogenesis*, 21, 953-958.

*Nature*, 359, 693-699.

*Genes Dev*, 13, 1309-1321.

of carcinogenic. *J Int Soc Sports Nutr*, 6, 17.

mutagenesis. *Pharmacogenet Genomics*, 16, 159-169.

ApcMin/+ mice. *Cancer Res*, 66, 8430-8438.

*Cancer Prev Res (Phila)*, 2, 951-956.

homolog of the APC gene. *Science*, 256, 668-670.

strain with a truncated Apc(Delta474) by a novel COX-2 inhibitor, JTE-522.

iron stores without enhancing colon carcinogenesis in murine models of ulcerative

Takano, H.; Akiyama, T.; Toyoshima, K.; Kanamaru, R.; Kanegae, Y.; Saito, I.; Nakamura, Y.; Shiba, K. & Noda, T. (1997). Rapid colorectal adenoma formation

Proetzel, G.; Calvin, D. & et, a. (1992). Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease.

exercise on formation of pre-neoplastic lesions in rat colons in a short-term model

inhibits intestinal adenoma formation in APCMin/+ mice, potentially through

Cools, AR.; Ellenbroek, BA.; Plasterk, RH. & Cuppen, E. (2006). Generation of gene knockouts and mutant models in the laboratory rat by ENU-driven target-selected

Hofland, N.; van Dijk, J.; White, R.; Edelmann, W.; Kucherlapati, R.; Khan, PM. & Fodde, R. (1999). Apc1638T: a mouse model delineating critical domains of the adenomatous polyposis coli protein involved in tumorigenesis and development.

Laird, PW. (2006). Smad3 deficiency promotes tumorigenesis in the distal colon of

HJ. & van Kreijl, CF. (1998). The effect of soy isoflavones on the development of

Grubbs, CJ. & Lubet, RA. (2009). Chemopreventive efficacy of naproxen and nitric oxide-naproxen in rodent models of colon, urinary bladder, and mammary cancers.

Dove, WF. (1992). Multiple intestinal neoplasia caused by a mutation in the murine

SL.; McManus, EJ.; Schutz, G.; Watson, CJ.; Chodosh, LA.; Lindeman, GJ. &

Visvader, JE. (2006). c-myc as a mediator of accelerated apoptosis and involution in mammary glands lacking Socs3. *EMBO J*, 25, 5805-5815.


**16** 

*USA* 

**The Stem Cell Environment:** 

George D. Wilson1,2, John M. Robertson1, Harry Wasvary3 and Bryan J. Thibodeau2

*1Department of Radiation Oncology,* 

*3Department of Colon and Rectal Surgery,* 

*2Beaumont BioBank,* 

*William Beaumont Hospital, Royal Oak, Michigan,* 

**Kinetics, Signaling and Markers** 

Colorectal cancer (CRC) is one of the commonest cancers and the third leading cause of cancer death. In the developed world more than 1 million individuals will develop colorectal cancer every year (Parkin et al. 2005). According to the Surveillance, Epidemiology and End Results (SEER) Program database, the prognosis of CRC has an improving trend. 5-year survival rates have risen from 56.5% for patients diagnosed in the early 1980s to as much as 63.2% for those diagnosed in the early 1990s and most recently to 64.9%. Currently, rectal cancer is usually discussed together with colon cancer and historically accounted for more than 50% of CRCs. However, this has now decreased to less than colon cancers; in a recent review the incidence in the European Union was approximately 35% of the total CRC incidence (Glimelius and Oliveira 2009). The prognosis of rectal cancers is worse than that of colon cancers (Enblad et al. 1988), and the clinical treatment of rectal cancers is different from that of colon cancers (Vo et al. 2010). Rectal cancer is a very different tumor from colon cancer because of the anatomical narrow confines of the pelvis, the proximity of the genitourinary organs and nerves, and the anal sphincter mechanism. Oncological cure remains the primary aim of treatment for rectal cancer, but sparing of the anal sphincters with adequate bowel, genitourinary, and sexual

Pre-operative staging is crucial to stratify patients into one of three treatment strategies: patients whose tumors are superficial require surgery alone, patients with operable tumors but at an increased risk of local recurrence require short course radiotherapy and then optimal surgery, and thirdly those with more locally advanced rectal cancers with close or involved circumferential resection margins require neoadjuvant chemoradiation (CRT) followed by surgery. Early detection and treatment is vital to better survival. The five year survival rate of patients diagnosed with early stage CRC is approximately 90% as opposed to close to 10% for those diagnosed with locally advanced or metastatic disease. Indeed, the median survival of patients with metastatic CRC is only two years despite multiple available treatment modalities, including surgical resection, chemoradiation, monoclonal

**1. Introduction** 

function is also taken into consideration.


### **The Stem Cell Environment: Kinetics, Signaling and Markers**

George D. Wilson1,2, John M. Robertson1,

Harry Wasvary3 and Bryan J. Thibodeau2 *1Department of Radiation Oncology, 2Beaumont BioBank, 3Department of Colon and Rectal Surgery, William Beaumont Hospital, Royal Oak, Michigan, USA* 

#### **1. Introduction**

300 Colorectal Cancer – From Prevention to Patient Care

Wald, D.; Qin, J.; Zhao, Z.; Qian, Y.; Naramura, M.; Tian, L.; Towne, J.; Sims, JE.; Stark, GR.

Wang, QS.; Papanikolaou, A.; Sabourin, CL. & Rosenberg, DW. (1998). Altered expression of

Wirtz, S. & Neurath, MF. (2007). Mouse models of inflammatory bowel disease. *Adv Drug* 

Xiao, H.; Gulen, MF.; Qin, J.; Yao, J.; Bulek, K.; Kish, D.; Altuntas, CZ.; Wald, D.; Ma, C.;

Yang, W.; Velcich, A.; Lozonschi, I.; Liang, J.; Nicholas, C.; Zhuang, M.; Bancroft, L. &

Yang, YA.; Dukhanina, O.; Tang, B.; Mamura, M.; Letterio, JJ.; MacGregor, J.; Patel, SC.;

Yang, YA.; Tang, B.; Robinson, G.; Hennighausen, L.; Brodie, SG.; Deng, CX. & Wakefield,

Zan, Y.; Haag, JD.; Chen, KS.; Shepel, LA.; Wigington, D.; Wang, YR.; Hu, R.; Lopez-

mutagenesis and a yeast-based screening assay. *Nat Biotechnol*, 21, 645-651. Zhu, Y.; Richardson, JA.; Parada, LF. & Graff, JM. (1998). Smad3 mutant mice develop

by transforming growth factor-beta. *Cell Growth Differ*, 13, 123-130.

homeostasis, inflammation, and tumorigenesis. *Immunity*, 26, 461-475. Yamaguchi, K.; Cekanova, M.; McEntee, MF.; Yoon, JH.; Fischer, SM.; Renes, IB.; Van

receptor signaling. *Nat Immunol*, 4, 920-927.

tumorigenesis. *Carcinogenesis*, 19, 2001-2006.

pathways. *Mol Cancer Ther*, 7, 2779-2787.

formation in Muc2-/- mice. *Am J Pathol*, 166, 1239-1246.

without adverse side effects. *J Clin Invest*, 109, 1607-1615.

metastatic colorectal cancer. *Cell*, 94, 703-714.

*Deliv Rev*, 59, 1073-1083.

& Li, X. (2003). SIGIRR, a negative regulator of Toll-like receptor-interleukin 1

cyclin D1 and cyclin-dependent kinase 4 in azoxymethane-induced mouse colon

Zhou, H.; Tuohy, VK.; Fairchild, RL.; de la Motte, C.; Cua, D.; Vallance, BA. & Li, X. (2007). The Toll-interleukin-1 receptor member SIGIRR regulates colonic epithelial

Seuningen, I. & Baek, SJ. (2008). Peroxisome proliferator-activated receptor ligand MCC-555 suppresses intestinal polyps in ApcMin/+ mice via extracellular signalregulated kinase and peroxisome proliferator-activated receptor-dependent

Augenlicht, LH. (2005). Inactivation of p21WAF1/cip1 enhances intestinal tumor

Khozin, S.; Liu, ZY.; Green, J.; Anver, MR.; Merlino, G. & Wakefield, LM. (2002a). Lifetime exposure to a soluble TGF-beta antagonist protects mice against metastasis

LM. (2002b). Smad3 in the mammary epithelium has a nonredundant role in the induction of apoptosis, but not in the regulation of proliferation or differentiation

Guajardo, CC.; Brose, HL.; Porter, KI.; Leonard, RA.; Hitt, AA.; Schommer, SL.; Elegbede, AF. & Gould, MN. (2003). Production of knockout rats using ENU Colorectal cancer (CRC) is one of the commonest cancers and the third leading cause of cancer death. In the developed world more than 1 million individuals will develop colorectal cancer every year (Parkin et al. 2005). According to the Surveillance, Epidemiology and End Results (SEER) Program database, the prognosis of CRC has an improving trend. 5-year survival rates have risen from 56.5% for patients diagnosed in the early 1980s to as much as 63.2% for those diagnosed in the early 1990s and most recently to 64.9%. Currently, rectal cancer is usually discussed together with colon cancer and historically accounted for more than 50% of CRCs. However, this has now decreased to less than colon cancers; in a recent review the incidence in the European Union was approximately 35% of the total CRC incidence (Glimelius and Oliveira 2009). The prognosis of rectal cancers is worse than that of colon cancers (Enblad et al. 1988), and the clinical treatment of rectal cancers is different from that of colon cancers (Vo et al. 2010). Rectal cancer is a very different tumor from colon cancer because of the anatomical narrow confines of the pelvis, the proximity of the genitourinary organs and nerves, and the anal sphincter mechanism. Oncological cure remains the primary aim of treatment for rectal cancer, but sparing of the anal sphincters with adequate bowel, genitourinary, and sexual function is also taken into consideration.

Pre-operative staging is crucial to stratify patients into one of three treatment strategies: patients whose tumors are superficial require surgery alone, patients with operable tumors but at an increased risk of local recurrence require short course radiotherapy and then optimal surgery, and thirdly those with more locally advanced rectal cancers with close or involved circumferential resection margins require neoadjuvant chemoradiation (CRT) followed by surgery. Early detection and treatment is vital to better survival. The five year survival rate of patients diagnosed with early stage CRC is approximately 90% as opposed to close to 10% for those diagnosed with locally advanced or metastatic disease. Indeed, the median survival of patients with metastatic CRC is only two years despite multiple available treatment modalities, including surgical resection, chemoradiation, monoclonal

The Stem Cell Environment: Kinetics, Signaling and Markers 303

Fig. 1. The cancer stem cell hypothesis. In normal tissues, cell numbers are regulated by asymmetric division of stem cells to replace the loss of functional differentiated cells. In cancer, the mutated stem cell drives tumor heterogeneity through asymmetric division and aberrant proliferation/differentiation. Only the CSC has the capacity to form new tumors

population in normal tissues. These progenitors give rise to more or less partially differentiated bulk tumor cells through a combination of proliferation and abortive

The existence of CSCs has been supported by seminal research performed on acute myeloid leukemia (AML), where it has been demonstrated that only a specific cellular subset that express antigenic markers similar to hematopoietic stem cells has clonogenic activity in immunocompromised mice (Lapidot et al. 1994; Bonnet and Dick 1997). The cell responsible for tumor initiation was identified as having a CD34+ CD38− phenotype; interestingly the majority of AML cells tend to be CD34−. It was observed that as few as 5 ×103 CD34+ CD38<sup>−</sup> cells could successfully engraft an immunocompromised mouse, while 100 times more CD34− or CD34+ CD38+ cells were not tumorigenic (Bonnet and Dick 1997). Significantly, the resulting tumors were heterogeneous and composed of a mixture of tumorigenic and nontumorigenic cells similar to the donor leukemia. Subsequently, studies on tumors of epithelial origin, such as breast cancer, also provided evidence for the presence of stem-like cells within the cancer (Dontu et al. 2005). Ponti and colleagues demonstrated that only CD44+/CD24- cells, isolated from breast cancer, were able to produce tumors in immunocompromised mice (Ponti et al. 2005). The initial publications in leukemia and breast cancer were followed by reports showing the prospective isolation of CSCs in numerous malignancies including: brain, colon, head and neck, pancreatic, melanoma, mesenchymal, hepatic, lung, prostate, and ovarian tumors (O'Brien et al. 2010). The

and accumulate further mutations leading to tumor progression.

differentiation.

antibodies to tumor growth factors, and liver-directed therapies for metastatic disease. Few patients are sensitive to these therapies and even fewer are cured.

The early pathway to CRC tumorigenesis has been well elucidated by the seminal work of Vogelstein and colleagues in which a single colorectal epithelial cell acquires a mutation in the tumor suppressor APC gene (Jones et al. 2008). The cells subsequently acquire a complex array of molecular mutations and quickly acquire the potential to metastasize (Jones et al. 2008). The concept of clonal evolution which postulated that tumor progression results from acquired genetic variability within the original mutated clone allowing sequential selection of more aggressive sub-lines provided a ready explanation for the relentless advance toward ever more malignant behavior within established tumors (Nowell 1976) including colorectal cancer (Fearon and Vogelstein 1990). However, prior to the theory of clonal evolution, the cancer stem cell concept had been formulated to account for the heterogeneity, resistance to treatment, and dormancy exhibited by many solid tumors (Pierce and Speers 1988). The CSC concept postulated that, similar to the growth of normal proliferative tissues such as bone marrow, skin or intestinal epithelium, the growth of tumors is driven by limited numbers of dedicated stem cells that are capable of self-renewal. More recently, the CSC concept has gained more momentum due to studies in leukemia. These studies showed that engraftment of tumors in an immunodeficient mouse could only be initiated from a specific subpopulation of CD34+CD38− cells and led to the identification of a CSC in acute myeloid leukemia (AML) (Bonnet and Dick 1997). In 2003, Clarke and colleagues (Al-Hajj et al. 2003) applied the same concepts and experimental approaches to a solid breast cancer tumor and showed that as few as 100 CD44+CD24−/low cells were tumorigenic, whereas tens of thousands of cells with alternate phenotypes were not. CSC theory and clonal evolution are not mutually exclusive, and it is likely that a single tumor may contain multiple cancer stem cell clones that are genetically distinct. However, they will always have a common ancestor in the stem cell that sustained the first oncogenic mutation and became the origin of the tumor.

#### **2. Cancer stem cell theory**

Stem cell concepts and their application to cancer is not a new subject (Clevers 2011; Wicha et al. 2006). As far back as the nineteenth century, it was recognized that tumors exhibit profound histological heterogeneity. In the 1930's, it was discovered that a single cell from a mouse tumor could initiate a new tumor in a recipient mouse. Subsequently, several studies showed that the number of cells with tumor-initiating properties in solid tumors and leukemias was found to be variable but low (103 to 107 cells). The resulting tumors typically showed the morphologic heterogeneity of the original tumor.

CSCs possess several key properties of normal tissue stem cells including self-renewal, unlimited proliferative potential, infrequent or slow replication, resistance to toxic xenobiotics, enhanced DNA repair capacity, and the ability to give rise to daughter cells that differentiate. However, unlike highly regulated tissue stem cells, CSCs demonstrate deregulated self-renewal/differentiation processes and generate daughter cells that arrest at various stages of differentiation. The progeny of the stem cells make up the bulk of the tumor and are characterized by rapid replication, limited proliferative potential, and the inability to form a new tumor. Only the CSC is able to initiate tumor formation as it is solely capable of self-renewal (Figure 1). CSCs are thought to maintain their numbers by slow selfreplication and produce other tumor cells by asymmetric cell division. In this process, cell division of a CSC generates a CSC and a transformed "progenitor-like" cell, which has limited self-renewal ability but is highly proliferative, similar to a transit-amplifying

antibodies to tumor growth factors, and liver-directed therapies for metastatic disease. Few

The early pathway to CRC tumorigenesis has been well elucidated by the seminal work of Vogelstein and colleagues in which a single colorectal epithelial cell acquires a mutation in the tumor suppressor APC gene (Jones et al. 2008). The cells subsequently acquire a complex array of molecular mutations and quickly acquire the potential to metastasize (Jones et al. 2008). The concept of clonal evolution which postulated that tumor progression results from acquired genetic variability within the original mutated clone allowing sequential selection of more aggressive sub-lines provided a ready explanation for the relentless advance toward ever more malignant behavior within established tumors (Nowell 1976) including colorectal cancer (Fearon and Vogelstein 1990). However, prior to the theory of clonal evolution, the cancer stem cell concept had been formulated to account for the heterogeneity, resistance to treatment, and dormancy exhibited by many solid tumors (Pierce and Speers 1988). The CSC concept postulated that, similar to the growth of normal proliferative tissues such as bone marrow, skin or intestinal epithelium, the growth of tumors is driven by limited numbers of dedicated stem cells that are capable of self-renewal. More recently, the CSC concept has gained more momentum due to studies in leukemia. These studies showed that engraftment of tumors in an immunodeficient mouse could only be initiated from a specific subpopulation of CD34+CD38− cells and led to the identification of a CSC in acute myeloid leukemia (AML) (Bonnet and Dick 1997). In 2003, Clarke and colleagues (Al-Hajj et al. 2003) applied the same concepts and experimental approaches to a solid breast cancer tumor and showed that as few as 100 CD44+CD24−/low cells were tumorigenic, whereas tens of thousands of cells with alternate phenotypes were not. CSC theory and clonal evolution are not mutually exclusive, and it is likely that a single tumor may contain multiple cancer stem cell clones that are genetically distinct. However, they will always have a common ancestor in the stem cell that

patients are sensitive to these therapies and even fewer are cured.

sustained the first oncogenic mutation and became the origin of the tumor.

showed the morphologic heterogeneity of the original tumor.

Stem cell concepts and their application to cancer is not a new subject (Clevers 2011; Wicha et al. 2006). As far back as the nineteenth century, it was recognized that tumors exhibit profound histological heterogeneity. In the 1930's, it was discovered that a single cell from a mouse tumor could initiate a new tumor in a recipient mouse. Subsequently, several studies showed that the number of cells with tumor-initiating properties in solid tumors and leukemias was found to be variable but low (103 to 107 cells). The resulting tumors typically

CSCs possess several key properties of normal tissue stem cells including self-renewal, unlimited proliferative potential, infrequent or slow replication, resistance to toxic xenobiotics, enhanced DNA repair capacity, and the ability to give rise to daughter cells that differentiate. However, unlike highly regulated tissue stem cells, CSCs demonstrate deregulated self-renewal/differentiation processes and generate daughter cells that arrest at various stages of differentiation. The progeny of the stem cells make up the bulk of the tumor and are characterized by rapid replication, limited proliferative potential, and the inability to form a new tumor. Only the CSC is able to initiate tumor formation as it is solely capable of self-renewal (Figure 1). CSCs are thought to maintain their numbers by slow selfreplication and produce other tumor cells by asymmetric cell division. In this process, cell division of a CSC generates a CSC and a transformed "progenitor-like" cell, which has limited self-renewal ability but is highly proliferative, similar to a transit-amplifying

**2. Cancer stem cell theory** 

Fig. 1. The cancer stem cell hypothesis. In normal tissues, cell numbers are regulated by asymmetric division of stem cells to replace the loss of functional differentiated cells. In cancer, the mutated stem cell drives tumor heterogeneity through asymmetric division and aberrant proliferation/differentiation. Only the CSC has the capacity to form new tumors and accumulate further mutations leading to tumor progression.

population in normal tissues. These progenitors give rise to more or less partially differentiated bulk tumor cells through a combination of proliferation and abortive differentiation.

The existence of CSCs has been supported by seminal research performed on acute myeloid leukemia (AML), where it has been demonstrated that only a specific cellular subset that express antigenic markers similar to hematopoietic stem cells has clonogenic activity in immunocompromised mice (Lapidot et al. 1994; Bonnet and Dick 1997). The cell responsible for tumor initiation was identified as having a CD34+ CD38− phenotype; interestingly the majority of AML cells tend to be CD34−. It was observed that as few as 5 ×103 CD34+ CD38<sup>−</sup> cells could successfully engraft an immunocompromised mouse, while 100 times more CD34− or CD34+ CD38+ cells were not tumorigenic (Bonnet and Dick 1997). Significantly, the resulting tumors were heterogeneous and composed of a mixture of tumorigenic and nontumorigenic cells similar to the donor leukemia. Subsequently, studies on tumors of epithelial origin, such as breast cancer, also provided evidence for the presence of stem-like cells within the cancer (Dontu et al. 2005). Ponti and colleagues demonstrated that only CD44+/CD24- cells, isolated from breast cancer, were able to produce tumors in immunocompromised mice (Ponti et al. 2005). The initial publications in leukemia and breast cancer were followed by reports showing the prospective isolation of CSCs in numerous malignancies including: brain, colon, head and neck, pancreatic, melanoma, mesenchymal, hepatic, lung, prostate, and ovarian tumors (O'Brien et al. 2010). The

The Stem Cell Environment: Kinetics, Signaling and Markers 305

Much of the detailed research on intestinal stem cells has been carried out in the setting of the small intestine where the localization of putative intestinal stem cells (ISCs) was initially studied using indirect means. Based on the premise that stem cells may be slowly cycling, Potten and colleagues used long-term label retaining techniques (tritiated thymidine or bromodeoxyuridine) to mark putative ISCs in the small intestine (Potten et al. 1997). They detected long-term label-retaining cells (LRCs) in an annulus four cells up (+4 LRCs) from the crypt base. More recently, using in vivo lineage tracing, it was shown that cells expressing BMI1 predominantly mark +4 LRC position and are able to give rise to all four epithelial lineages (Sangiorgi and Capecchi 2009). BMI1 encodes a chromatin remodeling protein of the polycomb group that has essential roles in self-renewal of hematopoietic and neural stem cells. However, an alternative hypothesis was put forward by Barker and colleagues, who identified a WNT target gene, LGR5/GPR49, which is expressed exclusively in crypt base columnar cells (CBCs) at crypt positions 1–4 (Barker et al. 2007). They elegantly showed that LGR5- expressing CBCs fulfill all criteria of putative ISCs in that they can persist for a long time, self-renew, give rise to all mature intestinal epithelial cells, and are also apoptosis-resistant. However, in contrast to the putative +4 LRC/BMI1 cells, they are highly proliferative. This raises the possibility that there are two types of intestinal stem cells: quiescent stem cells +4 LRC/BMI1 reflecting their inhibitory microenvironment, and the active CBC/LGR5-positive stem cells, representing a population of stem cells able to respond to stimulating signals generated from adjacent mesenchymal cells (Scoville et al. 2008). Very recently, work from Clevers laboratory has implicated the

Paneth cell as an important component of the CBC/LGR5 niche (Sato et al. 2011).

average, 19 ISCs per crypt (Potten et al. 2003).

There are clear differences between the small and large intestine, apart from the lack of villi. Paneth cells are not generated in the large intestine, and there are differences in the enteroendocrine cell types. However, colonic stem cells have also been shown to reside in the base of the crypts within the stem-cell niche, which is formed by the stem cells themselves and mesenchymal cells that surround the crypt base (Potten 1998). Using bromodeoxyuridine injections into patients with various colorectal cancers, we were able to show marked differences in the proliferation characteristics of "normal" ileum, colon and rectal mucosa (Potten et al. 1992). The mean crypt height in sections of the human colon was 81.9 and 79.5 cells for the rectum. The mean crypt circumference was 41.6 cells in the colon, and 46.0 cells in the rectum. This gave a total of 2044 cells per crypt in the colon and 2194 cells per crypt in the rectum. In the colonic crypts 10% of cells were in S phase and 0 4% in mitosis. Ninety per cent of labeled cells were found between cell positions 4 and 43; we showed that the maximum labeling index was about 30% and occurred at cell position 15. The labeling index at the crypt base, the putative stem cell zone, was about 14%. The rectum showed significant differences. The rectal mucosal crypts contain approximately 30% fewer S phase and mitotic cells (Figure 3). This may indicate either that the cell cycle time of rectal mucosa cells is longer than in the colon, or that there are fewer proliferating cells in the rectum. Extrapolating from biologic studies in rodents suggests that ISCs in the human colonic and rectal crypts represent only a small proportion of crypt cells (approximately 20 cells per crypt, or approximately 1%) (Potten and Loeffler 1990). This finding is consistent with recent immunostaining studies in human colonic crypts for Musashi-1 protein, a putative ISC marker, indicating that there are, on

A key component of tissue architecture that is involved in the regulation of stem cells has been termed the "stem cell niche " (Spradling et al. 2001). The stem cell niche has been well characterized in hematopoietic and neural systems and is an intricate and dynamic milieu

existence of CSCs in rectal cancer has yet to be verified by stem cell isolation and xenotransplantation into immunodeficient mice.

#### **3. Intestinal structure and the stem cell niche**

The large bowel consists of a rapidly proliferating and perpetually differentiating epithelium. Unlike the small intestine, the mucosal surface of the colon has no villi. The ileocecal junction marks an abrupt transition from the villi of the small intestine to the smooth glandular pattern of colon. The crypts of Leiberkuhn continue, and these straight tubular glands are lined with simple columnar epithelium for the reabsorbtion of water and electrolytes, numerous goblet cells for mucus secretion, stem cells for replication, and occasional enteroendocrine cells. Stem cells located in the crypts of Lieberkuhn give rise to proliferating progenitor or transit amplifying cells that differentiate into the four major epithelial cell types (Figure 2). These include columnar absorptive cells or enterocytes, mucous-secreting goblet cells, enteroendocrine cells, and Paneth cells. Enterocytic, goblet, and enteroendocrine cell differentiation takes place during migration upward from the crypt to the surface epithelium, whereas Paneth cells complete their differentiation at the crypt base. The crypt is surrounded by the supporting lamina propria which contains cells of mesenchymal origin, the pericryptal myofibroblasts, which are derived from a mesenchymal lineage.

Fig. 2. The intestinal stem cell structure. Two putative populations of stem cells exist: (1) a quiescent/reserved population that consists of label retaining cells located above the basally situated Paneth cell and (2) an actively cycling/primed population that consists of crypt base columnar cells.

existence of CSCs in rectal cancer has yet to be verified by stem cell isolation and

The large bowel consists of a rapidly proliferating and perpetually differentiating epithelium. Unlike the small intestine, the mucosal surface of the colon has no villi. The ileocecal junction marks an abrupt transition from the villi of the small intestine to the smooth glandular pattern of colon. The crypts of Leiberkuhn continue, and these straight tubular glands are lined with simple columnar epithelium for the reabsorbtion of water and electrolytes, numerous goblet cells for mucus secretion, stem cells for replication, and occasional enteroendocrine cells. Stem cells located in the crypts of Lieberkuhn give rise to proliferating progenitor or transit amplifying cells that differentiate into the four major epithelial cell types (Figure 2). These include columnar absorptive cells or enterocytes, mucous-secreting goblet cells, enteroendocrine cells, and Paneth cells. Enterocytic, goblet, and enteroendocrine cell differentiation takes place during migration upward from the crypt to the surface epithelium, whereas Paneth cells complete their differentiation at the crypt base. The crypt is surrounded by the supporting lamina propria which contains cells of mesenchymal

origin, the pericryptal myofibroblasts, which are derived from a mesenchymal lineage.

Fig. 2. The intestinal stem cell structure. Two putative populations of stem cells exist: (1) a quiescent/reserved population that consists of label retaining cells located above the basally situated Paneth cell and (2) an actively cycling/primed population that consists of

crypt base columnar cells.

xenotransplantation into immunodeficient mice.

**3. Intestinal structure and the stem cell niche** 

Much of the detailed research on intestinal stem cells has been carried out in the setting of the small intestine where the localization of putative intestinal stem cells (ISCs) was initially studied using indirect means. Based on the premise that stem cells may be slowly cycling, Potten and colleagues used long-term label retaining techniques (tritiated thymidine or bromodeoxyuridine) to mark putative ISCs in the small intestine (Potten et al. 1997). They detected long-term label-retaining cells (LRCs) in an annulus four cells up (+4 LRCs) from the crypt base. More recently, using in vivo lineage tracing, it was shown that cells expressing BMI1 predominantly mark +4 LRC position and are able to give rise to all four epithelial lineages (Sangiorgi and Capecchi 2009). BMI1 encodes a chromatin remodeling protein of the polycomb group that has essential roles in self-renewal of hematopoietic and neural stem cells. However, an alternative hypothesis was put forward by Barker and colleagues, who identified a WNT target gene, LGR5/GPR49, which is expressed exclusively in crypt base columnar cells (CBCs) at crypt positions 1–4 (Barker et al. 2007). They elegantly showed that LGR5- expressing CBCs fulfill all criteria of putative ISCs in that they can persist for a long time, self-renew, give rise to all mature intestinal epithelial cells, and are also apoptosis-resistant. However, in contrast to the putative +4 LRC/BMI1 cells, they are highly proliferative. This raises the possibility that there are two types of intestinal stem cells: quiescent stem cells +4 LRC/BMI1 reflecting their inhibitory microenvironment, and the active CBC/LGR5-positive stem cells, representing a population of stem cells able to respond to stimulating signals generated from adjacent mesenchymal cells (Scoville et al. 2008). Very recently, work from Clevers laboratory has implicated the Paneth cell as an important component of the CBC/LGR5 niche (Sato et al. 2011).

There are clear differences between the small and large intestine, apart from the lack of villi. Paneth cells are not generated in the large intestine, and there are differences in the enteroendocrine cell types. However, colonic stem cells have also been shown to reside in the base of the crypts within the stem-cell niche, which is formed by the stem cells themselves and mesenchymal cells that surround the crypt base (Potten 1998). Using bromodeoxyuridine injections into patients with various colorectal cancers, we were able to show marked differences in the proliferation characteristics of "normal" ileum, colon and rectal mucosa (Potten et al. 1992). The mean crypt height in sections of the human colon was 81.9 and 79.5 cells for the rectum. The mean crypt circumference was 41.6 cells in the colon, and 46.0 cells in the rectum. This gave a total of 2044 cells per crypt in the colon and 2194 cells per crypt in the rectum. In the colonic crypts 10% of cells were in S phase and 0 4% in mitosis. Ninety per cent of labeled cells were found between cell positions 4 and 43; we showed that the maximum labeling index was about 30% and occurred at cell position 15. The labeling index at the crypt base, the putative stem cell zone, was about 14%. The rectum showed significant differences. The rectal mucosal crypts contain approximately 30% fewer S phase and mitotic cells (Figure 3). This may indicate either that the cell cycle time of rectal mucosa cells is longer than in the colon, or that there are fewer proliferating cells in the rectum. Extrapolating from biologic studies in rodents suggests that ISCs in the human colonic and rectal crypts represent only a small proportion of crypt cells (approximately 20 cells per crypt, or approximately 1%) (Potten and Loeffler 1990). This finding is consistent with recent immunostaining studies in human colonic crypts for Musashi-1 protein, a putative ISC marker, indicating that there are, on average, 19 ISCs per crypt (Potten et al. 2003).

A key component of tissue architecture that is involved in the regulation of stem cells has been termed the "stem cell niche " (Spradling et al. 2001). The stem cell niche has been well characterized in hematopoietic and neural systems and is an intricate and dynamic milieu

The Stem Cell Environment: Kinetics, Signaling and Markers 307

This also highlights the importance of members of the BMP pathway as important contributors to the colorectal epithelial stem cell niche by modulation of the Wnt pathway. Bone morphogenetic proteins (BMPs) bind to BMP receptor types I or II (BMPR1 or BMPR2) leading to phosphorylation of SMAD1, 5 or 8, which then form a heterodimer with SMAD4, translocate to the nucleus, and act as transcriptional activators (von Bubnoff and Cho 2001). Active BMP signaling, indicated by phosphorylated SMADs, is found predominantly in differentiated intestinal epithelial cells. Several lines of evidence support the postulated inhibitory role of BMP signaling on stem cell self-renewal. These include the observations that conditional mutation of BMPR1A resulted in *de novo* crypt formation and a juvenile polyposis phenotype (He et al. 2004) and leads to reduced differentiation into the three secretory cells types, Paneth cells, goblet cells and enteroendocrine cells in mice (Auclair et al. 2007). Also, human juvenile polyposis has been shown to be associated with mutations in

Several other pathways have been shown to be important in intestinal stem cell regulation including Notch signaling, hedgehog signaling, and the PTEN–PI3K–Akt pathway. Notch has been implicated in the control of cell fate in many tissues. The binding of the ligands Jagged or Delta to the Notch receptor induces proteolytic cleavage by *γ*–secretase which releases a fragment, NCID. This fragment translocates to the nucleus and acts as a transcription factor after dimerization with RBP-j*κ*/CSL. One of the key genes stimulated by this activation is a bHLH transcription factor termed hairy/enhancer of split (Hes), which has been shown to activate factors involved in the control of proliferation and differentiation (Bray 2006). Knocking out RBP-j*κ* or Hes1 leads to increased numbers of secretory epithelial cells (Jensen et al. 2000) whereas mutation in ATOK1, a transcription factor repressed by Notch signaling, leads to depletion of all three secretory lineages (Yang et al. 2001). Therefore, Notch functions by stimulating proliferation of crypt progenitor cells in the transit-amplifying units, and suppression of Notch signaling induces specific differentiation into the intestinal epithelial lineages. A role of the PTEN–PI3K–Akt in enhancing stem cell self-renewal in the intestine has been suggested as a result of the connection between this pathway and the Wnt pathway. p-Akt can phosphorylate *β*-catenin and thus enhance the transcriptional activity of *β*-catenin whilst PTEN exhibits strongest expression in lumenal epithelial cells and might be involved in the restriction of strong Wnt signaling to the crypt base (Persad et al. 2001). In contrast to the other signal pathways which seem to be regulated in response to ligands originating from the niche cells, the morphogens of the hedgehog (HH) pathway, sonic hedgehog (Shh), and Indian hedgehog (Ihh) are secreted by epithelial cells. Their receptor, Patched (PTCH), is expressed on the pericryptal myofibroblasts. Accordingly, HH signaling is not directly concerned with the fate of the epithelial cells but is important in shaping and regulating the proper overall structure of the intestinal mucosa

the SMAD4/DPC4 or BMPR1A genes (Sayed et al. 2002).

into crypts and villi (Madison et al. 2005).

**4. Stem cells and the development of colorectal cancer** 

of the respiratory chain, are also relatively common (Taylor et al. 2003).

The mechanisms underlying colorectal cancer initiation have yet to be fully elucidated (see section 5). It is clear that the APC gene is involved as APC mutations are found in 75% to 80% of sporadic CRCs (Powell et al. 1992). However, mutations in mitochondrial DNA and mutations in the genes encoding cytochrome c oxidase (COX), a component of complex IV

Fig. 3. A comparison of the bromodeoxyuridine labeling index frequency plots for the human colon and the rectum.

that adapts in response to environmental signals. The niche consists of a stromal microenvironment surrounding the stem cell population that can contain neural cells, lymphocytes, macrophages, endothelial cells, fibroblasts, smooth muscle cells and myofibroblasts in an extracellular matrix (Li and Xie 2005). As mentioned previously, the crypt is surrounded by the supporting lamina propria which contains pericryptal myofibroblasts. These myofibroblasts are thought to be strategic cells in the regulation of stem cell behavior through growth factor and cytokine signaling (Mills and Gordon 2001). Several key signaling pathways are common to stem cells and their niche including Wnt, BMP and Notch.

The discovery that mutations in the APC gene, the most important tumor suppressor in intestinal tumorigenesis, affects the control of Wnt signaling, indicated the importance of this pathway in intestinal stem cell regulation (Korinek et al. 1997). It is thought that the pericryptal myofibroblasts produce the Wnt signaling ligands that access Frizzled receptors on the basal epithelial stem cells (Fevr et al. 2007), and this prevents degradation of the main effector, *β*-catenin by a destruction complex containing APC and AXIN1/2. *β*-Catenin translocates to the nucleus, where it acts as a transcriptional activator after binding to TCF/LEF family members. Korinek *et al* showed that lack of the intestine specific *β*-catenin partner, TCF4, resulted in the depletion of the epithelial stem cell compartment in the small intestine (Korinek et al. 1998). Wnt signaling varies across the crypt (Kosinski et al. 2007) in that the crypt top is characterized by APC, WNT5B, and TCF4 whilst the crypt bottom, the putative stem cell niche, expresses AXIN2, TCF3, and several secreted Wnt inhibitors including DKK3, SFRP1, SFRP2, FZD2, FZD3, FZD7, and FZDB. The identification of many different Wnt/ *β*-catenin target genes indicates that Wnt signaling has different effects in different cell types depending on their localization along the crypt axis. The microenvironment surrounding the LGR5/GPR49 stem cell is characterized by prominent Wnt activity and inhibition of BMP signaling with the presence of BMP inhibitors noggin and gremlin whereas the microenvironment surrounding the +4 LRC is characterized by expression of the Wnt inhibitor sFRP5 and BMP4 (Pages et al. 2009).

Fig. 3. A comparison of the bromodeoxyuridine labeling index frequency plots for the

that adapts in response to environmental signals. The niche consists of a stromal microenvironment surrounding the stem cell population that can contain neural cells, lymphocytes, macrophages, endothelial cells, fibroblasts, smooth muscle cells and myofibroblasts in an extracellular matrix (Li and Xie 2005). As mentioned previously, the crypt is surrounded by the supporting lamina propria which contains pericryptal myofibroblasts. These myofibroblasts are thought to be strategic cells in the regulation of stem cell behavior through growth factor and cytokine signaling (Mills and Gordon 2001). Several key signaling pathways are common to stem cells and their niche including Wnt,

The discovery that mutations in the APC gene, the most important tumor suppressor in intestinal tumorigenesis, affects the control of Wnt signaling, indicated the importance of this pathway in intestinal stem cell regulation (Korinek et al. 1997). It is thought that the pericryptal myofibroblasts produce the Wnt signaling ligands that access Frizzled receptors on the basal epithelial stem cells (Fevr et al. 2007), and this prevents degradation of the main effector, *β*-catenin by a destruction complex containing APC and AXIN1/2. *β*-Catenin translocates to the nucleus, where it acts as a transcriptional activator after binding to TCF/LEF family members. Korinek *et al* showed that lack of the intestine specific *β*-catenin partner, TCF4, resulted in the depletion of the epithelial stem cell compartment in the small intestine (Korinek et al. 1998). Wnt signaling varies across the crypt (Kosinski et al. 2007) in that the crypt top is characterized by APC, WNT5B, and TCF4 whilst the crypt bottom, the putative stem cell niche, expresses AXIN2, TCF3, and several secreted Wnt inhibitors including DKK3, SFRP1, SFRP2, FZD2, FZD3, FZD7, and FZDB. The identification of many different Wnt/ *β*-catenin target genes indicates that Wnt signaling has different effects in different cell types depending on their localization along the crypt axis. The microenvironment surrounding the LGR5/GPR49 stem cell is characterized by prominent Wnt activity and inhibition of BMP signaling with the presence of BMP inhibitors noggin and gremlin whereas the microenvironment surrounding the +4 LRC is characterized by

expression of the Wnt inhibitor sFRP5 and BMP4 (Pages et al. 2009).

human colon and the rectum.

BMP and Notch.

This also highlights the importance of members of the BMP pathway as important contributors to the colorectal epithelial stem cell niche by modulation of the Wnt pathway. Bone morphogenetic proteins (BMPs) bind to BMP receptor types I or II (BMPR1 or BMPR2) leading to phosphorylation of SMAD1, 5 or 8, which then form a heterodimer with SMAD4, translocate to the nucleus, and act as transcriptional activators (von Bubnoff and Cho 2001). Active BMP signaling, indicated by phosphorylated SMADs, is found predominantly in differentiated intestinal epithelial cells. Several lines of evidence support the postulated inhibitory role of BMP signaling on stem cell self-renewal. These include the observations that conditional mutation of BMPR1A resulted in *de novo* crypt formation and a juvenile polyposis phenotype (He et al. 2004) and leads to reduced differentiation into the three secretory cells types, Paneth cells, goblet cells and enteroendocrine cells in mice (Auclair et al. 2007). Also, human juvenile polyposis has been shown to be associated with mutations in the SMAD4/DPC4 or BMPR1A genes (Sayed et al. 2002).

Several other pathways have been shown to be important in intestinal stem cell regulation including Notch signaling, hedgehog signaling, and the PTEN–PI3K–Akt pathway. Notch has been implicated in the control of cell fate in many tissues. The binding of the ligands Jagged or Delta to the Notch receptor induces proteolytic cleavage by *γ*–secretase which releases a fragment, NCID. This fragment translocates to the nucleus and acts as a transcription factor after dimerization with RBP-j*κ*/CSL. One of the key genes stimulated by this activation is a bHLH transcription factor termed hairy/enhancer of split (Hes), which has been shown to activate factors involved in the control of proliferation and differentiation (Bray 2006). Knocking out RBP-j*κ* or Hes1 leads to increased numbers of secretory epithelial cells (Jensen et al. 2000) whereas mutation in ATOK1, a transcription factor repressed by Notch signaling, leads to depletion of all three secretory lineages (Yang et al. 2001). Therefore, Notch functions by stimulating proliferation of crypt progenitor cells in the transit-amplifying units, and suppression of Notch signaling induces specific differentiation into the intestinal epithelial lineages. A role of the PTEN–PI3K–Akt in enhancing stem cell self-renewal in the intestine has been suggested as a result of the connection between this pathway and the Wnt pathway. p-Akt can phosphorylate *β*-catenin and thus enhance the transcriptional activity of *β*-catenin whilst PTEN exhibits strongest expression in lumenal epithelial cells and might be involved in the restriction of strong Wnt signaling to the crypt base (Persad et al. 2001). In contrast to the other signal pathways which seem to be regulated in response to ligands originating from the niche cells, the morphogens of the hedgehog (HH) pathway, sonic hedgehog (Shh), and Indian hedgehog (Ihh) are secreted by epithelial cells. Their receptor, Patched (PTCH), is expressed on the pericryptal myofibroblasts. Accordingly, HH signaling is not directly concerned with the fate of the epithelial cells but is important in shaping and regulating the proper overall structure of the intestinal mucosa into crypts and villi (Madison et al. 2005).

#### **4. Stem cells and the development of colorectal cancer**

The mechanisms underlying colorectal cancer initiation have yet to be fully elucidated (see section 5). It is clear that the APC gene is involved as APC mutations are found in 75% to 80% of sporadic CRCs (Powell et al. 1992). However, mutations in mitochondrial DNA and mutations in the genes encoding cytochrome c oxidase (COX), a component of complex IV of the respiratory chain, are also relatively common (Taylor et al. 2003).

The Stem Cell Environment: Kinetics, Signaling and Markers 309

alterations are responsible for the transition to more tumorigenic phenotypes. While alterations in the APC (adenomatous polyposis coli gene)/β-catenin pathways as well as inactivation of mismatch repair proteins generally occur early, modifications to p53 and DCC/SMAD4/SMAD2 occur as one of the final steps in the progression to carcinoma. The step-wise progression of colorectal carcinogenesis through transitional dysplastic and adenoma stages is demonstrated by the high rate of success seen in preventing the development of colorectal cancer by removing polyps. Polyps represent the dysplastic and adenoma stages of colorectal carcinogenesis, and their removal prevents the development of

There are three main pathways that lead to the genetic alterations responsible for colorectal tumorigenesis, the chromosomal instability (CIN) pathway, the mismatch repair (MMR) pathway and the hypermethylation phenotype. The CIN pathway is characterized by alteration of APC tumor suppressor gene signaling. A germline mutation of the APC gene results in the development of familial adenomatous polyposis (FAP) which is typified by hundreds to thousands of colorectal polyps by age 20-30. Tumors that developed via the CIN pathway have a high level of chromosomal instability that results in large numbers of deletions, insertions, and loss of heterozygosity. The MMR pathway to colorectal carcinogenesis results from a failure of DNA repair genes, in particular MLH1 and MSH2. This malfunction in DNA repair results in an accumulation of errors throughout the genome, particularly in areas called microsatellites. Microsatellites are short nucleotide regions that are repeated hundreds of times within the genome; thus, the MMR pathway is characterized by microsatellite instability. Germline mutations in one of the MMR genes results in hereditary nonpolyposis colorectal cancer (HNPCC). Finally, the hypermethylation pathway is characterized by a high incidence of methylation of CpG

In addition to these major pathways in colorectal carcinogenesis, there are specific genetic pathways that have a role in colorectal carcinogenesis. The gene K-ras is mutated in ~50% of sporadic colorectal cancer and in 50% of adenomas larger than 1 cm while rarely in smaller adenomas (Vogelstein et al. 1988). The lack of mutations in the smaller adenomas implies that the K-ras mutation is relevant to a later stage of progression. However, the presence of K-ras mutations in both nondysplastic aberrant crypt foci and hyperplastic polyps makes

Another commonly mutated gene in colorectal cancer is p53. In response to DNA damage and other stress, p53 induces responses ranging from cell cycle arrest and senescence to differentiation and is inactivated in 50-70% of colorectal cancers. The p53 gene is located on chromosome 17p which is lost in up to 75% of colorectal cancers, but it is lost rarely in adenomas. This suggests that the loss occurs late in the progression (Baker et al. 1990). One more common feature of colorectal cancers is the loss of chromosome 18q. The deletion is seen in 73% of sporadic colorectal cancers and 47% of large adenomas but less in smaller adenomas (Vogelstein et al. 1988). This divergence implies the chromosomal loss occurs later in tumorigenesis. Chromosome 18q contains three significant genes: DCC ("deleted in colon cancer"), SMAD4, and SMAD2. DCC functions as a tumor suppressor and has a role in cell-cell interactions. Loss of DCC expression is associated with a worse overall survival in colorectal cancer patients (Popat et al. 2007). As mentioned previously, SMAD4 and SMAD2 are both involved in BMP signaling as well as in the TGFβ signaling cascade which

islands which may result in gene silencing of the MMR genes.

modulates cell proliferation, apoptosis, and differentiation.

the role of these mutations unclear.

carcinoma.

When an initial mutation occurs in a basally situated crypt stem cell, it gives rise to a clone that migrates up the crypt, expanding as it progresses. At this stage the proliferation pattern of crypts are shifted toward the crypt top with a maximum labeling index (LI) at approximately crypt level 20. The mutant clone then begins to colonize the base of the crypt, in effect taking over and replacing the non-mutant cells in the stem-cell niche in a process that has been termed niche succession. Eventually the entire niche will be colonized with mutant stem cells and the crypt filled with their progeny, a result termed monoclonal conversion. Interestingly, crypts containing this proliferative abnormality do not show any discernible histological changes. Crypts only begin to show obvious abnormalities when they become dysplastic during later formation of premalignant adenomas. Boman and colleagues have postulated that only an increase in crypt SC number and not changes in cell cycle proliferation, differentiation, or apoptosis of non-SC populations, could explain the LI shift in these crypts (Boman et al. 2001). This led to the hypothesis that the link between APC mutation and the LI shift is crypt SC overpopulation caused by a decrease in the rate of degradation of cytoplasmic β-catenin and alteration of TCF-4 transcriptional activity and survivin expression leading to inhibition of apoptosis and promotion of mitosis (Boman and Huang 2008).

Inactivation of the second APC allele occurs during the development of intestinal adenomas, and it has been proposed that the next critical event may be the movement of the progeny of the mutated stem cell moving from the niche into the proliferative zone of the crypt where they are freed from the constraints of the niche cells and are able to undergo further symmetric divisions and clonal expansion forming monocryptal adenomas (Humphries and Wright 2008). These are the earliest histologically detectable precursor lesions of tumor development and are thought to precede adenoma development. Further clonal expansion seems to be through crypt fission. Crypt fission is a normal process that leads to crypt replication. This process is responsible for the increased number of new crypts that arise during a short postnatal period, after which the total number of crypts increases only gradually with age. During crypt fission, development of a fissure bisects the crypt base and ascends longitudinally. This bifurcation results in the symmetric creation of two identical daughter crypts and must therefore be a process that results from symmetric ISC division. Experimental evidence suggests that mutant APC and an increasing rate of crypt fission, leads to abnormal, asymmetric crypt fissioning during adenoma development resulting in characteristic crypt branching and budding (Wasan et al. 1998). Crypt fission is also the mechanism that leads to the spread of mutant crypt populations in normal colonic epithelium. Further development of the lesion may be through random collision between neighboring neoplastic clones or through clonal interaction in which active cooperation between multiple initiated clones promotes continued survival and growth of the adenoma leading to genetic heterogeneity (Axelrod et al. 2006). Indeed, it has been shown that genetic alterations occur in the stroma from an early stage of carcinogenesis and that these may induce microregional differences in tumor susceptibility promoting loss of heterozygosity in the associated epithelium (Thliveris et al. 2005).

#### **5. Genetic differences between colon and rectal tumors**

The most well known model for colorectal carcinogenesis (the Vogelstein model) describes the progression of normal epithelium into adenomatous polyps and neoplasia and finally into metastatic carcinoma (Fearon and Vogelstein 1990). A series of specific genetic

When an initial mutation occurs in a basally situated crypt stem cell, it gives rise to a clone that migrates up the crypt, expanding as it progresses. At this stage the proliferation pattern of crypts are shifted toward the crypt top with a maximum labeling index (LI) at approximately crypt level 20. The mutant clone then begins to colonize the base of the crypt, in effect taking over and replacing the non-mutant cells in the stem-cell niche in a process that has been termed niche succession. Eventually the entire niche will be colonized with mutant stem cells and the crypt filled with their progeny, a result termed monoclonal conversion. Interestingly, crypts containing this proliferative abnormality do not show any discernible histological changes. Crypts only begin to show obvious abnormalities when they become dysplastic during later formation of premalignant adenomas. Boman and colleagues have postulated that only an increase in crypt SC number and not changes in cell cycle proliferation, differentiation, or apoptosis of non-SC populations, could explain the LI shift in these crypts (Boman et al. 2001). This led to the hypothesis that the link between APC mutation and the LI shift is crypt SC overpopulation caused by a decrease in the rate of degradation of cytoplasmic β-catenin and alteration of TCF-4 transcriptional activity and survivin expression leading to inhibition of apoptosis and promotion of mitosis (Boman and

Inactivation of the second APC allele occurs during the development of intestinal adenomas, and it has been proposed that the next critical event may be the movement of the progeny of the mutated stem cell moving from the niche into the proliferative zone of the crypt where they are freed from the constraints of the niche cells and are able to undergo further symmetric divisions and clonal expansion forming monocryptal adenomas (Humphries and Wright 2008). These are the earliest histologically detectable precursor lesions of tumor development and are thought to precede adenoma development. Further clonal expansion seems to be through crypt fission. Crypt fission is a normal process that leads to crypt replication. This process is responsible for the increased number of new crypts that arise during a short postnatal period, after which the total number of crypts increases only gradually with age. During crypt fission, development of a fissure bisects the crypt base and ascends longitudinally. This bifurcation results in the symmetric creation of two identical daughter crypts and must therefore be a process that results from symmetric ISC division. Experimental evidence suggests that mutant APC and an increasing rate of crypt fission, leads to abnormal, asymmetric crypt fissioning during adenoma development resulting in characteristic crypt branching and budding (Wasan et al. 1998). Crypt fission is also the mechanism that leads to the spread of mutant crypt populations in normal colonic epithelium. Further development of the lesion may be through random collision between neighboring neoplastic clones or through clonal interaction in which active cooperation between multiple initiated clones promotes continued survival and growth of the adenoma leading to genetic heterogeneity (Axelrod et al. 2006). Indeed, it has been shown that genetic alterations occur in the stroma from an early stage of carcinogenesis and that these may induce microregional differences in tumor susceptibility promoting loss of heterozygosity in

Huang 2008).

the associated epithelium (Thliveris et al. 2005).

**5. Genetic differences between colon and rectal tumors** 

The most well known model for colorectal carcinogenesis (the Vogelstein model) describes the progression of normal epithelium into adenomatous polyps and neoplasia and finally into metastatic carcinoma (Fearon and Vogelstein 1990). A series of specific genetic alterations are responsible for the transition to more tumorigenic phenotypes. While alterations in the APC (adenomatous polyposis coli gene)/β-catenin pathways as well as inactivation of mismatch repair proteins generally occur early, modifications to p53 and DCC/SMAD4/SMAD2 occur as one of the final steps in the progression to carcinoma. The step-wise progression of colorectal carcinogenesis through transitional dysplastic and adenoma stages is demonstrated by the high rate of success seen in preventing the development of colorectal cancer by removing polyps. Polyps represent the dysplastic and adenoma stages of colorectal carcinogenesis, and their removal prevents the development of carcinoma.

There are three main pathways that lead to the genetic alterations responsible for colorectal tumorigenesis, the chromosomal instability (CIN) pathway, the mismatch repair (MMR) pathway and the hypermethylation phenotype. The CIN pathway is characterized by alteration of APC tumor suppressor gene signaling. A germline mutation of the APC gene results in the development of familial adenomatous polyposis (FAP) which is typified by hundreds to thousands of colorectal polyps by age 20-30. Tumors that developed via the CIN pathway have a high level of chromosomal instability that results in large numbers of deletions, insertions, and loss of heterozygosity. The MMR pathway to colorectal carcinogenesis results from a failure of DNA repair genes, in particular MLH1 and MSH2. This malfunction in DNA repair results in an accumulation of errors throughout the genome, particularly in areas called microsatellites. Microsatellites are short nucleotide regions that are repeated hundreds of times within the genome; thus, the MMR pathway is characterized by microsatellite instability. Germline mutations in one of the MMR genes results in hereditary nonpolyposis colorectal cancer (HNPCC). Finally, the hypermethylation pathway is characterized by a high incidence of methylation of CpG islands which may result in gene silencing of the MMR genes.

In addition to these major pathways in colorectal carcinogenesis, there are specific genetic pathways that have a role in colorectal carcinogenesis. The gene K-ras is mutated in ~50% of sporadic colorectal cancer and in 50% of adenomas larger than 1 cm while rarely in smaller adenomas (Vogelstein et al. 1988). The lack of mutations in the smaller adenomas implies that the K-ras mutation is relevant to a later stage of progression. However, the presence of K-ras mutations in both nondysplastic aberrant crypt foci and hyperplastic polyps makes the role of these mutations unclear.

Another commonly mutated gene in colorectal cancer is p53. In response to DNA damage and other stress, p53 induces responses ranging from cell cycle arrest and senescence to differentiation and is inactivated in 50-70% of colorectal cancers. The p53 gene is located on chromosome 17p which is lost in up to 75% of colorectal cancers, but it is lost rarely in adenomas. This suggests that the loss occurs late in the progression (Baker et al. 1990).

One more common feature of colorectal cancers is the loss of chromosome 18q. The deletion is seen in 73% of sporadic colorectal cancers and 47% of large adenomas but less in smaller adenomas (Vogelstein et al. 1988). This divergence implies the chromosomal loss occurs later in tumorigenesis. Chromosome 18q contains three significant genes: DCC ("deleted in colon cancer"), SMAD4, and SMAD2. DCC functions as a tumor suppressor and has a role in cell-cell interactions. Loss of DCC expression is associated with a worse overall survival in colorectal cancer patients (Popat et al. 2007). As mentioned previously, SMAD4 and SMAD2 are both involved in BMP signaling as well as in the TGFβ signaling cascade which modulates cell proliferation, apoptosis, and differentiation.

The Stem Cell Environment: Kinetics, Signaling and Markers 311

These differences in mutational and gene expression patterns demonstrate a fundamental difference in the carcinogenesis of rectal and colon cancer. With the low rate of MMR disruption seen in rectal cancer, the incidence of MSI is quite low. However, the incidence of CIN is high in rectal cancer as exemplified by the mutations seen in APC. The importance of genomic instability in rectal cancer is further illustrated by examining gene expression studies which were done to differentiate rectal cancer patients that respond to radiation therapy (Watanabe et al. 2006; Ojima et al. 2007; Rimkus et al. 2008) or patients that have local recurrence (Kalady et al. 2010). These four studies identify a total of 120 genes that are differentially expressed between rectal cancer patient populations. Of these genes, 30 are associated with cell survival, and 10 have been linked to genome instability (Figure 4).

Various cell surface markers (Table 1) have been used for the identification of cancer stem cells. These markers are used to isolate sub-populations of cells that are characterized by the ability to reconstitute the original tumor by xenotransplantation using a limited number of cells. Breast cancer stem cells that are CD44+ CD24-/low Lin- are able to form tumors in mice using as few as 100 cells while injection of tens of thousands of cells with different phenotypes fail to form tumors (Al-Hajj et al. 2003). Similarly, CD133+ cells from brain tumors are able to form tumors with an original xenograft of only 100 cells (Singh et al. 2004). Even more astonishing, human ovarian cancer cells that have high ALDH activity and express the cell surface marker CD133 are able to form tumors in mice using as few as 11

CD133+ (Singh et al. 2004), ALDH1 activity (Wang et al. 2011)

**HNSCC** CD44+ (Harper et al. 2007), CD133+ (Yang et al. 2011), ALDH activity (Clay et al. 2010), Hoescht exclusion (Sun et al. 2010) **Lung** CD133+ (Bertolini et al. 2009), ALDH activity (Liang and Shi 2011) **Melanoma** CD20+ (Zabierowski and Herlyn 2008), CD133+ (Gazzaniga et al. 2010) **Ovary** ALDH activity (Silva et al. 2011), Hoescht exclusion (Hosonuma et al. 2011)

**Pancreas** CD44+ CD24+ ESA+ (Lee et al. 2008), ALDH activity (Kim et al. 2011)

**Colon** CD44+ ESAhigh (Dalerba et al. 2007), CD133+ (Ricci-Vitiani et al. 2010), CD166+

**Rectum** CD44+ (Nagata et al. 2011), CD133+ (Nagata et al. 2011), CD133+ESA+ (Yang et

Table 1. Stem cell markers. CNS: central nervous system; HNSCC: Head and neck squamous

cell carcinoma; ALDH: aldehyde dehydrogenase-1; ESA: epithelial-specific antigen

(Dalerba et al. 2007), ALDH activity (Huang et al. 2009), Lgr5+ve (Takahashi et

**Breast** CD44+ CD24-/low Lin- (Al-Hajj et al. 2003), ALD1 activity (Marcato et al. 2011)

**6. Current candidates as stem markers in rectal tumors** 

**Esophagus** Hoescht exclusion (Kalabis et al., 2008)

**Prostate** CD44+ α2β1high CD133+ (Miki et al. 2007)

al. 2011)

al. 2010)

cells (Silva et al. 2011).

**Site Markers** 

**Brain / CNS** 

However, while cancers of the colon and rectum have generally been grouped into the single category of "colorectal" cancer, it has long been speculated that cancers that develop in different anatomical areas of the colon and rectum should be considered as separate diseases. Differences in the biology and function and risk factors between proximal colon, distal colon, and rectum may lend to these divergent disease entities. Genetic evidence also supports etiological evidence that colon and rectal tumors are different entities. These differences between colon and rectal tumors include incidence of certain gene mutations, change in gene expression, even differences in the mechanisms of carcinogenesis. The frequency of mutations to K-ras and APC differs between sites. Mutations to K-ras were more widespread in tumors of the colon. Rectal cancer more often has mutations restricted to APC while colon cancers often contain mutations in several genes (Frattini et al. 2004). In fact, the number of genetic mutations regardless of gene were higher in colon than the rectum (Li and Lai 2009).

In addition to mutations, there are several genes that show different levels of expression between rectal and colon cancers including β-catenin, MMR proteins, p53, and COX2 (cyclooxygenase 2). β-catenin binds the APC protein and is involved in regulating cell growth and adhesion between cells. The cellular localization of β-catenin is altered between colon and rectal cancers. Nuclear β-catenin expression was found more often in cancers of the rectum than colon (Kapiteijn et al. 2001), and reduced membranous and cytoplasmic staining was associated with increased metastatic disease in rectal cancer (Fernebro et al. 2004). Unlike colon cancer, rectal tumors rarely show a loss of expression in MMR proteins including MLH1 and MSH2 (Fernebro et al. 2004). Over-expression of p53 was more common in rectal cancers than colon cancer; however, this may indicate a higher level of p53 mutation in rectal cancer (Kapiteijn et al. 2001). In addition, 90% of rectal tumors demonstrate up-regulation of COX2 while only 20% of colon cancers had increased levels (Li and Lai 2009).

Fig. 4. Gene expression linked cell survival and genome instability in rectal cancer. Genes identified in 4 gene expression studies of rectal cancer were highly represented by genes involved in cell survival and genome instability. Pathway generated by Ariadne Pathway Studio.

However, while cancers of the colon and rectum have generally been grouped into the single category of "colorectal" cancer, it has long been speculated that cancers that develop in different anatomical areas of the colon and rectum should be considered as separate diseases. Differences in the biology and function and risk factors between proximal colon, distal colon, and rectum may lend to these divergent disease entities. Genetic evidence also supports etiological evidence that colon and rectal tumors are different entities. These differences between colon and rectal tumors include incidence of certain gene mutations, change in gene expression, even differences in the mechanisms of carcinogenesis. The frequency of mutations to K-ras and APC differs between sites. Mutations to K-ras were more widespread in tumors of the colon. Rectal cancer more often has mutations restricted to APC while colon cancers often contain mutations in several genes (Frattini et al. 2004). In fact, the number of genetic mutations regardless of gene were higher in colon than the

In addition to mutations, there are several genes that show different levels of expression between rectal and colon cancers including β-catenin, MMR proteins, p53, and COX2 (cyclooxygenase 2). β-catenin binds the APC protein and is involved in regulating cell growth and adhesion between cells. The cellular localization of β-catenin is altered between colon and rectal cancers. Nuclear β-catenin expression was found more often in cancers of the rectum than colon (Kapiteijn et al. 2001), and reduced membranous and cytoplasmic staining was associated with increased metastatic disease in rectal cancer (Fernebro et al. 2004). Unlike colon cancer, rectal tumors rarely show a loss of expression in MMR proteins including MLH1 and MSH2 (Fernebro et al. 2004). Over-expression of p53 was more common in rectal cancers than colon cancer; however, this may indicate a higher level of p53 mutation in rectal cancer (Kapiteijn et al. 2001). In addition, 90% of rectal tumors demonstrate up-regulation of COX2 while only 20% of colon cancers had increased levels

Fig. 4. Gene expression linked cell survival and genome instability in rectal cancer. Genes identified in 4 gene expression studies of rectal cancer were highly represented by genes

involved in cell survival and genome instability. Pathway generated by Ariadne Pathway Studio.

rectum (Li and Lai 2009).

(Li and Lai 2009).

These differences in mutational and gene expression patterns demonstrate a fundamental difference in the carcinogenesis of rectal and colon cancer. With the low rate of MMR disruption seen in rectal cancer, the incidence of MSI is quite low. However, the incidence of CIN is high in rectal cancer as exemplified by the mutations seen in APC. The importance of genomic instability in rectal cancer is further illustrated by examining gene expression studies which were done to differentiate rectal cancer patients that respond to radiation therapy (Watanabe et al. 2006; Ojima et al. 2007; Rimkus et al. 2008) or patients that have local recurrence (Kalady et al. 2010). These four studies identify a total of 120 genes that are differentially expressed between rectal cancer patient populations. Of these genes, 30 are associated with cell survival, and 10 have been linked to genome instability (Figure 4).

#### **6. Current candidates as stem markers in rectal tumors**

Various cell surface markers (Table 1) have been used for the identification of cancer stem cells. These markers are used to isolate sub-populations of cells that are characterized by the ability to reconstitute the original tumor by xenotransplantation using a limited number of cells. Breast cancer stem cells that are CD44+ CD24-/low Lin- are able to form tumors in mice using as few as 100 cells while injection of tens of thousands of cells with different phenotypes fail to form tumors (Al-Hajj et al. 2003). Similarly, CD133+ cells from brain tumors are able to form tumors with an original xenograft of only 100 cells (Singh et al. 2004). Even more astonishing, human ovarian cancer cells that have high ALDH activity and express the cell surface marker CD133 are able to form tumors in mice using as few as 11 cells (Silva et al. 2011).


Table 1. Stem cell markers. CNS: central nervous system; HNSCC: Head and neck squamous cell carcinoma; ALDH: aldehyde dehydrogenase-1; ESA: epithelial-specific antigen

The Stem Cell Environment: Kinetics, Signaling and Markers 313

and other genes and cellular processes. Cellular processes that are associated with CD133 include those that are expected to be involved with cancer stem cells: cell death, morphogenesis, apoptosis, cell differentiation, cell proliferation, and drug response (response to drug). Others may provide hints at important processes that have yet to be investigated: lipid metabolism, glucose metabolism, and vascularization. In addition, CD133 is linked with well-known cancer-related genes such as p53, Myc, Src, and transforming growth factor β1 and the TGFβ-associated SMAD6 and SMAD7. Expression of the gene HIF-1α in conjunction with CD133 is associated with tumor recurrence following chemoradiation (Saigusa et al. 2011). Also associated with recurrence of rectal cancer after chemoradiation is putative stem cell marker POU class 5 homeobox 1 (POU5F1). Expression of CD133, POU5F1, and the SOX2 gene following treatment was associated with poor disease-free

The addition of other known cancer stem cell markers such as CD44 and ESA (EPCAM) to the diagram exemplifies the level of similarity in signaling. Figure 5 shows CD133 associated with 61 other genes and cellular processes. While ESA is associated with 10 of these entities, CD44 has been linked with almost half. CD44 has been associated with the inhibition of apoptosis, cell differentiation, and p53 signaling/expression. Both CD133 and CD44 are linked to CXCL12 (chemokine (C-X-C motif) ligand 12). CXCL12 and its receptor CXCR4 have previously been associated with the mobilization and homing of hematopoetic stem cells (Juarez and Bendall 2004). Interestingly, CXCR4 overlaps much of the CD133 signaling and has been implicated in cancer stem cell signaling at several other sites. Sustained CXCR4/CXCL12 signaling occurs in prostate cancer stem cells (Mimeault and Batra 2011) and is involved in the up-regulation of stem cell-related gene expression in breast cancer cells (Zhang et al. 2011). Additionally, CXCR4 was identified as a therapeutic target of glioblastoma stem cell-like cell lines (Schulte et al. 2011). This implies that, while many markers may be used to identify cancer stem cells, much of the signaling behind these

The mainline cancer therapies of conventional chemo- and radiotherapy target rapidly cycling cancer cells and can cause impressive, but usually temporary, clinical remissions. This initial remission followed by local recurrence would support the argument for the existence of a small subpopulation of resistant CSCs, while at the same time the majority of the non-CSCs being responsive to the treatment. Treatment failure could be explained by several CSC characteristics that would make them difficult to eradicate by conventional agents. First, they may be slow-cycling or quiescent rendering them less sensitive to agents that target actively cycling cells. Second, a characteristic of many normal stem cells is the increased activity of ABC transporter proteins as a protective mechanism against environmental toxins; these are also up-regulated in CSCs (Dean et al. 2005). Third, there are data suggesting that CSCs may be more resistant to radiation (Bao et al. 2006) although this

CD133 has been extensively studied in the context of radiation sensitivity in the setting of glioma. An increase of the CD133+ fraction following irradiation of human glioma cells has been shown in vitro as well as in tumors in nude mice (Bao et al. 2006). The CD133+ cell fraction was found to have a reduced sensitivity to radiation-induced apoptosis. Interestingly, when CD133+ cells were irradiated with 3Gy, they were able to initiate tumors

superficial membrane markers may be quite similar.

has not been universally found (McCord et al. 2009).

**7. Stem cells and treatment response** 

survival.

As mentioned previously, there are several critical pathways that are involved in the maintenance of cancer stem cells including the Wnt, Sonic Hedgehog (SHH), Notch, phosphoinosital-3-kinase (PI3K), and bone morphogenic protein (BMP) pathways (Brabletz et al. 2009). Interestingly, most of the known cancer stem cell markers are not directly related to these pathways; however, these signaling pathways are up-regulated in these cancer stem cell-enriched sub-populations. Pancreatic cancer stem cells that are CD44+ CD24+ ESA+ show an up-regulation of SHH and BMI-1 signaling. These pancreatic cancer stem cells represent less than 1% of all pancreatic cancer cells (Lee et al. 2008), and 100 of these pancreatic stem cells are able to form tumors that are indistinguishable from the original tumor (Chen et al. 2011).

In colorectal cancer, ALDH+ cells are rare in the normal colorectal epithelium and located exclusively in the normal crypt base which is the proposed location for colorectal stem cells. As colorectal carcinogenesis progresses from normal through adenoma and carcinoma, the number of ALDH+ cells increases as well as being distributed more extensively (Huang et al. 2009). Colorectal tumor cells that express both CD44 and ESA are able produce tumors and reproduce the full heterogeneity of the original tissue. CD133+ colorectal cancer stem cells constitute 2.5% of all cells in the tumor (Ricci-Vitiani et al. 2010). These CD133+ cells are able to reproduce the original tumor while the CD133- cells cannot form tumors. However, it is worth noting that a study using fractionating dilution revealed that 1 in 262 CD133+ cells are able to form tumors (O'Brien et al. 2007). While significantly enriched compared to the unfractionated cell population which form tumors at a rate of 1 in 57,000, it still illustrates that not all CD133+ cells are able to reconstitute the original tumors.

Although CSCs have been studied in colon cancer, the existence and implication of stem cells has not been extensively studied in rectal cancer. In a case study, Yang et. al. found elevated levels of CD133+ ESA+ cancer stem cells circulating in the blood of a 75-year old rectal cancer patient who later developed liver metastasis (Yang et al. 2010). Immunohistological analysis of rectal cancer tissue demonstrated that local recurrence was greater for patients that were positive for either CD133 or CD44 (Nagata et al. 2011). Yasuda and colleagues (Yasuda et al. 2009) showed that elevated CD133, but not VEGF or EGFR, was a predictive marker of distant recurrence after preoperative chemoradiotherapy in rectal cancer while Wang et. al. (Wang et al. 2009) showed that the proportion of CD133+ cells was a significant prognostic factor for adverse disease-free survival and overall survival independent of TNM stage, tumour differentiation or lymphovascular invasion. More recently, Kojima and colleagues (Kojima et al. 2010) studied 92 cases of rectal cancer of which 43 patients received preoperative chemoradiation therapy and 49 patients underwent surgery alone. Forty pretreatment biopsy specimens from 43 patients in the preoperative chemoradiation therapy group were also analyzed. CD133-positive cases were more common in the preoperative chemoradiation therapy group than in the surgery-alone group. Furthermore, CD133-positive cases were more common in the preoperative chemoradiation therapy group than in pretreatment biopsy specimens. In the preoperative chemoradiation therapy group, the CD133-positive cases showed poorer prognosis than the CD133-negative cases. These studies suggest that the CD133+ population is important for outcome and that chemoradiation enriches this population.

The biological function of CD133 remains unknown. It is a transmembrane pentaspan protein that was initially described as a surface antigen which was specific to human hematopoietic stem cells (Yin et al. 1997). Utilizing the literature mining software found in Ariadne Pathway Studio, Figure 5 illustrates the known relationships found between CD133

As mentioned previously, there are several critical pathways that are involved in the maintenance of cancer stem cells including the Wnt, Sonic Hedgehog (SHH), Notch, phosphoinosital-3-kinase (PI3K), and bone morphogenic protein (BMP) pathways (Brabletz et al. 2009). Interestingly, most of the known cancer stem cell markers are not directly related to these pathways; however, these signaling pathways are up-regulated in these cancer stem cell-enriched sub-populations. Pancreatic cancer stem cells that are CD44+ CD24+ ESA+ show an up-regulation of SHH and BMI-1 signaling. These pancreatic cancer stem cells represent less than 1% of all pancreatic cancer cells (Lee et al. 2008), and 100 of these pancreatic stem cells are able to form tumors that are indistinguishable from the

In colorectal cancer, ALDH+ cells are rare in the normal colorectal epithelium and located exclusively in the normal crypt base which is the proposed location for colorectal stem cells. As colorectal carcinogenesis progresses from normal through adenoma and carcinoma, the number of ALDH+ cells increases as well as being distributed more extensively (Huang et al. 2009). Colorectal tumor cells that express both CD44 and ESA are able produce tumors and reproduce the full heterogeneity of the original tissue. CD133+ colorectal cancer stem cells constitute 2.5% of all cells in the tumor (Ricci-Vitiani et al. 2010). These CD133+ cells are able to reproduce the original tumor while the CD133- cells cannot form tumors. However, it is worth noting that a study using fractionating dilution revealed that 1 in 262 CD133+ cells are able to form tumors (O'Brien et al. 2007). While significantly enriched compared to the unfractionated cell population which form tumors at a rate of 1 in 57,000, it still illustrates

Although CSCs have been studied in colon cancer, the existence and implication of stem cells has not been extensively studied in rectal cancer. In a case study, Yang et. al. found elevated levels of CD133+ ESA+ cancer stem cells circulating in the blood of a 75-year old rectal cancer patient who later developed liver metastasis (Yang et al. 2010). Immunohistological analysis of rectal cancer tissue demonstrated that local recurrence was greater for patients that were positive for either CD133 or CD44 (Nagata et al. 2011). Yasuda and colleagues (Yasuda et al. 2009) showed that elevated CD133, but not VEGF or EGFR, was a predictive marker of distant recurrence after preoperative chemoradiotherapy in rectal cancer while Wang et. al. (Wang et al. 2009) showed that the proportion of CD133+ cells was a significant prognostic factor for adverse disease-free survival and overall survival independent of TNM stage, tumour differentiation or lymphovascular invasion. More recently, Kojima and colleagues (Kojima et al. 2010) studied 92 cases of rectal cancer of which 43 patients received preoperative chemoradiation therapy and 49 patients underwent surgery alone. Forty pretreatment biopsy specimens from 43 patients in the preoperative chemoradiation therapy group were also analyzed. CD133-positive cases were more common in the preoperative chemoradiation therapy group than in the surgery-alone group. Furthermore, CD133-positive cases were more common in the preoperative chemoradiation therapy group than in pretreatment biopsy specimens. In the preoperative chemoradiation therapy group, the CD133-positive cases showed poorer prognosis than the CD133-negative cases. These studies suggest that the CD133+ population is important for

The biological function of CD133 remains unknown. It is a transmembrane pentaspan protein that was initially described as a surface antigen which was specific to human hematopoietic stem cells (Yin et al. 1997). Utilizing the literature mining software found in Ariadne Pathway Studio, Figure 5 illustrates the known relationships found between CD133

that not all CD133+ cells are able to reconstitute the original tumors.

outcome and that chemoradiation enriches this population.

original tumor (Chen et al. 2011).

and other genes and cellular processes. Cellular processes that are associated with CD133 include those that are expected to be involved with cancer stem cells: cell death, morphogenesis, apoptosis, cell differentiation, cell proliferation, and drug response (response to drug). Others may provide hints at important processes that have yet to be investigated: lipid metabolism, glucose metabolism, and vascularization. In addition, CD133 is linked with well-known cancer-related genes such as p53, Myc, Src, and transforming growth factor β1 and the TGFβ-associated SMAD6 and SMAD7. Expression of the gene HIF-1α in conjunction with CD133 is associated with tumor recurrence following chemoradiation (Saigusa et al. 2011). Also associated with recurrence of rectal cancer after chemoradiation is putative stem cell marker POU class 5 homeobox 1 (POU5F1). Expression of CD133, POU5F1, and the SOX2 gene following treatment was associated with poor disease-free survival.

The addition of other known cancer stem cell markers such as CD44 and ESA (EPCAM) to the diagram exemplifies the level of similarity in signaling. Figure 5 shows CD133 associated with 61 other genes and cellular processes. While ESA is associated with 10 of these entities, CD44 has been linked with almost half. CD44 has been associated with the inhibition of apoptosis, cell differentiation, and p53 signaling/expression. Both CD133 and CD44 are linked to CXCL12 (chemokine (C-X-C motif) ligand 12). CXCL12 and its receptor CXCR4 have previously been associated with the mobilization and homing of hematopoetic stem cells (Juarez and Bendall 2004). Interestingly, CXCR4 overlaps much of the CD133 signaling and has been implicated in cancer stem cell signaling at several other sites. Sustained CXCR4/CXCL12 signaling occurs in prostate cancer stem cells (Mimeault and Batra 2011) and is involved in the up-regulation of stem cell-related gene expression in breast cancer cells (Zhang et al. 2011). Additionally, CXCR4 was identified as a therapeutic target of glioblastoma stem cell-like cell lines (Schulte et al. 2011). This implies that, while many markers may be used to identify cancer stem cells, much of the signaling behind these superficial membrane markers may be quite similar.

#### **7. Stem cells and treatment response**

The mainline cancer therapies of conventional chemo- and radiotherapy target rapidly cycling cancer cells and can cause impressive, but usually temporary, clinical remissions. This initial remission followed by local recurrence would support the argument for the existence of a small subpopulation of resistant CSCs, while at the same time the majority of the non-CSCs being responsive to the treatment. Treatment failure could be explained by several CSC characteristics that would make them difficult to eradicate by conventional agents. First, they may be slow-cycling or quiescent rendering them less sensitive to agents that target actively cycling cells. Second, a characteristic of many normal stem cells is the increased activity of ABC transporter proteins as a protective mechanism against environmental toxins; these are also up-regulated in CSCs (Dean et al. 2005). Third, there are data suggesting that CSCs may be more resistant to radiation (Bao et al. 2006) although this has not been universally found (McCord et al. 2009).

CD133 has been extensively studied in the context of radiation sensitivity in the setting of glioma. An increase of the CD133+ fraction following irradiation of human glioma cells has been shown in vitro as well as in tumors in nude mice (Bao et al. 2006). The CD133+ cell fraction was found to have a reduced sensitivity to radiation-induced apoptosis. Interestingly, when CD133+ cells were irradiated with 3Gy, they were able to initiate tumors

The Stem Cell Environment: Kinetics, Signaling and Markers 315

with almost the same efficacy than the non-irradiated CD133+ cells. A clinical study in atypical teratoid/rhabdoid tumors demonstrated a correlation of the amount of immunohistochemically detected CD133+ cells with resistance to combined chemoradiation and decreased survival (Chiou et al. 2008). These data are supported by evaluations of glioma (Murat et al. 2008) and rectal cancer patients (Wang et al. 2009). All these studies show correlations between CD133 expression and efficacy of radiotherapy or combined treatment. Similarly, studies have also shown that CD133 positivity is associated with chemoresistance in glioma (Nakai et al. 2009), oral squamous cancer (Zhang et al.), and mesothelioma (Cortes-Dericks et al. 2010) amongst others. A large clinical study of 501 cases of human colorectal cancers showed that CD133-overexpressing tumors were more resistant to 5-FU–based chemotherapy and that CD133 expression was associated with poor prognosis (Ong et al. 2010). A recent study showed that treatment of human HT-29 colorectal cancer cells with high doses of 5-FU or oxaliplatin resulted in enrichment of CD133+ and CD44+ CSCs, which also exhibited decreased in vitro proliferation rate (Dallas et al. 2009). Interestingly in another colorectal cell line study, a recent publication has suggested that although CD133+ cells had higher *in vivo* tumor-forming ability than CD133− cells, it was the CD133− cells that were more resistant to 5-fluorouracil (FU) treatment

Although newer targeted agents such as cetuximab and bevacizumab are being tested in both frontline (Minsky et al. 2010) therapy and the metastatic setting (Cunningham et al. 2004; Hurwitz et al. 2004), they have modest effects on disease-free survival and overall survival. It would seem that the current combined modality therapies for rectal cancer will not be effective against CSCs no matter what combination is used. However, the most effective method to target CSCs has yet to be elucidated, but a number of possibilities exist including the administration of differentiating agents such as salinomycin (Gupta et al. 2009), targeting the specific signaling pathways of the CSCs (hedgehog, wnt, Notch) with drugs like cyclopamine (Merchant and Matsui 2010; Pannuti et al. 2010; Takahashi-Yanaga and Kahn 2010), targeting the microenvironmental niche of CSCs (LaBarge 2010), targeting the DNA checkpoint response (Frosina 2009), or using normal stem cells to home to the region of tumor (Hu et al. 2010). It is likely that future therapies will include inhibitors of survival pathways, along with immune cells, differentiation agents, and cytotoxic drugs, as

Immunological approaches have been demonstrated to be effective against CSCs in colorectal cancer. It has been shown that inhibiting the IL-4 signaling transduction pathway with an anti–IL-4 neutralizing antibody or an IL-4 receptor α antagonist sensitized CSCs to chemotherapeutics through down-regulation of anti-apoptotic proteins, such as cFLIP, BclxL, and PED (Todaro et al. 2007). The same group has also shown that incubation of colon CSCs with the bisphosphonate zoledronate induced an efficient γδ T-cell response. These immune cells have been shown to be effective at killing different tumor cells in vitro, but

Another important area of future investigation will be to determine the optimal timing of CSC-targeted therapies with other modalities, i.e. should it be co-administration of agents for newly diagnosed tumor or sequential scheduling after a remission to standard treatment has been obtained or at the time of progression after a standard treatment (Al-Hajj et al. 2004). A crucial element in optimizing timing will be the development of novel imaging probes to develop strategies for robust and efficient tracking and validation of CSCs and their niche under in vivo conditions. This will pave the way to better elucidate the

this was the first report of using γδ T-cell to target CSCs (Todaro et al. 2009).

(Hongo et al. 2011).

a combination.

Fig. 5. Genes and cellular processes associated with CD133 expression. The literature mining software in Ariadne Pathway Studio software identifies genes and cellular processes that have been shown to be associated with CD133. This figure was supplemented by adding the stem cell markers CD44, ESA (EPCAM), and CXCR4.

Fig. 5. Genes and cellular processes associated with CD133 expression. The literature mining software in Ariadne Pathway Studio software identifies genes and cellular processes that have been shown to be associated with CD133. This figure was supplemented by adding the

stem cell markers CD44, ESA (EPCAM), and CXCR4.

with almost the same efficacy than the non-irradiated CD133+ cells. A clinical study in atypical teratoid/rhabdoid tumors demonstrated a correlation of the amount of immunohistochemically detected CD133+ cells with resistance to combined chemoradiation and decreased survival (Chiou et al. 2008). These data are supported by evaluations of glioma (Murat et al. 2008) and rectal cancer patients (Wang et al. 2009). All these studies show correlations between CD133 expression and efficacy of radiotherapy or combined treatment. Similarly, studies have also shown that CD133 positivity is associated with chemoresistance in glioma (Nakai et al. 2009), oral squamous cancer (Zhang et al.), and mesothelioma (Cortes-Dericks et al. 2010) amongst others. A large clinical study of 501 cases of human colorectal cancers showed that CD133-overexpressing tumors were more resistant to 5-FU–based chemotherapy and that CD133 expression was associated with poor prognosis (Ong et al. 2010). A recent study showed that treatment of human HT-29 colorectal cancer cells with high doses of 5-FU or oxaliplatin resulted in enrichment of CD133+ and CD44+ CSCs, which also exhibited decreased in vitro proliferation rate (Dallas et al. 2009). Interestingly in another colorectal cell line study, a recent publication has suggested that although CD133+ cells had higher *in vivo* tumor-forming ability than CD133− cells, it was the CD133− cells that were more resistant to 5-fluorouracil (FU) treatment (Hongo et al. 2011).

Although newer targeted agents such as cetuximab and bevacizumab are being tested in both frontline (Minsky et al. 2010) therapy and the metastatic setting (Cunningham et al. 2004; Hurwitz et al. 2004), they have modest effects on disease-free survival and overall survival. It would seem that the current combined modality therapies for rectal cancer will not be effective against CSCs no matter what combination is used. However, the most effective method to target CSCs has yet to be elucidated, but a number of possibilities exist including the administration of differentiating agents such as salinomycin (Gupta et al. 2009), targeting the specific signaling pathways of the CSCs (hedgehog, wnt, Notch) with drugs like cyclopamine (Merchant and Matsui 2010; Pannuti et al. 2010; Takahashi-Yanaga and Kahn 2010), targeting the microenvironmental niche of CSCs (LaBarge 2010), targeting the DNA checkpoint response (Frosina 2009), or using normal stem cells to home to the region of tumor (Hu et al. 2010). It is likely that future therapies will include inhibitors of survival pathways, along with immune cells, differentiation agents, and cytotoxic drugs, as a combination.

Immunological approaches have been demonstrated to be effective against CSCs in colorectal cancer. It has been shown that inhibiting the IL-4 signaling transduction pathway with an anti–IL-4 neutralizing antibody or an IL-4 receptor α antagonist sensitized CSCs to chemotherapeutics through down-regulation of anti-apoptotic proteins, such as cFLIP, BclxL, and PED (Todaro et al. 2007). The same group has also shown that incubation of colon CSCs with the bisphosphonate zoledronate induced an efficient γδ T-cell response. These immune cells have been shown to be effective at killing different tumor cells in vitro, but this was the first report of using γδ T-cell to target CSCs (Todaro et al. 2009).

Another important area of future investigation will be to determine the optimal timing of CSC-targeted therapies with other modalities, i.e. should it be co-administration of agents for newly diagnosed tumor or sequential scheduling after a remission to standard treatment has been obtained or at the time of progression after a standard treatment (Al-Hajj et al. 2004). A crucial element in optimizing timing will be the development of novel imaging probes to develop strategies for robust and efficient tracking and validation of CSCs and their niche under in vivo conditions. This will pave the way to better elucidate the

The Stem Cell Environment: Kinetics, Signaling and Markers 317

Bao, S., Q. Wu, R. E. McLendon, Y. Hao, Q. Shi, A. B. Hjelmeland, M. W. Dewhirst, D. D.

Barker, N., J. H. van Es, J. Kuipers, P. Kujala, M. van den Born, M. Cozijnsen, A. Haegebarth,

Bertolini, G., L. Roz, P. Perego, M. Tortoreto, E. Fontanella, L. Gatti, G. Pratesi, A. Fabbri, F.

Boman, B. M., J. Z. Fields, O. Bonham-Carter and O. A. Runquist (2001). Computer

Boman, B. M. and E. Huang (2008). Human colon cancer stem cells: a new paradigm in

Bonnet, D. and J. E. Dick (1997). Human acute myeloid leukemia is organized as a hierarchy

Brabletz, S., O. Schmalhofer and T. Brabletz (2009). Gastrointestinal stem cells in

Bray, S. J. (2006). Notch signalling: a simple pathway becomes complex. *Nature Reviews* 

Chen, S. Y., Y. C. Huang, S. P. Liu, F. J. Tsai, W. C. Shyu and S. Z. Lin (2011). An overview of

Chiou, S. H., C. L. Kao, Y. W. Chen, C. S. Chien, S. C. Hung, J. F. Lo, Y. J. Chen, H. H. Ku, M.

Clay, M. R., M. Tabor, J. H. Owen, T. E. Carey, C. R. Bradford, G. T. Wolf, M. S. Wicha and

Clevers, H. The cancer stem cell: premises, promises and challenges (2011). *Nature Medicine*,

Cortes-Dericks, L., G. L. Carboni, R. A. Schmid and G. Karoubi (2010). Putative cancer stem

*Molecular Cell Biology*, Vol.7, No.(9), pp. 678-689, ISSN 1471-0072

Vol.50, No.(23), pp. 7717-7722, ISSN 0008-5472

pp. 756-760, ISSN 1476-4687

1003-1007, ISSN 1476-4687

2838, ISSN 0732-183X

0022-3417

ISSN 0963-6897

6203

pp. 730-737, ISSN 1078-8956

No.(9), pp. 1195-1201, ISSN 1043-3074

Vol.17, No.(3), pp. 313-319, ISSN 1078-8956

No.(38), pp. 16281-16286, ISSN 0027-8424

*Research*, Vol.61, No.(23), pp. 8408-8411, ISSN 0008-5472

17p allelic deletions as late events in colorectal tumorigenesis. *Cancer Research*,

Bigner and J. N. Rich (2006). Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. *Nature*, Vol.444, No.(7120),

J. Korving, H. Begthel, P. J. Peters and H. Clevers (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. *Nature*, Vol.449, No.(7165), pp.

Andriani, S. Tinelli, E. Roz, R. Caserini, S. Lo Vullo, T. Camerini, L. Mariani, D. Delia, E. Calabro, U. Pastorino and G. Sozzi (2009). Highly tumorigenic lung cancer CD133+ cells display stem-like features and are spared by cisplatin treatment. *Proceedings of the National Academy of Sciences of the United States of America*, Vol.106,

modeling implicates stem cell overproduction in colon cancer initiation. *Cancer* 

gastrointestinal oncology. *Journal of Clinical Oncology*, Vol.26, No.(17), pp. 2828-

that originates from a primitive hematopoietic cell. *Nature Medicine*, Vol.3, No.(7),

development and cancer. *Journal of Pathology*, Vol.217, No.(2), pp. 307-317, ISSN

concepts for cancer stem cells. *Cell Transplantation*, Vol.20, No.(1), pp. 113-120, 1555-

T. Hsu and T. T. Wong (2008). Identification of CD133-positive radioresistant cells in atypical teratoid/rhabdoid tumor. *PLoS One*, Vol.3, No.(5), pp. e2090, ISSN 1932-

M. E. Prince (2010). Single-marker identification of head and neck squamous cell carcinoma cancer stem cells with aldehyde dehydrogenase. *Head and Neck*, Vol.32,

cells in malignant pleural mesothelioma show resistance to cisplatin and

underlying regulatory mechanisms of CSC and develop platforms for targeted theragnostics.

#### **8. Conclusions**

The CSC model, if generally correct, has important implications for the current paradigm of treatment for rectal cancer. The practice of small, successive improvements in survival by the refinement of current schedules and addition of newer agents is unlikely to result in significant advances in treatment outcomes. Gene expression analyses of CSCs populations have the possibility to identify novel diagnostic markers and novel therapeutic targets. A recent study identified EGR1 to the be the most highly expressed gene in CD133 positive colorectal cancer cells (Ernst et al. 2011). EGR1 is known to regulate Wnt through upregulation of TCF4, which induces stem cell marker LGR5. Previous studies identified 100 candidate-genes, which were differentially expressed in the CD133 positive fraction (Regenbrecht et al. 2008) of which 10 genes were shown to be differentially regulated between the different studies. However, 9 of these 10 genes were shown to form an interactive network with each other and that these genes were positioned at the interface between proliferative pathways (JAK/STAT) and differentiating pathways (HOX, PBX, MEIS, GATA2). Of importance in this cascade was the gene KIT which encodes the receptor for stem cell factor (SCF). It is clear from Figure 5 that CD133 is involved in many signaling pathways and identifying the candidate pathways for future drug development will be challenging. Equally challenging will be targeting pathways of stem cell self renewal without affecting this crucial process in normal stem cells. Thus, the elucidation of the mechanisms regulating the survival, self-renewal, and differentiation of normal and CSCs could potentially lead to significant advances in the treatment of neoplastic diseases including rectal cancer.

#### **9. References**


underlying regulatory mechanisms of CSC and develop platforms for targeted

The CSC model, if generally correct, has important implications for the current paradigm of treatment for rectal cancer. The practice of small, successive improvements in survival by the refinement of current schedules and addition of newer agents is unlikely to result in significant advances in treatment outcomes. Gene expression analyses of CSCs populations have the possibility to identify novel diagnostic markers and novel therapeutic targets. A recent study identified EGR1 to the be the most highly expressed gene in CD133 positive colorectal cancer cells (Ernst et al. 2011). EGR1 is known to regulate Wnt through upregulation of TCF4, which induces stem cell marker LGR5. Previous studies identified 100 candidate-genes, which were differentially expressed in the CD133 positive fraction (Regenbrecht et al. 2008) of which 10 genes were shown to be differentially regulated between the different studies. However, 9 of these 10 genes were shown to form an interactive network with each other and that these genes were positioned at the interface between proliferative pathways (JAK/STAT) and differentiating pathways (HOX, PBX, MEIS, GATA2). Of importance in this cascade was the gene KIT which encodes the receptor for stem cell factor (SCF). It is clear from Figure 5 that CD133 is involved in many signaling pathways and identifying the candidate pathways for future drug development will be challenging. Equally challenging will be targeting pathways of stem cell self renewal without affecting this crucial process in normal stem cells. Thus, the elucidation of the mechanisms regulating the survival, self-renewal, and differentiation of normal and CSCs could potentially lead to significant advances in the treatment of neoplastic diseases

Al-Hajj, M., M. W. Becker, M. Wicha, I. Weissman and M. F. Clarke (2004). Therapeutic

Al-Hajj, M., M. S. Wicha, A. Benito-Hernandez, S. J. Morrison and M. F. Clarke (2003).

Auclair, B. A., Y. D. Benoit, N. Rivard, Y. Mishina and N. Perreault (2007). Bone

Axelrod, R., D. E. Axelrod and K. J. Pienta (2006). Evolution of cooperation among tumor

Baker, S. J., A. C. Preisinger, J. M. Jessup, C. Paraskeva, S. Markowitz, J. K. Willson, S.

Vol.14, No.(1), pp. 43-47, ISSN 0959-437X

Vol.103, No.(36), pp. 13474-13479, ISSN 0027-8424

3988, ISSN 0027-8424

0016-5085

implications of cancer stem cells. *Current Opinion in Genetics and Development*,

Prospective identification of tumorigenic breast cancer cells. *Proceedings of the National Academy of Sciences of the United States of America*, Vol.100, No.(7), pp. 3983-

morphogenetic protein signaling is essential for terminal differentiation of the intestinal secretory cell lineage. *Gastroenterology*, Vol.133, No.(3), pp. 887-896, ISSN

cells. *Proceedings of the National Academy of Sciences of the United States of America*,

Hamilton and B. Vogelstein (1990). p53 gene mutations occur in combination with

theragnostics.

**8. Conclusions** 

including rectal cancer.

**9. References** 

17p allelic deletions as late events in colorectal tumorigenesis. *Cancer Research*, Vol.50, No.(23), pp. 7717-7722, ISSN 0008-5472


The Stem Cell Environment: Kinetics, Signaling and Markers 319

Gazzaniga, P., E. Cigna, V. Panasiti, V. Devirgiliis, U. Bottoni, B. Vincenzi, C. Nicolazzo, A.

Glimelius, B. and J. Oliveira (2009). Rectal cancer: ESMO clinical recommendations for

Gupta, P. B., T. T. Onder, G. Jiang, K. Tao, C. Kuperwasser, R. A. Weinberg and E. S. Lander

Harper, L. J., K. Piper, J. Common, F. Fortune and I. C. Mackenzie (2007). Stem cell patterns

He, X. C., J. Zhang, W. G. Tong, O. Tawfik, J. Ross, D. H. Scoville, Q. Tian, X. Zeng, X. He, L.

Hongo, K., J. Tanaka, N. H. Tsuno, K. Kawai, T. Nishikawa, Y. Shuno, K. Sasaki, M. Kaneko,

Hosonuma, S., Y. Kobayashi, S. Kojo, H. Wada, K. Seino, K. Kiguchi and B. Ishizuka (2011).

Hu, Y. L., Y. H. Fu, Y. Tabata and J. Q. Gao (2010). Mesenchymal stem cells: a promising

Huang, E. H., M. J. Hynes, T. Zhang, C. Ginestier, G. Dontu, H. Appelman, J. Z. Fields, M. S.

Humphries, A. and N. A. Wright (2008). Colonic crypt organization and tumorigenesis.

Hurwitz, H., L. Fehrenbacher, W. Novotny, T. Cartwright, J. Hainsworth, W. Heim, J. Berlin,

Jensen, J., E. E. Pedersen, P. Galante, J. Hald, R. S. Heller, M. Ishibashi, R. Kageyama, F.

*Nature Reviews Cancer*, Vol.8, No.(6), pp. 415-424, ISSN 1474-175X

*Pathology and Medicine*, Vol.36, No.(10), pp. 594-603, ISSN 0904-2512

screening. *Cell*, Vol.138, No.(4), pp. 645-659, ISSN 0092-8674

*Genetics*, Vol.36, No.(10), pp. 1117-1121, ISSN 1061-4036

Signaling. *Journal of Surgical Research*, ISSN 0022-4804

No.(1), pp. 9-12, ISSN 0914-7470

2335-2342, ISSN 0028-4793

Vol.147, No.(2), pp. 154-162, ISSN 0168-3659

Vol.36, No.(12), pp. 1211-1214, ISSN 0748-7983

ISSN 0923-7534

Petracca and A. Gradilone (2010). CD133 and ABCB5 as stem cell markers on sentinel lymph node from melanoma patients. *European Journal of Surgical Oncology*,

diagnosis, treatment and follow-up. *Annals of Oncology*, Vol.20 Suppl 4, pp. 54-56,

(2009). Identification of selective inhibitors of cancer stem cells by high-throughput

in cell lines derived from head and neck squamous cell carcinoma. *Journal of Oral* 

M. Wiedemann, Y. Mishina and L. Li (2004). BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-beta-catenin signaling. *Nature* 

M. Hiyoshi, E. Sunami, J. Kitayama, K. Takahashi and H. Nagawa (2011). CD133(-) Cells, Derived From a Single Human Colon Cancer Cell Line, Are More Resistant to 5-Fluorouracil (FU) Than CD133(+) Cells, Dependent on the beta1-Integrin

Clinical significance of side population in ovarian cancer cells. *Human Cell*, Vol.24,

targeted-delivery vehicle in cancer gene therapy. *Journal of Controlled Release*,

Wicha and B. M. Boman (2009). Aldehyde dehydrogenase 1 is a marker for normal and malignant human colonic stem cells (SC) and tracks SC overpopulation during colon tumorigenesis. *Cancer Research*, Vol.69, No.(8), pp. 3382-3389, ISSN 0008-5472

A. Baron, S. Griffing, E. Holmgren, N. Ferrara, G. Fyfe, B. Rogers, R. Ross and F. Kabbinavar (2004). Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. *New England Journal of Medicine*, Vol.350, No.(23), pp.

Guillemot, P. Serup and O. D. Madsen (2000). Control of endodermal endocrine development by Hes-1. *Nature Genetics*, Vol.24, No.(1), pp. 36-44, ISSN 1061-4036 Jones, S., W. D. Chen, G. Parmigiani, F. Diehl, N. Beerenwinkel, T. Antal, A. Traulsen, M. A.

Nowak, C. Siegel, V. E. Velculescu, K. W. Kinzler, B. Vogelstein, J. Willis and S. D. Markowitz (2008). Comparative lesion sequencing provides insights into tumor

pemetrexed. *International Journal of Oncology*, Vol.37, No.(2), pp. 437-444, ISSN 1019- 6439


Cunningham, D., Y. Humblet, S. Siena, D. Khayat, H. Bleiberg, A. Santoro, D. Bets, M.

Dalerba, P., S. J. Dylla, I. K. Park, R. Liu, X. Wang, R. W. Cho, T. Hoey, A. Gurney, E. H.

Dallas, N. A., L. Xia, F. Fan, M. J. Gray, P. Gaur, G. van Buren, 2nd, S. Samuel, M. P. Kim, S.

inhibition. *Cancer Research*, Vol.69, No.(5), pp. 1951-1957, ISSN 0008-5472 Dean, M., T. Fojo and S. Bates (2005). Tumour stem cells and drug resistance. *Nature Reviews* 

Dontu, G., S. Liu and M. S. Wicha (2005). Stem cells in mammary development and

Enblad, P., H. O. Adami, R. Bergstrom, B. Glimelius, U. Krusemo and L. Pahlman (1988).

Fearon, E. R. and B. Vogelstein (1990). A genetic model for colorectal tumorigenesis. *Cell*,

Fernebro, E., P. O. Bendahl, M. Dictor, A. Persson, M. Ferno and M. Nilbert (2004).

Fevr, T., S. Robine, D. Louvard and J. Huelsken (2007). Wnt/beta-catenin is essential for

Frattini, M., D. Balestra, S. Suardi, M. Oggionni, P. Alberici, P. Radice, A. Costa, M. G.

Frosina, G. (2009). DNA repair and resistance of gliomas to chemotherapy and radiotherapy. *Molecular Cancer Research*, Vol.7, No.(7), pp. 989-999, ISSN 1541-7786

*Cellular Biology*, Vol.27, No.(21), pp. 7551-7559, ISSN 0270-7306

Vol.10, No.(12 Pt 1), pp. 4015-4021, ISSN 1078-0432

*National Cancer Institute*, Vol.80, No.(8), pp. 586-591, ISSN 0027-8874 Ernst, A., M. Aigner, S. Nakata, F. Engel, M. Schlotter, M. Kloor, K. Brand, S. Schmitt, G.

6439

0028-4793

No.(24), pp. 10158-10163, ISSN 0027-8424

No.(3), pp. 207-213, ISSN 1550-8943

Vol.61, No.(5), pp. 759-767, ISSN 0092-8674

227, ISSN 0031-3025

ISSN 0020-7136

*Cancer*, Vol.5, No.(4), pp. 275-284, ISSN 1474-175X

pemetrexed. *International Journal of Oncology*, Vol.37, No.(2), pp. 437-444, ISSN 1019-

Mueser, A. Harstrick, C. Verslype, I. Chau and E. Van Cutsem (2004). Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. *New England Journal of Medicine*, Vol.351, No.(4), pp. 337-345, ISSN

Huang, D. M. Simeone, A. A. Shelton, G. Parmiani, C. Castelli and M. F. Clarke (2007). Phenotypic characterization of human colorectal cancer stem cells. *Proceedings of the National Academy of Sciences of the United States of America*, Vol.104,

J. Lim and L. M. Ellis (2009). Chemoresistant colorectal cancer cells, the cancer stem cell phenotype, and increased sensitivity to insulin-like growth factor-I receptor

carcinogenesis: implications for prevention and treatment. *Stem Cell Reviews*, Vol.1,

Improved survival of patients with cancers of the colon and rectum? *Journal of the* 

Steinert, N. Rahbari, M. Koch, B. Radlwimmer, J. Weitz and P. Lichter (2011). A gene signature distinguishing CD133hi from CD133- colorectal cancer cells: essential role for EGR1 and downstream factors. *Pathology*, Vol.43, No.(3), pp. 220-

Immunohistochemical patterns in rectal cancer: application of tissue microarray with prognostic correlations. *International Journal of Cancer*, Vol.111, No.(6), pp. 921-

intestinal homeostasis and maintenance of intestinal stem cells. *Molecular and* 

Daidone, E. Leo, S. Pilotti, L. Bertario and M. A. Pierotti (2004). Different genetic features associated with colon and rectal carcinogenesis. *Clinical Cancer Research*,


The Stem Cell Environment: Kinetics, Signaling and Markers 321

Li, L. and T. Xie (2005). Stem cell niche: structure and function. *Annual Review of Cell and* 

Liang, D. and Y. Shi (2011). Aldehyde dehydrogenase-1 is a specific marker for stem cells in

Madison, B. B., K. Braunstein, E. Kuizon, K. Portman, X. T. Qiao and D. L. Gumucio (2005).

Marcato, P., C. A. Dean, D. Pan, R. Araslanova, M. Gillis, M. Joshi, L. Helyer, L. Pan, A.

McCord, A. M., M. Jamal, E. S. Williams, K. Camphausen and P. J. Tofilon (2009). CD133+

Merchant, A. A. and W. Matsui (2010). Targeting Hedgehog--a cancer stem cell pathway. *Clinical Cancer Research*, Vol.16, No.(12), pp. 3130-3140, ISSN 1078-0432 Miki, J., B. Furusato, H. Li, Y. Gu, H. Takahashi, S. Egawa, I. A. Sesterhenn, D. G. McLeod, S.

Mills, J. C. and J. I. Gordon (2001). The intestinal stem cell niche: there grows the

Mimeault, M. and S. K. Batra (2011). Frequent Gene Products and Molecular Pathways

Murat, A., E. Migliavacca, T. Gorlia, W. L. Lambiv, T. Shay, M. F. Hamou, N. de Tribolet, L.

Nagata, T., C. Sakakura, S. Komiyama, A. Miyashita, M. Nishio, Y. Murayama, S. Komatsu,

cancer. *Anticancer Research*, Vol.31, No.(2), pp. 495-500, ISSN 0250-7005 Nakai, E., K. Park, T. Yawata, T. Chihara, A. Kumazawa, H. Nakabayashi and K. Shimizu

*Cancer Research*, Vol.67, No.(7), pp. 3153-3161, ISSN 0008-5472

cancer. *Cancer Journal*, Vol.16, No.(3), pp. 253-261, ISSN 1528-9117

*America*, Vol.98, No.(22), pp. 12334-12336, ISSN 0027-8424

Vol.26, No.(18), pp. 3015-3024, ISSN 0732-183X

Epithelial hedgehog signals pattern the intestinal crypt-villus axis. *Development*,

Leidal, S. Gujar, C. A. Giacomantonio and P. W. Lee (2011). Aldehyde dehydrogenase activity of breast cancer stem cells is primarily due to isoform ALDH1A3 and its expression is predictive of metastasis. *Stem Cells*, Vol.29, No.(1),

glioblastoma stem-like cells are radiosensitive with a defective DNA damage response compared with established cell lines. *Clinical Cancer Research*, Vol.15,

Srivastava and J. S. Rhim (2007). Identification of putative stem cell markers, CD133 and CXCR4, in hTERT-immortalized primary nonmalignant and malignant tumorderived human prostate epithelial cell lines and in prostate cancer specimens.

neighborhood. *Proceedings of the National Academy of Sciences of the United States of* 

Altered in Prostate Cancer- and Metastasis-Initiating Cells and their Progenies and Novel Promising Multitargeted Therapies. *Molecular Medicine*, ISSN 1076-1551 Minsky, B. D., C. Roedel and V. Valentini (2010). Combined modality therapy for rectal

Regli, W. Wick, M. C. Kouwenhoven, J. A. Hainfellner, F. L. Heppner, P. Y. Dietrich, Y. Zimmer, J. G. Cairncross, R. C. Janzer, E. Domany, M. Delorenzi, R. Stupp and M. E. Hegi (2008). Stem cell-related "self-renewal" signature and high epidermal growth factor receptor expression associated with resistance to concomitant chemoradiotherapy in glioblastoma. *Journal of Clinical Oncology*,

A. Shiozaki, Y. Kuriu, H. Ikoma, M. Nakanishi, D. Ichikawa, H. Fujiwara, K. Okamoto, T. Ochiai, Y. Kokuba, T. Sonoyama and E. Otsuji (2011). Expression of cancer stem cell markers CD133 and CD44 in locoregional recurrence of rectal

(2009). Enhanced MDR1 expression and chemoresistance of cancer stem cells

*Developmental Biology*, Vol.21, pp. 605-631, ISSN 1081-0706

Vol.132, No.(2), pp. 279-289, ISSN 0950-1991

No.(16), pp. 5145-5153, ISSN 1078-0432

pp. 32-45, ISSN 1066-5099

human lung adenocarcinoma. *Medical Oncology*, ISSN 1357-0560

evolution. *Proceedings of the National Academy of Sciences of the United States of America*, Vol.105, No.(11), pp. 4283-4288, ISSN 0027-8424


Juarez, J. and L. Bendall (2004). SDF-1 and CXCR4 in normal and malignant hematopoiesis. *Histology and Histopathology*, Vol.19, No.(1), pp. 299-309, ISSN 0213-3911 Kalabis, J., K. Oyama, T. Okawa, H. Nakagawa, C. Z. Michaylira, D. B. Stairs, J. L.

Kalady, M. F., K. Dejulius, J. M. Church, I. C. Lavery, V. W. Fazio and H. Ishwaran (2010).

*American College of Surgeons*, Vol.211, No.(2), pp. 187-195, ISSN 1072-7515 Kapiteijn, E., G. J. Liefers, L. C. Los, E. K. Kranenbarg, J. Hermans, R. A. Tollenaar, Y.

Kim, M. P., J. B. Fleming, H. Wang, J. L. Abbruzzese, W. Choi, S. Kopetz, D. J. McConkey, D.

Korinek, V., N. Barker, P. Moerer, E. van Donselaar, G. Huls, P. J. Peters and H. Clevers

lacking Tcf-4. *Nature Genetics*, Vol.19, No.(4), pp. 379-383, ISSN 1061-4036 Korinek, V., N. Barker, P. J. Morin, D. van Wichen, R. de Weger, K. W. Kinzler, B. Vogelstein

Kosinski, C., V. S. Li, A. S. Chan, J. Zhang, C. Ho, W. Y. Tsui, T. L. Chan, R. C. Mifflin, D. W.

LaBarge, M. A. (2010). The difficulty of targeting cancer stem cell niches. *Clinical Cancer* 

Lapidot, T., C. Sirard, J. Vormoor, B. Murdoch, T. Hoang, J. Caceres-Cortes, M. Minden, B.

Lee, C. J., J. Dosch and D. M. Simeone (2008). Pancreatic cancer stem cells. *Journal of Clinical* 

Li, F. Y. and M. D. Lai (2009). Colorectal cancer, one entity or three. *Journal of Zhejiang* 

*America*, Vol.104, No.(39), pp. 15418-15423, ISSN 0027-8424

*Research*, Vol.16, No.(12), pp. 3121-3129, ISSN 1078-0432

*Oncology*, Vol.26, No.(17), pp. 2806-2812, ISSN 0732-183X

*University Science B*, Vol.10, No.(3), pp. 219-229, ISSN 1673-1581

*America*, Vol.105, No.(11), pp. 4283-4288, ISSN 0027-8424

Vol.118, No.(12), pp. 3860-3869, ISSN 0021-9738

Vol.101, No.(4), pp. 906-912, ISSN 1347-9032

171-178, ISSN 0022-3417

ISSN 0036-8075

pp. 645-648, ISSN 0028-0836

evolution. *Proceedings of the National Academy of Sciences of the United States of* 

Figueiredo, U. Mahmood, J. A. Diehl, M. Herlyn and A. K. Rustgi (2008). A subpopulation of mouse esophageal basal cells has properties of stem cells with the capacity for self-renewal and lineage specification. *Journal of Clinical Investigation*,

Gene signature is associated with early stage rectal cancer recurrence. *Journal of the* 

Moriya, C. J. van de Velde and J. H. van Krieken (2001). Mechanisms of oncogenesis in colon versus rectal cancer. *Journal of Pathology*, Vol.195, No.(2), pp.

B. Evans and G. E. Gallick (2011). ALDH Activity Selectively Defines an Enhanced Tumor-Initiating Cell Population Relative to CD133 Expression in Human Pancreatic Adenocarcinoma. *PLoS One*, Vol.6, No.(6), pp. e20636, ISSN 1932-6203 Kojima, M., G. Ishii, N. Atsumi, Y. Nishizawa, N. Saito and A. Ochiai (2010). CD133

expression in rectal cancer after preoperative chemoradiotherapy. *Cancer Science*,

(1998). Depletion of epithelial stem-cell compartments in the small intestine of mice

and H. Clevers (1997). Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma. *Science*, Vol.275, No.(5307), pp. 1784-1787,

Powell, S. T. Yuen, S. Y. Leung and X. Chen (2007). Gene expression patterns of human colon tops and basal crypts and BMP antagonists as intestinal stem cell niche factors. *Proceedings of the National Academy of Sciences of the United States of* 

Paterson, M. A. Caligiuri and J. E. Dick (1994). A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. *Nature*, Vol.367, No.(6464),


The Stem Cell Environment: Kinetics, Signaling and Markers 323

Potten, C. S., C. Booth and D. M. Pritchard (1997). The intestinal epithelial stem cell: the

Potten, C. S., C. Booth, G. L. Tudor, D. Booth, G. Brady, P. Hurley, G. Ashton, R. Clarke, S.

Potten, C. S., M. Kellett, S. A. Roberts, D. A. Rew and G. D. Wilson (1992). Measurement of

Potten, C. S. and M. Loeffler (1990). Stem cells: attributes, cycles, spirals, pitfalls and

Powell, S. M., N. Zilz, Y. Beazer-Barclay, T. M. Bryan, S. R. Hamilton, S. N. Thibodeau, B.

Ricci-Vitiani, L., R. Pallini, M. Biffoni, M. Todaro, G. Invernici, T. Cenci, G. Maira, E. A.

Rimkus, C., J. Friederichs, A. L. Boulesteix, J. Theisen, J. Mages, K. Becker, H. Nekarda, R.

Saigusa, S., K. Tanaka, Y. Toiyama, T. Yokoe, Y. Okugawa, Y. Koike, H. Fujikawa, Y. Inoue,

Sangiorgi, E. and M. R. Capecchi (2009). Bmi1 lineage tracing identifies a self-renewing

Sato, T., J. H. van Es, H. J. Snippert, D. E. Stange, R. G. Vries, M. van den Born, N. Barker, N.

Sayed, M. G., A. F. Ahmed, J. R. Ringold, M. E. Anderson, J. L. Bair, F. A. Mitros, H. T.

*America*, Vol.106, No.(17), pp. 7101-7106, ISSN 0027-8424

tumorigenesis. *Nature*, Vol.359, No.(6392), pp. 235-237, ISSN 0028-0836 Regenbrecht, C. R., H. Lehrach and J. Adjaye (2008). Stemming cancer: functional genomics

219-243, ISSN 0959-9673

1001-1020, ISSN 0950-1991

ISSN 1550-8943

53-61, ISSN 1542-3565

ISSN 0028-0836

pp. 323-332, ISSN 0936-6555

Vol.33, No.(1), pp. 71-78, ISSN 0017-5749

No.(7325), pp. 824-828, ISSN 0028-0836

0301-4681

mucosal governor. *International Journal of Experimental Pathology*, Vol.78, No.(4), pp.

Sakakibara and H. Okano (2003). Identification of a putative intestinal stem cell and early lineage marker; musashi-1. *Differentiation*, Vol.71, No.(1), pp. 28-41, ISSN

in vivo proliferation in human colorectal mucosa using bromodeoxyuridine. *Gut*,

uncertainties. Lessons for and from the crypt. *Development*, Vol.110, No.(4), pp.

Vogelstein and K. W. Kinzler (1992). APC mutations occur early during colorectal

of cancer stem cells in solid tumors. *Stem Cell Reviews*, Vol.4, No.(4), pp. 319-328,

Parati, G. Stassi, L. M. Larocca and R. De Maria (2010). Tumour vascularization via endothelial differentiation of glioblastoma stem-like cells. *Nature*, Vol.468,

Rosenberg, K. P. Janssen and J. R. Siewert (2008). Microarray-based prediction of tumor response to neoadjuvant radiochemotherapy of patients with locally advanced rectal cancer. *Clinical Gastroenterology and Hepatology*, Vol.6, No.(1), pp.

C. Miki and M. Kusunoki (2011). Clinical significance of CD133 and hypoxia inducible factor-1alpha gene expression in rectal cancer after preoperative chemoradiotherapy. *Clinical Oncology (Royal College of Radiologists)*, Vol.23, No.(5),

pancreatic acinar cell subpopulation capable of maintaining pancreatic organ homeostasis. *Proceedings of the National Academy of Sciences of the United States of* 

F. Shroyer, M. van de Wetering and H. Clevers (2011). Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts. *Nature*, Vol.469, No.(7330), pp. 415-418,

Lynch, S. T. Tinley, G. M. Petersen, F. M. Giardiello, B. Vogelstein and J. R. Howe (2002). Germline SMAD4 or BMPR1A mutations and phenotype of juvenile polyposis. *Annals of Surgical Oncology*, Vol.9, No.(9), pp. 901-906, ISSN 1068-9265

derived from glioblastoma. *Cancer Investigation*, Vol.27, No.(9), pp. 901-908, ISSN 0735-7907


Nowell, P. C. (1976). The clonal evolution of tumor cell populations. *Science*, Vol.194,

O'Brien, C. A., A. Kreso and C. H. Jamieson (2010). Cancer stem cells and self-renewal. *Clinical Cancer Research*, Vol.16, No.(12), pp. 3113-3120, ISSN 1078-0432 O'Brien, C. A., A. Pollett, S. Gallinger and J. E. Dick (2007). A human colon cancer cell

Ojima, E., Y. Inoue, C. Miki, M. Mori and M. Kusunoki (2007). Effectiveness of gene

Ong, C. W., L. G. Kim, H. H. Kong, L. Y. Low, B. Iacopetta, R. Soong and M. Salto-Tellez

Pannuti, A., K. Foreman, P. Rizzo, C. Osipo, T. Golde, B. Osborne and L. Miele (2010).

Parkin, D. M., F. Bray, J. Ferlay and P. Pisani (2005). Global cancer statistics, 2002. *CA: A Cancer Journal for Clinicians*, Vol.55, No.(2), pp. 74-108, ISSN 0007-9235 Persad, S., A. A. Troussard, T. R. McPhee, D. J. Mulholland and S. Dedhar (2001). Tumor

Pierce, G. B. and W. C. Speers (1988). Tumors as caricatures of the process of tissue renewal:

Ponti, D., A. Costa, N. Zaffaroni, G. Pratesi, G. Petrangolini, D. Coradini, S. Pilotti, M. A.

Popat, S., D. Zhao, Z. Chen, H. Pan, Y. Shao, I. Chandler and R. S. Houlston (2007).

Potten, C. S. (1998). Stem cells in gastrointestinal epithelium: numbers, characteristics and

cancer. *Modern Pathology*, Vol.23, No.(3), pp. 450-457, ISSN 0893-3952 Pages, F., A. Kirilovsky, B. Mlecnik, M. Asslaber, M. Tosolini, G. Bindea, C. Lagorce, P.

0735-7907

No.(4260), pp. 23-28, ISSN 0036-8075

No.(7123), pp. 106-110, ISSN 0028-0836

No.(35), pp. 5944-5951, ISSN 0732-183X

*Biology*, Vol.153, No.(6), pp. 1161-1174, ISSN 0021-9525

*Research*, Vol.65, No.(13), pp. 5506-5511, ISSN 0008-5472

*Sciences*, Vol.353, No.(1370), pp. 821-830, ISSN 0962-8436

pp. 3141-3152, ISSN 1078-0432

pp. 1996-2004, ISSN 0008-5472

pp. 627-633, ISSN 0250-7005

derived from glioblastoma. *Cancer Investigation*, Vol.27, No.(9), pp. 901-908, ISSN

capable of initiating tumour growth in immunodeficient mice. *Nature*, Vol.445,

expression profiling for response prediction of rectal cancer to preoperative radiotherapy. *Journal of Gastroenterology*, Vol.42, No.(9), pp. 730-736, ISSN 0944-1174

(2010). CD133 expression predicts for non-response to chemotherapy in colorectal

Wind, F. Marliot, P. Bruneval, K. Zatloukal, Z. Trajanoski, A. Berger, W. H. Fridman and J. Galon (2009). In situ cytotoxic and memory T cells predict outcome in patients with early-stage colorectal cancer. *Journal of Clinical Oncology*, Vol.27,

Targeting Notch to target cancer stem cells. *Clinical Cancer Research*, Vol.16, No.(12),

suppressor PTEN inhibits nuclear accumulation of beta-catenin and T cell/lymphoid enhancer factor 1-mediated transcriptional activation. *Journal of Cell* 

prospects for therapy by directing differentiation. *Cancer Research*, Vol.48, No.(8),

Pierotti and M. G. Daidone (2005). Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. *Cancer* 

Relationship between chromosome 18q status and colorectal cancer prognosis: a prospective, blinded analysis of 280 patients. *Anticancer Research*, Vol.27, No.(1B),

death. *Philosophical Transactions of the Royal Society of London. Series B: Biological* 


The Stem Cell Environment: Kinetics, Signaling and Markers 325

Vo, D. M., L. A. Julien and A. G. Thorson (2010). Current controversies in colon and rectal cancer. *Minerva Chirurgica*, Vol.65, No.(6), pp. 677-693, ISSN 0026-4733 Vogelstein, B., E. R. Fearon, S. R. Hamilton, S. E. Kern, A. C. Preisinger, M. Leppert, Y.

von Bubnoff, A. and K. W. Cho (2001). Intracellular BMP signaling regulation in vertebrates:

Wang, Q., Z. G. Chen, C. Z. Du, H. W. Wang, L. Yan and J. Gu (2009). Cancer stem cell

Wang, X., C. Venugopal, B. Manoranjan, N. McFarlane, E. O'Farrell, S. Nolte, T. Gunnarsson,

Wasan, H. S., H. S. Park, K. C. Liu, N. K. Mandir, A. Winnett, P. Sasieni, W. F. Bodmer, R. A.

Watanabe, T., Y. Komuro, T. Kiyomatsu, T. Kanazawa, Y. Kazama, J. Tanaka, T. Tanaka, Y.

Wicha, M. S., S. Liu and G. Dontu (2006). Cancer stem cells: an old idea--a paradigm shift.

Yang, C. Y., C. H. Lin and J. K. Jiang (2010). Early relapse in a rectal cancer patient: possible

Yang, J. P., Y. Liu, W. Zhong, D. Yu, L. J. Wen and C. S. Jin (2011). Chemoresistance of

Yang, Q., N. A. Bermingham, M. J. Finegold and H. Y. Zoghbi (2001). Requirement of Math1

Yasuda, H., K. Tanaka, S. Saigusa, Y. Toiyama, Y. Koike, Y. Okugawa, T. Yokoe, A.

Yin, A. H., S. Miraglia, E. D. Zanjani, G. Almeida-Porada, M. Ogawa, A. G. Leary, J. Olweus,

*Journal of Pathology*, Vol.185, No.(3), pp. 246-255, ISSN 0022-3417

525-532, ISSN 0028-4793

Vol.55, No.(3), pp. 284-293, ISSN 0309-0167

initiating cells. *Oncogene*, ISSN 0950-9232

No.(2), pp. 196-197, ISSN 0022-4790

Vol.124, No.(7), pp. 1055-1060, ISSN 0366-6999

No.(5549), pp. 2155-2158, ISSN 0036-8075

1606

ISSN 0008-5472

ISSN 1021-335X

5472

Nakamura, R. White, A. M. Smits and J. L. Bos (1988). Genetic alterations during colorectal-tumor development. *New England Journal of Medicine*, Vol.319, No.(9), pp.

pathway or network? *Developmental Biology*, Vol.239, No.(1), pp. 1-14, ISSN 0012-

marker CD133+ tumour cells and clinical outcome in rectal cancer. *Histopathology*,

R. Hollenberg, J. Kwiecien, P. Northcott, M. D. Taylor, C. Hawkins and S. K. Singh (2011). Sonic hedgehog regulates Bmi1 in human medulloblastoma brain tumor-

Goodlad and N. A. Wright (1998). APC in the regulation of intestinal crypt fission.

Yamamoto, M. Shirane, T. Muto and H. Nagawa (2006). Prediction of sensitivity of rectal cancer cells in response to preoperative radiotherapy by DNA microarray analysis of gene expression profiles. *Cancer Research*, Vol.66, No.(7), pp. 3370-3374,

*Cancer Research*, Vol.66, No.(4), pp. 1883-1890; discussion 1895-1886, ISSN 0008-

implication of circulating cancer stem cell. *Journal of Surgical Oncology*, Vol.102,

CD133+ cancer stem cells in laryngeal carcinoma. *Chinese Medical Journal (English)*,

for secretory cell lineage commitment in the mouse intestine. *Science*, Vol.294,

Kawamoto, Y. Inoue, C. Miki and M. Kusunoki (2009). Elevated CD133, but not VEGF or EGFR, as a predictive marker of distant recurrence after preoperative chemoradiotherapy in rectal cancer. *Oncology Reports*, Vol.22, No.(4), pp. 709-717,

J. Kearney and D. W. Buck (1997). AC133, a novel marker for human hematopoietic stem and progenitor cells. *Blood*, Vol.90, No.(12), pp. 5002-5012, ISSN 0006-4971


Schulte, A., H. S. Gunther, H. S. Phillips, D. Kemming, T. Martens, S. Kharbanda, R. H.

Scoville, D. H., T. Sato, X. C. He and L. Li (2008). Current view: intestinal stem cells and signaling. *Gastroenterology*, Vol.134, No.(3), pp. 849-864, ISSN 0016-5085 Silva, I. A., S. Bai, K. McLean, K. Yang, K. Griffith, D. Thomas, C. Ginestier, C. Johnston, A.

Singh, S. K., C. Hawkins, I. D. Clarke, J. A. Squire, J. Bayani, T. Hide, R. M. Henkelman, M.

Spradling, A., D. Drummond-Barbosa and T. Kai (2001). Stem cells find their niche. *Nature*,

Sun, G., M. Fujii, A. Sonoda, Y. Tokumaru, T. Matsunaga and N. Habu (2010). Identification

Takahashi-Yanaga, F. and M. Kahn (2010). Targeting Wnt signaling: can we safely eradicate

Takahashi, H., H. Ishii, N. Nishida, I. Takemasa, T. Mizushima, M. Ikeda, T. Yokobori, K.

Taylor, R. W., M. J. Barron, G. M. Borthwick, A. Gospel, P. F. Chinnery, D. C. Samuels, G.

Thliveris, A. T., R. B. Halberg, L. Clipson, W. F. Dove, R. Sullivan, M. K. Washington, S.

Todaro, M., M. P. Alea, A. B. Di Stefano, P. Cammareri, L. Vermeulen, F. Iovino, C. Tripodo,

Todaro, M., M. D'Asaro, N. Caccamo, F. Iovino, M. G. Francipane, S. Meraviglia, V. Orlando,

cells. *Nature*, Vol.432, No.(7015), pp. 396-401, ISSN 0028-0836

Vol.414, No.(6859), pp. 98-104, ISSN 0028-0836

Vol.18, No.(4), pp. 1166-1174, ISSN 1068-9265

*America*, Vol.102, No.(19), pp. 6960-6965, ISSN 0027-8424

*Immunology,* Vol. 182, No. 11, pp. 7287-7296, ISSN 1550-6606

(4), pp. 389-402, ISSN 1875-9777

No.(6), pp. 2005-2010, ISSN 0250-7005

Vol.59, No.(4), pp. 590-602, ISSN 0894-1491

3991-4001, ISSN 0008-5472

1078-0432

9738

Soriano, Z. Modrusan, S. Zapf, M. Westphal and K. Lamszus (2011). A distinct subset of glioma cell lines with stem cell-like properties reflects the transcriptional phenotype of glioblastomas and overexpresses CXCR4 as therapeutic target. *Glia*,

Kueck, R. K. Reynolds, M. S. Wicha and R. J. Buckanovich (2011). Aldehyde Dehydrogenase in Combination with CD133 Defines Angiogenic Ovarian Cancer Stem Cells That Portend Poor Patient Survival. *Cancer Research*, Vol.71, No.(11), pp.

D. Cusimano and P. B. Dirks (2004). Identification of human brain tumour initiating

of stem-like cells in head and neck cancer cell lines. *Anticancer Research*, Vol.30,

cancer stem cells? *Clinical Cancer Research*, Vol.16, No.(12), pp. 3153-3162, ISSN

Mimori, H. Yamamoto, M. Sekimoto, Y. Doki and M. Mori (2011). Significance of Lgr5(+ve) cancer stem cells in the colon and rectum. *Annals of Surgical Oncology*,

A. Taylor, S. M. Plusa, S. J. Needham, L. C. Greaves, T. B. Kirkwood and D. M. Turnbull (2003). Mitochondrial DNA mutations in human colonic crypt stem cells. *Journal of Clinical Investigation*, Vol.112, No.(9), pp. 1351-1360, ISSN 0021-

Stanhope and M. A. Newton (2005). Polyclonality of familial murine adenomas: analyses of mouse chimeras with low tumor multiplicity suggest short-range interactions. *Proceedings of the National Academy of Sciences of the United States of* 

A. Russo, G. Gulotta, J. P. Medema and G. Stassi. (2007). *Cell Stem Cell,* Vol. 1, No.

C. La Mendola, G. Gulotta, A. Salerno, F. Dieli and G. Stassi (2009). Efficient killing of human colon cancer stem cells by gammadelta T lymphocytes. *The Journal of* 


**17** 

*Japan* 

**Endoscopic Diagnosis and** 

Rieko Yamada and Kiyo Watanabe

**Treatment for Colorectal Cancer** 

Hiroyuki Kato, Teruhiko Sakamoto, Hiroko Otsuka,

Colonoscopy plays an important role in the medical care of patients with colorectal cancer. It is generally used for both the diagnosis of different stages of colorectal cancer and the treatment of early colorectal cancer and its precursors. The recent progress in colonoscopy has been remarkable. Endoscopes with variable rigidity and small diameters provide efficient insertion to the cecum and result in lower distress for patients. Trained colonoscopists can insert endoscopes into the cecum within a few minutes, and it is not

We can obtain good-quality pictures and special images to assist in diagnosis by using highvision endoscopes, magnifying endoscopes, dye spray, and narrow-band imaging (NBI). Determining whether a colorectal carcinoma can be curatively resected by endoscopic treatment or whether the carcinoma has a risk of lymph node metastasis is a very delicate and important task. In particular, the depth of cancer invasion is related to lymph node metastasis; therefore, endoscopic ultrasonography and the classification of pit patterns, capillary patterns via NBI, and the lesion-lifted condition are used to diagnose the depth of

Treatment for colorectal neoplastic lesions begins with hot biopsy and snare polypectomy, and recently, endoscopic submucosal resection (EMR), piecemeal EMR (EPMR), and endoscopic submucosal dissection (ESD) have become available for large and flat lesions of the colon and rectum. Early colorectal carcinoma is defined as a carcinoma within the submucosal layer that is not invading the muscularis propria. Carcinoma in situ (mucosal carcinoma) and carcinoma that slightly invades the submucosa and without risk factors for metastasis do not metastasize into lymph nodes or distant organs. Nonmetastatic carcinoma is cured by local resection with colonoscope. It is important to make an accurate diagnosis by endoscopy and to perform confident resection for pathological

In this chapter, we describe endoscopic diagnosis for colorectal carcinoma and differential diagnosis, and treatment options for early colorectal cancer without metastasis and for adenoma which is regarded as a precancerous condition. In addition, we briefly discuss risk

necessary to anesthetize patients without severe peritoneal adhesion.

factors for lymph node metastasis in early colorectal carcinoma.

cancer invasion (Kato, 2001, Sano, 2008).

**1. Introduction** 

evaluation.

*Tokyo Women's Medical University, Medical Center East, Department of Clinical Laboratory and Endoscopy, Tokyo,* 


### **Endoscopic Diagnosis and Treatment for Colorectal Cancer**

Hiroyuki Kato, Teruhiko Sakamoto, Hiroko Otsuka, Rieko Yamada and Kiyo Watanabe *Tokyo Women's Medical University, Medical Center East, Department of Clinical Laboratory and Endoscopy, Tokyo, Japan* 

**1. Introduction** 

326 Colorectal Cancer – From Prevention to Patient Care

Zabierowski, S. E. and M. Herlyn (2008). Melanoma stem cells: the dark seed of melanoma. *Journal of Clinical Oncology*, Vol.26, No.(17), pp. 2890-2894, ISSN 0732-183X Zhang, F., C. Song, Y. Ma, L. Tang, Y. Xu and H. Wang (2011). Effect of fibroblasts on breast

Zhang, Q., S. Shi, Y. Yen, J. Brown, J. Q. Ta and A. D. Le A subpopulation of CD133(+)

ISSN 1107-3756

0304-3835

cancer cell mammosphere formation and regulation of stem cell-related gene expression. *International Journal of Molecular Medicine*, Vol.28, No.(3), pp. 365-371,

cancer stem-like cells characterized in human oral squamous cell carcinoma confer resistance to chemotherapy (2010). *Cancer Letters*, Vol.289, No.(2), pp. 151-160, ISSN

> Colonoscopy plays an important role in the medical care of patients with colorectal cancer. It is generally used for both the diagnosis of different stages of colorectal cancer and the treatment of early colorectal cancer and its precursors. The recent progress in colonoscopy has been remarkable. Endoscopes with variable rigidity and small diameters provide efficient insertion to the cecum and result in lower distress for patients. Trained colonoscopists can insert endoscopes into the cecum within a few minutes, and it is not necessary to anesthetize patients without severe peritoneal adhesion.

> We can obtain good-quality pictures and special images to assist in diagnosis by using highvision endoscopes, magnifying endoscopes, dye spray, and narrow-band imaging (NBI). Determining whether a colorectal carcinoma can be curatively resected by endoscopic treatment or whether the carcinoma has a risk of lymph node metastasis is a very delicate and important task. In particular, the depth of cancer invasion is related to lymph node metastasis; therefore, endoscopic ultrasonography and the classification of pit patterns, capillary patterns via NBI, and the lesion-lifted condition are used to diagnose the depth of cancer invasion (Kato, 2001, Sano, 2008).

> Treatment for colorectal neoplastic lesions begins with hot biopsy and snare polypectomy, and recently, endoscopic submucosal resection (EMR), piecemeal EMR (EPMR), and endoscopic submucosal dissection (ESD) have become available for large and flat lesions of the colon and rectum. Early colorectal carcinoma is defined as a carcinoma within the submucosal layer that is not invading the muscularis propria. Carcinoma in situ (mucosal carcinoma) and carcinoma that slightly invades the submucosa and without risk factors for metastasis do not metastasize into lymph nodes or distant organs. Nonmetastatic carcinoma is cured by local resection with colonoscope. It is important to make an accurate diagnosis by endoscopy and to perform confident resection for pathological evaluation.

> In this chapter, we describe endoscopic diagnosis for colorectal carcinoma and differential diagnosis, and treatment options for early colorectal cancer without metastasis and for adenoma which is regarded as a precancerous condition. In addition, we briefly discuss risk factors for lymph node metastasis in early colorectal carcinoma.

Endoscopic Diagnosis and Treatment for Colorectal Cancer 329

of Type 2 is occasionally observed as a stenosis due to the tumor of the large intestine, and ulceration is not always detected by endoscopy. Pathological examination of biopsy specimen from the stenosis or edge of ulceration reveals adenocarcinoma. Type 4 lesion is observed like hard mucosal stenosis and neither obvious tumor nor ulceration is always recognized by endoscopy. And pathological diagnosis from biopsy is very difficult because

Early colorectal carcinoma is defined as a carcinoma that is confined to the mucosa (M) and submucosa (SM). Early colon carcinoma may occur in an adenomatous polyp or may be difficult to distinguish from a nonmalignant adenomatous polyp by colonoscopy. For example, a 2-cm-wide villous adenoma has an approximately 40% chance of harboring cancer (Kim, 1997). Polyp risk factors for malignancy include villous rather than tubular histology, large size, sessile morphology, and high numbers of colonic polyps (Morson, 1972). Another route of carcinogenesis is "de novo" carcinogenesis, which produces small, aggressive carcinomas that do not appear to develop from adenomas (Kudo, 1997; Mueller, 2002). Macroscopic depressed type is the most common type of this carcinoma. This type is difficult to diagnose early using colonoscopy let alone barium enema; therefore, it is important to observe these lesions by endoscope extremely carefully. Early colorectal carcinoma is asymptomatic. It is usually revealed by screening colonoscopy or a positive stool occult blood test followed by colonoscopy. Colorectal screenings are important for

detecting early colorectal carcinoma, which may be curatively treated by endoscopy.

Regarding the classification of macroscopic-type lesions, the Japanese colorectal cancer handling protocol and Paris classification are representative classification systems. Both are used to judge endoscopic findings. Early colorectal carcinoma is classified as any Type 0

Type 0 is subclassified into Type 0-I (tubercle type) and 0-II (surface type). Type 0-I is further subclassified as Ip (pedunculated), Isp (subpedunculated), and Is (sessile), whereas Type 0-II is further subclassified as IIa (surface tubercle), IIb (surface flatness), and IIc

Fig. 3. Classification of superficial colorectal carcinoma (Japanese protocol)

**2.2.1 Macroscopic type of early colorectal carcinoma** 

**2.2.2 Japanese classification of colorectal carcinoma** 

lesion judged to be a superficial carcinoma.

(surface depressed) (Fig. 3).

the carcinoma is covered with normal mucosa.

**2.2 Early colorectal carcinoma** 

#### **2. Endoscopic diagnosis of colorectal carcinoma**

Colorectal carcinomas are the most common malignancies in industrialized countries, and are classified as early or advanced according to the depth of invasion. In advanced cancers, the invasion reaches the muscularis propria (MP) or the deeper layers. In endoscopic diagnosis, macroscopic classification is the most basic information. In this section, endoscopic diagnosis of colon carcinoma is discussed.

#### **2.1 Macroscopic classification of colorectal carcinoma**

Colonoscopy is a valuable tool in the diagnosis and management of colorectal neoplasms. Advanced colorectal carcinoma can be divided into 4 groups based on endoscopic appearances (Fig. 1).

#### Fig. 1. Macroscopic types of advanced colorectal carcinoma

Type 1 lesion, protuberant tumor with fold convergence

Type 2 lesion, showing an irregular ulceration and clear marginal swelling

Type 3 lesion, showing an irregular ulceration and unclear marginal swelling

Type 4 lesion, showing an irregularly edematous mucosa with luminal stenosis due to diffuse infiltration, Ulceration is not pointed out on the lesion.

Fig. 2. Endoscopic view of each type of advanced colorectal carcinoma

Localized carcinoma is classified as a polypoid- (protuberant type) or ulcerative–type lesion. More than 90% of colorectal carcinomas are ulcerative-type lesions. To further distinguish early colorectal carcinoma and unclassified advanced colorectal carcinoma, the macroscopic type is subclassified from Type 0 to Type 5 (Sugihara, 2009). Early cancers are defined by the depth of cancer invasion into mucosal or submucosal layers. In this manual, early carcinoma is classified as Type 0, and advanced carcinoma is classified as Type 1 to 5.

The subclassification of Type 0 carcinoma is described in further detail in the next section. Type 1 lesions are protuberant. Type 2 lesions include ulcerative-type lesions with clear margins, and Type 3 lesions include ulcerative-type lesions with infiltration. Diffusely infiltrating lesions are classified as Type 4 lesions (Fig. 2). Type 5 lesions are an unclassified type. Type 2 is the most common type of advanced colorectal carcinoma. Circular carcinoma of Type 2 is occasionally observed as a stenosis due to the tumor of the large intestine, and ulceration is not always detected by endoscopy. Pathological examination of biopsy specimen from the stenosis or edge of ulceration reveals adenocarcinoma. Type 4 lesion is observed like hard mucosal stenosis and neither obvious tumor nor ulceration is always recognized by endoscopy. And pathological diagnosis from biopsy is very difficult because the carcinoma is covered with normal mucosa.

#### **2.2 Early colorectal carcinoma**

328 Colorectal Cancer – From Prevention to Patient Care

Colorectal carcinomas are the most common malignancies in industrialized countries, and are classified as early or advanced according to the depth of invasion. In advanced cancers, the invasion reaches the muscularis propria (MP) or the deeper layers. In endoscopic diagnosis, macroscopic classification is the most basic information. In this section,

Colonoscopy is a valuable tool in the diagnosis and management of colorectal neoplasms. Advanced colorectal carcinoma can be divided into 4 groups based on endoscopic

Type 1 Type 2 Type 3 Type 4

Type 1 Type 2 Type 3 Type 4

Localized carcinoma is classified as a polypoid- (protuberant type) or ulcerative–type lesion. More than 90% of colorectal carcinomas are ulcerative-type lesions. To further distinguish early colorectal carcinoma and unclassified advanced colorectal carcinoma, the macroscopic type is subclassified from Type 0 to Type 5 (Sugihara, 2009). Early cancers are defined by the depth of cancer invasion into mucosal or submucosal layers. In this manual, early carcinoma

The subclassification of Type 0 carcinoma is described in further detail in the next section. Type 1 lesions are protuberant. Type 2 lesions include ulcerative-type lesions with clear margins, and Type 3 lesions include ulcerative-type lesions with infiltration. Diffusely infiltrating lesions are classified as Type 4 lesions (Fig. 2). Type 5 lesions are an unclassified type. Type 2 is the most common type of advanced colorectal carcinoma. Circular carcinoma

Type 4 lesion, showing an irregularly edematous mucosa with luminal stenosis due to diffuse

**2. Endoscopic diagnosis of colorectal carcinoma** 

endoscopic diagnosis of colon carcinoma is discussed.

appearances (Fig. 1).

**2.1 Macroscopic classification of colorectal carcinoma** 

Fig. 1. Macroscopic types of advanced colorectal carcinoma

Type 1 lesion, protuberant tumor with fold convergence

infiltration, Ulceration is not pointed out on the lesion.

Type 2 lesion, showing an irregular ulceration and clear marginal swelling Type 3 lesion, showing an irregular ulceration and unclear marginal swelling

Fig. 2. Endoscopic view of each type of advanced colorectal carcinoma

is classified as Type 0, and advanced carcinoma is classified as Type 1 to 5.

Early colorectal carcinoma is defined as a carcinoma that is confined to the mucosa (M) and submucosa (SM). Early colon carcinoma may occur in an adenomatous polyp or may be difficult to distinguish from a nonmalignant adenomatous polyp by colonoscopy. For example, a 2-cm-wide villous adenoma has an approximately 40% chance of harboring cancer (Kim, 1997). Polyp risk factors for malignancy include villous rather than tubular histology, large size, sessile morphology, and high numbers of colonic polyps (Morson, 1972). Another route of carcinogenesis is "de novo" carcinogenesis, which produces small, aggressive carcinomas that do not appear to develop from adenomas (Kudo, 1997; Mueller, 2002). Macroscopic depressed type is the most common type of this carcinoma. This type is difficult to diagnose early using colonoscopy let alone barium enema; therefore, it is important to observe these lesions by endoscope extremely carefully. Early colorectal carcinoma is asymptomatic. It is usually revealed by screening colonoscopy or a positive stool occult blood test followed by colonoscopy. Colorectal screenings are important for detecting early colorectal carcinoma, which may be curatively treated by endoscopy.

#### **2.2.1 Macroscopic type of early colorectal carcinoma**

Regarding the classification of macroscopic-type lesions, the Japanese colorectal cancer handling protocol and Paris classification are representative classification systems. Both are used to judge endoscopic findings. Early colorectal carcinoma is classified as any Type 0 lesion judged to be a superficial carcinoma.

#### **2.2.2 Japanese classification of colorectal carcinoma**

Type 0 is subclassified into Type 0-I (tubercle type) and 0-II (surface type). Type 0-I is further subclassified as Ip (pedunculated), Isp (subpedunculated), and Is (sessile), whereas Type 0-II is further subclassified as IIa (surface tubercle), IIb (surface flatness), and IIc (surface depressed) (Fig. 3).

Fig. 3. Classification of superficial colorectal carcinoma (Japanese protocol)

Endoscopic Diagnosis and Treatment for Colorectal Cancer 331

Type 0-IIc (A) (B) Type 0-IIa+IIc (C) (D) Fig. 6. (A) Type 0-IIc, Ordinary colonoscopic picture showing a depressed area with erosion (B) Type 0-IIc, Indigocarmine dye spraying view (C) Type 0-IIa+IIc, Ordinary colonoscopic picture showing a flat elevated lesion with irregular depressed area (D) Type 0-IIa+IIc,

The importance of an endoscopic classification system for superficial lesions is that it permits endoscopic staging. In other words, we can predict the depth of invasion of a superficial carcinoma and predict the risk of lymph node metastasis, both of which assist in treatment selection (endoscopic treatment or surgical resection). Regarding Type 0-I lesions, if a lesion becomes large size, the risk of submucosal invasion is increased gradually. Conversely, Type 0-IIc lesions have deep invasion tendencies despite their small size. In addition, Type 0-IIa+IIc lesions frequently infiltrate the deep stratum submucosum, and

**2.3 Endoscopic ultrasonography (EUS) and diagnosis of depth invasion for colorectal** 

EUS is an imaging technique for ultrasound scanning of the gastrointestinal tract lumen. It can depict lesions as vertical tomographic images. EUS can be used to evaluate the depth of invasion of epithelial tumors and carcinomas, as well as for qualitative diagnosis, such as the differential diagnosis of extramural lesions in patients with submucosal tumors. EUS is an important diagnostic procedure for deciding the treatment policy and assessing the status of diseases involving the lower gastrointestinal tract. This section focuses on the

The lower gastrointestinal tract has the highest incidence of colorectal cancer; EUS is indicated for the diagnosis of the depth of wall invasion and lymph node metastases. EUS is also indicated for the evaluation of submucosal tumors. Malignant lymphomas, gastrointestinal stromal tumors (GIST), lymphangiomas, and lipomas arise at a relatively

Ultrasonographic instruments specifically for the colorectum and ultrasonic probes are available for EUS of the colorectal region. Endoscopic three-dimensional ultrasonic probes

An ultrasonic probe is attached to the tip of a direct-viewing electronic endoscope to perform mechanical radial ultrasonic scanning. One advantage of using a specialized device is the excellent ultrasonic resolution, allowing distinct tomographic images to be obtained throughout the entire intestine. Scanning can be performed at either of two frequencies (7.5 MHz, 20 MHz), and the frequency best suited for a given lesion can be selected. The higher

their potential for progression is higher than that of other types.

Indigocarmine dye spraying view

diagnosis of colorectal cancer.

are also commercially available.

high frequency in the lower gastrointestinal tract.

**2.3.1 Instruments and ultrasonic probe (USP) of EUS** 

**carcinoma** 

Carcinomas can also be mixed-type lesions, which are lesions possessing elements of both Type 0-I and 0-II. Mixed-type lesions include types 0-IIc+IIa, 0-IIa+IIc, 0-IIc+Is, and 0-Is+IIc.

#### **2.2.3 Paris classification**

Terms used in the Paris classification of macroscopic-type lesions are unified by the terms used in a paper by Schlemper in 2002 (Fig. 4).

Fig. 4. Paris classification of superficial colorectal carcinoma

Firstly, lesions are divided into polypoid (Type 0-I) and non-polypoid (Type 0-IIa, IIb, IIc) lesions, and Type 0-I is subclassified as Type 0-Ip (pedunculated) and Type 0-Is (sessile). Type 0-III carcinomas comprise excavated-type lesions in the original classification, but these lesions are rare in the colon and rectum. Isp lesions are classified in the Japanese colorectal carcinoma handling protocol as Type 0-Is. Type 0-IIa lesions include those in which their height does not exceed that of closed biopsy forceps (about 2.5 mm), and lesions with heights exceeding this threshold are classified as type0-Is. Mixed-type lesions include Type 0-IIa+IIc, 0-IIc+IIa, 0-IIc+Is, and 0-Is+IIc (Fig. 5,6).

Fig. 5. Classification of mixed-type lesions (Paris classification)

Carcinomas can also be mixed-type lesions, which are lesions possessing elements of both Type 0-I and 0-II. Mixed-type lesions include types 0-IIc+IIa, 0-IIa+IIc, 0-IIc+Is, and 0-Is+IIc.

Terms used in the Paris classification of macroscopic-type lesions are unified by the terms

Firstly, lesions are divided into polypoid (Type 0-I) and non-polypoid (Type 0-IIa, IIb, IIc) lesions, and Type 0-I is subclassified as Type 0-Ip (pedunculated) and Type 0-Is (sessile). Type 0-III carcinomas comprise excavated-type lesions in the original classification, but these lesions are rare in the colon and rectum. Isp lesions are classified in the Japanese colorectal carcinoma handling protocol as Type 0-Is. Type 0-IIa lesions include those in which their height does not exceed that of closed biopsy forceps (about 2.5 mm), and lesions with heights exceeding this threshold are classified as type0-Is. Mixed-type lesions include

**2.2.3 Paris classification** 

used in a paper by Schlemper in 2002 (Fig. 4).

Fig. 4. Paris classification of superficial colorectal carcinoma

Type 0-IIa+IIc, 0-IIc+IIa, 0-IIc+Is, and 0-Is+IIc (Fig. 5,6).

Fig. 5. Classification of mixed-type lesions (Paris classification)

Fig. 6. (A) Type 0-IIc, Ordinary colonoscopic picture showing a depressed area with erosion (B) Type 0-IIc, Indigocarmine dye spraying view (C) Type 0-IIa+IIc, Ordinary colonoscopic picture showing a flat elevated lesion with irregular depressed area (D) Type 0-IIa+IIc, Indigocarmine dye spraying view

The importance of an endoscopic classification system for superficial lesions is that it permits endoscopic staging. In other words, we can predict the depth of invasion of a superficial carcinoma and predict the risk of lymph node metastasis, both of which assist in treatment selection (endoscopic treatment or surgical resection). Regarding Type 0-I lesions, if a lesion becomes large size, the risk of submucosal invasion is increased gradually. Conversely, Type 0-IIc lesions have deep invasion tendencies despite their small size. In addition, Type 0-IIa+IIc lesions frequently infiltrate the deep stratum submucosum, and their potential for progression is higher than that of other types.

#### **2.3 Endoscopic ultrasonography (EUS) and diagnosis of depth invasion for colorectal carcinoma**

EUS is an imaging technique for ultrasound scanning of the gastrointestinal tract lumen. It can depict lesions as vertical tomographic images. EUS can be used to evaluate the depth of invasion of epithelial tumors and carcinomas, as well as for qualitative diagnosis, such as the differential diagnosis of extramural lesions in patients with submucosal tumors. EUS is an important diagnostic procedure for deciding the treatment policy and assessing the status of diseases involving the lower gastrointestinal tract. This section focuses on the diagnosis of colorectal cancer.

The lower gastrointestinal tract has the highest incidence of colorectal cancer; EUS is indicated for the diagnosis of the depth of wall invasion and lymph node metastases. EUS is also indicated for the evaluation of submucosal tumors. Malignant lymphomas, gastrointestinal stromal tumors (GIST), lymphangiomas, and lipomas arise at a relatively high frequency in the lower gastrointestinal tract.

#### **2.3.1 Instruments and ultrasonic probe (USP) of EUS**

Ultrasonographic instruments specifically for the colorectum and ultrasonic probes are available for EUS of the colorectal region. Endoscopic three-dimensional ultrasonic probes are also commercially available.

An ultrasonic probe is attached to the tip of a direct-viewing electronic endoscope to perform mechanical radial ultrasonic scanning. One advantage of using a specialized device is the excellent ultrasonic resolution, allowing distinct tomographic images to be obtained throughout the entire intestine. Scanning can be performed at either of two frequencies (7.5 MHz, 20 MHz), and the frequency best suited for a given lesion can be selected. The higher

Endoscopic Diagnosis and Treatment for Colorectal Cancer 333

(A) (B) (C)

unclearly demarcated.

colorectal cancer.

**2.3.4 Diagnosis of invasion depth of SM cancer** 

**2.3.5 Diagnosis of lymph node metastasis** 

Fig. 8. Endoscopic view and EUS image, (A) Mucosal cancer, (B) SM cancer, (C)MP cancer In SS to SE cancer (cancer invading the sabserosa or serosa) or A cancer (cancer invading the tunica adventitia), the fifth layer is narrowed or ruptured by the mass, but the border with the adjacent organ remains intact. In SI or AI cancer (cancer invading the adjacent organ), up to the fifth layer is destroyed by the mass, and the border with the adjacent organ is

Intramucosal cancer and SM cancer with mild invasion to a vertical depth of less than 1000 µm from the lower border of the muscularis mucosae have virtually no risk of metastasis, and cure can be expected after endoscopic treatment. Endoscopic therapy is thus indicated for such lesions (Sugihara, 2009). On EUS, SM cancers are classified into lesions with shallow invasion and those with deep invasion at the time of diagnosis based on Kudo's classification of SM cancer (Kudo, 2000). The third layer, corresponding to the normal submucosa adjacent to cancer, is subdivided into 3 equal layers, and the location of deepest region of the hypoechoic mass is determined. Masses that are confined to the shallowest third of the submucosa are classified as sm1, those that invade the second third are classified as sm2, and those that invade deeper than the second third, but do not extend beyond the region near the medial border of the fourth layer are classified as sm3. SM cancers with shallow invasion correspond to sm1, and those with deep invasion correspond to sm2 and sm3. EUS diagnosis thus plays an important role in the selection of treatment for early

Normal lymph nodes are not depicted on EUS. Lymph nodes visualized on EUS that have a shortest diameter of ≥5 mm, a hyperechoic and homogeneous internal echo, and are either

frequency is better suited for low and superficial lesions, whereas the lower frequency is recommended for the assessment of high and deep lesions and the examination of lymph node metastases and other lesions around the intestine. A disadvantage of specialized devices is the large outer diameter of the scope and the long hard tip, often precluding insertion into the proximal side of the sigmoid colon.

A USP can usually be inserted in the forceps channel of the endoscope. Either a 12-, 20-, or 30-MHz probe is selected, depending on the lesion. A USP is inferior to a specialized device in terms of lateral resolution and durability, but excels with respect to targeting because ultrasonic procedures can be done while directly viewing a lesion. Lesions associated with stenosis are also good indications for a USP.

#### **2.3.2 Diagnostic technique of EUS for colorectal carcinoma**

The colorectal wall is fundamentally depicted as a 5-layer structure on EUS. Starting from the lumen, the first, hyperechoic layer and the second, hypoechoic layer correspond to the mucosa, the third, hyperechoic layer to the submucosa, the fourth, hypoechoic layer to the muscularis propria, and the fifth, hyperechoic layer to the sabserosa and serosa (the tunica adventitia at sites with no serosa) (Fig. 7.).

Fig. 7. Layer structure of the normal intestinal wall

With the use of high-frequency devices (20-30 MHz), the muscularis mucosae is depicted as a thin hypoechoic layer with a hyperechoic border at the upper margin of the third layer. In the fourth layer, the connective tissue of the muscularis is depicted as a thin hyperechoic layer, and the colorectal wall is sometimes depicted as a 9-layer structure. An understanding of these characteristics is essential for the diagnosis of lesions by comparison with the layer structure of the intestinal wall.

#### **2.3.3 Diagnosis of colorectal carcinoma on EUS**

On EUS, the depth of wall invasion is evaluated on the basis of what layers are preserved or destroyed by a hypoechoic mass. In M cancer (intramucosal cancer), the mass is confined to the first to second layers. In SM cancer (cancer invading the submucosa), the third layer is narrowed or ruptured by the mass, but the fourth layer remains intact. In MP cancer (cancer invading the muscularis propria), the fourth layer is narrowed or ruptured by the mass, but the fifth layer remains intact (Fig. 8.).

frequency is better suited for low and superficial lesions, whereas the lower frequency is recommended for the assessment of high and deep lesions and the examination of lymph node metastases and other lesions around the intestine. A disadvantage of specialized devices is the large outer diameter of the scope and the long hard tip, often precluding

A USP can usually be inserted in the forceps channel of the endoscope. Either a 12-, 20-, or 30-MHz probe is selected, depending on the lesion. A USP is inferior to a specialized device in terms of lateral resolution and durability, but excels with respect to targeting because ultrasonic procedures can be done while directly viewing a lesion. Lesions associated with

The colorectal wall is fundamentally depicted as a 5-layer structure on EUS. Starting from the lumen, the first, hyperechoic layer and the second, hypoechoic layer correspond to the mucosa, the third, hyperechoic layer to the submucosa, the fourth, hypoechoic layer to the muscularis propria, and the fifth, hyperechoic layer to the sabserosa and serosa (the tunica

With the use of high-frequency devices (20-30 MHz), the muscularis mucosae is depicted as a thin hypoechoic layer with a hyperechoic border at the upper margin of the third layer. In the fourth layer, the connective tissue of the muscularis is depicted as a thin hyperechoic layer, and the colorectal wall is sometimes depicted as a 9-layer structure. An understanding of these characteristics is essential for the diagnosis of lesions by comparison with the layer

On EUS, the depth of wall invasion is evaluated on the basis of what layers are preserved or destroyed by a hypoechoic mass. In M cancer (intramucosal cancer), the mass is confined to the first to second layers. In SM cancer (cancer invading the submucosa), the third layer is narrowed or ruptured by the mass, but the fourth layer remains intact. In MP cancer (cancer invading the muscularis propria), the fourth layer is narrowed or ruptured by the mass, but

insertion into the proximal side of the sigmoid colon.

**2.3.2 Diagnostic technique of EUS for colorectal carcinoma** 

stenosis are also good indications for a USP.

adventitia at sites with no serosa) (Fig. 7.).

Fig. 7. Layer structure of the normal intestinal wall

**2.3.3 Diagnosis of colorectal carcinoma on EUS** 

structure of the intestinal wall.

the fifth layer remains intact (Fig. 8.).

Fig. 8. Endoscopic view and EUS image, (A) Mucosal cancer, (B) SM cancer, (C)MP cancer

In SS to SE cancer (cancer invading the sabserosa or serosa) or A cancer (cancer invading the tunica adventitia), the fifth layer is narrowed or ruptured by the mass, but the border with the adjacent organ remains intact. In SI or AI cancer (cancer invading the adjacent organ), up to the fifth layer is destroyed by the mass, and the border with the adjacent organ is unclearly demarcated.

#### **2.3.4 Diagnosis of invasion depth of SM cancer**

Intramucosal cancer and SM cancer with mild invasion to a vertical depth of less than 1000 µm from the lower border of the muscularis mucosae have virtually no risk of metastasis, and cure can be expected after endoscopic treatment. Endoscopic therapy is thus indicated for such lesions (Sugihara, 2009). On EUS, SM cancers are classified into lesions with shallow invasion and those with deep invasion at the time of diagnosis based on Kudo's classification of SM cancer (Kudo, 2000). The third layer, corresponding to the normal submucosa adjacent to cancer, is subdivided into 3 equal layers, and the location of deepest region of the hypoechoic mass is determined. Masses that are confined to the shallowest third of the submucosa are classified as sm1, those that invade the second third are classified as sm2, and those that invade deeper than the second third, but do not extend beyond the region near the medial border of the fourth layer are classified as sm3. SM cancers with shallow invasion correspond to sm1, and those with deep invasion correspond to sm2 and sm3. EUS diagnosis thus plays an important role in the selection of treatment for early colorectal cancer.

#### **2.3.5 Diagnosis of lymph node metastasis**

Normal lymph nodes are not depicted on EUS. Lymph nodes visualized on EUS that have a shortest diameter of ≥5 mm, a hyperechoic and homogeneous internal echo, and are either

Endoscopic Diagnosis and Treatment for Colorectal Cancer 335

tubular or roundish pits that are larger than normal pits. Almost all of Type IIIL lesions are tubular adenomas in pathology, which can be treated by polypectomy. Type IV includes branch-like or gyrus-like pits, most of which are tubulovillous adenoma. Mucosal carcinoma is present in 35% of these pits and can be treated by endoscopy. Type VI includes irregularly arranged pits that may be submucosal invasive carcinoma, for which the proper treatment straddles the borderline between endoscopic and surgical therapy. Lastly, type VN includes nonstructured pits, which indicate massive submucosal invasive carcinoma and require

Kudo reported that small round pit patterns (type IIIs) and non-pit patterns (type V) were common in depressed lesions and that these depressed lesions had invaded the deeper

The NBI system involves modifying spectral features by narrowing the bandwidth of spectral transmittance using various optical filters (Sano, 2001). This modification provides a unique image that emphasizes the capillary pattern, as well as the surface structure, by simple operation of a button on the control panel of the endoscope. Because of its similarity to chromoendoscopy, NBI can be referred to as optical or digital chromoendoscopy. Sano et al. (2006) classified 3 types and names (CP types I, II, and III) of microvascular architectures based on the magnified NBI pattern. CP type I has no meshed capillary vessels. CP type II has meshed capillary vessels surrounding the mucosal glands. CP type III lesions were further classified into 2 groups: types IIIA and IIIB. CP type IIIA has irregular meshed capillary vessels, whereas irregular meshed capillary vessels disappear or loosen in CP type IIIB (Fig.10). CP types I, II, IIIA, and IIIB are observed in nonneoplastic lesions, adenomas, mucosal or slightly invasive submucosal carcinoma, and massive invasive submucosal carcinoma, respectively. Capillary patterns, as assessed by magnifying NBI, are useful for differentiating small colorectal nonneoplastic polyps from neoplastic ones (accuracy, 95.3%; sensitivity, 96.4%; and specificity, 92.3%) (Sano, 2008), and they are highly accurate for distinguishing low-grade dysplasia from high-grade dysplasia/invasive cancer (accuracy, 95.5%; sensitivity, 90.3%; and specificity, 97.1%) (Katagiri, 2008). Therefore, capillary

surgical resection with lymph node dissection.

layers more rapidly than had protruding lesions.

Fig. 10. Capillary pattern by magnified NBI

**2.4.2 Classification of capillary pattern by magnified NBI** 

patterns can be used to predict the histopathology of colorectal neoplasia.

CP type I CP type II CP type IIIA CP type IIIB

Lymph node metastasis is reported to occur in approximately 10% of SM cancers. SM cancer is a boundary lesion, and its treatment plan involves endoscopic treatment or surgical

**2.5 Risk factor for lymph node metastasis of submucosal invasive carcinoma** 

circular or irregularly shaped are considered positive for metastasis. However, differential diagnosis from enlarged lymph nodes associated with inflammation is challenging. The rate of correctly diagnosing lymph node metastases has been reported to be 70% to 80% (Tio, 1991, Cho E, et al. 1993). The diagnostic ability of EUS is thus not considered good.

EUS is useful for diagnosis of the invasion depth of colorectal cancer because it can depict lesions as vertical tomographic images. It is thus an important diagnostic procedure for deciding treatment policy and evaluating disease status.

#### **2.4 Diagnosis of colorectal neoplastic lesions by chromoendoscopy and imageenhanced endoscopy**

If a colorectal lesion is detected by conventional endoscopy, the location, size, macroscopic type, color, surface pattern, presence of fold conversion, and air-induced deformation can be observed. The indigo carmine dye-spraying method more clearly reveals the extent and surface pattern. In addition, magnifying endoscopy after staining with indigo carmine or crystal violet is useful for pit pattern classification (Kudo et al., 1994), as it enables the differentiation of neoplasms as well as histological grading and depth evaluation of early cancers. This leads to the selection of endoscopic therapy or surgery.

The combination of image-enhanced endoscopic techniques such as NBI with magnification is used to observe the capillary pattern of the tumor surface, and these techniques can also improve diagnosis (Sano et al., 2006).

#### **2.4.1 Classification of pit pattern**

Pit patterns in the large intestine were classified into 7 types by Kudo (Fig. 9). Type I includes round pits that are observed in normal mucosa. Type II includes stellar or papillary pits, and these pits always indicate hyperplasia. Type IIIs includes small tubular or round pit that are smaller than normal pits, and they indicate neoplastic lesions, occasionally including carcinoma that can be resected by endoscopy. Type IIIL includes

Fig. 9. Classification of pit pattern

circular or irregularly shaped are considered positive for metastasis. However, differential diagnosis from enlarged lymph nodes associated with inflammation is challenging. The rate of correctly diagnosing lymph node metastases has been reported to be 70% to 80% (Tio,

EUS is useful for diagnosis of the invasion depth of colorectal cancer because it can depict lesions as vertical tomographic images. It is thus an important diagnostic procedure for

If a colorectal lesion is detected by conventional endoscopy, the location, size, macroscopic type, color, surface pattern, presence of fold conversion, and air-induced deformation can be observed. The indigo carmine dye-spraying method more clearly reveals the extent and surface pattern. In addition, magnifying endoscopy after staining with indigo carmine or crystal violet is useful for pit pattern classification (Kudo et al., 1994), as it enables the differentiation of neoplasms as well as histological grading and depth evaluation of early

The combination of image-enhanced endoscopic techniques such as NBI with magnification is used to observe the capillary pattern of the tumor surface, and these techniques can also

Pit patterns in the large intestine were classified into 7 types by Kudo (Fig. 9). Type I includes round pits that are observed in normal mucosa. Type II includes stellar or papillary pits, and these pits always indicate hyperplasia. Type IIIs includes small tubular or round pit that are smaller than normal pits, and they indicate neoplastic lesions, occasionally including carcinoma that can be resected by endoscopy. Type IIIL includes

Type I Type II Type IIIs Type IIIL

Type IV Type VI Type VN

1991, Cho E, et al. 1993). The diagnostic ability of EUS is thus not considered good.

**2.4 Diagnosis of colorectal neoplastic lesions by chromoendoscopy and image-**

deciding treatment policy and evaluating disease status.

cancers. This leads to the selection of endoscopic therapy or surgery.

**enhanced endoscopy** 

improve diagnosis (Sano et al., 2006).

**2.4.1 Classification of pit pattern** 

Fig. 9. Classification of pit pattern

tubular or roundish pits that are larger than normal pits. Almost all of Type IIIL lesions are tubular adenomas in pathology, which can be treated by polypectomy. Type IV includes branch-like or gyrus-like pits, most of which are tubulovillous adenoma. Mucosal carcinoma is present in 35% of these pits and can be treated by endoscopy. Type VI includes irregularly arranged pits that may be submucosal invasive carcinoma, for which the proper treatment straddles the borderline between endoscopic and surgical therapy. Lastly, type VN includes nonstructured pits, which indicate massive submucosal invasive carcinoma and require surgical resection with lymph node dissection.

Kudo reported that small round pit patterns (type IIIs) and non-pit patterns (type V) were common in depressed lesions and that these depressed lesions had invaded the deeper layers more rapidly than had protruding lesions.

#### **2.4.2 Classification of capillary pattern by magnified NBI**

The NBI system involves modifying spectral features by narrowing the bandwidth of spectral transmittance using various optical filters (Sano, 2001). This modification provides a unique image that emphasizes the capillary pattern, as well as the surface structure, by simple operation of a button on the control panel of the endoscope. Because of its similarity to chromoendoscopy, NBI can be referred to as optical or digital chromoendoscopy. Sano et al. (2006) classified 3 types and names (CP types I, II, and III) of microvascular architectures based on the magnified NBI pattern. CP type I has no meshed capillary vessels. CP type II has meshed capillary vessels surrounding the mucosal glands. CP type III lesions were further classified into 2 groups: types IIIA and IIIB. CP type IIIA has irregular meshed capillary vessels, whereas irregular meshed capillary vessels disappear or loosen in CP type IIIB (Fig.10). CP types I, II, IIIA, and IIIB are observed in nonneoplastic lesions, adenomas, mucosal or slightly invasive submucosal carcinoma, and massive invasive submucosal carcinoma, respectively. Capillary patterns, as assessed by magnifying NBI, are useful for differentiating small colorectal nonneoplastic polyps from neoplastic ones (accuracy, 95.3%; sensitivity, 96.4%; and specificity, 92.3%) (Sano, 2008), and they are highly accurate for distinguishing low-grade dysplasia from high-grade dysplasia/invasive cancer (accuracy, 95.5%; sensitivity, 90.3%; and specificity, 97.1%) (Katagiri, 2008). Therefore, capillary patterns can be used to predict the histopathology of colorectal neoplasia.

Fig. 10. Capillary pattern by magnified NBI

#### **2.5 Risk factor for lymph node metastasis of submucosal invasive carcinoma**

Lymph node metastasis is reported to occur in approximately 10% of SM cancers. SM cancer is a boundary lesion, and its treatment plan involves endoscopic treatment or surgical

Endoscopic Diagnosis and Treatment for Colorectal Cancer 337

Vessel invasions contain 2 types of lymphatic invasion and venous invasion. The distinction is not so easy. When cancer nests are located in lymphatic ducts lined by flat endothelial cells, it is considered positive lymphatic invasion. Cancer nests existing near an artery are very likely to represent venous invasion, which can be confirmed by determining the presence of an internal elastic membrane and plain muscle around the cancer nests. Occasionally, it is difficult to detect venous invasion by hematoxylin and eosin staining.

Budding is defined as cancer nests comprising less than 5 cancer cells and invading the interstitial tissue of the cancer growth front. The area where budding appears most frequently is selected and the number of instances is counted in a ×200 field. Budding is classified into 3 groups (grade 1, 0–4 pieces; grade 2, 5–9 pieces; and grade 3, ≥10 pieces),

There are many diseases that must be differentiated from colorectal carcinoma. The characteristic appearance of colorectal carcinoma does not cause interpretive difficulties

**Metastatic lesions in the large intestine:** Cancers that frequently metastasize to the large intestine include those of the stomach, pancreas, ovaries, lung, and breasts in descending order of frequency. The endoscopic appearance of metastatic lesions generally includes (1) extraluminal masses with or without hyperemia, (2) wall thickening, and (3) hyperemia. Extraluminal masses are often smooth, but the base is not distinct. Mucosal hyperemias are often multiple, and the border is obscure. Ulcerated tumors are also the findings of metastatic lesions, but the marginal elevation typical of primary cancers is rarely observed

**Peritonitis carcinomatosa**: Peritonitis carcinomatosa does not affect the mucosal surface. **Malignant lymphoma and sarcoma**: Polypoid-type lymphomas are often smooth-surfaced. Colon cancer ulcers are usually accompanied by irregular margins, but that of ulcerativetype malignant lymphoma often contains a smooth margin (Fig. 12B, C). Sarcomas are likely

(A) (B) (C) (D)

**Carcinoids**: Carcinoids are covered with the normal mucosa initially. In addition, carcinoids are usually elastic, hard, and yellowish. Endoscopic ultrasound staging (EUS) is useful for

A, metastatic colon cancer from gastric cancer (poorly differentiated adenocarcinoma); B, ulcerative type of malignant lymphoma; C, elevated type of malignant lymphoma;

Elastica van Gieson staining, however, is useful for detecting venous invasion.

and grades 2 and 3 are risk factors for lymph node metastasis.

**2.6 Differential diagnosis** 

to be malignant lymphomas.

Fig. 12. Colorectal neoplasmas

generally.

(Fig. 12A).

D, carcinoid

resection with lymph node dissection. Therefore, investigation of the risk factors for lymph node metastasis is important. The risk factors for lymph node metastasis are described on the "Colorectal Cancer Treatment Guideline 2009," written by the Japanese Society for Cancer of the Colon and Rectum; these guidelines are listed below. When a risk factor for metastasis is revealed upon examination of an endoscopically resected SM cancer specimen, additional surgical resection with lymph node dissection is recommended.


In classifying the histological type of a carcinoma, the predominant pattern is adopted as its representative histological type in the Japanese classification of colorectal carcinoma (Sugihara, 2009). For example, for a tumor consisting mainly of well-differentiated carcinoma with a small portion of moderately differentiated carcinoma, a diagnosis of "welldifferentiated carcinoma" should be made. High-grade carcinomas that are poorly differentiated adenocarcinomas, signet ring cell adenocarcinomas, or mucinous carcinomas have strong proliferative and metastatic capabilities.

When it is possible to identify the muscularis mucosae, the depth of submucosal invasion is the distance between the deepest edge of the muscularis mucosae and the deepest invasion. If the muscularis mucosae cannot be identified, the depth of submucosal invasion is the distance between the surface of the tumor and the deepest invasion. In Ip lesions with disrupted muscularis mucosae, the depth of submucosal invasion is the distance between the deepest invasion and the reference line, and it is defined as the boundary between the tumor head and the pedicle. When cancer does not invade beyond the reference line, it is defined as head invasion (Fig. 11). According to Kitamura, for pedunculated SM cancer, the rate of lymph node metastasis was 0% in cases of head and stalk invasion with depths <3000 μm if lymphatic invasion was not observed. For nonpedunculated SM cancer, the rate of lymph node metastasis was also 0% if SM depth was <1000μm (Kitajima, 2004).

Fig. 11. Depth of submucosal invasion in SM cancers

Vessel invasions contain 2 types of lymphatic invasion and venous invasion. The distinction is not so easy. When cancer nests are located in lymphatic ducts lined by flat endothelial cells, it is considered positive lymphatic invasion. Cancer nests existing near an artery are very likely to represent venous invasion, which can be confirmed by determining the presence of an internal elastic membrane and plain muscle around the cancer nests. Occasionally, it is difficult to detect venous invasion by hematoxylin and eosin staining. Elastica van Gieson staining, however, is useful for detecting venous invasion.

Budding is defined as cancer nests comprising less than 5 cancer cells and invading the interstitial tissue of the cancer growth front. The area where budding appears most frequently is selected and the number of instances is counted in a ×200 field. Budding is classified into 3 groups (grade 1, 0–4 pieces; grade 2, 5–9 pieces; and grade 3, ≥10 pieces), and grades 2 and 3 are risk factors for lymph node metastasis.

#### **2.6 Differential diagnosis**

336 Colorectal Cancer – From Prevention to Patient Care

resection with lymph node dissection. Therefore, investigation of the risk factors for lymph node metastasis is important. The risk factors for lymph node metastasis are described on the "Colorectal Cancer Treatment Guideline 2009," written by the Japanese Society for Cancer of the Colon and Rectum; these guidelines are listed below. When a risk factor for metastasis is revealed upon examination of an endoscopically resected SM cancer specimen,

1. Carcinoma that is histologically classified as poorly differentiated adenocarcinoma,

In classifying the histological type of a carcinoma, the predominant pattern is adopted as its representative histological type in the Japanese classification of colorectal carcinoma (Sugihara, 2009). For example, for a tumor consisting mainly of well-differentiated carcinoma with a small portion of moderately differentiated carcinoma, a diagnosis of "welldifferentiated carcinoma" should be made. High-grade carcinomas that are poorly differentiated adenocarcinomas, signet ring cell adenocarcinomas, or mucinous carcinomas

When it is possible to identify the muscularis mucosae, the depth of submucosal invasion is the distance between the deepest edge of the muscularis mucosae and the deepest invasion. If the muscularis mucosae cannot be identified, the depth of submucosal invasion is the distance between the surface of the tumor and the deepest invasion. In Ip lesions with disrupted muscularis mucosae, the depth of submucosal invasion is the distance between the deepest invasion and the reference line, and it is defined as the boundary between the tumor head and the pedicle. When cancer does not invade beyond the reference line, it is defined as head invasion (Fig. 11). According to Kitamura, for pedunculated SM cancer, the rate of lymph node metastasis was 0% in cases of head and stalk invasion with depths <3000 μm if lymphatic invasion was not observed. For nonpedunculated SM cancer, the rate of

> mm mp

> > mm mp

mm

lymph node metastasis was also 0% if SM depth was <1000μm (Kitajima, 2004).

additional surgical resection with lymph node dissection is recommended.

signet ring cell adenocarcinoma, or mucinous carcinoma

2. Depth of submucosal invasion >1000 μm

have strong proliferative and metastatic capabilities.

mm, muscularis mucosae; mp, muscularis propria

Fig. 11. Depth of submucosal invasion in SM cancers

3. Vascular invasion is positive

4. Grade 2/3 budding

There are many diseases that must be differentiated from colorectal carcinoma. The characteristic appearance of colorectal carcinoma does not cause interpretive difficulties generally.

**Metastatic lesions in the large intestine:** Cancers that frequently metastasize to the large intestine include those of the stomach, pancreas, ovaries, lung, and breasts in descending order of frequency. The endoscopic appearance of metastatic lesions generally includes (1) extraluminal masses with or without hyperemia, (2) wall thickening, and (3) hyperemia. Extraluminal masses are often smooth, but the base is not distinct. Mucosal hyperemias are often multiple, and the border is obscure. Ulcerated tumors are also the findings of metastatic lesions, but the marginal elevation typical of primary cancers is rarely observed (Fig. 12A).

**Peritonitis carcinomatosa**: Peritonitis carcinomatosa does not affect the mucosal surface.

**Malignant lymphoma and sarcoma**: Polypoid-type lymphomas are often smooth-surfaced. Colon cancer ulcers are usually accompanied by irregular margins, but that of ulcerativetype malignant lymphoma often contains a smooth margin (Fig. 12B, C). Sarcomas are likely to be malignant lymphomas.

A, metastatic colon cancer from gastric cancer (poorly differentiated adenocarcinoma); B, ulcerative type of malignant lymphoma; C, elevated type of malignant lymphoma; D, carcinoid

Fig. 12. Colorectal neoplasmas

**Carcinoids**: Carcinoids are covered with the normal mucosa initially. In addition, carcinoids are usually elastic, hard, and yellowish. Endoscopic ultrasound staging (EUS) is useful for

Endoscopic Diagnosis and Treatment for Colorectal Cancer 339

through the middle or deep submucosa. It is a less invasive treatment option for colorectal lesions even if the lesion is flat and difficult to remove by snare polypectomy. When we want to resect early colorectal carcinoma surrounded by the normal mucosa, EMR is a suitable procedure. After observation of the lesion, hypertonic saline solution with epinephrine is injected into the submucosal layer. At this time, the lesion-lifted condition is observed as follows (see section 3.3). When the lifted condition is complete, the bulging lesion is captured in a surgical snare and removed by cauterization with a high-frequency current. A lesion less than 2.0 cm in diameter can generally be resected en bloc. The relationship between tumor size and the en bloc resection rate in our medical center between 2000 and 2010 is presented in Fig. 13. Lesions smaller than 20 mm in diameter were resected en bloc in more than 90% of colorectal carcinoma cases. Most lesions larger than 26 mm in diameter

Fig. 13. Relationship between lesion size and en bloc resection (EMR) vs. EPMP

currently resected using ESD at our medical center.

considered.

stage carcinomas can be safely and effectively resected by EMR.

Among 281 cases of early colorectal carcinomas treated by EMR, only 2 cases recurred during the observation period of 5 years (recurrence rate, 0.8%). The recurrent cases could be retreated by endoscopic methods. Conversely, 8 cases among 148 cases treated by EPMR recurred (recurrence rate, 5.4%), and 2 of these cases were treated surgically. Therefore, the recurrence rate of EMR is significantly lower than that of EPMR (p<0.01). However, these methods are used to treat lesions of different sizes, and thus, simple comparisons of recurrence rates can be misleading. However, we believe that en bloc resection is better than piecemeal resection from the point of view of accurate pathological diagnosis. Large superficial carcinomas without lymph node metastasis requiring en bloc resection are

Bergmann (2003) reported that local recurrence after EMR was observed in 2 of 59 completely resected adenomas and in 0 of 6 early-stage carcinomas during a mean followup of 18 months. He concluded that advanced non-polypoid colorectal adenomas and early-

Jin et al. (2009) reported that recurrence was found to be related to piecemeal resection and diameters larger than 20 mm and that >20-mm-diameter is an independent risk factor for laterally spreading tumors (LSTs) treated by EMR. They stated that for the LSTs larger than 20 mm in diameter, another method, such as ESD or even a major operation, should be

are treated by the piecemeal method.

diagnosing the internal state and present layer. If they invade mucosa, vessels and ulcer formation occur (Fig.12D)

**Inflammatory disease**: Inflammatory colonic diseases, such as Crohn disease (CD), ulcerative colitis (UC), and colonic tuberculosis (TBc), are sometimes likely to resemble colonic carcinoma regarding its endoscopic features. CD often includes longitudinal ulceration, but the ulcer is sharp and smooth. UC is rarely displayed as a self-limited ulcer, but ulcers of UC are soft and thin. Colonic TBc is rarely similar to Type 2 cancer, but the ulcers of TBc are not hard or irregular.

#### **3. Endoscopic treatment for colorectal carcinoma and its precursors**

There are many methods of treating colorectal tumors, such as hot biopsy and snare polypectomy, EMR, EPMR, and ESD. Advances in endoscopic instruments and techniques have led to a large increase in the number of endoscopically resected lesions. Safe and reliable endoscopic treatment for colorectal carcinoma requires diagnostic ability and skill in colonoscopy.

#### **3.1 Hot biopsy and polypectomy**

Colonic polyps less than 0.8 cm in diameter are usually removed by hot biopsy, particularly when they are sessile, whereas polyps more than 0.8 cm in diameter are usually removed by snare polypectomy, particularly when they are pedunculated (Mann, 1999). Hot biopsy involves grasping the top part of the polyp upward and moderately cauterizing the polyp. Hot biopsy is performed cautiously in the cecum using a low amplitude and brief duration of current because the colonic wall is the thinnest and most vulnerable to transmural necrosis in this region (Weston, 1995). One limitation of hot biopsy is that only a portion of the polyp can be examined pathologically because the polyp cannot be completed removed. Therefore, we also perform snare polypectomy for even polyps smaller than 0.8 cm in diameter.

Snare polypectomy is chiefly applied to pedunculated and sessile lesions of 0.5 to 2.0 cm in diameter. Sessile polyps between 2 and 3 cm in diameter may be removed by snare polypectomy after creating a pseudopedicle by injecting normal saline or other solution into the polyp base, as described in the following section (Waye, 1997; Kato, 2001, 2008). Sessile polyps more than 3 cm in diameter may be unamenable to conventional snare polypectomy but can be removed by sequential piecemeal polypectomy over several colonoscopies (Dell'Abate, 2001).

The complication rate of therapeutic colonoscopy is 1.4–2.0% (Jentschura, 1994; Nelson, 2002; Kato, 2008). The most common postpolypectomy complications are gastrointestinal bleeding, colonic perforation, and local peritonitis. In local peritonitis, a patient develops abdominal pain, leukocytosis, and localized peritoneal irritation from an almost transmural burn occurring during polypectomy. This occurs in up to 1% of polypectomies (Waye, 1996). This syndrome is usually managed medically by the cessation of oral intake, intravenous hydration, and antibiotic administration (Waye, 1993).

#### **3.2 EMR and EPMR**

EMR combines the classic principles of conventional snare polypectomy with submucosal injection to remove more deeply affected mucosa or submucosa by resecting the lesion are treated by the piecemeal method.

338 Colorectal Cancer – From Prevention to Patient Care

diagnosing the internal state and present layer. If they invade mucosa, vessels and ulcer

**Inflammatory disease**: Inflammatory colonic diseases, such as Crohn disease (CD), ulcerative colitis (UC), and colonic tuberculosis (TBc), are sometimes likely to resemble colonic carcinoma regarding its endoscopic features. CD often includes longitudinal ulceration, but the ulcer is sharp and smooth. UC is rarely displayed as a self-limited ulcer, but ulcers of UC are soft and thin. Colonic TBc is rarely similar to Type 2 cancer, but the

There are many methods of treating colorectal tumors, such as hot biopsy and snare polypectomy, EMR, EPMR, and ESD. Advances in endoscopic instruments and techniques have led to a large increase in the number of endoscopically resected lesions. Safe and reliable endoscopic treatment for colorectal carcinoma requires diagnostic ability and skill in

Colonic polyps less than 0.8 cm in diameter are usually removed by hot biopsy, particularly when they are sessile, whereas polyps more than 0.8 cm in diameter are usually removed by snare polypectomy, particularly when they are pedunculated (Mann, 1999). Hot biopsy involves grasping the top part of the polyp upward and moderately cauterizing the polyp. Hot biopsy is performed cautiously in the cecum using a low amplitude and brief duration of current because the colonic wall is the thinnest and most vulnerable to transmural necrosis in this region (Weston, 1995). One limitation of hot biopsy is that only a portion of the polyp can be examined pathologically because the polyp cannot be completed removed. Therefore, we also perform snare polypectomy for even polyps smaller than 0.8 cm in

Snare polypectomy is chiefly applied to pedunculated and sessile lesions of 0.5 to 2.0 cm in diameter. Sessile polyps between 2 and 3 cm in diameter may be removed by snare polypectomy after creating a pseudopedicle by injecting normal saline or other solution into the polyp base, as described in the following section (Waye, 1997; Kato, 2001, 2008). Sessile polyps more than 3 cm in diameter may be unamenable to conventional snare polypectomy but can be removed by sequential piecemeal polypectomy over several colonoscopies

The complication rate of therapeutic colonoscopy is 1.4–2.0% (Jentschura, 1994; Nelson, 2002; Kato, 2008). The most common postpolypectomy complications are gastrointestinal bleeding, colonic perforation, and local peritonitis. In local peritonitis, a patient develops abdominal pain, leukocytosis, and localized peritoneal irritation from an almost transmural burn occurring during polypectomy. This occurs in up to 1% of polypectomies (Waye, 1996). This syndrome is usually managed medically by the cessation of oral intake, intravenous

EMR combines the classic principles of conventional snare polypectomy with submucosal injection to remove more deeply affected mucosa or submucosa by resecting the lesion

**3. Endoscopic treatment for colorectal carcinoma and its precursors** 

formation occur (Fig.12D)

colonoscopy.

diameter.

(Dell'Abate, 2001).

**3.2 EMR and EPMR** 

hydration, and antibiotic administration (Waye, 1993).

ulcers of TBc are not hard or irregular.

**3.1 Hot biopsy and polypectomy** 

through the middle or deep submucosa. It is a less invasive treatment option for colorectal lesions even if the lesion is flat and difficult to remove by snare polypectomy. When we want to resect early colorectal carcinoma surrounded by the normal mucosa, EMR is a suitable procedure. After observation of the lesion, hypertonic saline solution with epinephrine is injected into the submucosal layer. At this time, the lesion-lifted condition is observed as follows (see section 3.3). When the lifted condition is complete, the bulging lesion is captured in a surgical snare and removed by cauterization with a high-frequency current. A lesion less than 2.0 cm in diameter can generally be resected en bloc. The relationship between tumor size and the en bloc resection rate in our medical center between 2000 and 2010 is presented in Fig. 13. Lesions smaller than 20 mm in diameter were resected en bloc in more than 90% of colorectal carcinoma cases. Most lesions larger than 26 mm in diameter

Fig. 13. Relationship between lesion size and en bloc resection (EMR) vs. EPMP

Among 281 cases of early colorectal carcinomas treated by EMR, only 2 cases recurred during the observation period of 5 years (recurrence rate, 0.8%). The recurrent cases could be retreated by endoscopic methods. Conversely, 8 cases among 148 cases treated by EPMR recurred (recurrence rate, 5.4%), and 2 of these cases were treated surgically. Therefore, the recurrence rate of EMR is significantly lower than that of EPMR (p<0.01). However, these methods are used to treat lesions of different sizes, and thus, simple comparisons of recurrence rates can be misleading. However, we believe that en bloc resection is better than piecemeal resection from the point of view of accurate pathological diagnosis. Large superficial carcinomas without lymph node metastasis requiring en bloc resection are currently resected using ESD at our medical center.

Bergmann (2003) reported that local recurrence after EMR was observed in 2 of 59 completely resected adenomas and in 0 of 6 early-stage carcinomas during a mean followup of 18 months. He concluded that advanced non-polypoid colorectal adenomas and earlystage carcinomas can be safely and effectively resected by EMR.

Jin et al. (2009) reported that recurrence was found to be related to piecemeal resection and diameters larger than 20 mm and that >20-mm-diameter is an independent risk factor for laterally spreading tumors (LSTs) treated by EMR. They stated that for the LSTs larger than 20 mm in diameter, another method, such as ESD or even a major operation, should be considered.

Endoscopic Diagnosis and Treatment for Colorectal Cancer 341

to fall into the CLH category. On the other hand, type IIa+IIc is relatively common type of morphology among ICL and NL lesions. The relationship between lifted conditions from CLS to ICL lesions and their corresponding macroscopic types are statistically significant

 Ip Isp Is IIa IIc IIa+IIc LST CLS 4 30 66 324 24 9 105 CLH 25 73 90 77 13 8 18 ICL 0 10 22 21 7 12 13 NL 0 0 8 2 1 5 2

Classification of submucosal invasion is based on the division of the submucosa into 3 layers from sm1 to sm3. sm1, sm2, and sm3 are lesions that are limited to the upper, middle, and lower thirds of the submucosal layer, respectively. sm1 lesions are further subdivided into 3 categories (a, b, and c) with regard to the degree of horizontal involvement of the upper submucosal layer (ratio of involved part and non-involved part). Whereas sm1a+sm1b lesions have a very low risk for metastasis, the malignant potential increases with increasing depth of submucosal invasion (Kudo, 1997; Kashida, 2006). The relationship between the lesion-lifted condition and the depth of invasion is shown in Table 2. We used Kudo's classification to subclassify the depth of SM cancer. All CLS lesions are found to be sm1 or shallower, whereas the CLH category included 20 sm2 and 14 sm3 lesions. The rate of SM massive cancer (sm2 or sm3) among CLH lesions was 11.0%. ICL lesions range from sm1 to sm3, and most of the NL lesions exhibit invasion to sm3 or deeper. Four noncancerous cases and 1 mucosal cancer case of NL lesions were recurred adenomas and cancers that had previously been treated by endoscopic therapy. The rates of SM massive cancer among ICL and NL lesions are 38.5% and 74.1%, respectively. The lesion-lifted condition well correlates

 nc cia m sm1 sm2 sm3- CLS 451 65 34 3 0 0 CLH 123 75 58 18 20 14 ICL 20 14 7 7 10 20 NL 4 0 1 0 1 13

Table 2. Relationship between the lesion-lifted condition and the depth of invasion (Number

When the lifted condition of a lesion is CLS or CLH, it is a good indication that endoscopic resection will be successful. When the lifted condition is ICL, however, endoscopic resection has a smaller chance of success. And almost all of NL lesion without previous endoscopic

nc, noncancerous lesion; cia, cancer in adenoma; m, mucosal cancer; sm1, shallow SM cancer;

Table 1. Relationship between lifted condition and macroscopic type (p<0.001)

(p<0.001).

with the depth of invasion.

Chi-square test, p < 0.0001

of the lesions)

sm2, moderate SM cancer; sm3, deep SM cancer

therapy had better receive surgical treatment.

#### **3.3 Classification of lesion-lifted condition**

Special findings such as depression, ulceration, fold convergence, bleeding tendency, irregular shape, and a non-lifting sign indicate a deep invasion (Uno, 1994; Kobayashi, 2007). The "non-lifting sign" is a simple yes/no classification, and compared to EUS, it is a much easier method to determine whether EMR is indicated. However, because it is by no means rare for submucosal invasion to be found among lesions that exhibit a negative lifting sign, we have created a more detailed classification of the lesion-lifted condition. In a previous study, we focused on the tumor's lifted condition after submucosal injection and classified lesions into 4 types (Kato, 2001).

Fig. 14. Classification of the lesion-lifted condition (Kato, 2001)

This classification is closely related to the depth of invasion, and has proved to be particularly useful in the identification of early colorectal cancers that are good candidates for endoscopic treatment without requiring any special apparatus. We classified the lesionlifted condition at the time of submucosal fluid injection into 4 categories (Fig. 14): (1) completely lifted/soft (CLS), (2) completely lifted/hard (CLH), (3) incompletely lifted (ICL), and (4) non-lifting (NL). A CLS lesion is completely lifted by submucosal injection, and it stretches softly like a dome. A CLH lesion is completely lifted, but it is rigid and maintains its original form. An ICL lesion is slightly lifted, but the surrounding mucosa lifts higher than the lesion. An NL lesion is not lifted, and only the surrounding mucosa is elevated (Fig. 15).

Fig. 15. Classification of the lesion-lifted condition (Kato, 2001)

Lesion-lifted conditions are related to tumor pathology and the extent of tumor invasion, and they often correspond to particular macroscopic types of tumors. The relationship between the lifted condition and macroscopic type is shown in Table 1. Type IIa predominates among CLS lesions, whereas elevated lesions such as types Ip, Isp, and Is tend

Special findings such as depression, ulceration, fold convergence, bleeding tendency, irregular shape, and a non-lifting sign indicate a deep invasion (Uno, 1994; Kobayashi, 2007). The "non-lifting sign" is a simple yes/no classification, and compared to EUS, it is a much easier method to determine whether EMR is indicated. However, because it is by no means rare for submucosal invasion to be found among lesions that exhibit a negative lifting sign, we have created a more detailed classification of the lesion-lifted condition. In a previous study, we focused on the tumor's lifted condition after submucosal injection and

This classification is closely related to the depth of invasion, and has proved to be particularly useful in the identification of early colorectal cancers that are good candidates for endoscopic treatment without requiring any special apparatus. We classified the lesionlifted condition at the time of submucosal fluid injection into 4 categories (Fig. 14): (1) completely lifted/soft (CLS), (2) completely lifted/hard (CLH), (3) incompletely lifted (ICL), and (4) non-lifting (NL). A CLS lesion is completely lifted by submucosal injection, and it stretches softly like a dome. A CLH lesion is completely lifted, but it is rigid and maintains its original form. An ICL lesion is slightly lifted, but the surrounding mucosa lifts higher than the lesion. An NL lesion is not lifted, and only the surrounding mucosa is elevated (Fig. 15).

CLS CLH ICL NL

Lesion-lifted conditions are related to tumor pathology and the extent of tumor invasion, and they often correspond to particular macroscopic types of tumors. The relationship between the lifted condition and macroscopic type is shown in Table 1. Type IIa predominates among CLS lesions, whereas elevated lesions such as types Ip, Isp, and Is tend

**3.3 Classification of lesion-lifted condition** 

classified lesions into 4 types (Kato, 2001).

Fig. 14. Classification of the lesion-lifted condition (Kato, 2001)

Fig. 15. Classification of the lesion-lifted condition (Kato, 2001)

to fall into the CLH category. On the other hand, type IIa+IIc is relatively common type of morphology among ICL and NL lesions. The relationship between lifted conditions from CLS to ICL lesions and their corresponding macroscopic types are statistically significant (p<0.001).


Table 1. Relationship between lifted condition and macroscopic type (p<0.001)

Classification of submucosal invasion is based on the division of the submucosa into 3 layers from sm1 to sm3. sm1, sm2, and sm3 are lesions that are limited to the upper, middle, and lower thirds of the submucosal layer, respectively. sm1 lesions are further subdivided into 3 categories (a, b, and c) with regard to the degree of horizontal involvement of the upper submucosal layer (ratio of involved part and non-involved part). Whereas sm1a+sm1b lesions have a very low risk for metastasis, the malignant potential increases with increasing depth of submucosal invasion (Kudo, 1997; Kashida, 2006). The relationship between the lesion-lifted condition and the depth of invasion is shown in Table 2. We used Kudo's classification to subclassify the depth of SM cancer. All CLS lesions are found to be sm1 or shallower, whereas the CLH category included 20 sm2 and 14 sm3 lesions. The rate of SM massive cancer (sm2 or sm3) among CLH lesions was 11.0%. ICL lesions range from sm1 to sm3, and most of the NL lesions exhibit invasion to sm3 or deeper. Four noncancerous cases and 1 mucosal cancer case of NL lesions were recurred adenomas and cancers that had previously been treated by endoscopic therapy. The rates of SM massive cancer among ICL and NL lesions are 38.5% and 74.1%, respectively. The lesion-lifted condition well correlates with the depth of invasion.


nc, noncancerous lesion; cia, cancer in adenoma; m, mucosal cancer; sm1, shallow SM cancer; sm2, moderate SM cancer; sm3, deep SM cancer Chi-square test, p < 0.0001

Table 2. Relationship between the lesion-lifted condition and the depth of invasion (Number of the lesions)

When the lifted condition of a lesion is CLS or CLH, it is a good indication that endoscopic resection will be successful. When the lifted condition is ICL, however, endoscopic resection has a smaller chance of success. And almost all of NL lesion without previous endoscopic therapy had better receive surgical treatment.

Endoscopic Diagnosis and Treatment for Colorectal Cancer 343

the submucosal layer, grasping and soft coagulation are performed using coagulation forceps. Furthermore, a surrounding incision is made, and submucosal dissection is performed while lifting up the dissected part of the tumor with the edge of the transparent cap at the tip of the scope. Finally, hemostasis and the lack of a weak point of the muscularis

(1) (2) (3) (4)

(5) (6) (7) (8)

The usefulness of new grasping-type scissor forceps (GSF) such as ClutchCutter™ (Fujifilm Corp., Tokyo, Japan) was reported by Akahoshi et al. (2010). ESD using GSF is a safe (no intraoperative complication) and technically efficient (curative en bloc resection rate, 92%) method for the dissection of early gastrointestinal tumors. The use of GSF is a promising option for performing ESD in early-stage GI tract tumors both safely and effectively. We typically use GSF on the lesions that are difficult to approach or control by endoscopy. The ability to confirm that GSF is not grasping the muscle layer before coagulation or cutting is a

We investigated 116 patients with colorectal lesions for whom ESD was performed between Jan 2005 and Mar 2011. The tumors were entirely located in the large intestine (27 in the

Fig. 17. ESD using FlushKnife BTTM for rectal mucosal carcinoma of 80mm in diameter

Fig. 18. ESD using Clutchcutter™ (Fujifilm Corp., Tokyo, Japan)

propria are confirmed after resection.

point of safety (Fig.18).

**3.4.2 Cases of ESD** 

#### **3.4 ESD**

ESD is a resection technique for superficial neoplastic lesions of the gastrointestinal tract without the use of snaring. It was developed for en bloc resection of large superficial mucosal tumors, and it was initially used in the stomach and later in the esophagus, colon, and rectum. ESD is superior to EMR for a more reliable en bloc resection of a targeted area of the mucosa. It also provides a higher complete resection rate with a lower recurrence rate compared with EPMR (Saito, 2010). The drawbacks of ESD include that it is a timeconsuming procedure, has greater technical demands, and has a higher rate of perforation. For these reasons, ESD for colorectal tumor is performed as an advanced medical treatment because it is currently not recognized as treatment covered by the national health insurance system of Japan in 2011.

#### **3.4.1 Method of ESD**

Method of ESD is incising mucosa around a lesion lifted by injected fluid and dissecting the submucosal space under the lesion (Fig.16). ESD for colorectal tumors is considered more technically demanding than ESD in the stomach for a variety of reasons including the following: (1) the colonic wall is thinner and softer than the gastric wall; (2) endoscopic control is difficult in some parts of the colon because of its meandering form; and (3) there are limitations in the retroflex approach due to the narrow lumen of the colon, and tumors can be located on or behind a prominent fold of the colon.

Fig. 16. Method of ESD

First, the borders of the tumors are determined by chromoendoscopy with indigo carmine spraying for enhanced or magnified observation using NBI. Marking around the tumor is not necessary in most cases because colorectal neoplasms typically have clear margins.

The use of 0.4% sodium hyaluronate solution for submucosal injection keeps the tumor lifted for long periods (Yamamoto, 1999). For successful ESD, the position of the patient should be selected such that the lesion is located at the top of the colonic lumen with regard to gravity. Because the lesion is naturally pulled down and blood flows down from the bleeding point by gravity, good visualization of submucosal space can be maintained.

Next, the mucosal incision in front of the tumor is made with a short needle knife such as FlushKnife BTTM (1.5 mm; Fujifilm Corp., Tokyo, Japan) (Fig.17). Only the needle part should be used for the incision, keeping the tip of the sheath touching the surface of the mucosa without pushing the sheath into the submucosal layer. We use the endcut mode of electric surgical unit for the mucosal incision. After repeated submucosal injection, submucosal dissection is performed parallel to the muscular layer by sliding the knife from the center to the side while hooking submucosal fibers with the knife. We use the swift coagulation mode of electric surgical unit at this time. When thick vessels can be observed in

ESD is a resection technique for superficial neoplastic lesions of the gastrointestinal tract without the use of snaring. It was developed for en bloc resection of large superficial mucosal tumors, and it was initially used in the stomach and later in the esophagus, colon, and rectum. ESD is superior to EMR for a more reliable en bloc resection of a targeted area of the mucosa. It also provides a higher complete resection rate with a lower recurrence rate compared with EPMR (Saito, 2010). The drawbacks of ESD include that it is a timeconsuming procedure, has greater technical demands, and has a higher rate of perforation. For these reasons, ESD for colorectal tumor is performed as an advanced medical treatment because it is currently not recognized as treatment covered by the national health insurance

Method of ESD is incising mucosa around a lesion lifted by injected fluid and dissecting the submucosal space under the lesion (Fig.16). ESD for colorectal tumors is considered more technically demanding than ESD in the stomach for a variety of reasons including the following: (1) the colonic wall is thinner and softer than the gastric wall; (2) endoscopic control is difficult in some parts of the colon because of its meandering form; and (3) there are limitations in the retroflex approach due to the narrow lumen of the colon, and tumors

First, the borders of the tumors are determined by chromoendoscopy with indigo carmine spraying for enhanced or magnified observation using NBI. Marking around the tumor is not necessary in most cases because colorectal neoplasms typically have clear margins. The use of 0.4% sodium hyaluronate solution for submucosal injection keeps the tumor lifted for long periods (Yamamoto, 1999). For successful ESD, the position of the patient should be selected such that the lesion is located at the top of the colonic lumen with regard to gravity. Because the lesion is naturally pulled down and blood flows down from the bleeding point by gravity, good visualization of submucosal space can be maintained. Next, the mucosal incision in front of the tumor is made with a short needle knife such as FlushKnife BTTM (1.5 mm; Fujifilm Corp., Tokyo, Japan) (Fig.17). Only the needle part should be used for the incision, keeping the tip of the sheath touching the surface of the mucosa without pushing the sheath into the submucosal layer. We use the endcut mode of electric surgical unit for the mucosal incision. After repeated submucosal injection, submucosal dissection is performed parallel to the muscular layer by sliding the knife from the center to the side while hooking submucosal fibers with the knife. We use the swift coagulation mode of electric surgical unit at this time. When thick vessels can be observed in

can be located on or behind a prominent fold of the colon.

**3.4 ESD** 

system of Japan in 2011.

**3.4.1 Method of ESD** 

Fig. 16. Method of ESD

the submucosal layer, grasping and soft coagulation are performed using coagulation forceps. Furthermore, a surrounding incision is made, and submucosal dissection is performed while lifting up the dissected part of the tumor with the edge of the transparent cap at the tip of the scope. Finally, hemostasis and the lack of a weak point of the muscularis propria are confirmed after resection.

Fig. 17. ESD using FlushKnife BTTM for rectal mucosal carcinoma of 80mm in diameter

The usefulness of new grasping-type scissor forceps (GSF) such as ClutchCutter™ (Fujifilm Corp., Tokyo, Japan) was reported by Akahoshi et al. (2010). ESD using GSF is a safe (no intraoperative complication) and technically efficient (curative en bloc resection rate, 92%) method for the dissection of early gastrointestinal tumors. The use of GSF is a promising option for performing ESD in early-stage GI tract tumors both safely and effectively. We typically use GSF on the lesions that are difficult to approach or control by endoscopy. The ability to confirm that GSF is not grasping the muscle layer before coagulation or cutting is a point of safety (Fig.18).

Fig. 18. ESD using Clutchcutter™ (Fujifilm Corp., Tokyo, Japan)

#### **3.4.2 Cases of ESD**

We investigated 116 patients with colorectal lesions for whom ESD was performed between Jan 2005 and Mar 2011. The tumors were entirely located in the large intestine (27 in the

Endoscopic Diagnosis and Treatment for Colorectal Cancer 345

The mean operation times of the 3 methods were also compared. ESD required about 75 minutes to perform and longer than the other techniques. Regarding ESD complications, postoperative hemorrhage is not frequent, but perforation and muscularis propria incision are more common with ESD than with EMR or EPMR. However, all perforations and muscle incisions could be closed by endoscopic clipping, and there was no negative effect in

endoscopic hemostasis 16(1.5%) 6(4.1%) 3(2.6%)

Perforation: endoscopic closure 1(0.1) 0 8(6.9)

Perforation: surgical closure 1(0.1) 0 0

endoscopic clipping 0 0 7(6.0)

Local peritonitis: conservative therapy 2(0.2) 2(1.4) 2(1.7)

Colorectal ESD can be performed in all sites of the large intestine, and even a large lesion could be resected en broc using ESD. However, the procedure was lengthy and involved more complications than did other treatments. Further technical proficiency and

New methods of endoscopic diagnosis and treatment have been recently developed. Patients with early-stage colorectal carcinoma can be diagnosed by colonoscopy. Endoscopic treatment facilitates healing, and the method is less invasive, more cost-effective, and less time-consuming for patients. Endoscopic apparatuses, devices, and techniques must be further improved in the near future. Endoscopy for colorectal carcinoma will remain

Akahoshi, K. & Akahane, H. (2010). A new breakthrough: ESD using a newly developed

Bergmann, U. & Beger, HG. (2003). Endoscopic mucosal resection for advanced non-

 *Gastrointest Endosc,* Vol. 2. No. 3. pp. 90-96, ISSN 1948-5190

17. No. 3. pp. 475-479, DOI: 10.1007/s00464-002-8931-6

grasping type scissor forceps for early gastrointestinal tract neoplasms. *World J* 

polypoid colorectal adenoma and early stage carcinoma. Surgical Endoscopy, Vol.

Table 5. Complications of endoscopic treatments for colorectal tumor

instrumental improvements are expected in the future.

important in medical education and practice.

(n=1039)

EPMR (n=147)

ESD (n=116)

the clinical course (Table 5).

Postoperative hemorrhage:

Incision of muscularis propria:

**4. Conclusions** 

**5. References** 

EMR

transverse colon and 25 in the rectum) (Table 3). Type IIa was the most common macroscopic type (Table 4). The average diameter was approximately 30 mm (range, 4–82 mm). The average operation time was 75 min. Regarding complications, an incision in the muscularis propria was found in 6% cases. Perforation was experienced in 7%. But all of the perforation hole could be closed by endoscopic clip without surgical procedure.

Approximately half the lesions were adenomas, and the rest were carcinomas. One patient with carcinoma in situ exhibited recurrence in the mucosa and received endoscopic treatment. Additional colectomy was performed in 8 patients with submucosal invasion. There are pathological residual cancer nests in 2 cases. One patient had persistent carcinoma in the colonic wall, and another had lymph node metastasis.


C, cecum; A, ascending colon; T, transverse colon; D, descending colon; S, sigmoid colon; R, rectum

Table 3. location of the ESD lesions


SMT, submucosal tumor

Table 4. Macroscopic types of ESD lesions

#### **3.4.3 Comparisons among EMR, EPMR and ESD**

Clinicopathological data were compared among EMR, EPMR, and ESD between 2000 and 2011. The size in diameter of the lesions that were treated by each technique was compared. Very large lesions can be treated by EPMR and ESD. The mean sizes of lesions treated by EMR, EPMR, and EMR were 13.1 (range, 2–45), 24.6 (4–69), and 29.7 mm (4–82), respectively (Fig. 19).

Fig. 19. (A)Mean diameter of the lesions treated by each method , (B) Mean operation time

transverse colon and 25 in the rectum) (Table 3). Type IIa was the most common macroscopic type (Table 4). The average diameter was approximately 30 mm (range, 4–82 mm). The average operation time was 75 min. Regarding complications, an incision in the muscularis propria was found in 6% cases. Perforation was experienced in 7%. But all of the

Approximately half the lesions were adenomas, and the rest were carcinomas. One patient with carcinoma in situ exhibited recurrence in the mucosa and received endoscopic treatment. Additional colectomy was performed in 8 patients with submucosal invasion. There are pathological residual cancer nests in 2 cases. One patient had persistent carcinoma

C A T D S R

27 (23)

Is IIa IIc SMT

Clinicopathological data were compared among EMR, EPMR, and ESD between 2000 and 2011. The size in diameter of the lesions that were treated by each technique was compared. Very large lesions can be treated by EPMR and ESD. The mean sizes of lesions treated by EMR, EPMR, and

(A) (B)

Fig. 19. (A)Mean diameter of the lesions treated by each method , (B) Mean operation time

EMR were 13.1 (range, 2–45), 24.6 (4–69), and 29.7 mm (4–82), respectively (Fig. 19).

29.7

82 (71)

8 (7)

9.5

28.2

EMR EPMR ESD

4 (3)

18 (16)

25 (22)

6 (5)

75.1

perforation hole could be closed by endoscopic clip without surgical procedure.

19 (16)

24 (21)

C, cecum; A, ascending colon; T, transverse colon; D, descending colon; S, sigmoid colon; R, rectum

in the colonic wall, and another had lymph node metastasis.

12 (10)

No of Lesions (%)

Table 3. location of the ESD lesions

Table 4. Macroscopic types of ESD lesions

**3.4.3 Comparisons among EMR, EPMR and ESD** 

24.6

EMR EPMR ESD

No of Lesions (%)

13.1

SMT, submucosal tumor

The mean operation times of the 3 methods were also compared. ESD required about 75 minutes to perform and longer than the other techniques. Regarding ESD complications, postoperative hemorrhage is not frequent, but perforation and muscularis propria incision are more common with ESD than with EMR or EPMR. However, all perforations and muscle incisions could be closed by endoscopic clipping, and there was no negative effect in the clinical course (Table 5).


Table 5. Complications of endoscopic treatments for colorectal tumor

Colorectal ESD can be performed in all sites of the large intestine, and even a large lesion could be resected en broc using ESD. However, the procedure was lengthy and involved more complications than did other treatments. Further technical proficiency and instrumental improvements are expected in the future.

#### **4. Conclusions**

New methods of endoscopic diagnosis and treatment have been recently developed. Patients with early-stage colorectal carcinoma can be diagnosed by colonoscopy. Endoscopic treatment facilitates healing, and the method is less invasive, more cost-effective, and less time-consuming for patients. Endoscopic apparatuses, devices, and techniques must be further improved in the near future. Endoscopy for colorectal carcinoma will remain important in medical education and practice.

#### **5. References**


Endoscopic Diagnosis and Treatment for Colorectal Cancer 347

Morson, BC. & Dawson, IMP. (1972). Gastrointestinal pathology. Oxford: Blackwell

Mueller, JD.; Bethke, B. & Stolte, M. (2002). Colorectal de novo carcinoma: a review of its

Participants in the Paris Workshop. (2003). The Paris endoscopic classification of superficial

Saito, Y.; Fukuzawa, M.; Matsuda, T.; Fukunaga, S.; Sakamoto, T.; Uraoka, T.; Nakajima, T.;

Sano, Y.; Kobayashi, M.; Hamamoto, Y. & et al.(2001). New diagnostic method based on

Sano, Y.; Horimatsu T.; Fu, KI.; Katagiri, A.; Muto, M. & Ishikawa, H. (2006). Magnifying

Sano, Y.; Ikematsu, H.; Fu, KI.; Emura, F.; Katagiri, A.; Horimatsu, T.; Kaneko, K.;

Schlemper, R. J., Hirata, I., Dixon, MF. (2002). The macroscopic classification of early

Sugihara, K.; Kusunoki, M.; Watanabe, T.; Sakai, Y.; Sekimoto, M. & Ajioka, Y. (2009).

Uno, Y.; Munakata, A. (1994). The non-lifting sign of invasive colon cancer. *Gastrointestinal* 

Waye, JD. (1993). Management of complications of colonoscopic polypectomy.

Waye, JD.; Kahn, O. & Auerbach, ME. (1996). Complications of colonoscopy and flexible

Waye, JD. (1997). New methods of polypectomy. *Gastrointest Endosc Clin N Am,* Vol. 7. pp.

Weston, AP. & Campbell, DR. (1995). Diminutive colonic polyps: histopathology, spatial

distribution, concomitant significant lesions, and treatment complications. *Am J* 

sigmoidoscopy. *Gastrointest Endosc Clin N Am*, Vol. 6. pp. 343–377

neoplasia of the digestive tract. *Endoscopy,* Vol 34, pp. 163―168

 *Archiv,* Vol. 440, No. 5, pp. 453-460, DOI: 10.1007/s00428-002-0623-z Nelson, DB.; McQuaid, KR.; Bond, JH.; Lieberman, DA.; Weiss, DG.& Johnston, TK. (2002).

*Gastrointest Endosc*, Vol. 55, pp. 307–314

2002. *Gastrointest Endosc,* Vol. 58, No6;S3―S43

2. pp. 343-352, DOI: 10.1007/s00464-009-0562-8

imaging system. *Digest Endosc,* Vol. 18**.** S44–51

endosonography. *Radiology*, Vol. 179. pp. 165-70

 *Endoscopy*, Vol. 40, pp. 485-489.

 *Gastroenterol,* Vol. 90. pp. 24–8

*Gastroenterologist,* Vol.1. pp.158–164

*Gastrointest Endosc,* Vol. 53. AB125.

Vol.23. pp. 278-283.

413–422

diagnosis, histopathology, molecular biology, and clinical relevance .*Virchows* 

Procedural success and complications of large-scale screening colonoscopy.

neoplastic lesions:esophagus, stomach, and colon―November 30 to December 1,

 Ikehara, H.; Fu, KI.; Takao Itoi, T. & Fujii, T. Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection. (2010). *Surgical Endoscopy*, Vol. 24. No.

colour imaging using narrow band imaging (NBI) system for gastrointestinal tract.

observation of microvascular architecture of colorectal lesions using a narrow band

Soetikno, R. & Yoshida, S. (2008). Meshed capillary vessels using narrow band imaging for differential diagnosis of small colorectal polyps. *Gastrointest Endosc,*

*Japanese classification of colorectal carcinoma,* Japanese society for the cancer of the colon and rectum (2nd ed) Kanehara & Co.Ltd. Tokyo. ISBN978-4-307-20244-2 Tio, TL. et al. (1991).Colorectal carcinoma: Preoperative TNM classification with

Scientific


Cho E, et al. (1993). Endoscopic ultrasonography in the diagnosis of colorectal cancer

Dell'Abate, P.; Iosca, A.; Galimberti, A.; Piccolo, P. Soliani, P. & Foggi, E. (2001). Endoscopic

Jentschura, D.; Raute, M.; Winter, J.; Henkel, T.; Kraus, M. & Manegold, BC. (1994).

Jin, HY.; Wu, K.; Ye, H.; Zhu, Y.; Zhang, J. & Ding, Y. (2009). Size over 20mm is an

Kashida, H. & Kudo, SE. (2006). Early colorectal cancer: concept, diagnosis and,

Katagiri, A.; Fu, KI.; Sano, Y.; Ikematsu, H.; Horimatsu, T.; Kaneko, K.; Muto, M. &

Kato, H.; Haga, S.; Endo, S. & et al. (2001). Lifting of lesions during EMR of early

Kato, H.; Sakamoto, T.; Yamada, R.; Tsunoda, C.; Haga, S. (2008). Endoscopic Mucosal

Kitajima, K.; Fujimori,T.; Fujii, S. & et al. (2004). Correlations between lymph node

Kobayashi, N.; Saito, Y.; Sano, Y. et al. (2007). Determining the treatment strategy for

Kudo, S. (1993). Endoscopic mucosal resection of flat and depressed types of early colorectal

Kudo, S.; Hirota, S.; Nakajima, T.; Hosobe, S.; Kusaka, H.; Kobayashi, T.; Himori, M. &

Kudo, S.; Kashida, H.; Nakajima, T.; Tamura, S. & Nakajo, K. (1997). Endoscopic

Kudo, S.; Kashida, H. Tamura, T.; Kogure, E.; Imai, Y.; Yamano, H. & et al. (2000).

diagnosing invasion depth? *Endoscopy*, Vol. 39. pp. 701-705

 of the Tumor Invasion. *Ann. Cancer Res. Therap,* Vol. 16, No. 1, pp. 25-30 Kim, EC. & Lance, P. (1997). Colorectal polyps and their relationship to cancer. *Gastroenterol* 

treatment of colorectal benign-appearing lesions 3 cm or larger: techniques and

Complications in endoscopy of the lower gastrointestinal tract: therapy and

 independent risk factor of endoscopic mucosa resection (EMR) for colorectal lateral spread tumor (LST): A prospective study and multivariate analysis. *Cancer Therapy,*

Yoshida, S. (2008). Narrow band imaging with magnifying colonoscopy as a diagnostic tool for predicting the histology of early colorectal neoplasia. *Aliment* 

colorectal cancer: implications for the assessment resectabolity. *Endoscopy,* Vol.33,

Resection (EMR) for Colorectal Lesions and Lesion-lifted Condition as an Indicator

 metastasis and depth of submucosal invasion in submucosal invasive colorectal carcinoma:a Japanese collaborative study, *J Gastroenterol*, Vol.39, pp. 534–543, DOI

colorectal neoplastic lesions: endoscopic assessment or the non-lifting sign for

Yagyuu, A. (1994). Colorectal tumours and pit pattern. *Journal of Clinical Pathology*,

diagnosis and treatment of early colorectal cancer. *World J. Surg,* Vol. 21. No. 7.

Colonoscopic diagnosis and management of nonpolypoidearly colorectal cancer.

invasion. *Gastrointest Endosc*, Vol. 39 pp. 521-527

outcome. *Dis Colon Rectum*, Vol. 44. pp. 112–118

management. *Int. J. Clin. Oncol,* Vol. 11. No. 1. pp. 1-8

prognosis. *Surg Endosc*, Vol. 8. pp. 672–676

*Pharmacol Ther,* Vol. 27. pp. 1269-1274

 *Clin North Am,* Vol. 26. pp. 1–17

cancer. *Endoscopy*, Vol.25. pp.455-461.

*World J Surg*, Vol. 24. pp.1081-1090.

10.1007/s00535-004-1339-4

Vol. 47. pp. 880-885

pp. 694-701

Vol. 7. pp. 27-30

pp. 568-573


**18** 

**Peri-Operative Care in Colorectal** 

Under conventional circumstances, colorectal cancer resection has been associated with an often protracted recovery. Large published studies, randomized trials and meta-analyses suggest an average length of hospital stay of about ten days (Bokey et al., 1995; Abraham et al., 2004 & 2007). In an attempt to mimic the success of laparoscopic gall bladder surgery, laparoscopic colorectal resection was introduced in 1991 as a proposed less invasive alternative to the open technique (Jacobs, 1991; Redwine, 1991). Under conventional circumstances, in the first published series of 20 laparoscopic sigmoid colectomies, the

However, subsequent larger studies including randomized trials reported an average length of stay of about eight days which is still an improvement of about 20% compared with conventional open resections (Abraham et al., 2004 & 2007; Schwenk et al., 2005). The last published large randomized controlled trial of the topic (The ALCCaS) showed no statistically significant difference in postoperative complications, reoperation rate, or perioperative mortality between laparoscopic and open resections (Allardyce et al., 2010). However, a recent meta-analysis showed that laparoscopic colorectal resections were associated with higher intra-operative complication rates than open resections (Sammour et al., 2011). The ALCCaS group also reported that reviews show that the short-term advantages for laparoscopic resection for colorectal cancer are arguably relatively minor and

In 1999, in a series of 16 open colectomies, the authors reported using a Fast Track (Enhanced Recovery after Surgery (ERAS)) Program with a median postoperative length of hospital stay of two days (Kehlet & Mogensen, 1999). However, subsequent larger studies reported a median length of stay of about five days (Abraham & Albayati, 2011). ERAS programs challenge the conventional approaches to peri-operative care in colorectal surgery in an evidence-based manner. These include conventional bowel preparation, peri-operative starvation, routine nasogastric decompression, routine prophylactic drainage, defunctioning ileostomy, vigorous intravenous hydration, narcotic analgesia, etc ... These traditional protocols and practices are replaced with evidence based protocols that enhance

authors reported that a five-day hospital stay was achieved in 70% of patients.

**1. Introduction** 

often subjective (Allardyce et al., 2010).

postoperative recovery.

**Surgery in the Twenty-**

**First Century** 

*University of New South Wales,* 

Ned Abraham

*Australia* 

*Coffs Harbour, NSW,* 

Yamamoto, H.; Yube, T.; Isoda, N.; Sato, Y.; Sekine, Y.; Higashizawa, T.; Ido, K.; Kimura, K. & Kanai, N. (1999). a novel method of endoscopic mucosal resection using sodium hyaluronate. *Gastroint Endosc,* Vol. 50. pp. 251–256

### **Peri-Operative Care in Colorectal Surgery in the Twenty-First Century**

Ned Abraham *University of New South Wales, Coffs Harbour, NSW, Australia* 

#### **1. Introduction**

348 Colorectal Cancer – From Prevention to Patient Care

Yamamoto, H.; Yube, T.; Isoda, N.; Sato, Y.; Sekine, Y.; Higashizawa, T.; Ido, K.; Kimura, K.

hyaluronate. *Gastroint Endosc,* Vol. 50. pp. 251–256

& Kanai, N. (1999). a novel method of endoscopic mucosal resection using sodium

Under conventional circumstances, colorectal cancer resection has been associated with an often protracted recovery. Large published studies, randomized trials and meta-analyses suggest an average length of hospital stay of about ten days (Bokey et al., 1995; Abraham et al., 2004 & 2007). In an attempt to mimic the success of laparoscopic gall bladder surgery, laparoscopic colorectal resection was introduced in 1991 as a proposed less invasive alternative to the open technique (Jacobs, 1991; Redwine, 1991). Under conventional circumstances, in the first published series of 20 laparoscopic sigmoid colectomies, the authors reported that a five-day hospital stay was achieved in 70% of patients.

However, subsequent larger studies including randomized trials reported an average length of stay of about eight days which is still an improvement of about 20% compared with conventional open resections (Abraham et al., 2004 & 2007; Schwenk et al., 2005). The last published large randomized controlled trial of the topic (The ALCCaS) showed no statistically significant difference in postoperative complications, reoperation rate, or perioperative mortality between laparoscopic and open resections (Allardyce et al., 2010). However, a recent meta-analysis showed that laparoscopic colorectal resections were associated with higher intra-operative complication rates than open resections (Sammour et al., 2011). The ALCCaS group also reported that reviews show that the short-term advantages for laparoscopic resection for colorectal cancer are arguably relatively minor and often subjective (Allardyce et al., 2010).

In 1999, in a series of 16 open colectomies, the authors reported using a Fast Track (Enhanced Recovery after Surgery (ERAS)) Program with a median postoperative length of hospital stay of two days (Kehlet & Mogensen, 1999). However, subsequent larger studies reported a median length of stay of about five days (Abraham & Albayati, 2011). ERAS programs challenge the conventional approaches to peri-operative care in colorectal surgery in an evidence-based manner. These include conventional bowel preparation, peri-operative starvation, routine nasogastric decompression, routine prophylactic drainage, defunctioning ileostomy, vigorous intravenous hydration, narcotic analgesia, etc ... These traditional protocols and practices are replaced with evidence based protocols that enhance postoperative recovery.

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 351

Preoperative starvation for eight hours or more before receiving an anesthetic has been implemented as an unchallenged rule for a very long time. This was meant to reduce the risk of aspiration pneumonitis if gastric contents were regurgitated during the course of induction of the general anesthetic. There is currently Level I evidence that shows that drinking clear fluids up to two hours prior to surgery does not increase the risk of aspiration or regurgitation as it does not increase gastric acidity or the amount of gastric secretions that

Patients were generally instructed to completely fast from midnight the night before their procedures are due to take place. This used to further complicate the semi-starvation state associated with bowel preparation. This could result in an increased catabolic state, dehydration and electrolyte imbalance especially if the procedure took place later in the morning or in the afternoon. This catabolic state is aggravated and complicated further by the surgery itself with the added negative nitrogen balance, insulin resistance and the release of stress hormones such as catecholamines, glucagon and cortisol (Nygren et al., 2001). Starvation also compromises the physiological response to hemorrhage and infection (Brady et al., 2003; Nygren et al., 2001). Patients receiving oral preoperative carbohydrate loading are more likely to have physiological postoperative insulin levels compared with those receiving glucose via the intravenous route and those fasting overnight and not

In an ERAS protocol, patients are typically allowed clear fluids up to two hours before the anesthetic and routinely "loaded" with oral carbohydrate and protein drinks and symbiotics

Again, patients have traditionally been "fasted" postoperatively until they passed flatus. Even then, they were only allowed clear fluids until they had passed a bowel motion. It was believed that such practice would minimize the risk of an anastomotic leak or make such a leak more easily manageable than if the patient were allowed to eat. This further complicated any pre-operative malnutrition (Garth et al., 2010). Bowel preparation, the strict diet that goes with it and perioperative starvation further increase the catabolic state. Furthermore, the increased immediate postoperative need for nutrients is not met resulting

There is now Level I evidence that shows that there is no benefit in postoperative starvation in terms of reducing anastomotic leak rates (Lewis et al., 2009). It is likely that enteral nutrition reduces the overall risks of wound infection and intra-abdominal sepsis, probably

In a standard ERAS protocol, patients are allowed clear fluids the evening after the procedure, free fluids on postoperative day one and a soft diet on postoperative day two regardless of the type of resection performed. We find this protocol to be well tolerated. We warn patients beforehand of the small risk of vomiting but reassure them that this would

The traditional aim of routine nasogastric intubation is to achieve gastric decompression in order to reduce the risk of postoperative ileus, vomiting and abdominal distension. This

in proteolysis, negative nitrogen balance and increased insulin resistance.

through improving the capillary-intestinal barrier (Lewis et al., 2009).

not be of serious consequence if it took place.

**3.4 No nasogastric decompression** 

**3.2 Minimal preoperative starvation** 

could be regurgitated (Brady et al., 2003; Ljungkvist et al., 2003).

receiving any carbohydrate loading (Kaska et al., 2010).

preoperatively.

**3.3 No postoperative starvation** 

#### **2. General outline of an ERAS protocol**

In an ERAS program, all the small steps that form the care package in colorectal surgery are "optimized" to achieve the best possible outcome. These include targeted preoperative interview to educate the patient on what to expect, preoperative nutritional assessment if required, minimal peri-operative starvation, preoperative carbohydrate and protein loading, no routine bowel preparation, transverse or oblique incision if seen fit by the operating surgeon, high oxygen concentrations and normothermia.

They also include avoiding excessive intravenous hydration and the routine use of nasogastric tubes and drains. Other important postoperative management issues include multimodal analgesia (epidural analgesia if seen fit by the anaesthetist, subcostal nerve block when possible, continuous wound infiltration with a local anaesthetic agent (wound soaker) and regular oral non-narcotic analgesia with minimal or no morphia and only using patient controlled applications.

Other elements of an ERAS protocol include the routline use of regular prokinetic agents, the routine use of regular anti-emetic drugs, a structured early postoperative mobilization program and early oral feeding (clear fluid intake on the evening of surgery, free fluid intake on day one and a soft diet on day two). This is all achieved through the cooperation of a team of clinicians, nursing staff, physiotherapists, stoma therapists, dieticians, etc … The general aim is to have the patient ready for discharge by postoperative day four or five.

#### **3. Supportive evidence for the main ERAS practices**

#### **3.1 No routine bowel preparation**

Mechanical bowel preparation was used for almost a century to cleanse the colon prior to surgery. The aim was to evacuate the colon, reduce the fecal load in the hope that this would – in a plausible way - reduce the bacterial load thus reducing the risk of postoperative infection and anastomotic leak rates. It was also believed that bowel preparation allowed better visualization of the lumen as well as making the anastomosis technically easier. Mechanical bowel preparation became "traditional". However, microbiological testing showed that bowel preparation did not reduce the microbial count in colonic mucosa (Jung et al., 2010). For a few decades, right hemicolectomies have been performed without bowel preparation.

Avoiding routine mechanical bowel preparation is an important component of any ERAS program. In the early seventies, Hughes showed that receiving preoperative bowel preparation made no difference to outcomes including anastomotic leak rates (Hughes, 1972). Multiple studies addressing the same questions have since been conducted (Scabini et al., 2010). A meta-analysis of outcomes following close to five thousand colorectal resections showed no evidence to suggest that bowel preparation reduced the incidence of anastomotic leakage (Guenaga et al., 2009). In fact there was a suggestion that routine bowel preparation could actually increase the risk of infections and overall complication rates associated with colorectal resections.

In an ERAS protocol, patients admitted for right sided resections receive no bowel preparation. For left sided resection, we use enema preparation the night before and the day of surgery to evacuate the rectum and the left colon to facilitate the surgery from the technical point of view. Others use normal saline enemas.

#### **3.2 Minimal preoperative starvation**

350 Colorectal Cancer – From Prevention to Patient Care

In an ERAS program, all the small steps that form the care package in colorectal surgery are "optimized" to achieve the best possible outcome. These include targeted preoperative interview to educate the patient on what to expect, preoperative nutritional assessment if required, minimal peri-operative starvation, preoperative carbohydrate and protein loading, no routine bowel preparation, transverse or oblique incision if seen fit by the operating

They also include avoiding excessive intravenous hydration and the routine use of nasogastric tubes and drains. Other important postoperative management issues include multimodal analgesia (epidural analgesia if seen fit by the anaesthetist, subcostal nerve block when possible, continuous wound infiltration with a local anaesthetic agent (wound soaker) and regular oral non-narcotic analgesia with minimal or no morphia and only using

Other elements of an ERAS protocol include the routline use of regular prokinetic agents, the routine use of regular anti-emetic drugs, a structured early postoperative mobilization program and early oral feeding (clear fluid intake on the evening of surgery, free fluid intake on day one and a soft diet on day two). This is all achieved through the cooperation of a team of clinicians, nursing staff, physiotherapists, stoma therapists, dieticians, etc … The general aim is to have the patient ready for discharge by

Mechanical bowel preparation was used for almost a century to cleanse the colon prior to surgery. The aim was to evacuate the colon, reduce the fecal load in the hope that this would – in a plausible way - reduce the bacterial load thus reducing the risk of postoperative infection and anastomotic leak rates. It was also believed that bowel preparation allowed better visualization of the lumen as well as making the anastomosis technically easier. Mechanical bowel preparation became "traditional". However, microbiological testing showed that bowel preparation did not reduce the microbial count in colonic mucosa (Jung et al., 2010). For a few decades, right hemicolectomies have been performed without bowel

Avoiding routine mechanical bowel preparation is an important component of any ERAS program. In the early seventies, Hughes showed that receiving preoperative bowel preparation made no difference to outcomes including anastomotic leak rates (Hughes, 1972). Multiple studies addressing the same questions have since been conducted (Scabini et al., 2010). A meta-analysis of outcomes following close to five thousand colorectal resections showed no evidence to suggest that bowel preparation reduced the incidence of anastomotic leakage (Guenaga et al., 2009). In fact there was a suggestion that routine bowel preparation could actually increase the risk of infections and overall complication rates associated with

In an ERAS protocol, patients admitted for right sided resections receive no bowel preparation. For left sided resection, we use enema preparation the night before and the day of surgery to evacuate the rectum and the left colon to facilitate the surgery from the

**2. General outline of an ERAS protocol** 

patient controlled applications.

postoperative day four or five.

**3.1 No routine bowel preparation** 

preparation.

colorectal resections.

surgeon, high oxygen concentrations and normothermia.

**3. Supportive evidence for the main ERAS practices** 

technical point of view. Others use normal saline enemas.

Preoperative starvation for eight hours or more before receiving an anesthetic has been implemented as an unchallenged rule for a very long time. This was meant to reduce the risk of aspiration pneumonitis if gastric contents were regurgitated during the course of induction of the general anesthetic. There is currently Level I evidence that shows that drinking clear fluids up to two hours prior to surgery does not increase the risk of aspiration or regurgitation as it does not increase gastric acidity or the amount of gastric secretions that could be regurgitated (Brady et al., 2003; Ljungkvist et al., 2003).

Patients were generally instructed to completely fast from midnight the night before their procedures are due to take place. This used to further complicate the semi-starvation state associated with bowel preparation. This could result in an increased catabolic state, dehydration and electrolyte imbalance especially if the procedure took place later in the morning or in the afternoon. This catabolic state is aggravated and complicated further by the surgery itself with the added negative nitrogen balance, insulin resistance and the release of stress hormones such as catecholamines, glucagon and cortisol (Nygren et al., 2001). Starvation also compromises the physiological response to hemorrhage and infection (Brady et al., 2003; Nygren et al., 2001). Patients receiving oral preoperative carbohydrate loading are more likely to have physiological postoperative insulin levels compared with those receiving glucose via the intravenous route and those fasting overnight and not receiving any carbohydrate loading (Kaska et al., 2010).

In an ERAS protocol, patients are typically allowed clear fluids up to two hours before the anesthetic and routinely "loaded" with oral carbohydrate and protein drinks and symbiotics preoperatively.

#### **3.3 No postoperative starvation**

Again, patients have traditionally been "fasted" postoperatively until they passed flatus. Even then, they were only allowed clear fluids until they had passed a bowel motion. It was believed that such practice would minimize the risk of an anastomotic leak or make such a leak more easily manageable than if the patient were allowed to eat. This further complicated any pre-operative malnutrition (Garth et al., 2010). Bowel preparation, the strict diet that goes with it and perioperative starvation further increase the catabolic state. Furthermore, the increased immediate postoperative need for nutrients is not met resulting in proteolysis, negative nitrogen balance and increased insulin resistance.

There is now Level I evidence that shows that there is no benefit in postoperative starvation in terms of reducing anastomotic leak rates (Lewis et al., 2009). It is likely that enteral nutrition reduces the overall risks of wound infection and intra-abdominal sepsis, probably through improving the capillary-intestinal barrier (Lewis et al., 2009).

In a standard ERAS protocol, patients are allowed clear fluids the evening after the procedure, free fluids on postoperative day one and a soft diet on postoperative day two regardless of the type of resection performed. We find this protocol to be well tolerated. We warn patients beforehand of the small risk of vomiting but reassure them that this would not be of serious consequence if it took place.

#### **3.4 No nasogastric decompression**

The traditional aim of routine nasogastric intubation is to achieve gastric decompression in order to reduce the risk of postoperative ileus, vomiting and abdominal distension. This

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 353

resections within 5 cm of the anal verge. They reported clinically significant anastomotic leaks in 12 out of 120 (10%) in the "no ileostomy" group with two patients requiring Hartman's procedures. In the "ileostomy" group, clinically significant anastomotic leaks occurred in three out of 136 (2.2%) with no patients requiring a re-operation. The authors recommended the routine use of loop ileostomy for all anastomoses within five cm of the anal verge. Experience shows that this is particularly relevant if the patient has received

These results were confirmed in a Cochrane systematic review of six randomized trials of routine ileostomies for rectal resections with anastomoses within five cm of the anal verge (Montedori et al., 2010). A defunctioning ileostomy was associated with a reduced risk of reoperation for an anastomotic leak. In another systematic review of 27 retrospective studies and four randomized trials, the authors reported that the use of a defunctioning ileostomy after low rectal resections did not reduce the incidence of an anastomotic leak but was associated with improved outcomes in terms of a reduction in clinically significant leak rates (OR=0.32(0.17-0.59); (P<0.001)) and a reduction in associated reoperation rates (OR=0.27

The electrolyte imbalance, dehydration and hypotension resulting from preoperative starvation and the use of bowel preparation are often over-compensated for with the liberal use of perioperative intravenous isotonic fluids. However, this liberal use of perioperative intravenous rehydration has been shown to be associated with an increased risk of cardiopulmonary complications, a delay in the return of gastrointestinal function and an increased length of postoperative stay in hospital (Lobo et al., 2002). On the other hand, restricting perioperative intravenous fluid therapy has been shown to hasten gastrointestinal recovery, reduce postoperative complication rates and shorten the length of

Routine spinal anesthesia was used in the original ERAS protocol described by Kehlet and his group in 1999 (Kehlet et al., 1999). However, this has evolved into the concept of multimodal analgesia as an integral part of the ERAS approach. The use of epidural analgesia with general anesthesia for major abdominal surgery has been shown to be associated with a reduced incidence of postoperative nausea and vomiting as well as lower rates of respiratory complications compared with intravenous narcotic analgesia, whether as

The use of a local anesthetic agent administered via an epidural catheter (usually as a continuous infusion and patient controlled boluses) following major abdominal surgery has also been shown to be associated with faster return of gastrointestinal function compared to intravenous and epidural narcotic analgesia to achieve the same analgesic effect (Jorgensen et al., 2000). Autonomic reflexes activated through a painful laparotomy incision cause inhibition of gastrointestinal functions. This is further aggravated by the use of narcotic

Hypothermia is quite common with general anesthesia and abdominal surgical procedures. This is due to the combination of impaired thermoregulation, exposure and the use of air

a continuous infusion and/or patient-controlled boluses (White et al., 2007).

neoadjuvant radiotherapy.

(95% CI 0.14-0.51); (p<0.001)) (Huser et al., 2008).

hospital stay (Nisanevich et al., 2005; Holte & Kehlet, 2006).

analgesia and the nausea and vomiting associated with it.

**3.10 Normothermia** 

**3.9 Multimodal postoperative analgesia** 

**3.8 Intravenous fluid restriction** 

seemed to be a plausible means of improving postoperative peristalsis in an attempt to achieve an early return to bowel function. However, Cheatham et al showed in 1995 that routine nasogastric intubation did not reduce the risk of complications or length of postoperative stay in hospital following abdominal surgery (Cheatham et al., 1995). They also showed that for every patient requiring nasogastric intubation, 20 patients will not need it.

The results of this meta-analysis were reinforced by a recent Cochrane review of studies of close to six thousand patients (Nelson et al., 2007). Routine nasogastric decompression slowed the return of bowel function and did not reduce the risk of an anastomotic leak compared with no decompression. It was also associated with a more prolonged length of stay in hospital.

#### **3.5 Consideration for a transverse or oblique incision**

It has been suggested that a transverse or an oblique incision is an important part of the practices contributing to a quick recovery (Kehlet et al., 1999). A Cochrane review suggested an overall advantage in adopting a transverse over a midline incision (Brown & Goodfellow, 2005). This finding was supported by the results of a randomized controlled trial of transverse versus longitudinal incisions for cholecystectomy (Halm et al., 2009). Right hemicolectomies have probably been more commonly performed through a transverse rather than a vertical incision for a few decades.

A recent randomized trial suggested that there was no advantage in using a transverse incision over a longitudinal incision in terms of required analgesia, pain, pulmonary complications, median length of stay, median time to tolerating a diet or one year incisional hernia rates (Seiler et al., 2009). The sample size was small and a type II error could not be excluded. However, more wound infections occurred in the transverse incision group (15% vs. 5%, P = 0.02). It is the author's experience that left sided resections are overall easier to perform through a midline incision compared with an oblique incision. The choice between midline and transverse incisions may continue to be debated for some time yet.

#### **3.6 No routine prophylactic drainage**

It has been thought that prophylactic drainage of colorectal anastomoses would reduce the risk of anastomotic leakage. This was thought to be by a process of reducing the likelihood of a postoperative collection forming near the anastomosis with the plausible risk of infection and a subsequent anastomotic leak. The presence of a drain could also make it easy to detect an anastomotic leak guided by the amount and quality of drain output.

However, multiple randomized trials and a subsequent meta-analysis failed to demonstrate a benefit for routine drainage in colorectal surgery. The systematic review referred to above include the results of 1140 colorectal resections (Jesus et al., 2004). It showed no statistically significant difference between outcomes in patients receiving routine prophylactic drainage or no drainage for colorectal resections in terms of anastomotic leakage, wound infection and all complication rates (Qadan et al., 2009). There is probably no advantage for routine prophylactic drainage of low rectal or colo-anal anastomoses either (Merad et al., 1999; Yeh et al., 2005).

#### **3.7 The limited role of a defunctioning ileostomy**

A relevant randomised trial was published in 2008 (Chude et al., 2008). The authors compared routine defunctioning loop ileostomy versus no ileostomy for low rectal resections within 5 cm of the anal verge. They reported clinically significant anastomotic leaks in 12 out of 120 (10%) in the "no ileostomy" group with two patients requiring Hartman's procedures. In the "ileostomy" group, clinically significant anastomotic leaks occurred in three out of 136 (2.2%) with no patients requiring a re-operation. The authors recommended the routine use of loop ileostomy for all anastomoses within five cm of the anal verge. Experience shows that this is particularly relevant if the patient has received neoadjuvant radiotherapy.

These results were confirmed in a Cochrane systematic review of six randomized trials of routine ileostomies for rectal resections with anastomoses within five cm of the anal verge (Montedori et al., 2010). A defunctioning ileostomy was associated with a reduced risk of reoperation for an anastomotic leak. In another systematic review of 27 retrospective studies and four randomized trials, the authors reported that the use of a defunctioning ileostomy after low rectal resections did not reduce the incidence of an anastomotic leak but was associated with improved outcomes in terms of a reduction in clinically significant leak rates (OR=0.32(0.17-0.59); (P<0.001)) and a reduction in associated reoperation rates (OR=0.27 (95% CI 0.14-0.51); (p<0.001)) (Huser et al., 2008).

#### **3.8 Intravenous fluid restriction**

352 Colorectal Cancer – From Prevention to Patient Care

seemed to be a plausible means of improving postoperative peristalsis in an attempt to achieve an early return to bowel function. However, Cheatham et al showed in 1995 that routine nasogastric intubation did not reduce the risk of complications or length of postoperative stay in hospital following abdominal surgery (Cheatham et al., 1995). They also showed that for every patient requiring nasogastric intubation, 20 patients will not need it. The results of this meta-analysis were reinforced by a recent Cochrane review of studies of close to six thousand patients (Nelson et al., 2007). Routine nasogastric decompression slowed the return of bowel function and did not reduce the risk of an anastomotic leak compared with no decompression. It was also associated with a more prolonged length of

It has been suggested that a transverse or an oblique incision is an important part of the practices contributing to a quick recovery (Kehlet et al., 1999). A Cochrane review suggested an overall advantage in adopting a transverse over a midline incision (Brown & Goodfellow, 2005). This finding was supported by the results of a randomized controlled trial of transverse versus longitudinal incisions for cholecystectomy (Halm et al., 2009). Right hemicolectomies have probably been more commonly performed through a transverse

A recent randomized trial suggested that there was no advantage in using a transverse incision over a longitudinal incision in terms of required analgesia, pain, pulmonary complications, median length of stay, median time to tolerating a diet or one year incisional hernia rates (Seiler et al., 2009). The sample size was small and a type II error could not be excluded. However, more wound infections occurred in the transverse incision group (15% vs. 5%, P = 0.02). It is the author's experience that left sided resections are overall easier to perform through a midline incision compared with an oblique incision. The choice between

It has been thought that prophylactic drainage of colorectal anastomoses would reduce the risk of anastomotic leakage. This was thought to be by a process of reducing the likelihood of a postoperative collection forming near the anastomosis with the plausible risk of infection and a subsequent anastomotic leak. The presence of a drain could also make it easy

However, multiple randomized trials and a subsequent meta-analysis failed to demonstrate a benefit for routine drainage in colorectal surgery. The systematic review referred to above include the results of 1140 colorectal resections (Jesus et al., 2004). It showed no statistically significant difference between outcomes in patients receiving routine prophylactic drainage or no drainage for colorectal resections in terms of anastomotic leakage, wound infection and all complication rates (Qadan et al., 2009). There is probably no advantage for routine prophylactic drainage of low rectal or colo-anal anastomoses either (Merad et al., 1999; Yeh

A relevant randomised trial was published in 2008 (Chude et al., 2008). The authors compared routine defunctioning loop ileostomy versus no ileostomy for low rectal

midline and transverse incisions may continue to be debated for some time yet.

to detect an anastomotic leak guided by the amount and quality of drain output.

stay in hospital.

et al., 2005).

**3.5 Consideration for a transverse or oblique incision** 

rather than a vertical incision for a few decades.

**3.6 No routine prophylactic drainage** 

**3.7 The limited role of a defunctioning ileostomy** 

The electrolyte imbalance, dehydration and hypotension resulting from preoperative starvation and the use of bowel preparation are often over-compensated for with the liberal use of perioperative intravenous isotonic fluids. However, this liberal use of perioperative intravenous rehydration has been shown to be associated with an increased risk of cardiopulmonary complications, a delay in the return of gastrointestinal function and an increased length of postoperative stay in hospital (Lobo et al., 2002). On the other hand, restricting perioperative intravenous fluid therapy has been shown to hasten gastrointestinal recovery, reduce postoperative complication rates and shorten the length of hospital stay (Nisanevich et al., 2005; Holte & Kehlet, 2006).

#### **3.9 Multimodal postoperative analgesia**

Routine spinal anesthesia was used in the original ERAS protocol described by Kehlet and his group in 1999 (Kehlet et al., 1999). However, this has evolved into the concept of multimodal analgesia as an integral part of the ERAS approach. The use of epidural analgesia with general anesthesia for major abdominal surgery has been shown to be associated with a reduced incidence of postoperative nausea and vomiting as well as lower rates of respiratory complications compared with intravenous narcotic analgesia, whether as a continuous infusion and/or patient-controlled boluses (White et al., 2007).

The use of a local anesthetic agent administered via an epidural catheter (usually as a continuous infusion and patient controlled boluses) following major abdominal surgery has also been shown to be associated with faster return of gastrointestinal function compared to intravenous and epidural narcotic analgesia to achieve the same analgesic effect (Jorgensen et al., 2000). Autonomic reflexes activated through a painful laparotomy incision cause inhibition of gastrointestinal functions. This is further aggravated by the use of narcotic analgesia and the nausea and vomiting associated with it.

#### **3.10 Normothermia**

Hypothermia is quite common with general anesthesia and abdominal surgical procedures. This is due to the combination of impaired thermoregulation, exposure and the use of air

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 355

approach would complement ERAS rehabilitation programmes. In a small, 2:1 design, randomised trial of 62 patients (43 laparoscopic and 19 open resections), the authors reported an added benefit for adopting the laparoscopic technique in an ERAS protocol in

However, the results of a systematic review of two randomised controlled trials and three controlled clinical trials of laparoscopic versus open colorectal surgery under ERAS rehabilitation programs were inconclusive as no clear advantage for laparoscopic over open resection was demonstrated under ERAS protocols (Vlug et al., 2009). Further research was

Another recently published large review of 11 studies (four randomised trials and 11 controlled clinical trials) including 1021 patients reported a clear advantage for patients enrolled in an ERAS rehabilitation program in terms of length of hospital stay compared with those who were not (Gouvas et al., 2009). Although the authors reported that an added benefit to recovery rates in adopting the laparoscopic over the open approach was assumed, such a benefit could not be established. The authors concluded that ERAS programs should

In a systematic review of three randomised trials and seven non-randomised studies of laparoscopic versus open colorectal resections under an ERAS protocol, Khan and colleagues reported that the currently available limited evidence suggests that the inclusion of laparoscopic surgery in ERAS protocols for colorectal resections does not confer an added benefit in terms of postoperative recovery rates and postoperative length of stay (Khan et

A consensus statement on ERAS was published in 2005 (Fearon et al., 2005). The statement was written by colorectal surgeons and other specialists and professionals from five universities or tertiary hospitals in five European countries (Denmark, Scotland, Sweden, Norway and The Netherlands). The authors presented their methodology in the published article with a specific focus on colorectal resections. They also recommended their protocol as one that may provide a standard of care against which current and future novel elements of an ERAS approach can be tested or added to. Members of the same group published the outcomes of 169 colorectal resections under an ERAS protocol with very good results

Figures 1-4 show a summary of an ERAS colorectal program adopted at the Coffs Harbour Health Campus, a regional hospital in New South Wales, Australia in July 2006. The summary results of 111 ERAS consecutive open colorectal resections performed at that hospital by one surgeon have been recently published with outcomes similar to those in the

The Australian Safety and Efficacy Register of New Interventional Procedures - Surgical (ASERNIP-S) under the auspices of the Royal Australasian College of Surgeons and the Department of Health and Aging – Victoria, assessed the experience of Australian and New Zealand surgeons with colorectal resection under ERAS protocols (Strum & Cameron, 2009). They concluded that ERAS programs can result in beneficial outcomes for patients by reducing the length of hospital stay with no significant increase in readmission rates. They also indicated that further work is required to assist in standardisation and implementation

terms of a reduced postoperative length of stay (King et al., 2006).

become a mainstay of elective colorectal surgery.

**6. Standardisation of a colorectal ERAS protocol** 

North European experience (Abraham & Albayati, 2011).

recommended.

al., 2009).

(Nygren et al., 2009).

of ERAS protocols.

conditioning and negative pressure ventilation in the operating rooms (Qadan et al., 2009). Hypothermia in a surgical setting is associated with an increased risk of bleeding due to coagulopathy as well as arrhythmias, myocardial ischemia and overall risks of complications (Diaz & Becker, 2010).

Under an ERAS protocol, hypothermia is actively prevented using warm and space blankets, warm intravenous infusions, avoiding unnecessary exposure, etc … The patient does not leave the recovery ward until normothermic.

#### **4. Supportive evidence for ERAS protocols**

As pointed out above, in the first published series of 16 open sigmoid colectomies under an ERAS (Fast Track) protocol, the authors reported a median postoperative length of hospital stay of two days (Kehlet & Mogensen, 1999). However, subsequent larger studies reported a median length of stay of about five days, three days longer than what was reported in the first series (Abraham & Albayati, 2011; Nygren et al., 2009).

Multiple published trials and systematic reviews have reported that ERAS protocols were associated with a faster recovery, reduced primary and overall lengths of hospital stay and complication rates after colorectal resections compared with the traditional approach. Wind et al reported that the use of an ERAS protocol in the care of patients having elective colorectal resections was associated with a reduced length of hospital stay by about one-and-half days as well as significantly reduced postoperative morbidity rates with no significant increase in readmission rates compared with conventional care (Wind et al., 2006).

These results have been further confirmed in a number of other meta-analyses. These reported a reduced overall length of postoperative hospital stay after elective colorectal resections by about 2.5 days with a reduced overall risk of postoperative complications with adopting an ERAS protocol compared with the traditional approach (Gouvas et al., 2009 & Eskicioglu et al., 2009).

#### **5. Laparoscopic surgery under ERAS protocols**

As pointed out above, meta-analyses of laparoscopic versus open colorectal resections showed that the postoperative length of hospital stay was reduced by about 20% by adopting the laparoscopic approach (Abraham et al., 2004 & 2007; Schwenk et al., 2005). The ALCCaS trial showed no statistically significant difference in postoperative complication, reoperation or peri-operative mortality rates between laparoscopic and open resections (Allardyce et al., 2010). The ALCCaS group also reported that reviews show that the shortterm advantages of laparoscopic resection for colorectal cancer are arguably relatively minor and often subjective (Allardyce et al., 2010). They also reported that the benefit in adopting a laparoscopic approach in colorectal resections may be limited mainly to patients 70 years of age or older in whom the procedure was completed laparoscopically. An average length of stay of about eight days is common between those trials and meta-analyses. This is three days longer than what was initially reported in the first published series of laparoscopic colorectal resections (Jacobs 1991). A recent meta-analysis of the topic showed that laparoscopic colorectal resections were associated with higher intra-operative complication rates than open resections (Sammour et al., 2011).

To date, the role of laparoscopic resection within an ERAS protocol has not been established. Multiple studies have been conducted to assess whether adopting the laparoscopic

conditioning and negative pressure ventilation in the operating rooms (Qadan et al., 2009). Hypothermia in a surgical setting is associated with an increased risk of bleeding due to coagulopathy as well as arrhythmias, myocardial ischemia and overall risks of

Under an ERAS protocol, hypothermia is actively prevented using warm and space blankets, warm intravenous infusions, avoiding unnecessary exposure, etc … The patient

As pointed out above, in the first published series of 16 open sigmoid colectomies under an ERAS (Fast Track) protocol, the authors reported a median postoperative length of hospital stay of two days (Kehlet & Mogensen, 1999). However, subsequent larger studies reported a median length of stay of about five days, three days longer than what was reported in the

Multiple published trials and systematic reviews have reported that ERAS protocols were associated with a faster recovery, reduced primary and overall lengths of hospital stay and complication rates after colorectal resections compared with the traditional approach. Wind et al reported that the use of an ERAS protocol in the care of patients having elective colorectal resections was associated with a reduced length of hospital stay by about one-and-half days as well as significantly reduced postoperative morbidity rates with no significant increase in

These results have been further confirmed in a number of other meta-analyses. These reported a reduced overall length of postoperative hospital stay after elective colorectal resections by about 2.5 days with a reduced overall risk of postoperative complications with adopting an ERAS protocol compared with the traditional approach (Gouvas et al., 2009 &

As pointed out above, meta-analyses of laparoscopic versus open colorectal resections showed that the postoperative length of hospital stay was reduced by about 20% by adopting the laparoscopic approach (Abraham et al., 2004 & 2007; Schwenk et al., 2005). The ALCCaS trial showed no statistically significant difference in postoperative complication, reoperation or peri-operative mortality rates between laparoscopic and open resections (Allardyce et al., 2010). The ALCCaS group also reported that reviews show that the shortterm advantages of laparoscopic resection for colorectal cancer are arguably relatively minor and often subjective (Allardyce et al., 2010). They also reported that the benefit in adopting a laparoscopic approach in colorectal resections may be limited mainly to patients 70 years of age or older in whom the procedure was completed laparoscopically. An average length of stay of about eight days is common between those trials and meta-analyses. This is three days longer than what was initially reported in the first published series of laparoscopic colorectal resections (Jacobs 1991). A recent meta-analysis of the topic showed that laparoscopic colorectal resections were associated with higher intra-operative complication

To date, the role of laparoscopic resection within an ERAS protocol has not been established. Multiple studies have been conducted to assess whether adopting the laparoscopic

complications (Diaz & Becker, 2010).

Eskicioglu et al., 2009).

does not leave the recovery ward until normothermic.

**4. Supportive evidence for ERAS protocols** 

first series (Abraham & Albayati, 2011; Nygren et al., 2009).

**5. Laparoscopic surgery under ERAS protocols** 

rates than open resections (Sammour et al., 2011).

readmission rates compared with conventional care (Wind et al., 2006).

approach would complement ERAS rehabilitation programmes. In a small, 2:1 design, randomised trial of 62 patients (43 laparoscopic and 19 open resections), the authors reported an added benefit for adopting the laparoscopic technique in an ERAS protocol in terms of a reduced postoperative length of stay (King et al., 2006).

However, the results of a systematic review of two randomised controlled trials and three controlled clinical trials of laparoscopic versus open colorectal surgery under ERAS rehabilitation programs were inconclusive as no clear advantage for laparoscopic over open resection was demonstrated under ERAS protocols (Vlug et al., 2009). Further research was recommended.

Another recently published large review of 11 studies (four randomised trials and 11 controlled clinical trials) including 1021 patients reported a clear advantage for patients enrolled in an ERAS rehabilitation program in terms of length of hospital stay compared with those who were not (Gouvas et al., 2009). Although the authors reported that an added benefit to recovery rates in adopting the laparoscopic over the open approach was assumed, such a benefit could not be established. The authors concluded that ERAS programs should become a mainstay of elective colorectal surgery.

In a systematic review of three randomised trials and seven non-randomised studies of laparoscopic versus open colorectal resections under an ERAS protocol, Khan and colleagues reported that the currently available limited evidence suggests that the inclusion of laparoscopic surgery in ERAS protocols for colorectal resections does not confer an added benefit in terms of postoperative recovery rates and postoperative length of stay (Khan et al., 2009).

#### **6. Standardisation of a colorectal ERAS protocol**

A consensus statement on ERAS was published in 2005 (Fearon et al., 2005). The statement was written by colorectal surgeons and other specialists and professionals from five universities or tertiary hospitals in five European countries (Denmark, Scotland, Sweden, Norway and The Netherlands). The authors presented their methodology in the published article with a specific focus on colorectal resections. They also recommended their protocol as one that may provide a standard of care against which current and future novel elements of an ERAS approach can be tested or added to. Members of the same group published the outcomes of 169 colorectal resections under an ERAS protocol with very good results (Nygren et al., 2009).

Figures 1-4 show a summary of an ERAS colorectal program adopted at the Coffs Harbour Health Campus, a regional hospital in New South Wales, Australia in July 2006. The summary results of 111 ERAS consecutive open colorectal resections performed at that hospital by one surgeon have been recently published with outcomes similar to those in the North European experience (Abraham & Albayati, 2011).

The Australian Safety and Efficacy Register of New Interventional Procedures - Surgical (ASERNIP-S) under the auspices of the Royal Australasian College of Surgeons and the Department of Health and Aging – Victoria, assessed the experience of Australian and New Zealand surgeons with colorectal resection under ERAS protocols (Strum & Cameron, 2009). They concluded that ERAS programs can result in beneficial outcomes for patients by reducing the length of hospital stay with no significant increase in readmission rates. They also indicated that further work is required to assist in standardisation and implementation of ERAS protocols.

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 357

**Intraoperative Care:**  Thoracic Epidural:

 If planned General Anesthetic: (guide only) Induction Agent: Propofol *Narcotic:* Fentanyl

*Maintenance:* 80% oxygen with air Sevo Fentanyl as indicated *Muscle Relaxant:* Rocuronium or Atracurium

Dexamethasone 8mgs IVI

**Urinary catheter,** TEDS & SCD

Wound Soaker Placement:

Fluid Replacement:

through the sheath.

**Normothermia:** not less than 36ºC *No routine drains or NG tube used* 

Fig. 2. Intra-operative care in a typical ERAS program

*Antibiotics:* Ceftriaxone 1g, Metronidazole BP 500mg in 100mls

Prior to wound closure, the fascia is grasped with two Moynihan tissue forceps and elevated. On each side of the incision, the introducer is placed at the superior end of the incision and tracks into the preperitoneal space. The introducer should be inspected through the parietal peritoneum to ensure the catheter is not placed deep into the muscle. Care must be taken to place the introducer greater than 1cm from the fascial edge to avoid incorporation with the fascial sutures. The introducer should be placed to its fullest extent. The needle is withdrawn and a soaker catheter is placed

*Antiemetics:* Dolasetron (Anzemet) 12.5mg IVI stat plus

*NSAID:* Parecoxib sodium 40mgs IVI, single dose

Hartmann's Solution 1-2mls/kg/hour (don't over hydrate)

#### Protocol for Enhanced Recovery After Surgery (ERAS) Coffs Harbour Health Campus NSW Australia

#### **Preoperative Care:**

Surgeon's Rooms:

	- Nil *or*
	- Enema preparation *or*
	- Colonoscopy preparation

Preadmission Clinic:

	- Fleet enema the night before and the morning of procedure.
	- Carbohydrate loading: 6 tetra packs (4 between 9 and 10 pm the night before and 2 between 5 and 6 am the day of surgery)

Day Surgery Unit:


Fig. 1. Preoperative care in a typical ERAS program

Protocol for Enhanced Recovery After Surgery (ERAS) Coffs Harbour Health Campus NSW Australia

Referral to Cancer Co-ordinator, Dietician, Stoma Therapist as required

Fleet enema the night before and the morning of procedure.

and 2 between 5 and 6 am the day of surgery)

loading drinks, postoperative pain management, etc …

Other referrals: (cancer co0ordinator, stoma therapist, dietitian, etc )

Preoperative investigations: FBC, UEC, LFTs COAG, CEA and others as required

Carbohydrate loading: 6 tetra packs (4 between 9 and 10 pm the night before

Anesthetic consultation: Anesthetic assessment and explanation of postoperative

 Perioperative nurse consultation: Patient education regarding symbiotics (e.g. Inner Health Plus), bowel preparation / enemas, low residue diet, carbohydrate

**Preoperative Care:**  Surgeon's Rooms:

Nutritional screen

Nil *or*

Preadmission Clinic:

pain management.

Base line observations charted

Fig. 1. Preoperative care in a typical ERAS program

Normothermia maintained

Day Surgery Unit:

 Skin preparation Enema if ordered

 Clinical pathway commenced Education brochure given to patient

Bowel preparation specified

 Enema preparation *or* Colonoscopy preparation

Once only medications prescribed:

Preadmission process completed by RMO/RN.


*No routine drains or NG tube used* 

Fig. 2. Intra-operative care in a typical ERAS program

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 359

As is the case with most innovations, it will probably take some time for the ERAS approach

to be used widely.

Bowel chart

 Bowel chart Surgical Ward:

Bowel chart

 Remove IVC Bowel chart.

**8. Conclusion** 

High Dependency Unit: *Day of surgery (0-24h):* 

*Postoperative day 1 (24-48h):*  Daily weigh (day 1-4)

*Postoperative day 2 (48-72h):*  Remove epidural 0600

Maintain pain score <5

*Postoperative day 3 (72-86h):* 

*Postoperative Day 4: Discharge* 

Regular paracetamol & NSAID

PCEA management as per epidural orders (if present)

Out of bed 6 hours postoperatively for 2 hours with physiotherapist

Mobilize 8 hours, 100 meters of walking with physiotherapist

Remove wound soaker catheter when device is empty

Remove urinary catheter 0800 (2 hours after epidural removal)

Maintain pain management, mobilization, fluids and diet.

Early Discharge Planner review and appointments confirmed

*Postoperative Day 10:* Skin clips removed, histology Surgeon's Rooms

In this chapter, the evidence (mainly Level I and II) against traditional peri-operative colorectal surgical practices was presented. These practices included mechanical bowel preparation, peri-operative starvation, the routine use of nasogastric decompression and prophylactic drainage, defunctioning ileostomy, aggressive IV hydration and the routine

PCA/Wound soaker management (if present)

Oral fluids and 2 protein drinks to 1000mls

 Fluids: 2000mls including 4 protein drinks Normal diet and sit out of bed for all meals

 Fluids, 2000mls including 4 protein drinks Mobilize 100 meters and out of bed 8hrs

*Postoperative week 4:* Patient interview by phone

Fig. 4. Ward care in a typical ERAS program

use of postoperative narcotic analgesia.

**Ward Care:** 

#### **Postoperative Care:**

Recovery Ward (PACU):


Thoracic Epidural:


Wound Soaker:


Postoperative Medications:


Fig. 3. Early postoperative care in a typical ERAS program

### **7. Implementation of a colorectal ERAS protocol**

A transverse incision has been used for right sided colonic resections for a few decades. There is an observation that many of the other components of ERAS protocols such as multimodal analgesia have been incorporated in traditional colorectal surgical care without necessarily implementing a complete protocol. The implementation of a structured complete ERAS program is less common (Lassen et al., 2005). The implementation of such a protocol requires coordinated training and a team approach by anesthetic, surgical, nursing and other staff (Fearon et al., 2005). This could explain the somewhat delayed uptake of the approach despite the available supportive evidence. It has been suggested that an ERAS protocol should be routinely implemented in colorectal surgical care (Gouvas et al., 2009). As is the case with most innovations, it will probably take some time for the ERAS approach to be used widely.

#### **Ward Care:**

358 Colorectal Cancer – From Prevention to Patient Care

Observations documented as per relevant chart or Recovery Ward protocol

 Clinician Initiated Dose/ Loading Dose: 5mls repeat after 20mins. Maintenance postoperatively: 0.2% Naropin with 2 mcgs Fentanyl /ml

Patient Controlled (PCEA) Dose: 5mls with 20min lockout interval

Ibuprofen 400mgs TDS for first 2 days then PRN oral

Maxolon 10mgs TDS IV commence on arrival to ward

Fig. 3. Early postoperative care in a typical ERAS program

**7. Implementation of a colorectal ERAS protocol** 

Clexane 40mgs (at least 2 hours post epidural insertion) daily SC

1. Bilateral Pain Buster for open abdominal wound: Naropin 0.375%: 270mls/5mls per

2. Single Pain buster for laparoscopic wounds: Naropin 0.375%: 270mls/5mls per hour

A transverse incision has been used for right sided colonic resections for a few decades. There is an observation that many of the other components of ERAS protocols such as multimodal analgesia have been incorporated in traditional colorectal surgical care without necessarily implementing a complete protocol. The implementation of a structured complete ERAS program is less common (Lassen et al., 2005). The implementation of such a protocol requires coordinated training and a team approach by anesthetic, surgical, nursing and other staff (Fearon et al., 2005). This could explain the somewhat delayed uptake of the approach despite the available supportive evidence. It has been suggested that an ERAS protocol should be routinely implemented in colorectal surgical care (Gouvas et al., 2009).

**Postoperative Care:**  Recovery Ward (PACU):

Maintain SCD

Thoracic Epidural:

Wound Soaker:

hour each unit

Postoperative Medications: Movicol half sachet BD

 Paracetamol 1gm QID oral Dolasetron PRN 12.5mgs BD IV

Droperidol PRN 0.5 to 1.25mgs TDS IV

VAS score

 Maintain Normothermia Continue 80% Oxygen

Clexane as per anesthetic orders

 Commence i.v. PCA if no epidural Check Wound Soaker if no epidural

Dose Range: 2.5 to 5ml /hr continuous

High Dependency Unit:

*Day of surgery (0-24h):* 


*Postoperative day 1 (24-48h):* 


Surgical Ward:

*Postoperative day 2 (48-72h):* 


*Postoperative day 3 (72-86h):* 


*Postoperative Day 10:* Skin clips removed, histology Surgeon's Rooms *Postoperative week 4:* Patient interview by phone

Fig. 4. Ward care in a typical ERAS program

#### **8. Conclusion**

In this chapter, the evidence (mainly Level I and II) against traditional peri-operative colorectal surgical practices was presented. These practices included mechanical bowel preparation, peri-operative starvation, the routine use of nasogastric decompression and prophylactic drainage, defunctioning ileostomy, aggressive IV hydration and the routine use of postoperative narcotic analgesia.

Peri-Operative Care in Colorectal Surgery in the Twenty-First Century 361

Chude, G., Rayate, N. et al. (2008). Defunctioning loop ileostomy with low anterior resection

Fearon, K., Von Meyenfeldt, M. et al. (2005). Enhanced recovery after surgery: a consensus

Garth, A., Newsome, C. et al. (2010). Nutritional status, nutrition practices and postoperative complications in patients with gastrointestinal cancer. *J Hum Nutr Diet.* Mar 23 2010. Gouvas, N., Tan, E. et al. (2009). Fast-track vs standard care in colorectal surgery: a meta -

Guenaga, K., Matos, D. & Wille-Jorgensen, P. (2009). Mechanical bowel preparation for elective colorectal surgery. *Cochrane Database Syst Rev.* 2009(1):CD001544. Halm, J., Lip, H. et al. (2009). Incisional hernia after upper abdominal surgery: a randomised

Holte, K. & Kehlet, H. (2006). Fluid therapy and surgical outcomes in elective surgery: a

Hughes, E. (1972). Asepsis in large-bowel surgery. *Ann R Coll Surg Engl.* 1972;51(6):347-356. Huser, N., Michalski, C. et al. (2008). Systematic review and meta-analysis of the role of defunctioning stoma in low rectal cancer surgery. *Ann Surg.* Jul 2008;248(1):52-60. Jacobs, M., Verdeja, J. & Goldstein, H. (1991). Minimally invasive colon resection (laparoscopic colectomy). *Surg Laparosc Endosc.* Sep 1991;1(3):144-150. Jesus, E., Karliczek, A. et al. (2004). Prophylactic anastomotic drainage for colorectal surgery.

Jorgensen, H., Wetterslev, J. et al. (2000). Epidural local anaesthetics versus opioid based

after abdominal surgery. *Cochrane Database Syst Rev.* 2000(4):CD001893. Jung, B., Matthiessen, P. et al. (2010). Mechanical bowel preparation does not affect the intramucosal bacterial colony count. *Int J Colorectal Dis.* Apr 2010;25(4):439-442. Kaska, M., Grosmanova, T. et al. (2010). The impact and safety of preoperative oral or

randomized controlled trial. *Wien Klin Wochenschr.* Jan 2010;122(1-2):23-30. Kehlet, H. & Mogensen T. (1999). Hospital stay of 2 days after open sigmoidectomy with a multimodal rehabilitation programme. *Br J Surg.* Feb 1999;86(2):227-230. Khan, S., Gatt, M. & MacFie, J. (2009). Enhanced recovery programmes and colorectal

King, P., Blazeby, J. et al. (2006). Randomized clinical trial comparing laparoscopic and open

Lassen, K., Hannemann, P. et al. (2005). Patterns in current perioperative practice: survey of

analgesic regimens on postoperative gastrointestinal paralysis, PONV and pain

intravenous carbohydrate administration versus fasting in colorectal surgery--a

surgery: does the laparoscope confer additional advantages? *Colorectal Dis.* 2009

surgery for colorectal cancer within an enhanced recovery programme. *Br J Surg.* 

colorectal surgeons in five northern European countries. *BMJ.* Jun 18

analysis update. *Int J Colorectal Dis.* Oct 2009;24(10):1119-1131.

*Gastrointest Surg*. 2009 Dec;13(12):2321-9.

*Cochrane Database Syst Rev.* 2004(4):CD002100.

Nov; 11(9):902-8. Epub 2009 Jan 27.

Mar 2006;93(3):300-308.

2005;330(7505):1420-1421.

2005; 24: 466-477.

Epub 2009 Mar 4.

for distal rectal cancer: should we make an ileostomy as a routine procedure? A prospective randomized study. *Hepatogastroenterology* 55 (86-87):1562-7 (2008) Diaz, M. & Becker, D. (2010). Thermoregulation: physiological and clinical considerations during sedation and general anesthesia. *Anesth Prog.* Spring 2010;57(1):25-32; quiz 33-24. Eskicioglu, C., Forbes, S. et al. (2009). Enhanced recovery after surgery (ERAS) programs for

patients having colorectal surgery: a meta-analysis of randomized trials. *J* 

review of clinical care for patients undergoing colonic resection. *Clinical Nutrition* 

controlled trial of midline versus transverse incision. *Hernia*. 2009 Jun;13(3):275-80.

need for reassessment in fast-track surgery. *J Am Coll Surg.* Jun 2006;202(6):971-989.

At the same time, supportive evidence for the individual aspects of an ERAS protocol and for such a protocol as a whole was also presented. The main emphasis was on avoiding mechanical bowel preparation and peri-operative starvation, ensuring nutritional support including preoperative carbohydrate and protein loading, transverse or oblique incisions if deemed appropriate by the surgeon, high oxygen concentrations, normothermia, minimal intravenous hydration, multimodal analgesia including non-narcotic epidural catheter analgesia if deemed appropriate by the anesthetist, prokinetic agents, anti-emetic drugs and early mobilization, feeding and discharge.

ERAS programs for colorectal resections have been shown to be associated with a faster recovery and a shorter length of hospital stay compared with traditional practices. Furthermore, a number of studies showed that ERAS programmes are also associated with reduced complication rates. Although further research may be required, the current evidence suggests that under an ERAS programme, there is no added benefit in adopting a laparoscopic approach over the open approach.

As with most other innovations, the use of ERAS programs might take some time to become widely spread. However, an ERAS protocol is recommended as a mainstay in colorectal surgical practice.

#### **9. Acknowledgment**

The author would like to acknowledge the assistance of Moya Anderson (Clinical Nurse Consultant), Tracey Moore (Data Manager), Dr WB Ross, Dr A Sutherland, Dr M Smith and others for their assistance in establishing the ERAS protocol at the Coffs Harbour Health Campus, NSW, Australia. The author would also like to acknowledge the assistance of Dr John Blaxland in reviewing the text.

#### **10. References**


At the same time, supportive evidence for the individual aspects of an ERAS protocol and for such a protocol as a whole was also presented. The main emphasis was on avoiding mechanical bowel preparation and peri-operative starvation, ensuring nutritional support including preoperative carbohydrate and protein loading, transverse or oblique incisions if deemed appropriate by the surgeon, high oxygen concentrations, normothermia, minimal intravenous hydration, multimodal analgesia including non-narcotic epidural catheter analgesia if deemed appropriate by the anesthetist, prokinetic agents, anti-emetic drugs and

ERAS programs for colorectal resections have been shown to be associated with a faster recovery and a shorter length of hospital stay compared with traditional practices. Furthermore, a number of studies showed that ERAS programmes are also associated with reduced complication rates. Although further research may be required, the current evidence suggests that under an ERAS programme, there is no added benefit in adopting a

As with most other innovations, the use of ERAS programs might take some time to become widely spread. However, an ERAS protocol is recommended as a mainstay in colorectal

The author would like to acknowledge the assistance of Moya Anderson (Clinical Nurse Consultant), Tracey Moore (Data Manager), Dr WB Ross, Dr A Sutherland, Dr M Smith and others for their assistance in establishing the ERAS protocol at the Coffs Harbour Health Campus, NSW, Australia. The author would also like to acknowledge the assistance of Dr

Abraham N., Byrne, C. et al. (2007). MJ. Meta-analysis of non-randomized comparative

Abraham N., Young J. & Solomon, M. (2004). Meta-analysis of short-term outcomes after laparoscopic resection for colorectal cancer. *Br J Surg.* Sep 2004;91(9):1111-1124. Abraham, N. & Albayati, S. (2011). Enhanced Recovery After Surgery programs hasten

Allardyce, R., Bagshaw P. et al. (2010). Australasian Laparoscopic Colon Cancer Study

Brady, M., Kinn, S. & Stuart, P. (2003). Preoperative fasting for adults to prevent perioperative complications. *Cochrane Database Syst Rev.* 2003(4):CD004423. Brown, S. & Goodfellow, P. (2005). Transverse versus midline incisions for abdominal

Cheatham, M., Chapman, W. et al. (1995). A Meta-Analysis of Selective Versus Routine

surgery. *Cochrane Database Syst Rev.* 2005 Oct 19;(4):CD005199.

studies of the short-term outcomes of laparoscopic resection for colorectal cancer.

recovery after colorectal resections *World J Gastrointest Surg* 2011 January 27; 3(1): 1-6.

shows that elderly patients may benefit from lower postoperative complication rates following laparoscopic versus open resection. *Br J Surg*. 2010 Jan;97(1):86-91. Bokey, E., Chapuis, P. et al. (1995). Postoperative morbidity and mortality following

resection of the colon and rectum for cancer. *Dis Colon Rectum.* May 1995;38(5):480-

Nasogastric Decompression After Elective Laparotomy. *Ann Surg* Vol. 221, No. 5,

early mobilization, feeding and discharge.

laparoscopic approach over the open approach.

surgical practice.

**10. References** 

**9. Acknowledgment** 

John Blaxland in reviewing the text.

*ANZ J Surg.* Jul 2007;77(7):508-516.

486; discussion 486-487.

469-478 (1995) J. B. Lippincott Company.


**19** 

Miroslav Levy

*Czech Republic* 

**Follow Up and Recurrence** 

% %

*Surgical Department, Thomayer University Hospital, Charles University Prague,* 

Despite optimal primary treatment, with adequate surgery with or without adjuvant chemotherapy, ∼30%–50% of patients with colon cancer will relapse and die of their disease. The principal aim of follow-up programmes after curative resection of colorectal cancer is to improve survival (Gan et al., 2007). To achieve this goal, patients are screened for early recurrent disease with the intent of a second curative surgery. Patients with a history of colorectal cancer are also at risk to develop new primary colorectal cancers (CRC). The risk of development of new primary lesions has been estimated to 0,35% per year (Cali et al., 1993). Bouvier et al reported the incidence of metachronous cancer as being 1.8% at five years, 3.4% at 10 years, and 7.2% at 20 years with the greatest excess risk between one and

The main sites of colorectal cancer relapse are listed in table 1 (Figueredo et al., 1993).

Colorectal cancer relapse Colon cancer Rectum

Liver 35 30 Lung 20 30 Peritoneum 20 20 Retroperitoneum 15 5 Peripheral lymphonodes 2 7 Local relapse 15 35 Other ( brain, bones) <5 <5

To evaluate the stage of the disease, treatment strategy and prognosis a combination of investigations is necessary. In the past there was no strong evidence that regular follow-up could improve the outcome for patients radically resected for colon cancer. As follow-up can be expensive and resource consuming in terms of both money and procedures, an intensive

**1. Introduction**

five years post-surgery(Bouvier et al. 2008).

Table 1. Sites of colorectal cancer relapse

**of Colorectal Cancer** 


### **Follow Up and Recurrence of Colorectal Cancer**

Miroslav Levy

*Surgical Department, Thomayer University Hospital, Charles University Prague, Czech Republic* 

#### **1. Introduction**

362 Colorectal Cancer – From Prevention to Patient Care

Lewis, S., Andersen, H. & Thomas, S. (2009). Early enteral nutrition within 24 h of intestinal

Ljungqvist, O. & Soreide, E. (2003). Preoperative fasting. *Br J Surg.* Apr 2003;90(4):400-406. Lobo, D., Bostock, K. et al. (2002). Effect of salt and water balance on recovery of

Merad, F., Hay, J. et al. (1999). Is prophylactic pelvic drainage useful after elective rectal or

Montedori, A., Cirocchi, R. et al. (2010). Covering ileo- or colostomy in anterior resection for

Nelson, R., Edwards, S. & Tse, B. (2007). Prophylactic nasogastric decompression after abdominal surgery. *Cochrane Database of Syst Rev.* 2007(3):CD004929. Nisanevich, V., Felsenstein, I. et al. (2005). Effect of intraoperative fluid management on outcome after intraabdominal surgery. *Anesthesiology.* Jul 2005;103(1):25-32. Nygren, J., Soop, M., et al. (2009). An enhanced-recovery protocol improves outcome after

Nygren, J., Thorell, A. & Ljungqvist O. (2001). Preoperative oral carbohydrate nutrition: an

Qadan, M., Gardner, S. et al. (2009). Hypothermia and surgery: immunologic mechanisms=

Redwine, D. & Sharpe D. (1991). Laparoscopic segmental resection of the sigmoid colon for

Sammour, T., Kahokehr, A., et al. (2011). Laparoscopic colorectal surgery is associated with

Scabini, S., Rimini, E., et al. (2010). Colon and rectal surgery for cancer without mechanical bowel preparation: one-center randomized prospective trial. *World J Surg Oncol.* 2010;8:35. Schwenk, W., Haase, O., et al. (2005). Short term benefits for laparoscopic colorectal

Seiler, C., Deckert, A., et al. (2009). Midline versus transverse incision in major

Strum, L., & Cameron, A. (2009). Fast-track surgery and enhanced recovery after surgery (ERAS) programs. *ASERNIP-S Report No. 74.* March 2009 March 2009. Vlug, M., Wind, J., et al. (2009). Systematic review of laparoscopic vs open colonic surgery within an enhanced recovery programme. *Colorectal Dis.* May 2009;11(4):335-343. White, P., Kehlet, H., et al. (2007). The role of the anesthesiologist in fast-track surgery: from

Wind, J., Polle, S., et al. (2006). Systematic review of enhanced recovery programmes in

Yeh, C., Changchien, C., et al. (2005). Pelvic drainage and other risk factors for leakage after

analysis. *J Gastrointest Surg.* Mar 2009;13(3):569-575.

for Surgical Research. *Surgery.* May 1999;125(5):529-535.

rectal carcinoma. *Cochrane Database Syst Rev.* 2010;5:CD006878.

consecutive patients. *Dis Colon Rectum*. 2009 May;52(5):978-85.

update. *Curr Opin Clin Nutr Metab Care.* Jul 2001;4(4):255-259.

for current practice. *Ann Surg.* Jul 2009;250(1):134-140.

2011 – Volume 253 - Issue 1 - p 35–43

2007;104(6):1380-1396, table of contents.

patients. *Ann Surg.* Jan 2005;241(1):9-13.

colonic surgery. *Br J Surg.* Jul 2006;93(7):800-809.

endometriosis. *J Laparoendosc Surg.* Aug 1991;1(4):217-220.

resection. *Cochrane Database Syst Rev.* 2005(3):CD003145.

ISRCTN60734227). *Ann Surg.* 2009 Jun;249(6):913-20.

trial. *Lancet.* May 25 2002;359(9320):1812-1818.

surgery versus later commencement of feeding: a systematic review and meta-

gastrointestinal function after elective colonic resection: a randomised controlled

anal anastomosis? A multicenter controlled randomized trial. French Association

colorectal resection already during the first year: a single-center experience in 168

a higher intraoperative complication rate than open surgery. *Ann Surg.* January

abdominal surgery: a randomized, double-blind equivalence trial (POVATI:

multimodal analgesia to perioperative medical care. *Anesth Analg.* Jun

elective anterior resection in rectal cancer patients: a prospective study of 978

Despite optimal primary treatment, with adequate surgery with or without adjuvant chemotherapy, ∼30%–50% of patients with colon cancer will relapse and die of their disease. The principal aim of follow-up programmes after curative resection of colorectal cancer is to improve survival (Gan et al., 2007). To achieve this goal, patients are screened for early recurrent disease with the intent of a second curative surgery. Patients with a history of colorectal cancer are also at risk to develop new primary colorectal cancers (CRC). The risk of development of new primary lesions has been estimated to 0,35% per year (Cali et al., 1993). Bouvier et al reported the incidence of metachronous cancer as being 1.8% at five years, 3.4% at 10 years, and 7.2% at 20 years with the greatest excess risk between one and five years post-surgery(Bouvier et al. 2008).

The main sites of colorectal cancer relapse are listed in table 1 (Figueredo et al., 1993).


Table 1. Sites of colorectal cancer relapse

To evaluate the stage of the disease, treatment strategy and prognosis a combination of investigations is necessary. In the past there was no strong evidence that regular follow-up could improve the outcome for patients radically resected for colon cancer. As follow-up can be expensive and resource consuming in terms of both money and procedures, an intensive

Follow Up and Recurrence of Colorectal Cancer 365

 clinical examination, carcinoembrionic antigen (CEA), chest radiograph, ultrasonography of liver or computer tomography scan (CT) every 6 months for

first postoperative 3 years, 3 next years with a frequency of one year if the recurrence is detected the patient should be discussed in the multidisciplinary oncology team (surgeon, radiologist, oncologist) to consider the best course of

colonoscopy with the polypectomy 1 x per year, in the absence of polyps every 3-5

*Computed tomography (CT).* Patients who are at higher risk of recurrence, and who could be candidates for curative-intent surgery, should undergo annual computed tomography of the chest and abdomen for 3 years after primary therapy for colon and rectal cancer. A pelvic CT scan should be considered for surveillance after rectal cancer therapy, especially for

*Colonoscopy.* All patients with colon and rectal cancer should have a colonoscopy for the preor perioperative documentation of a cancer- and polyp-free colon. After the surgical treatment of colorectal cancer, ASCO recommends the surveillance guideline presented by the American Gastroenterological Association (AGA): a colonoscopy at 3 years and, if

*Flexible proctosigmoidoscopy (rectal cancer).* For patients who have not received pelvic radiation, flexible sigmoidoscopy of the rectum every 6 months for 5 years is recommended.

Tumor markers (Rocker et al., 2006): For colorectal cancer, it is recommended that carcinoembryonic antigen (CEA) be ordered preoperatively, if it would assist in staging and surgical planning. Postoperative CEA levels should be performed every 3 months for stage II and III disease for at least 3 years if the patient is a potential candidate for surgery or chemotherapy of metastatic disease. CEA is the marker of choice for monitoring the response of metastatic disease to systemic therapy. Data are insufficient to recommend the routine use of p53, ras, thymidine synthase, dihydropyrimidine dehydrogenase, thymidine phosphorylase, microsatellite instability, 18q loss of heterozygosity, or deleted in colon cancer (DCC) protein in the management of patients with colorectal cancer. For pancreatic cancer, carbohydrate antigen 19-9 (CA 19-9) can be measured every 1 to 3 months for patients with locally advanced or metastatic disease receiving active therapy. Elevations in serial CA 19-9 determinations suggest progressive disease but confirmation with other

for patients at high risk of relapse with comorbidities or other barriers

**2.2 American Society of Clinical Oncology (ASCO) recommendations** 

**2.1 Recommendations arising from the meta-analyzes**  for patients at high risk of relapse (stages IIb and III)

treatment

**Imaging Procedures:** 

**Endoscopic Techniques:** 

**Laboratory Tests** 

clinical examination every year

for all patients with resected colorectal cancer

years(Tjandra & Chan, 2007; Rosen et al., 1998).

patients who have not been treated with radiotherapy. *Chest x-ray.* Annual chest x-rays are *not* recommended.

normal, every 5 years thereafter (Winter et al, 2003).

surveillance needs to be justified with a good level of evidence. The more effective the treatment of patients with colorectal cancer, the less cost-effective is the follow-up with regard to diagnosis of a relapse.

Although most patients with relapsed colorectal cancer are inoperable at the time of diagnosis, one third of patients with isolated locoregional or distant metastases survive 5 years (Browne et al., 2005). Number of resected patients for relapse is increasing - about 20% of patients with liver metastases are indicated for surgery (Guyot et al., 2005). Other patients are operated after downstaging after chemotherapy or chemoradiotherapy. Long-term survival is also achieved after resection of pulmonary metastases, even when combined of liver and lung metastases resection (Ike et al., 2002). It is evident that high-risk patients (TNM II, III) with relapse diagnosed using imaging and endoscopic techniques have better survival than patients who had clinically evident relapse (Chau et al., 2004). Even patients, who are at the time of relapse diagnosis inoperable, had improved survival due to new palliative chemotherapy regimes (Ahmed et al., 2004).

Due to the different localization of possible recurrence, we cannot use one diagnostic tool, we need a combination of various imaging and laboratory methods. In the first two to three years after resection of primary tumor incidence of relapse increases exponentially, then passes into the plateau (Griffin et al., 1987; Scholefield et al., 2002). Therefore, diagnostic schedule must be adapted. Finally, the postoperative monitoring of patients after curative resection has psychological effects. It can be both positive and negative. Positive involves calming the patient and aid in further treatment. Negative effects includes a sense of false security when relapse is undetected.

Any follow-up system combines a number of tests, whether clinical, laboratory and imaging, as well as their frequency.

Meta-analyzes of studies concludes that intensive follow-up shows a statistically significant difference in overall 5-year survival rate of patients after curative surgery for colorectal cancer, and can diagnose relapse in curable stage, especially if located in the liver and lungs (Tjandra & Chan, 2007; Rosen et al., 1998). In the case of relapse in rectal cancer, studies indicate the minor importance of intensive follow-up (Secco et al., 2000). Analysis showed no significant difference in the incidence of relapse among patients in groups with minimal versus intensive monitoring system, however, closely monitored group had significantly higher number of surgical interventions for recurrence, which is given by an earlier diagnosis and thus a higher resectability of recurrence.

Studies previously conducted and their meta-analyses may be problematic because of nonstandard combinations of investigations and also non-standard frequency. Intensive followup study in one study is very similar to that in other studies considered to be less intense (Jeffery et al., 2007).

#### **2. Recomendations for follow-up**

Diagnostic tools used for follow-up can be dividend in to:


#### **2.1 Recommendations arising from the meta-analyzes**

	- clinical examination, carcinoembrionic antigen (CEA), chest radiograph, ultrasonography of liver or computer tomography scan (CT) every 6 months for first postoperative 3 years, 3 next years with a frequency of one year

#### **2.2 American Society of Clinical Oncology (ASCO) recommendations**

#### **Imaging Procedures:**

364 Colorectal Cancer – From Prevention to Patient Care

surveillance needs to be justified with a good level of evidence. The more effective the treatment of patients with colorectal cancer, the less cost-effective is the follow-up with

Although most patients with relapsed colorectal cancer are inoperable at the time of diagnosis, one third of patients with isolated locoregional or distant metastases survive 5 years (Browne et al., 2005). Number of resected patients for relapse is increasing - about 20% of patients with liver metastases are indicated for surgery (Guyot et al., 2005). Other patients are operated after downstaging after chemotherapy or chemoradiotherapy. Long-term survival is also achieved after resection of pulmonary metastases, even when combined of liver and lung metastases resection (Ike et al., 2002). It is evident that high-risk patients (TNM II, III) with relapse diagnosed using imaging and endoscopic techniques have better survival than patients who had clinically evident relapse (Chau et al., 2004). Even patients, who are at the time of relapse diagnosis inoperable, had improved survival due to new

Due to the different localization of possible recurrence, we cannot use one diagnostic tool, we need a combination of various imaging and laboratory methods. In the first two to three years after resection of primary tumor incidence of relapse increases exponentially, then passes into the plateau (Griffin et al., 1987; Scholefield et al., 2002). Therefore, diagnostic schedule must be adapted. Finally, the postoperative monitoring of patients after curative resection has psychological effects. It can be both positive and negative. Positive involves calming the patient and aid in further treatment. Negative effects includes a sense of false

Any follow-up system combines a number of tests, whether clinical, laboratory and

Meta-analyzes of studies concludes that intensive follow-up shows a statistically significant difference in overall 5-year survival rate of patients after curative surgery for colorectal cancer, and can diagnose relapse in curable stage, especially if located in the liver and lungs (Tjandra & Chan, 2007; Rosen et al., 1998). In the case of relapse in rectal cancer, studies indicate the minor importance of intensive follow-up (Secco et al., 2000). Analysis showed no significant difference in the incidence of relapse among patients in groups with minimal versus intensive monitoring system, however, closely monitored group had significantly higher number of surgical interventions for recurrence, which is given by an earlier

Studies previously conducted and their meta-analyses may be problematic because of nonstandard combinations of investigations and also non-standard frequency. Intensive followup study in one study is very similar to that in other studies considered to be less intense

regard to diagnosis of a relapse.

palliative chemotherapy regimes (Ahmed et al., 2004).

diagnosis and thus a higher resectability of recurrence.

Diagnostic tools used for follow-up can be dividend in to:

security when relapse is undetected.

imaging, as well as their frequency.

**2. Recomendations for follow-up** 

History and physical examination

(Jeffery et al., 2007).

 Imaging procedures Endoscopic techniques Laboratory tests

*Computed tomography (CT).* Patients who are at higher risk of recurrence, and who could be candidates for curative-intent surgery, should undergo annual computed tomography of the chest and abdomen for 3 years after primary therapy for colon and rectal cancer. A pelvic CT scan should be considered for surveillance after rectal cancer therapy, especially for patients who have not been treated with radiotherapy.

*Chest x-ray.* Annual chest x-rays are *not* recommended.

#### **Endoscopic Techniques:**

*Colonoscopy.* All patients with colon and rectal cancer should have a colonoscopy for the preor perioperative documentation of a cancer- and polyp-free colon. After the surgical treatment of colorectal cancer, ASCO recommends the surveillance guideline presented by the American Gastroenterological Association (AGA): a colonoscopy at 3 years and, if normal, every 5 years thereafter (Winter et al, 2003).

*Flexible proctosigmoidoscopy (rectal cancer).* For patients who have not received pelvic radiation, flexible sigmoidoscopy of the rectum every 6 months for 5 years is recommended.

#### **Laboratory Tests**

Tumor markers (Rocker et al., 2006): For colorectal cancer, it is recommended that carcinoembryonic antigen (CEA) be ordered preoperatively, if it would assist in staging and surgical planning. Postoperative CEA levels should be performed every 3 months for stage II and III disease for at least 3 years if the patient is a potential candidate for surgery or chemotherapy of metastatic disease. CEA is the marker of choice for monitoring the response of metastatic disease to systemic therapy. Data are insufficient to recommend the routine use of p53, ras, thymidine synthase, dihydropyrimidine dehydrogenase, thymidine phosphorylase, microsatellite instability, 18q loss of heterozygosity, or deleted in colon cancer (DCC) protein in the management of patients with colorectal cancer. For pancreatic cancer, carbohydrate antigen 19-9 (CA 19-9) can be measured every 1 to 3 months for patients with locally advanced or metastatic disease receiving active therapy. Elevations in serial CA 19-9 determinations suggest progressive disease but confirmation with other

Follow Up and Recurrence of Colorectal Cancer 367

History and physical examination every 3-6 months for 2 years, then every 6 months for

Ideally, colonoscopy should be performed pre-operatively. If this is not feasible, then it may be done three to six months post-operatively if no metastases were found. Air-contrast barium enema combined with sigmoidoscopy is an acceptable alternative where

After recovery from surgery, visits should only be scheduled as needed. The routine use of liver enzyme tests and abdominal ultrasound is not recommended in the absence of

The value of carcinoembryonic antigen (CEA) testing in the post-operative period is controversial and its usefulness is therefore limited. However, in individuals who would be candidates for resection of isolated hepatic or pulmonary metastases, serial measurement of CEA levels post-operatively (every three months for two years) may be of value in detecting

Repeat colonoscopy once every three years until no new adenomas are discovered. Thereafter, repeat colonoscopy every five years until the detection of new tumours is unlikely to influence the patient's lifespan. Air-contrast barium enema combined with sigmoidoscopy is an acceptable alternative where colonoscopy is not readily available.

For patients who have undergone low anterior resection of rectal cancers, digital rectal examinations and proctoscopy or sigmoidoscopy should be undertaken at three months, six months, one year and two years to look for anastomotic recurrence. Thereafter,

The Australian *Clinical Practice Guidelines for the prevention, early detection and management of CRC, 2nd edition, 2005* proposed that follow-up should be offered to all patients who have undergone curative surgery and are fit for further intervention if disease is detected. This includes patients who have had malignant polypectomy or curative endoscopic resection of Stage I CRC but excludes patients with Stage IV CRC if their treatment does not offer the

Patients with proved Lynch syndrome (HNPCC or hereditary non-polyposis colorectal cancer), should continue to have annual surveillance colonoscopy performed post-

CEA every 3-6 months for 2 years , then every 6 months for a total of 5 years

**2.5 National Comprehensive Cancer Network (NCCN) guidelines** 

Colonoscopy at 1 year , then as clinically indicated(NCCN, 2011)

**2.6 Guidelines British Columbia Medical Associattion (BCMA)** 

CT scan of abdomen and pelvis annually for 3 years

**Recommendation 1: Clearing colonoscopy** 

**Recommendation 2: Post-operative follow-up** 

recurrence that is treatable in up to 25 per cent of patients.

**Recommendation 4: Prevention of new cancers** 

recommendation 4 should be followed.(BCMA, 2011)

**2.7 Australian Clinical Practice Guidelines (CCA)** 

colonoscopy is not readily available.

**Recommendation 3: Tumour markers** 

**Recommendation 5: Low rectal cancer** 

possibility of cure.

symptoms.

a total of 5 years

studies should be sought. *Blood tests.* Routine blood tests (i.e., CBCs or liver fiction tests) are not recommended.

*Fecal occult blood test.* Periodic fecal occult blood testing is not recommended.

*Laboratory-derived prognostic and predictive factors.* Until prospective data are available, use of molecular or cellular markers should not influence the surveillance strategy (Desch et al., 2005).

#### **2.3 European Group On Tumour Markers (EGTM) recommendations**

For identifying recurrences in patients with previously diagnosed colorectal cancer, CEA has a sensitivity of about 80% (range 17-89%) and a specificity of approximately 70% (range 34-91%). Early studies showed that serial CEA levels could detect recurrent disease many months (usually 4-10 months) in advance of clinical evidence of disease (Fletscher, 1986). CEA testing was found to be most sensitive for diagnosing hepatic or retroperitoneal disease and relatively insensitive for either local, peritoneal or pulmonary involvement (Moertel et al., 1993). Some investigators have reported that a slowly rising CEA usually indicates a locoregional recurrence while rapidly increasing levels usually suggest hepatic metastasis (Begent, 1984).

In the follow-up of patients with colorectal cancer, the optimum interval between CEA measurements has not been established. In practice, most clinicians use intervals of 3 months, at least for the first 2 years after the initial diagnosis.

Clearly, further work is necessary to address the impact of CEA monitoring on patient survival, quality of life and cost of care. Ideally, this study should be carried out as part of a prospective randomised trial.

Although surgery remains the most effective therapy for colorectal cancer, chemotherapy is finding increasing use especially in patients with advanced disease. Administration of this therapy may however, cause transient elevations in CEA levels.

While CEA is the preferential biochemical test for colorectal cancer, a number of other markers such as CA19-9, CA242 and cytokeratins (e.g., TPA and TPS) have also been evaluated for this malignancy. While some of these markers have been found to complement CEA, further work will be required to see which marker is most complementary to CEA.

#### **2.4 European Society for Medical Oncology (ESMO) recommendations**


studies should be sought. *Blood tests.* Routine blood tests (i.e., CBCs or liver fiction tests) are

*Laboratory-derived prognostic and predictive factors.* Until prospective data are available, use of molecular or cellular markers should not influence the surveillance strategy (Desch et al.,

For identifying recurrences in patients with previously diagnosed colorectal cancer, CEA has a sensitivity of about 80% (range 17-89%) and a specificity of approximately 70% (range 34-91%). Early studies showed that serial CEA levels could detect recurrent disease many months (usually 4-10 months) in advance of clinical evidence of disease (Fletscher, 1986). CEA testing was found to be most sensitive for diagnosing hepatic or retroperitoneal disease and relatively insensitive for either local, peritoneal or pulmonary involvement (Moertel et al., 1993). Some investigators have reported that a slowly rising CEA usually indicates a locoregional recurrence while rapidly increasing levels usually suggest hepatic metastasis

In the follow-up of patients with colorectal cancer, the optimum interval between CEA measurements has not been established. In practice, most clinicians use intervals of 3

Clearly, further work is necessary to address the impact of CEA monitoring on patient survival, quality of life and cost of care. Ideally, this study should be carried out as part of a

Although surgery remains the most effective therapy for colorectal cancer, chemotherapy is finding increasing use especially in patients with advanced disease. Administration of this

While CEA is the preferential biochemical test for colorectal cancer, a number of other markers such as CA19-9, CA242 and cytokeratins (e.g., TPA and TPS) have also been evaluated for this malignancy. While some of these markers have been found to complement CEA, further work will be required to see which marker is most

 History and physical examination and CEA determination are advised every 3–6 months for 3 years and every 6–12 months at years 4 and 5 after surgery [II, B]. Colonoscopy must be performed at year 1 and thereafter every 3–5 years looking for

CT scan of chest and abdomen every 6–12 months for the first 3 years can be considered

Contrast enhanced ultrasound imaging (CEUS) could substitute for abdominal CT scan

Other laboratory and radiological examinations are of unproven benefit and must be

restricted to patients with suspicious symptoms.(Labianca et al., 2010)

*Fecal occult blood test.* Periodic fecal occult blood testing is not recommended.

**2.3 European Group On Tumour Markers (EGTM) recommendations** 

months, at least for the first 2 years after the initial diagnosis.

therapy may however, cause transient elevations in CEA levels.

metachronous adenomas and cancers [III, B].

in patients who are at higher risk for recurrence [II, B].

**2.4 European Society for Medical Oncology (ESMO) recommendations**  Intensive follow-up must be performed in colon cancer patients [ I, A].

not recommended.

2005).

(Begent, 1984).

prospective randomised trial.

complementary to CEA.

[III, C].

#### **2.5 National Comprehensive Cancer Network (NCCN) guidelines**


#### **2.6 Guidelines British Columbia Medical Associattion (BCMA)**

#### **Recommendation 1: Clearing colonoscopy**

Ideally, colonoscopy should be performed pre-operatively. If this is not feasible, then it may be done three to six months post-operatively if no metastases were found. Air-contrast barium enema combined with sigmoidoscopy is an acceptable alternative where colonoscopy is not readily available.

#### **Recommendation 2: Post-operative follow-up**

After recovery from surgery, visits should only be scheduled as needed. The routine use of liver enzyme tests and abdominal ultrasound is not recommended in the absence of symptoms.

#### **Recommendation 3: Tumour markers**

The value of carcinoembryonic antigen (CEA) testing in the post-operative period is controversial and its usefulness is therefore limited. However, in individuals who would be candidates for resection of isolated hepatic or pulmonary metastases, serial measurement of CEA levels post-operatively (every three months for two years) may be of value in detecting recurrence that is treatable in up to 25 per cent of patients.

#### **Recommendation 4: Prevention of new cancers**

Repeat colonoscopy once every three years until no new adenomas are discovered. Thereafter, repeat colonoscopy every five years until the detection of new tumours is unlikely to influence the patient's lifespan. Air-contrast barium enema combined with sigmoidoscopy is an acceptable alternative where colonoscopy is not readily available.

#### **Recommendation 5: Low rectal cancer**

For patients who have undergone low anterior resection of rectal cancers, digital rectal examinations and proctoscopy or sigmoidoscopy should be undertaken at three months, six months, one year and two years to look for anastomotic recurrence. Thereafter, recommendation 4 should be followed.(BCMA, 2011)

#### **2.7 Australian Clinical Practice Guidelines (CCA)**

The Australian *Clinical Practice Guidelines for the prevention, early detection and management of CRC, 2nd edition, 2005* proposed that follow-up should be offered to all patients who have undergone curative surgery and are fit for further intervention if disease is detected. This includes patients who have had malignant polypectomy or curative endoscopic resection of Stage I CRC but excludes patients with Stage IV CRC if their treatment does not offer the possibility of cure.

Patients with proved Lynch syndrome (HNPCC or hereditary non-polyposis colorectal cancer), should continue to have annual surveillance colonoscopy performed post-

Follow Up and Recurrence of Colorectal Cancer 369

The optimal combination and frequency of investigations in follow-up of patients after CRC resection has not been determined. Importantly, the performance of annual colonoscopy has

Studies comparing intensive and less intensive follow up programes were conducted in many countries around the World. In Finland( Makela et al., 1995) randomized more than 100 patients. Less intensive follow up system included either rigid sigmoideoscopy ( for rectal or sigmoid cancer) or barium enema ( for colon cancer) once a year. Patients in intensive arm underwent colonoscopy within 3 months of surgery and then yearly colonoscopy thereafter, liver ultrasound every 6 months and CT scans every year. Intensive follow up programe significantly earlier identified recurrence, there were no signifiant diference in resecability rates and five-year survival. Ohlsson et al. showed the same results on 107 patients (Ohlson et al., 1995). In Italy, (Pietra et al., 1998) randomized more than 200 consecutive patiens into low intensity follow up arm ( physical examination, liver ultrasound and CEA at 6 month and then yearly, colonoscopy and chest X ray annually) and intensive arm ( clinical controls, liver ultrasound, CEA each 3 months during the first 2 years, at 6 month interval for the next 3 years and then yearly, colonoscopy, chest X ray and CT scan yearly). Intensive group demonstrate statistically signifiant increase in five-year survival ( 73,1% vers. 58,3%). Kjeldsen and coleagues randomized big group up to 600 patients to frequent and minimal follow up arms and demonstrated significantly earlier detection of recurrence, but not improvement of overall or cancer-related five-year survival

Then, metaanalyses were conducted. Renehan and colleagues involved 1342 patients and demonstrated a signifiant improvement in overall five-year survival in intensivelly followed patiens. The intensive follow-up groups were also associated with significantly earlier detection of all recurrences and isolated local recurrences (Renehan et al., 2002). A metaanalysis by Tjandra et al concluded that intensive follow-up increased the re-resection rate for recurrent disease and improved overall survival but the survival advantage was not due to earlier detection of recurrence and cancer-related mortality was no better (Tjandra & Chan, 2007). Forty-one centers have participated in the GILDAtrial. There are 39 centers in Italy, one in Spain, and one in the United States. Both the less intensive follow-up group and the more intensive follow-up group are well matched for distribution of sex, age, cancer stage (Dukes B or C) and primary site of cancer (colon or rectum). This trial will allow us to quantify the lead-time provided by the specifically defined intensive follow-up regimen, and to compare the likelihood of uncovering recurrent disease amenable to salvage therapy

CTC has been introduced in the last decade for the identification of colorectal lesions, polyps and cancer (Reuterskiold et al., 2006). CTC is being increasingly used for the radiological evaluation of colorectal symptoms. There have been a few publications on the use of CTC in follow-up of these patients after surgery (Amitai et al., 2009). CTC has the advantage in demonstrating the inner surface of the colon tube simulating the endoscopic colonoscopic

**3. Discussion** 

(Kjeldsen et al., 1997).

(Grossmann et al., 2004).

**3.1 Imaging procedures – New possibilities** 

**3.1.1 Computed Tomographic Colonography (CTC)** 

not been shown to improve five-year survival.

operatively because of the apparent rapid progression of neoplasia from adenoma to carcinoma.

Patients including those whose initial diagnosis was made younger than 40 years of age, with probable or possible HNPCC (i.e. patients whose tumours are MSI-high and less than 50 years old at time of initial cancer diagnosis but not proved by genetic testing to have HNPCC), with hyperplastic polyposis and BRAF mutation and with multiple synchronous cancers or advanced adenomas at initial diagnosis should be considered following surgery to continuing with more frequent surveillance than would otherwise be recommended (e.g. initial post-operative colonoscopy at one year and then annually, second-yearly or thirdyearly. (CCA, 2011)

Summary in reccomendations for follow-up see in table 2.


Table 2. Summary of follow up recommendations

#### **3. Discussion**

368 Colorectal Cancer – From Prevention to Patient Care

operatively because of the apparent rapid progression of neoplasia from adenoma to

Patients including those whose initial diagnosis was made younger than 40 years of age, with probable or possible HNPCC (i.e. patients whose tumours are MSI-high and less than 50 years old at time of initial cancer diagnosis but not proved by genetic testing to have HNPCC), with hyperplastic polyposis and BRAF mutation and with multiple synchronous cancers or advanced adenomas at initial diagnosis should be considered following surgery to continuing with more frequent surveillance than would otherwise be recommended (e.g. initial post-operative colonoscopy at one year and then annually, second-yearly or third-

Metaanalyses ASCO EGTM ESMO NCCN BCMA

every 3 months

every 3-6 months

In 1 year, and thereafter every 3–5 years

months

every 6–12 months for the first 3 years

CEUS could substitute for abdominal CT scan

every 3-6 months for 2 years , then every 6 months

> at 1 year , then as clinically indicated

every 3-6 months for 2 years, then every 6 months

annually for 3 years

every 3 months for two years

3-6 months after surgery

as needed

carcinoma.

yearly. (CCA, 2011)

CEA every 6

Colonoscopy every 6

History, physical examination months

months

every 6

Chest X ray not

months

months

Table 2. Summary of follow up recommendations

CT every 6

Ultrasound every 6

Summary in reccomendations for follow-up see in table 2.

every 3 months

3 years after surgery

months every 3-6

reccomended

3 years after surgery

The optimal combination and frequency of investigations in follow-up of patients after CRC resection has not been determined. Importantly, the performance of annual colonoscopy has not been shown to improve five-year survival.

Studies comparing intensive and less intensive follow up programes were conducted in many countries around the World. In Finland( Makela et al., 1995) randomized more than 100 patients. Less intensive follow up system included either rigid sigmoideoscopy ( for rectal or sigmoid cancer) or barium enema ( for colon cancer) once a year. Patients in intensive arm underwent colonoscopy within 3 months of surgery and then yearly colonoscopy thereafter, liver ultrasound every 6 months and CT scans every year. Intensive follow up programe significantly earlier identified recurrence, there were no signifiant diference in resecability rates and five-year survival. Ohlsson et al. showed the same results on 107 patients (Ohlson et al., 1995). In Italy, (Pietra et al., 1998) randomized more than 200 consecutive patiens into low intensity follow up arm ( physical examination, liver ultrasound and CEA at 6 month and then yearly, colonoscopy and chest X ray annually) and intensive arm ( clinical controls, liver ultrasound, CEA each 3 months during the first 2 years, at 6 month interval for the next 3 years and then yearly, colonoscopy, chest X ray and CT scan yearly). Intensive group demonstrate statistically signifiant increase in five-year survival ( 73,1% vers. 58,3%). Kjeldsen and coleagues randomized big group up to 600 patients to frequent and minimal follow up arms and demonstrated significantly earlier detection of recurrence, but not improvement of overall or cancer-related five-year survival (Kjeldsen et al., 1997).

Then, metaanalyses were conducted. Renehan and colleagues involved 1342 patients and demonstrated a signifiant improvement in overall five-year survival in intensivelly followed patiens. The intensive follow-up groups were also associated with significantly earlier detection of all recurrences and isolated local recurrences (Renehan et al., 2002). A metaanalysis by Tjandra et al concluded that intensive follow-up increased the re-resection rate for recurrent disease and improved overall survival but the survival advantage was not due to earlier detection of recurrence and cancer-related mortality was no better (Tjandra & Chan, 2007). Forty-one centers have participated in the GILDAtrial. There are 39 centers in Italy, one in Spain, and one in the United States. Both the less intensive follow-up group and the more intensive follow-up group are well matched for distribution of sex, age, cancer stage (Dukes B or C) and primary site of cancer (colon or rectum). This trial will allow us to quantify the lead-time provided by the specifically defined intensive follow-up regimen, and to compare the likelihood of uncovering recurrent disease amenable to salvage therapy (Grossmann et al., 2004).

#### **3.1 Imaging procedures – New possibilities**

#### **3.1.1 Computed Tomographic Colonography (CTC)**

CTC has been introduced in the last decade for the identification of colorectal lesions, polyps and cancer (Reuterskiold et al., 2006). CTC is being increasingly used for the radiological evaluation of colorectal symptoms. There have been a few publications on the use of CTC in follow-up of these patients after surgery (Amitai et al., 2009). CTC has the advantage in demonstrating the inner surface of the colon tube simulating the endoscopic colonoscopic

Follow Up and Recurrence of Colorectal Cancer 371

90.8%. CEUS and CT were more sensitive than US also for detection of single lesions In 15 patients (13.8%), CEUS revealed more metastases than CT, while CT revealed more metastases than CEUS in 9 patients (8.2%) Piscaglia concluded that CEUS is more sensitive than conventional US in the detection of liver metastases and could be usefully employed in the staging of patients with gastrointestinal cancer. Findings at CEUS and CT appear to be

Colonoscopy: Surveillance colonoscopy after CRC resection has the theoretical potential to improve patient outcome by finding metachronous cancers at an early stage, detecting luminal/ anastomotic cancer recurrences and removing metachronous adenomas. Nevertheless, studies have differed in their conclusions about the overall effectiveness of colonoscopic surveillance. Recommendations about the timing of colonoscopy after CRC resection should be based upon the "natural history" of metachronous colonic neoplasia, in order to meet the objectives of surveillance, namely early detection of metachronous cancer

Patients undergoing either local excision (including transanal endoscopic microsurgery) of rectal cancer or advanced adenomas or ultra-low anterior resection for rectal cancer should be considered for periodic examination of the rectum at six monthly intervals for two or three years using either digital rectal examination, rigid proctoscopy, flexible proctoscopy, and/or rectal endoscopic ultrasound. These examinations are considered to be independent

Two recent case-control studies of colonoscopy suggest that colonoscopy is not as effective in decreasing mortality (Baxter et al., 2009) or incidence of right-sided cancer (Lakoff et al., 2008). A recent investigation of surveillance colonoscopy found that subjects were more likely to develop recurrent adenomas in the same colon segment, suggesting that particular attention be paid to where a previous adenoma has been removed (Pinsky

Leung and colleagues studied colorectal cancer risk despite surveilance colonoscopy and concluded that despite frequent colonoscopy there was a persistent ongoing risk of cancer in the years after the trial. Subjects with a history of advanced adenoma are at increased risk of subsequent cancer and should be followed closely with continued surveillance (Leung et al.,

Importantly, the performance of annual colonoscopy has not been shown to improve five-

Based on literature and also our results (Lipská et al., 2007), it can be concluded that monitoring of the tumor markers is valuable, mainly in those cases where preoperative CEA and/or CA19-9 were elevated. The level of CEA and CA19-9 increases according to the pTNM stage of the disease. CEA or CA19-9 below the cut-off level does not exclude even a very advanced colorectal cancer. To evaluate the stage of the disease, treatment strategy and prognosis a combination of investigations is necessary. Surveillance based only on CEA and/or CA19-9 is cost-effective, but does not disclose more than 1/3 of patients with relapse. In general practice CEA is often used as the only parameter in the follow-up

complementary in achieving maximum sensitivity.

and timely polypectomy for metachronous adenomas.

of the colonoscopic examination schedule (CCA, 2011).

**3.2 Endoscopic techniques** 

et al., 2009).

year survival.

**3.3 Laboratory tests** 

2010).

view and demonstrating the pericolonic structures at the same time. It has a high accuracy in detecting colonic neoplasia (Halligan et al., 2005).

#### **3.1.2 PET/CT scan**

In past 10 years PET/CT scan was introduced as a standard method for colorectal cancer imaging, especially for distant metastases diagnosis. Studies comparing PET/CT scan with standard methods showed superiory of this paging method. In these studies (Deleau et al., 2011; Han et al., 2011) data of patients with suspected CRC recurrence and in whom both FDG-PET/CT and CT were performed were analyzed. All detected lesions were characterized according to their number, size, and localization. In Deleau´s study 171 truepositive lesions were identified in 71 patients. CT scan was positive in 58 (82%) patients and FDG-PET/CT in 70 (98%) patients. In per lesion analysis, the global accuracy of FDG-PET/CT in detection of lesions was of 88% (sensitivity = 95%, specificity = 54%), which was higher than that of CT (53%, sensitivity = 55%, specificity = 43%), particularly in case of lymph nodes metastases (100 vs. 35%) and locoregional lesions (100 vs. 39%). FDG-PET/CT modified the clinical management in 31 patients.

At present, whole-body (18)F-FDG PET/CT is an advanced diagnostic imaging technique in detecting loco-regional recurrence and metastasis in postoperative patients with colorectal carcinoma for its higher sensitivity and specificity and also appears to be useful modality in evaluating chemotherapy response and can differentiate responders from nonresponders in recurrent CRC patients (Shamim et al., 2011).

#### **3.1.3 Contrast enhancement ultrasound scan**

There were few studies done to compare the sensitivity and specificity of contrast-enhanced ultrasonography (CEUS) and computed tomography (CT) in the detection of liver metastases in patients with colorectal cancer (Larsen et al., 2009). In this Denmark study 365 patients were included. All patients had undergone preoperative US, CEUS and Multidetector CT and 65.5% had received Intraoperative US. Multidetector CT found significantly more metastases than CEUS. In a patient-by-patient analysis MDCT had a nonsignificantly higher sensitivity in the detection of liver metastases compared to CEUS . The specificity of was slightly better than that of MDCT. Multidetector CT found significant more metastases than CEUS. In previous study, held by same authors (Larsen et al., 2007), sensitivity and specificity of contrast enhanced ultrasonography (CEUS) with conventional ultrasonography (US) in detection of liver metastases in patients with colorectal adenocarcinoma were compared. In 461 patients contrast enhanced ultrasonography improved the sensitivity significantly in detection of liver metastases from 0.69 by US to 0.80 (p=0.031). In 24 patients, CEUS found a higher number of metastases than US (p<0.001). The specificity (0.98) and the positive predictive value (0.86) was the same.

In Italian study (Piscaglia et al., 2007) a total of 109 patients with colorectal (n = 92) or gastric cancer prospectively underwent computed tomography (CT) scan and conventional US evaluation followed by real time CEUS. A diagnosis of metastases was made by CT or, for lesions not visible at CT, the diagnosis was achieved by histopathology or by a malignant behavior during follow-up.

Of 109 patients, 65 were found to have metastases at presentation. CEUS improved sensitivity in metastatic livers from 76.9% of patients (US) to 95.4% , while CT scan reached 90.8%. CEUS and CT were more sensitive than US also for detection of single lesions In 15 patients (13.8%), CEUS revealed more metastases than CT, while CT revealed more metastases than CEUS in 9 patients (8.2%) Piscaglia concluded that CEUS is more sensitive than conventional US in the detection of liver metastases and could be usefully employed in the staging of patients with gastrointestinal cancer. Findings at CEUS and CT appear to be complementary in achieving maximum sensitivity.

#### **3.2 Endoscopic techniques**

370 Colorectal Cancer – From Prevention to Patient Care

view and demonstrating the pericolonic structures at the same time. It has a high accuracy

In past 10 years PET/CT scan was introduced as a standard method for colorectal cancer imaging, especially for distant metastases diagnosis. Studies comparing PET/CT scan with standard methods showed superiory of this paging method. In these studies (Deleau et al., 2011; Han et al., 2011) data of patients with suspected CRC recurrence and in whom both FDG-PET/CT and CT were performed were analyzed. All detected lesions were characterized according to their number, size, and localization. In Deleau´s study 171 truepositive lesions were identified in 71 patients. CT scan was positive in 58 (82%) patients and FDG-PET/CT in 70 (98%) patients. In per lesion analysis, the global accuracy of FDG-PET/CT in detection of lesions was of 88% (sensitivity = 95%, specificity = 54%), which was higher than that of CT (53%, sensitivity = 55%, specificity = 43%), particularly in case of lymph nodes metastases (100 vs. 35%) and locoregional lesions (100 vs. 39%). FDG-PET/CT

At present, whole-body (18)F-FDG PET/CT is an advanced diagnostic imaging technique in detecting loco-regional recurrence and metastasis in postoperative patients with colorectal carcinoma for its higher sensitivity and specificity and also appears to be useful modality in evaluating chemotherapy response and can differentiate responders from nonresponders in

There were few studies done to compare the sensitivity and specificity of contrast-enhanced ultrasonography (CEUS) and computed tomography (CT) in the detection of liver metastases in patients with colorectal cancer (Larsen et al., 2009). In this Denmark study 365 patients were included. All patients had undergone preoperative US, CEUS and Multidetector CT and 65.5% had received Intraoperative US. Multidetector CT found significantly more metastases than CEUS. In a patient-by-patient analysis MDCT had a nonsignificantly higher sensitivity in the detection of liver metastases compared to CEUS . The specificity of was slightly better than that of MDCT. Multidetector CT found significant more metastases than CEUS. In previous study, held by same authors (Larsen et al., 2007), sensitivity and specificity of contrast enhanced ultrasonography (CEUS) with conventional ultrasonography (US) in detection of liver metastases in patients with colorectal adenocarcinoma were compared. In 461 patients contrast enhanced ultrasonography improved the sensitivity significantly in detection of liver metastases from 0.69 by US to 0.80 (p=0.031). In 24 patients, CEUS found a higher number of metastases than US (p<0.001). The

In Italian study (Piscaglia et al., 2007) a total of 109 patients with colorectal (n = 92) or gastric cancer prospectively underwent computed tomography (CT) scan and conventional US evaluation followed by real time CEUS. A diagnosis of metastases was made by CT or, for lesions not visible at CT, the diagnosis was achieved by histopathology or by a malignant

Of 109 patients, 65 were found to have metastases at presentation. CEUS improved sensitivity in metastatic livers from 76.9% of patients (US) to 95.4% , while CT scan reached

specificity (0.98) and the positive predictive value (0.86) was the same.

in detecting colonic neoplasia (Halligan et al., 2005).

modified the clinical management in 31 patients.

recurrent CRC patients (Shamim et al., 2011).

behavior during follow-up.

**3.1.3 Contrast enhancement ultrasound scan** 

**3.1.2 PET/CT scan** 

Colonoscopy: Surveillance colonoscopy after CRC resection has the theoretical potential to improve patient outcome by finding metachronous cancers at an early stage, detecting luminal/ anastomotic cancer recurrences and removing metachronous adenomas. Nevertheless, studies have differed in their conclusions about the overall effectiveness of colonoscopic surveillance. Recommendations about the timing of colonoscopy after CRC resection should be based upon the "natural history" of metachronous colonic neoplasia, in order to meet the objectives of surveillance, namely early detection of metachronous cancer and timely polypectomy for metachronous adenomas.

Patients undergoing either local excision (including transanal endoscopic microsurgery) of rectal cancer or advanced adenomas or ultra-low anterior resection for rectal cancer should be considered for periodic examination of the rectum at six monthly intervals for two or three years using either digital rectal examination, rigid proctoscopy, flexible proctoscopy, and/or rectal endoscopic ultrasound. These examinations are considered to be independent of the colonoscopic examination schedule (CCA, 2011).

Two recent case-control studies of colonoscopy suggest that colonoscopy is not as effective in decreasing mortality (Baxter et al., 2009) or incidence of right-sided cancer (Lakoff et al., 2008). A recent investigation of surveillance colonoscopy found that subjects were more likely to develop recurrent adenomas in the same colon segment, suggesting that particular attention be paid to where a previous adenoma has been removed (Pinsky et al., 2009).

Leung and colleagues studied colorectal cancer risk despite surveilance colonoscopy and concluded that despite frequent colonoscopy there was a persistent ongoing risk of cancer in the years after the trial. Subjects with a history of advanced adenoma are at increased risk of subsequent cancer and should be followed closely with continued surveillance (Leung et al., 2010).

Importantly, the performance of annual colonoscopy has not been shown to improve fiveyear survival.

#### **3.3 Laboratory tests**

Based on literature and also our results (Lipská et al., 2007), it can be concluded that monitoring of the tumor markers is valuable, mainly in those cases where preoperative CEA and/or CA19-9 were elevated. The level of CEA and CA19-9 increases according to the pTNM stage of the disease. CEA or CA19-9 below the cut-off level does not exclude even a very advanced colorectal cancer. To evaluate the stage of the disease, treatment strategy and prognosis a combination of investigations is necessary. Surveillance based only on CEA and/or CA19-9 is cost-effective, but does not disclose more than 1/3 of patients with relapse. In general practice CEA is often used as the only parameter in the follow-up

Follow Up and Recurrence of Colorectal Cancer 373

Results: At the time of relapse both markers were normal in 31% of the patients. When relapse was diagnosed, in patients with normal preoperative levels, CEA and CA19-9 were below cut-off in 48% and 79% of cases respectively, and in those with primary elevation,

Conclusion: The surveillance based only on CEA and /or CA19-9 was cost-effective, but failed to disclose 1/3 of patients suffering from relapse; markers must be combined with

The aim of the study was to investigate the clinical significance of serum tumor markers and biological activity markers - oncofetal tumormarker CEA, mucin tumormarkers CA19-9, CA242, proliferative tumor markers Thymidine kinase, soluble cytoceratines fragments TPS, TPA, adhesive molecules ICAM - 1, VCAM -1, IGF-1, and adipocytokinins Adiponectin, Leptin in patients with colorectal cancer before primary operation. The study included 142 patients between the ages of 35 - 89 years. We confirmed that CA19-9 is besides CEA an important marker in colorectal cancer. Comparing CA19-9 and CA242 in preoperative staging, CA242 is more specific. Statistical significant difference between early and metastatic stage of colorectal cancer was not confirmed in markers: ICAM-1, VCAM, adiponectin, leptin. Statistical significant difference between early and metastatic stage of colorectal cancer was confirmed in markers: CEA, CA19-9, CA242, TPS, TPA, TK, IGF-1. None of the used markers was able to distinguish stage II and III, in other words to identify patients with infiltration of lymph nodes. This fact is very important in our aspirations to find which marker from periferal blood could help to identify patiens at risk of lymphatic infiltration and select these patients for adjuvant therapy. Combination of CEA and either CA19-9 or CA242 can be recommended for preoperative investigation. CA 242 in this study seems to have slightly better results in preoperative staging (Levý et

Aims: To investigate presence of cell-free tumor DNA and its correlation to clinical status of the patient, especially metastatic liver disease. There has been increased interest in identifying biologic indicators that may help better define patients at risk for recurrence

Methods: In a prospective study cohort of 108 patients we have initially acquired a tissue samples from primary tumor (n=108). Where available, additional tissue was collected from nodes and liver metastases. For each patient, multiple plasma samples were acquired over a period covering initial examination, immediately preceding the surgery, at the surgery, post-surgery and during a subsequent follow-up. We have used the most frequent colorectal somatic mutations (APC, TP53, BRAF, PIK3CA and KRAS) detected in primary tumors to

Results: A total of 66 patients (66/108, 415%) had somatic mutation in primary tumor. From these 66 patients in 57 patients the plasma samples were examined. Where available, mutation from primary tumor was also confirmed in the metastatic liver tissue (4/4, 100%). Cell-free tumor DNA was detected in plasma according to TNM stages in 0%, 10%, 28% and 100% respectively. In 2 patients positivity was detected in subsequent plasma samples,

they were again elevated in 78% and 64% of cases respectively.

liver and chest imaging methods and colonoscopy.

after hepatic resection for colorectal metastases.

trace cell-free tumor DNA in plasma samples.

following the course of the disease development.

**3.4.2 Study 2** 

al., 2008).

**3.4.3 Study 3**

regimens. Based on this system, CEA seems highly effective, but when other investigations are included, 1/3 of relapsed patients are diagnosed by a method other than CEA.

Non-invasive screening of molecular biomarkers( such as cell-free tumor DNA ) may enable effective surgical intervention through an early diagnosis of the disease. To determine the most appropriate diagnostic and therapeutic strategy we need to know not only clinical and histological factors but also molecular factors of the tumor. New impromvent in molecular biology should open the way to new perspectives in research of carcinogenesis, medical (targeted therapy) and surgical treatment (in the appropriate moment and appropriate extent). There has been increased interest in identifying biologic indicators that may help better define patients at risk for recurrence.

#### **3.4 Author´s experience**

At Surgical Department of Thomayer University Hospital, Charles University Prague, there is for more than 15 years intensive system of follow-up applied. See table 3.


Table 3. Follow-up system in Thomayer University Hospital

Four follow-up studies (based on different tumor markers, CT scan, PET/CT scan, ultrasonography and other investigations) performed on author´s department are reported.

#### **3.4.1 Study 1**

Aim: To evaluate CEA and CA19-9 in a long-term follow-up after radical surgery for colorectal cancer.

Patients and Methods: A total of 1090 patients were operated on for colorectal cancer, 716 patients underwent R0 resection, 631 patients were under further surveillance, relapse was diagnosed in 122 patients (20%), 74 patients were indicated for reoperation The resectability of the relapse was 35%. An AxSYM instrument (Abbott) was used for analysis.

Results: At the time of relapse both markers were normal in 31% of the patients. When relapse was diagnosed, in patients with normal preoperative levels, CEA and CA19-9 were below cut-off in 48% and 79% of cases respectively, and in those with primary elevation, they were again elevated in 78% and 64% of cases respectively.

Conclusion: The surveillance based only on CEA and /or CA19-9 was cost-effective, but failed to disclose 1/3 of patients suffering from relapse; markers must be combined with liver and chest imaging methods and colonoscopy.

#### **3.4.2 Study 2**

372 Colorectal Cancer – From Prevention to Patient Care

regimens. Based on this system, CEA seems highly effective, but when other investigations

Non-invasive screening of molecular biomarkers( such as cell-free tumor DNA ) may enable effective surgical intervention through an early diagnosis of the disease. To determine the most appropriate diagnostic and therapeutic strategy we need to know not only clinical and histological factors but also molecular factors of the tumor. New impromvent in molecular biology should open the way to new perspectives in research of carcinogenesis, medical (targeted therapy) and surgical treatment (in the appropriate moment and appropriate extent). There has been increased interest in identifying biologic indicators that may help

At Surgical Department of Thomayer University Hospital, Charles University Prague, there

History X X X X X X X X X X X X X X X

CEA, CA19-9 X X X X X X X X X X X X X X X Ultrasonography X X X X X X X X X X X X X X X CT X X X X X X X X X Coloscopy X X X X X X X X X

examination <sup>X</sup> <sup>X</sup> X X X X X X X X X X X X

Chest X ray X X X X

Table 3. Follow-up system in Thomayer University Hospital

PET, PET/CT In the case of unsolved elevation in markers or before reoperation

Four follow-up studies (based on different tumor markers, CT scan, PET/CT scan, ultrasonography and other investigations) performed on author´s department are

Aim: To evaluate CEA and CA19-9 in a long-term follow-up after radical surgery for

Patients and Methods: A total of 1090 patients were operated on for colorectal cancer, 716 patients underwent R0 resection, 631 patients were under further surveillance, relapse was diagnosed in 122 patients (20%), 74 patients were indicated for reoperation The resectability

of the relapse was 35%. An AxSYM instrument (Abbott) was used for analysis.

Months after operation

120

108, 144, 166

X

3 6 12 18 24 30 36 42 48 54 60 72 84 96,

is for more than 15 years intensive system of follow-up applied. See table 3.

are included, 1/3 of relapsed patients are diagnosed by a method other than CEA.

better define patients at risk for recurrence.

**3.4 Author´s experience** 

Physical

reported.

**3.4.1 Study 1** 

colorectal cancer.

The aim of the study was to investigate the clinical significance of serum tumor markers and biological activity markers - oncofetal tumormarker CEA, mucin tumormarkers CA19-9, CA242, proliferative tumor markers Thymidine kinase, soluble cytoceratines fragments TPS, TPA, adhesive molecules ICAM - 1, VCAM -1, IGF-1, and adipocytokinins Adiponectin, Leptin in patients with colorectal cancer before primary operation. The study included 142 patients between the ages of 35 - 89 years. We confirmed that CA19-9 is besides CEA an important marker in colorectal cancer. Comparing CA19-9 and CA242 in preoperative staging, CA242 is more specific. Statistical significant difference between early and metastatic stage of colorectal cancer was not confirmed in markers: ICAM-1, VCAM, adiponectin, leptin. Statistical significant difference between early and metastatic stage of colorectal cancer was confirmed in markers: CEA, CA19-9, CA242, TPS, TPA, TK, IGF-1. None of the used markers was able to distinguish stage II and III, in other words to identify patients with infiltration of lymph nodes. This fact is very important in our aspirations to find which marker from periferal blood could help to identify patiens at risk of lymphatic infiltration and select these patients for adjuvant therapy. Combination of CEA and either CA19-9 or CA242 can be recommended for preoperative investigation. CA 242 in this study seems to have slightly better results in preoperative staging (Levý et al., 2008).

#### **3.4.3 Study 3**

Aims: To investigate presence of cell-free tumor DNA and its correlation to clinical status of the patient, especially metastatic liver disease. There has been increased interest in identifying biologic indicators that may help better define patients at risk for recurrence after hepatic resection for colorectal metastases.

Methods: In a prospective study cohort of 108 patients we have initially acquired a tissue samples from primary tumor (n=108). Where available, additional tissue was collected from nodes and liver metastases. For each patient, multiple plasma samples were acquired over a period covering initial examination, immediately preceding the surgery, at the surgery, post-surgery and during a subsequent follow-up. We have used the most frequent colorectal somatic mutations (APC, TP53, BRAF, PIK3CA and KRAS) detected in primary tumors to trace cell-free tumor DNA in plasma samples.

Results: A total of 66 patients (66/108, 415%) had somatic mutation in primary tumor. From these 66 patients in 57 patients the plasma samples were examined. Where available, mutation from primary tumor was also confirmed in the metastatic liver tissue (4/4, 100%). Cell-free tumor DNA was detected in plasma according to TNM stages in 0%, 10%, 28% and 100% respectively. In 2 patients positivity was detected in subsequent plasma samples, following the course of the disease development.

Follow Up and Recurrence of Colorectal Cancer 375

Amitai MM, Fišer H, Avidan B, Portnoy O, Apter S, Konen E, Hertz M. Contrast-enhanced

Baxter NN, Goldwasser MA, Paszat LF, et al. Association of colonoscopy and death from

Begent RHJ: The value of carcinoembryonic antigen measurement in clinical practice. *Ann* 

Bouvier AM, Latournerie M, Jooste V, Lepage C, Cottet V, Faivre J. The lifelong risk of

Bowne WB, Lee B, Wong WD, Ben-Porat L, Shia J, Cohen AM, Enker WE, Guillem JG,

Cali RL, Pitsch RM, Thomson AG, et al. Cumulative incidence of metachronous colorectal

CCA, Cancer Council Australia Colonoscopy Surveillance Working Party. Clinical

Deleau C, Buecher B, Rousseau C, Kraeber-Bodéré F, Flamant M, des Varannes SB, Frampas

Desch CE,. Benson III Al B,. Somerfield MR, Flynn PJ, Krause C,. Loprinzi CL,. Minsky BD,

Figueredo A, Rumble RB, Maroun J, Earle CC, Cummings B, McLeod R, Zuraw L, Zwaal C;

Gan S, Wilson K, Hollington P. Surveilance of patients following surgery with curative

Griffin MR, Bergtralh EJ, Coffey RJ, Beart RW and Melton LJ: Predictors of survival after

intent for colorectal cancer. *World J Gastronterol* 2007; 13: 3816-3823

cancer: a practiceguideline. *BMC Cancer*. 2003 Oct 6;3:26 Fletcher RH: Carcinoembryonic antigen. *Ann Int Med* 1986; 104: 66-73

metachronous colorectal cancer justifies long-term colonoscopic follow-up. *Eur J* 

Paty PB, Weiser MR. Operative salvage for locoregional recurrent colon cancer after curative resection: an analysis of 100 cases. *Dis Colon Rectum* 2005; 48: 897-

Practice Guidelines for Surveillance Colonoscopy – in adenoma follow-up; following curative resection of colorectal cancer; and for cancer surveillance in inflammatory bowel disease. Cancer Council Australia, Sydney 2011,

E, Galmiche JP, Matysiak-Budnik T.Clinical impact of fluorodeoxyglucose-positron emission tomography scan/computed tomography in comparison with computed tomography on the detection of colorectal cancer recurrence.*Eur J Gastroenterol* 

Pfister DG, Virgo KS, Petrelli NJ .American Society of Clinical OncologyColorectal Cancer Surveillance: 2005 Update of an American Society of Clinical Oncology

Gastrointestinal Cancer Disease Site Group of Cancer Care Ontario's Program in Evidence-based Care. Follow-up of patients with curatively resected colorectal

curative resection of carcinoma of the colon and rectum. *Cancer* 1987, 60:2318-

colorectal cancer. *Ann Intern Med* 2009;150:1-8

cancer. *Dis Colon Rectum* 1993; 36: 388-93

438

909

BCMA, http://www.bcguidelines.ca

*Clin Biochem* 1984; 21: 231-238

*Cancer* 2008;44(4):522-527.

www.cancer.org.au

Practice Guideline

2324

*Hepatol*. 2011 Mar;23(3):275-81

CT colonography with 64-slice MDCT compared to endoscopic colonoscopy in the follow-up of patients after colorectal cancer resection. *Clinical Imaging* 2009;33: 433–

Conclusion: Our results indicate a potential for the detection of cell-free tumor DNA as a non-invasive test of metastatic liver disease. Somatic mutations in additional genes (BRAF, PIK3CA) are now being explored as markers to increase the number of patients that can be evaluated from cell-free tumor DNA.

#### **3.4.4 Study 4**

Aim of the study:The conventional diagnostic techniques used to assess recurrence of colorectal cancer (CRC) often yield unspecific findings. Integrated FDG-PET/CT seems to offer promise for the differential diagnosis of benign and malignant lesions. The aim of this study was to compare the value of FDG-PET and PET/CT in the detection of CRCR subsequent to colonic resection or rectal amputation.

Methods:The population for this retrospective study comprised 84 patients with suspected CRC. The sensitivity, specificity and accuracy of PET and PET/CT were calculated for (a) intra-abdominal extrahepatic recurrences, (b) extra-abdominal and/or hepatic recurrences and (c) all recurrences, and tumour marker levels were analysed.

Results: The sensitivity, specificity and overall accuracy of PET in detecting intra-abdominal extrahepatic CRC were 82%, 88% and 86%, respectively, compared with 88%, 94% and 92%, respectively, for PET/CT. The corresponding figures for detection of extra-abdominal and/or hepatic CRC were 74%, 88% and 85% for PET and 95%, 100% and 99% for PET/CT. Considering the entire population, the sensitivity, specificity and overall accuracy of PET were 80%, 69% and 75%, respectively, compared with 89%, 92% and 90%, respectively, for PET/CT. FDG-PET/CT examination correctly detected 40 out of a total of 45 patients with CRC. Two of five patients with falsely negative FDG-PET/CT findings had local microscopic recurrences and one had miliary liver metastases. Of 39 patients without CRC, three showed false positive FDG-PET/CT results. Two of these cases were due to increased accumulation in inflammatory foci in the bowel wall, while one was due to haemorrhaging into the adrenal gland.

Conclusion:FDG-PET/CT appears to be a very promising method for distinguishing a viable tumour from fibrous changes, thereby avoiding unnecessary laparotomy.(Votrubova et al.,2006)

#### **4. Conclusion**

The goal of any surveillance program should be detection of recurrent disease at a early time to allow subsequent curative therapy. Periodic clinical examinations, laboratory tests, radiographic imaging, and markers testing is necessary. The optimal combination and frequency of investigations in follow-up of patients after CRC resection has not been determined. It seems that intensive follow-up increased the re-resection rate for recurrent disease and improved overall survival but the survival advantage was not due to earlier detection of recurrence and cancer-related mortality was no better.

All patients having recurrences should be assessed by a multidisciplinary team in a cancer centre.

#### **5. References**

Ahmed N, Ahmedzai S, Vora V, Hillam S, Paz S. Supportive care for patients with gastrointestinal cancer. *Cochrane Database Syst Rev* 2004: CD003445

Conclusion: Our results indicate a potential for the detection of cell-free tumor DNA as a non-invasive test of metastatic liver disease. Somatic mutations in additional genes (BRAF, PIK3CA) are now being explored as markers to increase the number of patients that can be

Aim of the study:The conventional diagnostic techniques used to assess recurrence of colorectal cancer (CRC) often yield unspecific findings. Integrated FDG-PET/CT seems to offer promise for the differential diagnosis of benign and malignant lesions. The aim of this study was to compare the value of FDG-PET and PET/CT in the detection of CRCR

Methods:The population for this retrospective study comprised 84 patients with suspected CRC. The sensitivity, specificity and accuracy of PET and PET/CT were calculated for (a) intra-abdominal extrahepatic recurrences, (b) extra-abdominal and/or hepatic recurrences

Results: The sensitivity, specificity and overall accuracy of PET in detecting intra-abdominal extrahepatic CRC were 82%, 88% and 86%, respectively, compared with 88%, 94% and 92%, respectively, for PET/CT. The corresponding figures for detection of extra-abdominal and/or hepatic CRC were 74%, 88% and 85% for PET and 95%, 100% and 99% for PET/CT. Considering the entire population, the sensitivity, specificity and overall accuracy of PET were 80%, 69% and 75%, respectively, compared with 89%, 92% and 90%, respectively, for PET/CT. FDG-PET/CT examination correctly detected 40 out of a total of 45 patients with CRC. Two of five patients with falsely negative FDG-PET/CT findings had local microscopic recurrences and one had miliary liver metastases. Of 39 patients without CRC, three showed false positive FDG-PET/CT results. Two of these cases were due to increased accumulation in inflammatory foci in the bowel wall, while one was due to haemorrhaging

Conclusion:FDG-PET/CT appears to be a very promising method for distinguishing a viable tumour from fibrous changes, thereby avoiding unnecessary laparotomy.(Votrubova

The goal of any surveillance program should be detection of recurrent disease at a early time to allow subsequent curative therapy. Periodic clinical examinations, laboratory tests, radiographic imaging, and markers testing is necessary. The optimal combination and frequency of investigations in follow-up of patients after CRC resection has not been determined. It seems that intensive follow-up increased the re-resection rate for recurrent disease and improved overall survival but the survival advantage was not due to earlier

All patients having recurrences should be assessed by a multidisciplinary team in a cancer

Ahmed N, Ahmedzai S, Vora V, Hillam S, Paz S. Supportive care for patients with

gastrointestinal cancer. *Cochrane Database Syst Rev* 2004: CD003445

evaluated from cell-free tumor DNA.

subsequent to colonic resection or rectal amputation.

and (c) all recurrences, and tumour marker levels were analysed.

detection of recurrence and cancer-related mortality was no better.

**3.4.4 Study 4** 

into the adrenal gland.

et al.,2006)

centre.

**5. References** 

**4. Conclusion** 


Follow Up and Recurrence of Colorectal Cancer 377

Lipská L, Visokai V, Levý M, Svobodová S, Kormunda S, Fínek J. Tumor markers in

Makela JT, Laitinen OS, Kairaluoma MI. Five-year follow-up after radical surgery for

Moertel CG, Fleming TR, Macdonald J, Haller DG, Laurie JA, Tangen C: An evaluation of

NCCN Clinical Practice Guidelines in Oncology, Colon Cancer, vision 3. 2011,

Ohlsson B, Breland U, Ekberg H, et al. Follow-up after curative surgery for colorectal

Pietra N, Sarli L, Costi R, et al. Role of follow-up in management of local recurrences of

Pinsky PF, Schoen RE, Weissfeld JL, et al. The yield of surveillance colonoscopy by adenoma history and time to examination. *Clin Gastroenterol Hepatol* 2009;7:86-92 Piscaglia F, Corradi F, Mancini M, Giangregorio F, Tamberi S, Ugolini G, Cola B, Bazzocchi

Renehan AG, Egger M, Saunders MP, et al. Impact on survival of intensive follow up after

Reuterskiold MH, Lasson A, Svenson E, Kilander A, Stoztzer PO, Hellstrom M. Diagnostic

Rosen M, Chan L, Beart RW, Vukasin P, Anthone G: Follow-up of colorectal cancer, a meta-

Secco GB, Fardelli R, Rovida S, Gianquinto D, Baldi E, Bonfante P, Derchi L, Ferraris R. Is

Shamim SA, Kumar R, Shandal V, Halanaik D, Kumar G, Bal CS, Malhotra A.FDG PET/CT

Scholefield JH, Steele RJ; British Society For Gastroenterology; Association of

resection of colorectal cancer. *Gut*. 2002 Oct;51 Suppl 5:V3-5

randomized trials. *British Medical Journal* 2002;324:1–8

Aug;27(4A):1901-5

1995;130:1062–7

www.nccn.org

1995;38:619–26

*Cancer*. 2007 Sep 3;7:171

Feb;7(1):32-7.

*of Clinical Oncology* 2006; 24: 5313-27

*Nucl Med*. 2011 Jan;36(1):11-6.

analysis. *Dis Colon Rectum 41:* 1116-1126, 1998.

33

colon cancer. *JAMA* 1993; 270: 943-947

patients with relapse of colorectal carcinoma. *Anticancer Res*. 2007 Jul-

colorectal cancer: results of a prospective randomized trial. *Archives of Surgery*

the carcinoembryonic antigen (CEA) test for monitoring patients with resected

carcinoma: randomized comparison with no follow-up. *Dis Colon Rectum*

colorectal cancer: a prospective randomized study. *Dis Colon Rectum* 1998;41:1127–

A, Righini R, Pini P, Fornari F, Bolondi L.Real time contrast enhanced ultrasonography in detection of liver metastases from gastrointestinal cancer.*BMC* 

curative resection for colorectal cancer: systematic review and meta-analysis of

performance of computed tomography colonography in symptomatic patients and in patients with increased risk for colorectal disease. *Acta Radiol* 2006;47:888–98 Rocker GY, Hamilton S, Haras J. American Society of Clinical Oncology 2006 Update of

Recommendations for the Use of Tumor Markers in Gastrointestinal Cancer, *Journal* 

intensive follow-up really able to improve prognosis of patients with local recurrence after curative surgery for rectal cancer? *Ann Surg Oncol*. 2000 Jan-

evaluation of treatment response in patients with recurrent colorectal cancer.*Clin* 

Coloproctology for Great Britain and Ireland. Guidelines for follow up after


Grossmann EM, Johnson FE, Virgo KS, Longo WE, Fossati R. Follow-up of colorectal cancer

Guyot F, Faivre J, Manfredi S, Meny B, Bonithon-Kopp C, Bouvier AM. Time trends in the

Halligan S, Altman DG, Taylor SA, Mallet S, Deeks JJ, Bartram CI, Atkin W. CT

Han A, Xue J, Zhu D, Zheng J, Yue J, Yu J.Clinical value of (18)F-FDG PET/CT in

Chau I, Allen MJ, Cunningham D, Norman AR, Brown G, Ford HE, Tebbutt N, Tait D, Hill

Ike H, Shimada H, Togo S, Yamaguchi S, Ichikawa Y, Tanaka K. Sequential resection of lung

Jeffery M, Hickey BE, Hider PN. Follow-up strategies for patients treated for non-metastatic

Kjeldsen BJ, Kronborg O, Fenger C, et al. A prospective randomized study of follow-up after

Labianca R, Nordlinger B, Berete GD, et al. Primary colon cancer: ESMO Clinical Practice

Lakoff J, Paszat LF, Saskin R, et al. Risk of developing proximal versus distal colorectal

Larsen LP, Rosenkilde M, Christensen H, Bang N, Bolvig L, Christiansen T, Laurberg S.Can

Larsen LP, Rosenkilde M, Christensen H, Bang N, Bolvig L, Christiansen T, Laurberg S.The

Leung K, Pinsky P, Laiyemo AO, Lanza E, Schatzkin A, Schoen RE. Ongoing colorectal

Levý M, Visokai V, Lipská L, Topolčan O. Tumor markers in staging and prognosis of

Follow-up Study. *Gastrointestinal Endoscopy* 2010; 71(1): 111-7

colorectal carcinoma. *Neoplasma*. 2008;55(2):138-42.

colorectal cancer (Review), *The Cochrane Library* 2007, Issue 4

radical surgery for colorectal cancer. *Br J Surg* 1997;84:666–91

2004; 13:119–124

2005;237: 893–904

2011 Feb 4[Epub ahead of print]

*Oncol* 2004; 22: 1420-1429

1164-1168

21(Suppl 5): 70-77

Feb;69(2):308-13

May;62(2):302-7

*Hepatol* 2008;6:1117-21

756-761

patients after resection with curative intent—the GILDA trial. *Surgical Oncology*

treatment and survival of reccurences from colorectal cancer. *Ann Oncol* 2005; 16:

colonography in the detection of colorectal polyps and cancer: systematic review, meta-analysis, and proposed minimum data set for study level reporting. *Radiology*

postoperative monitoring for patients with colorectal carcinoma.*Cancer Epidemiol*.

M, Ross PJ, Oates J. The value of routine serum CEA measurement and CT in the surveilance of patients after adjuvant chemotherapy for colorectal cancer. *J Clin* 

metastasis following partial hepatectomy for colorectal cancer. *Br J Surg* 2002; 89:

Guidelines for diagnosis, adjuvant treatment and follow-up. Ann Oncol 2010;

cancer after a negative colonoscopy: a populationbased study. *Clin Gastroenterol* 

contrast-enhanced ultrasonography replace multidetector-computed tomography in the detection of liver metastases from colorectal cancer? *Eur J Radiol*. 2009

value of contrast enhanced ultrasonography in detection of liver metastases from colorectal cancer: a prospective double-blinded study.*Eur J Radiol*. 2007

cancer risk despite surveillance colonoscopy: the Polyp Prevention Trial Continued


**Part 5** 

**Metastasis** 


## **Part 5**

**Metastasis** 

378 Colorectal Cancer – From Prevention to Patient Care

Tjandra JJ, Chan MK. Follow-Up After Curative Resection of Colorectal Cancer: A Meta-

Votrubova J, Belohlavek O, Jaruskova M, Oliverius M, Lohynska R, Trskova K, Sedlackova

Winter S, Fletcher R, Rex D, et al. Colorectal Cancer Screening and Surveillance: Clinical

colorectal cancer.Eur J *Nucl Med Mol Imaging*. 2006 Jul;33(7):779-84

E, Lipska L, Stahalova V.The role of FDG-PET/CT in the detection of recurrent

Guidelines and Rationale—Update Based on New Evidence, *Gastroenterology* 2003;

Analysis.*Dis Colon Rectum*. 2007;50(11):1783-99

124: 544-60

**20** 

**Panitumumab for the Treatment of** 

*2Kálmán Pándy Hospital of the Local Government of Békés County,* 

*3Kálmán Pándy Hospital of the Local Government of Békés County,* 

According to data by GLOBOCAN, the worldwide incidence of colorectal cancer in 2008 was 1,234,000 (with 663,000 male and 571,000 female cases). The number of deaths due to this disease was 608,000 (320,000 men and 288,000 women). Given these figures, colorectal cancer is the third and second leading cause of mortality among men and women. In the recent year in Hungary with a population around 10 million the annual incidence among males was 4,415, whereas the number of afflicted women was 3,690. Mortality data is similar with deaths among men and women being 2,563 and 2,190 respectively. Therefore, the disease is the second leading cause of death for both genders worldwide and in Hungary as well (Gaudi & Kásler, 2002; Ottó & Kásler, 2005; World Health Organization WHO – International Agency for Research on Cancer IARC, 2008). In international comparison Hungarian colorectal cancer mortality rates for 2008 were the highest in Europe for both men (31.4 per 100,000) and women (16.2 per 100,000). This is in striking contrast to comparable figures of Albanian men (6.2 per 100,000) and women (5.8 per 100,000), with the lowest registered numbers (WHO – IARC, 2008). Both frequencies of the disease and continuously improving treatment results highlight the accentuated place colorectal cancer takes in routine oncology practice and at the same time oblige health care services to

As a result of organizational efforts in the last decades to improve professional cooperation, leading to the development of new drugs and to a more conscious treatment planning with a closer to optimal use of combinations, metastatic colorectal cancer (mCRC) has become a chronic disease (Haller, 2007; Khan et al., 2008; Khan et al., 2010; Phillips & Currow, 2010;

**1. Introduction** 

provide the best possible treatment for patients.

van der Velden et al., 2009; van Kleffen et al., 2004).

*4University of Szeged, Albert Szent-Györgyi Clinical Center,* 

**Metastatic Colorectal Cancer** 

*1Kálmán Pándy Hospital of the Local Government of Békés County, County, Center of Oncology, Gyula,* 

Béla Pikó1, Ali Bassam1, Enikő Török2, Henriette Ócsai3 and Farkas Sükösd4

 *County Department of Radiology, Gyula,* 

*Department of Pathology, Szeged,* 

*Hungary* 

*Dermato-Oncology Outpatient Clinic, Gyula,* 

### **Panitumumab for the Treatment of Metastatic Colorectal Cancer**

Béla Pikó1, Ali Bassam1, Enikő Török2, Henriette Ócsai3 and Farkas Sükösd4

*1Kálmán Pándy Hospital of the Local Government of Békés County, County, Center of Oncology, Gyula, 2Kálmán Pándy Hospital of the Local Government of Békés County, County Department of Radiology, Gyula, 3Kálmán Pándy Hospital of the Local Government of Békés County, Dermato-Oncology Outpatient Clinic, Gyula, 4University of Szeged, Albert Szent-Györgyi Clinical Center, Department of Pathology, Szeged, Hungary* 

#### **1. Introduction**

According to data by GLOBOCAN, the worldwide incidence of colorectal cancer in 2008 was 1,234,000 (with 663,000 male and 571,000 female cases). The number of deaths due to this disease was 608,000 (320,000 men and 288,000 women). Given these figures, colorectal cancer is the third and second leading cause of mortality among men and women. In the recent year in Hungary with a population around 10 million the annual incidence among males was 4,415, whereas the number of afflicted women was 3,690. Mortality data is similar with deaths among men and women being 2,563 and 2,190 respectively. Therefore, the disease is the second leading cause of death for both genders worldwide and in Hungary as well (Gaudi & Kásler, 2002; Ottó & Kásler, 2005; World Health Organization WHO – International Agency for Research on Cancer IARC, 2008). In international comparison Hungarian colorectal cancer mortality rates for 2008 were the highest in Europe for both men (31.4 per 100,000) and women (16.2 per 100,000). This is in striking contrast to comparable figures of Albanian men (6.2 per 100,000) and women (5.8 per 100,000), with the lowest registered numbers (WHO – IARC, 2008). Both frequencies of the disease and continuously improving treatment results highlight the accentuated place colorectal cancer takes in routine oncology practice and at the same time oblige health care services to provide the best possible treatment for patients.

As a result of organizational efforts in the last decades to improve professional cooperation, leading to the development of new drugs and to a more conscious treatment planning with a closer to optimal use of combinations, metastatic colorectal cancer (mCRC) has become a chronic disease (Haller, 2007; Khan et al., 2008; Khan et al., 2010; Phillips & Currow, 2010; van der Velden et al., 2009; van Kleffen et al., 2004).

Panitumumab for the Treatment of Metastatic Colorectal Cancer 383

carcinoma, and lung cancer (excluding planocellular or small cell carcinoma-types) (EMA, 2011a). In addition, the U. S. Food and Drug Administration (FDA) has also approved its use in brain tumour recurrences following "traditional" treatment and in advanced brain tumour cases as well (glioblastoma multiforme) (U. S. Food and Drug Administration,

Beva binds to "vascular endothelial growth factor" (VEGF), one of the most important angiogenesis regulators. By doing this, beva inhibits the binding of VEGF to its receptors Flt-1 (VEGFR-1) and KDR (VEGFR-2) on the surface of endothelial cells. The neutralization of VEGF's biological activity lowers tumour vascularisation, normalizes the tumour's surviving vasculature and inhibits the development of a new vascular system for the tumour. By blocking tumour growth beva thus lowers intra-tumour pressure helping anticancer drug delivery to tumour tissue (Bergers & Benjamin, 2003; Borgstrom et al., 1999; EMA, 2011a;

One of its main side effects is high blood pressure (usually successfully treated with ACE inhibitors, calcium channel blockers, or diuretics), and this usually does not necessitate ending or suspending the use of the drug. Therapy-resistant chronic hypertension however may mean a treatment contraindication. The frequency of proteinuria can vary considerably. Its severity can range from laboratory value deviations to development of nephritic syndrome. The severity of the detrimental side effect congestive heart failure can also cover quite a wide spectrum. Reduced left ventricle ejection fraction may ensue without any clinical symptoms but can be represented in a life-threatening form too. A wide variety of arterial and venous thromboembolic complications, as well as bleeding of any grade can occur. Bleedings may represent in the gastrointestinal system, primarily as perforations in patients with metastatic colorectal cancer treated with bevacizumab. Inflammatory intestinal diseases render patients especially susceptible to such perforations. Fistulae can also develop in different areas; perforations of the nasal septum are detected rarely. Reversible posterior leukoencephalopathy syndrome is a rare, neurological disorder which can also develop during beva treatment. Differential diagnosis can be challenging in such cases to rule out headaches, mental disorders, and possible cortical blindness frequently caused by cerebral metastases. (Allen et al., 2006; BC Cancer Agency Cancer Management Guidelines, 2006; Benson et al., 2003; EMA, 2011a; Fakih & Lombardo, 2006; Giantonio et al., 2004; Hamilton, 2008; Kilickap et al., 2003; Martel et al., 2006; Pereg & Lishner, 2008; Scappaticci et

al., 2007; Traina et al., 2006; van Heeckeren et al., 2007; Widakowich et al., 2007).

**antibodies (cmab and pmab) and K-ras mutation status** 

**2.1.2 Correlation between the therapeutic effect of EGFR inhibitor monoclonal** 

Before presenting the mechanism of action of cmab and later that of pmab in details, it is necessary to understand the importance of EGFR status and K-ras mutation. Awareness of EGFR and K-ras mutation status has proven to be essential not only for an apt evaluation and interpretation of clinical trial results, but for adequate patient selection and diagnostics planning as well. A precise determination of both is a prerequisite for an effective treatment in everyday clinical routine. EGFR, a superficial structure of epithelial tumours and also CRC cells is a glycoprotein composed of three subunits. The exodomain receiving the ligand is outside the cell membrane, while the hydrophobic transmembrane domain provides proper cell membrane integration. The cytoplasmic "endodomain" is a catalytic subunit with tyrosine kinase activity. It transmits signals to other proteins by phosphorylating

2009).

Folkman, 1971, Kim al., 1993).

Today median survival of CRC-patients from the diagnosis of distant metastases can reach 36 months on overall. Even in disseminated illness the chances of surviving more than five years are above 12% now (Blaser, 2010; Chau & Cunningham, 2009; Goldberg, 2007; Grothey, 2007; Michael & Zalcberg, 2000; National Cancer Institute NCI, n. d.; Sudoyo, n. d.)

In 2004 Grothey and colleagues presented a diagram in the Journal of cilinical oncology which has been cited countless times ever since. The survival of mCRC patients was plotted on this diagram as a function of the proportion of patients treated with drug combinations considered "basic" (fluoropyrimidine, irinotecan, and oxaliplatin), and multiple linear regression was performed (Grothey et al., 2004) Based on the results it is clear that those the patients that had the greatest chance of survival who had received all three drugs during their treatment. Of course, it is not just "traditional" cytostatic remedies – antimetabolite fluoropyrimidines, the topoisomerase inhibitor irinotecan, and alkalizing agent oxaliplatin – that influence survival (Takimoto and Calvo, 2005). Based on new results, drugs aimed at biological targets do so, on their own and in different combinations with chemotherapy as well, which we will discuss later in detail.

#### **2. Biological targeted drugs**

#### **2.1 Brief description of drugs affecting biological targets**

Drugs currently in use in this category can be classified into two major groups.

A well known and characteristic representative of one of these groups is bevacizumab (Avastin®) (European Medicines Agency EMA, 2011a) inhibiting neoangiogenesis, i.e. this drug slows down the pathological vascularization of tumours and thus inhibits their provision of oxygen and nutrition.

The other group consists of cetuximab (Erbitux®) (EMA, 2010) and panitumumab (Vectibix®) (EMA, 2011b), both influencing the effect of "epidermal growth factor receptors" (EGFR) located on the surface of tumours and in this way both interfere with the regulation of cell division and proliferation (Helbling & Borner, 2007; Mayer, 2009; Siena et al., 2009; Willet et al., 2007).

These are all monoclonal antibodies. As a result of advances in manufacturing technology "chimeras" containing more non-human amino acid sequences (cetuximab – "cmab") were followed by "humanized" antibodies like bevacizumab ("beva") with increased proportion of human sequences within the molecule. The ultimate result of this process is the development of monoclonal antibodies containing exclusively human amino acid sequences (panitumumab – "pmab"). The ratio of human and non-human amino acid sequences within a given therapeutic antibody medication is crucial—the presence of the latter usually necessitates the use of saturating doses, while fully human substances can be administered using the same dose from the start of therapy. Human versus non human composition of complex protein molecules administered via infusion is also a key determinant of the frequency of infusion related and other side effects caused by "foreign proteins" (Eng, 2010; de Bono & Rowinsky, 2002; EMA, 2009; EMA, 2011a, b; Hochster, 2006; LoBuglio, 1989; Yang et al., 2001).

#### **2.1.1 Bevacizumab**

Generally used in combination with traditional cytostatic drugs, bevacizumab has been approved in Europe for many types of tumors: mCRC, breast cancer, clear cell renal cell

Today median survival of CRC-patients from the diagnosis of distant metastases can reach 36 months on overall. Even in disseminated illness the chances of surviving more than five years are above 12% now (Blaser, 2010; Chau & Cunningham, 2009; Goldberg, 2007; Grothey, 2007;

In 2004 Grothey and colleagues presented a diagram in the Journal of cilinical oncology which has been cited countless times ever since. The survival of mCRC patients was plotted on this diagram as a function of the proportion of patients treated with drug combinations considered "basic" (fluoropyrimidine, irinotecan, and oxaliplatin), and multiple linear regression was performed (Grothey et al., 2004) Based on the results it is clear that those the patients that had the greatest chance of survival who had received all three drugs during their treatment. Of course, it is not just "traditional" cytostatic remedies – antimetabolite fluoropyrimidines, the topoisomerase inhibitor irinotecan, and alkalizing agent oxaliplatin – that influence survival (Takimoto and Calvo, 2005). Based on new results, drugs aimed at biological targets do so, on their own and in different combinations with chemotherapy as

Michael & Zalcberg, 2000; National Cancer Institute NCI, n. d.; Sudoyo, n. d.)

well, which we will discuss later in detail.

**2.1 Brief description of drugs affecting biological targets** 

Drugs currently in use in this category can be classified into two major groups.

A well known and characteristic representative of one of these groups is bevacizumab (Avastin®) (European Medicines Agency EMA, 2011a) inhibiting neoangiogenesis, i.e. this drug slows down the pathological vascularization of tumours and thus inhibits their

The other group consists of cetuximab (Erbitux®) (EMA, 2010) and panitumumab (Vectibix®) (EMA, 2011b), both influencing the effect of "epidermal growth factor receptors" (EGFR) located on the surface of tumours and in this way both interfere with the regulation of cell division and proliferation (Helbling & Borner, 2007; Mayer, 2009; Siena et al., 2009; Willet et

These are all monoclonal antibodies. As a result of advances in manufacturing technology "chimeras" containing more non-human amino acid sequences (cetuximab – "cmab") were followed by "humanized" antibodies like bevacizumab ("beva") with increased proportion of human sequences within the molecule. The ultimate result of this process is the development of monoclonal antibodies containing exclusively human amino acid sequences (panitumumab – "pmab"). The ratio of human and non-human amino acid sequences within a given therapeutic antibody medication is crucial—the presence of the latter usually necessitates the use of saturating doses, while fully human substances can be administered using the same dose from the start of therapy. Human versus non human composition of complex protein molecules administered via infusion is also a key determinant of the frequency of infusion related and other side effects caused by "foreign proteins" (Eng, 2010; de Bono & Rowinsky, 2002; EMA, 2009; EMA, 2011a, b; Hochster, 2006; LoBuglio, 1989;

Generally used in combination with traditional cytostatic drugs, bevacizumab has been approved in Europe for many types of tumors: mCRC, breast cancer, clear cell renal cell

**2. Biological targeted drugs** 

provision of oxygen and nutrition.

al., 2007).

Yang et al., 2001).

**2.1.1 Bevacizumab** 

carcinoma, and lung cancer (excluding planocellular or small cell carcinoma-types) (EMA, 2011a). In addition, the U. S. Food and Drug Administration (FDA) has also approved its use in brain tumour recurrences following "traditional" treatment and in advanced brain tumour cases as well (glioblastoma multiforme) (U. S. Food and Drug Administration, 2009).

Beva binds to "vascular endothelial growth factor" (VEGF), one of the most important angiogenesis regulators. By doing this, beva inhibits the binding of VEGF to its receptors Flt-1 (VEGFR-1) and KDR (VEGFR-2) on the surface of endothelial cells. The neutralization of VEGF's biological activity lowers tumour vascularisation, normalizes the tumour's surviving vasculature and inhibits the development of a new vascular system for the tumour. By blocking tumour growth beva thus lowers intra-tumour pressure helping anticancer drug delivery to tumour tissue (Bergers & Benjamin, 2003; Borgstrom et al., 1999; EMA, 2011a; Folkman, 1971, Kim al., 1993).

One of its main side effects is high blood pressure (usually successfully treated with ACE inhibitors, calcium channel blockers, or diuretics), and this usually does not necessitate ending or suspending the use of the drug. Therapy-resistant chronic hypertension however may mean a treatment contraindication. The frequency of proteinuria can vary considerably. Its severity can range from laboratory value deviations to development of nephritic syndrome. The severity of the detrimental side effect congestive heart failure can also cover quite a wide spectrum. Reduced left ventricle ejection fraction may ensue without any clinical symptoms but can be represented in a life-threatening form too. A wide variety of arterial and venous thromboembolic complications, as well as bleeding of any grade can occur. Bleedings may represent in the gastrointestinal system, primarily as perforations in patients with metastatic colorectal cancer treated with bevacizumab. Inflammatory intestinal diseases render patients especially susceptible to such perforations. Fistulae can also develop in different areas; perforations of the nasal septum are detected rarely. Reversible posterior leukoencephalopathy syndrome is a rare, neurological disorder which can also develop during beva treatment. Differential diagnosis can be challenging in such cases to rule out headaches, mental disorders, and possible cortical blindness frequently caused by cerebral metastases. (Allen et al., 2006; BC Cancer Agency Cancer Management Guidelines, 2006; Benson et al., 2003; EMA, 2011a; Fakih & Lombardo, 2006; Giantonio et al., 2004; Hamilton, 2008; Kilickap et al., 2003; Martel et al., 2006; Pereg & Lishner, 2008; Scappaticci et al., 2007; Traina et al., 2006; van Heeckeren et al., 2007; Widakowich et al., 2007).

#### **2.1.2 Correlation between the therapeutic effect of EGFR inhibitor monoclonal antibodies (cmab and pmab) and K-ras mutation status**

Before presenting the mechanism of action of cmab and later that of pmab in details, it is necessary to understand the importance of EGFR status and K-ras mutation. Awareness of EGFR and K-ras mutation status has proven to be essential not only for an apt evaluation and interpretation of clinical trial results, but for adequate patient selection and diagnostics planning as well. A precise determination of both is a prerequisite for an effective treatment in everyday clinical routine. EGFR, a superficial structure of epithelial tumours and also CRC cells is a glycoprotein composed of three subunits. The exodomain receiving the ligand is outside the cell membrane, while the hydrophobic transmembrane domain provides proper cell membrane integration. The cytoplasmic "endodomain" is a catalytic subunit with tyrosine kinase activity. It transmits signals to other proteins by phosphorylating

Panitumumab for the Treatment of Metastatic Colorectal Cancer 385

affinity and specificity to human EGFR. It inhibits receptor autophosphorylation caused by all known EGFR ligands by attaching to the ligand-binding domain. Binding of pmab to EGFR results in the internalization of the receptor, inhibition of cell growth, induction of apoptosis, and decreased interleukin-8 and vascular endothelial growth factor production (Berardi et al., 2010; EMA, 2011b; Harari, 2004; Helbling & Borner, 2007; Keating, 2010;

The recommended dose of Vectibix is 6 mg/kg of bodyweight once every two weeks both in monotherapy and when combined with cytostatics. Prior to infusion Vectibix should be diluted in 100 mL of 0.9% sodium chloride solution to a final concentration not exceeding 10 mg/mL. Vectibix must be administered as an intravenous infusion via an infusion pump using a low protein binding 0.2 or 0.22 micrometer in-line filter through a peripheral line or indwelling catheter. The recommended infusion time is approximately 60 minutes (Alberta Health Services, 2010; EMA, 2011b). The first dose injected over 60 minutes was well tolerated in clinical trials where Vectibix was combined with cytostatic agents; subsequent treatments were allowed to be given over 30 minutes (Douillard et al., 2010; Peeters et al., 2010). Doses higher than 1,000 mg should be administered as a 150 mL solution over approx. 90 minutes. No incompatibilities have been observed with 0.9% sodium chloride injection in

The common pharmacological effect of EGFR inhibitors can lead to the following: EGFR inhibition in the skin, hair follicles, and periungual tissues can cause abnormal proliferation, migration and differentiation of target cells (i.e. basal keratinocytes), while changes in the skin structure attract inflammatory cells. Clinical symptoms emerge within 10 days following the introduction of pmab therapy and resolve in 28 days after the last injection on average. Skin symptoms are characteristic: papular skin rash, monomorphic pustular lesions, etc. presenting on skin areas exposed to the sun. Although signs may resemble those of acne for the first sight (labeled as "acneiform"), differentiation is easy and essential. Acne may manifest as non-inflammatory lesions on the basis of comedos or as inflammatory papules, pustules, or nodules. On the contrary, rash due to EGFR inhibitors is dominated by pustules. Non-inflammatory comedos are never seen in these cases. Skin rash is more widespread than classical acne as symptoms can often be observed on the upper and lower extremities and trunks of patients simultaneously. In order to prevent nail diseases it is important to avoid mechanical injuries (e.g. caused by tight shoes). Development of paronychia can be stopped by bathing the foot in diluted antiseptic agents and by using topical antiseptic ointments. Feet should not be soaked for a long time to prevent tissues from loosening. In some cases surgery cannot be avoided (Busam et al., 2001; Eaby, n. d.; EMA, 2011b; Moy & Goss, 2007; Pérez-Soler et al., 2005; Segaert & van Cutsem, 2005;

Conventional modalities to treat acne should not be used. On the contrary, advices and interventions are usually completely different from those applied during acne therapy. Sun bathing is prohibited, patients should protect themselves from any direct sunlight (hat, long-sleeved clothes, and sun screens are recommended). Dryness of the skin should be treated with neutral emollients. Caution is warranted if topical steroid drugs are used. Such

Martinelli et al., 2007; Peeters et al., 2008; Pikó, 2009; Rakkar, 2007).

polyvinyl chloride bags or polyolefin bags (EMA, 2011b; Knudson, 2007).

**3.2 Using Vectibix** 

**3.3. Side effects of panitumumab** 

**3.3.1 Skin toxicity** 

Winkeljohn, 2008).

messenger routes. In a complex mechanism, EGF activation initiates cell division following the reception of an adequate external signal. It also assures survival and inhibits apoptosis. The resulting effect is cell proliferation. While this mechanism is strictly controlled in healthy cells, EGF activation is uncontrolled in a considerable proportion of epithelial tumours. The signal is transmitted to other proteins via the biochemical route of tyrosine kinase by phosphorylation. EGF activation can initiate cell division, proliferation, development of metastases and inhibition of apoptosis. Apparently, this leads to tumour progression (Cohenuram & Saif, 2008; Coutinho & Rocha Lima, 2003; EMA, 2009; EMA 2011b; Harari, 2004; Hamilton, 2008; Herbst & Shin, 2002; (Ritter & Arteaga, 2003; van Cutsem et al., 2009).

EGFR inhibitors (cmab and pmab) are licensed for the treatment of mCRC patients. They bind to the extracellular ligand-binding domain and thus inhibit transmembrane signal transmission and prevent EGF dependent signal transduction within the cell as well. Although the mechanism of action has already been established in theory, EGFR inhibitors yield clinical improvement to not more than approximately 50% of mCRC patients. This observation led to the assumption that a biological factor could have prevented these monoclonal antibodies from being effective in tumours expressing EGFR. The K-ras ("Kirsten rat sarcoma 2 viral oncogene homolog") gene belongs to the family of RAS protooncogenes. The K-ras protein coded by this gene plays a central role in growth-inducing signal transmission routes. By doing so it affects cell reproduction, differentiation and survival. If a mitogenic signal reaches the EGF receptor, the signal is forwarded to the nucleus by the K-ras. It is essential that this close correlation applies only to the "normal" (i.e. non-mutated or "wild type") K-ras. Mutant types of K-ras escape receptorial regulation and thus they autonomously stimulate cell proliferation.

For this reason K-ras mutation is not a genetic failure with "function loss", on the contrary, in this case RAS remains in "on" status (i.e. phosphorylation is continuous) and acts independently from EGFR (and other physiological signaling pathways). As a consequence, despite the signals reaching the cell surface being "blocked" by monoclonal antibodies at the receptor level, signaling tracks regulated by EGFR under normal conditions remain (chronically) activated (Amado et al., 2008; Benvenuti et al., 2007; Dahabreh et al., 2011; De Roock et al., 2010; (Engstrom et al., 2011a, b; Esteller et al., 2001; EMA, 2009; EMA 2011b; Hamilton, 2008; Heinemann et al., 2009; Malumbres & Barbacid, 2003; Normanno et al., 2009).

As the estimated incidence of K-ras mutation in CRC is 30-50%, it is expected that in about half to two thirds of patients the regulation of signal effect and signal transmission are preserved and drugs acting via the K-ras route can be used with success. (Amado et al., 2008; Benvenuti et al., 2007; Bardelli & Sien, 2010; Esteller et al., 2001; Garcia-Sáenz et al., 2009; Malumbres & Barbacid, 2003; Nagasaka et al., 2004). In an interesting re-evaluation of their primary study population Hurwitz et al. found that though bev combined with IFL as a first line treatment of mCRC was effective in both K-ras wild type and mutant subgroups, efficacy was by large affected by K-ras status, underlining a mixed predictive and prognostic function of this mutation (Hurwitz et al., 2009).

#### **3. Characteristics, application and side effects of panitumumab**

#### **3.1 Characteristics of panitumumab (Vectibix® )**

Pmab is a recombinant fully human monoclonal IgG2 antibody produced in a mammalian cell line (Chinese Hamster Ovary, CHO) by recombinant DNA technology. Vectibix has high affinity and specificity to human EGFR. It inhibits receptor autophosphorylation caused by all known EGFR ligands by attaching to the ligand-binding domain. Binding of pmab to EGFR results in the internalization of the receptor, inhibition of cell growth, induction of apoptosis, and decreased interleukin-8 and vascular endothelial growth factor production (Berardi et al., 2010; EMA, 2011b; Harari, 2004; Helbling & Borner, 2007; Keating, 2010; Martinelli et al., 2007; Peeters et al., 2008; Pikó, 2009; Rakkar, 2007).

#### **3.2 Using Vectibix**

384 Colorectal Cancer – From Prevention to Patient Care

messenger routes. In a complex mechanism, EGF activation initiates cell division following the reception of an adequate external signal. It also assures survival and inhibits apoptosis. The resulting effect is cell proliferation. While this mechanism is strictly controlled in healthy cells, EGF activation is uncontrolled in a considerable proportion of epithelial tumours. The signal is transmitted to other proteins via the biochemical route of tyrosine kinase by phosphorylation. EGF activation can initiate cell division, proliferation, development of metastases and inhibition of apoptosis. Apparently, this leads to tumour progression (Cohenuram & Saif, 2008; Coutinho & Rocha Lima, 2003; EMA, 2009; EMA 2011b; Harari, 2004; Hamilton, 2008; Herbst & Shin, 2002; (Ritter & Arteaga, 2003; van

EGFR inhibitors (cmab and pmab) are licensed for the treatment of mCRC patients. They bind to the extracellular ligand-binding domain and thus inhibit transmembrane signal transmission and prevent EGF dependent signal transduction within the cell as well. Although the mechanism of action has already been established in theory, EGFR inhibitors yield clinical improvement to not more than approximately 50% of mCRC patients. This observation led to the assumption that a biological factor could have prevented these monoclonal antibodies from being effective in tumours expressing EGFR. The K-ras ("Kirsten rat sarcoma 2 viral oncogene homolog") gene belongs to the family of RAS protooncogenes. The K-ras protein coded by this gene plays a central role in growth-inducing signal transmission routes. By doing so it affects cell reproduction, differentiation and survival. If a mitogenic signal reaches the EGF receptor, the signal is forwarded to the nucleus by the K-ras. It is essential that this close correlation applies only to the "normal" (i.e. non-mutated or "wild type") K-ras. Mutant types of K-ras escape receptorial regulation

For this reason K-ras mutation is not a genetic failure with "function loss", on the contrary, in this case RAS remains in "on" status (i.e. phosphorylation is continuous) and acts independently from EGFR (and other physiological signaling pathways). As a consequence, despite the signals reaching the cell surface being "blocked" by monoclonal antibodies at the receptor level, signaling tracks regulated by EGFR under normal conditions remain (chronically) activated (Amado et al., 2008; Benvenuti et al., 2007; Dahabreh et al., 2011; De Roock et al., 2010; (Engstrom et al., 2011a, b; Esteller et al., 2001; EMA, 2009; EMA 2011b; Hamilton, 2008; Heinemann et al., 2009; Malumbres & Barbacid, 2003; Normanno et al., 2009). As the estimated incidence of K-ras mutation in CRC is 30-50%, it is expected that in about half to two thirds of patients the regulation of signal effect and signal transmission are preserved and drugs acting via the K-ras route can be used with success. (Amado et al., 2008; Benvenuti et al., 2007; Bardelli & Sien, 2010; Esteller et al., 2001; Garcia-Sáenz et al., 2009; Malumbres & Barbacid, 2003; Nagasaka et al., 2004). In an interesting re-evaluation of their primary study population Hurwitz et al. found that though bev combined with IFL as a first line treatment of mCRC was effective in both K-ras wild type and mutant subgroups, efficacy was by large affected by K-ras status, underlining a mixed predictive and

and thus they autonomously stimulate cell proliferation.

prognostic function of this mutation (Hurwitz et al., 2009).

**3.1 Characteristics of panitumumab (Vectibix®**

**3. Characteristics, application and side effects of panitumumab** 

**)**  Pmab is a recombinant fully human monoclonal IgG2 antibody produced in a mammalian cell line (Chinese Hamster Ovary, CHO) by recombinant DNA technology. Vectibix has high

Cutsem et al., 2009).

The recommended dose of Vectibix is 6 mg/kg of bodyweight once every two weeks both in monotherapy and when combined with cytostatics. Prior to infusion Vectibix should be diluted in 100 mL of 0.9% sodium chloride solution to a final concentration not exceeding 10 mg/mL. Vectibix must be administered as an intravenous infusion via an infusion pump using a low protein binding 0.2 or 0.22 micrometer in-line filter through a peripheral line or indwelling catheter. The recommended infusion time is approximately 60 minutes (Alberta Health Services, 2010; EMA, 2011b). The first dose injected over 60 minutes was well tolerated in clinical trials where Vectibix was combined with cytostatic agents; subsequent treatments were allowed to be given over 30 minutes (Douillard et al., 2010; Peeters et al., 2010). Doses higher than 1,000 mg should be administered as a 150 mL solution over approx. 90 minutes. No incompatibilities have been observed with 0.9% sodium chloride injection in polyvinyl chloride bags or polyolefin bags (EMA, 2011b; Knudson, 2007).

#### **3.3. Side effects of panitumumab**

#### **3.3.1 Skin toxicity**

The common pharmacological effect of EGFR inhibitors can lead to the following: EGFR inhibition in the skin, hair follicles, and periungual tissues can cause abnormal proliferation, migration and differentiation of target cells (i.e. basal keratinocytes), while changes in the skin structure attract inflammatory cells. Clinical symptoms emerge within 10 days following the introduction of pmab therapy and resolve in 28 days after the last injection on average. Skin symptoms are characteristic: papular skin rash, monomorphic pustular lesions, etc. presenting on skin areas exposed to the sun. Although signs may resemble those of acne for the first sight (labeled as "acneiform"), differentiation is easy and essential. Acne may manifest as non-inflammatory lesions on the basis of comedos or as inflammatory papules, pustules, or nodules. On the contrary, rash due to EGFR inhibitors is dominated by pustules. Non-inflammatory comedos are never seen in these cases. Skin rash is more widespread than classical acne as symptoms can often be observed on the upper and lower extremities and trunks of patients simultaneously. In order to prevent nail diseases it is important to avoid mechanical injuries (e.g. caused by tight shoes). Development of paronychia can be stopped by bathing the foot in diluted antiseptic agents and by using topical antiseptic ointments. Feet should not be soaked for a long time to prevent tissues from loosening. In some cases surgery cannot be avoided (Busam et al., 2001; Eaby, n. d.; EMA, 2011b; Moy & Goss, 2007; Pérez-Soler et al., 2005; Segaert & van Cutsem, 2005; Winkeljohn, 2008).

Conventional modalities to treat acne should not be used. On the contrary, advices and interventions are usually completely different from those applied during acne therapy. Sun bathing is prohibited, patients should protect themselves from any direct sunlight (hat, long-sleeved clothes, and sun screens are recommended). Dryness of the skin should be treated with neutral emollients. Caution is warranted if topical steroid drugs are used. Such

Panitumumab for the Treatment of Metastatic Colorectal Cancer 387

Fig. 3. Clearly visible inflammatory signs (pustules) differentiate EGFR-inhibitor therapy

Fig. 4. Nail lesions (paronychia and overgrowth) developed on 6th week of pmab therapy. The disease did not resolve on conservative therapy, surgical treatment (exploration and

It is important to modify or discontinue pmab administration according to the stage of rash. If the adverse events to (U. S. Department of Health And Human Services, U. S. National Institutes of Health, National Cancer Institute – Common Terminology Criteria for Adverse Events NCI-CTCAE) Grade 3 skin lesions emerge Vectibix should be suspended until the lesions resolve to Grade 2 or lower. In this case the product can be used by a 50% dose reduction; the dose can then be increased to the original in 25% increments every two weeks. If the rash persists or the symptoms recur in spite of dose reduction, pmab should be definitively discontinued (Alberta Health Services, 2010; EMA, 2011b; Pikó, 2009; Potthoff et

related rash from classical acne.

drainage) was necessary.

products are recommended solely to alleviate symptoms. Systemic antihistamines are more useful to cure itching. If rash is accompanied by superinfection, external use of either clindamycin or mupirocin, or internal use of tetracycline are to be considered (Eaby, n. d.; EMA, 2011b; Hoda et al., 2008; Lacouture, 2009; Lacouture et al., 2010; Melosky et al., 2009; Moy & Goss, 2007; Peeters et al., 2008; Pérez-Soler et al., 2005; Pikó, 2009; Potthoff et al., 2011; Saif & Cohenuram, 2006; Winkeljohn, 2008). Efforts to deal with skin toxicities via pre-emptive approach (i.e. applying emollients, hydrating and photoprotective creams, topical steroids and oral doxycyclin) in the STEPP ("Skin Toxicity Evaluation Protocol With Panitumumab") comparative clinical trial resulted in decreasing the frequency of Grade II or more severe forms already present from 62% to 28%. Quality of life improved significantly whereas the clinical efficacy of panitumumab treatment was unaffected. (Lacouture et al., 2010)

Fig. 1. 66-year-old male patient's acneiform rash after 2nd cycle (4th week) of pmab therapy for CRC with hepatic and pulmonary metastases.

Fig. 2. Similar but more pronounced symptoms are visible on the back of the above patient.

products are recommended solely to alleviate symptoms. Systemic antihistamines are more useful to cure itching. If rash is accompanied by superinfection, external use of either clindamycin or mupirocin, or internal use of tetracycline are to be considered (Eaby, n. d.; EMA, 2011b; Hoda et al., 2008; Lacouture, 2009; Lacouture et al., 2010; Melosky et al., 2009; Moy & Goss, 2007; Peeters et al., 2008; Pérez-Soler et al., 2005; Pikó, 2009; Potthoff et al., 2011; Saif & Cohenuram, 2006; Winkeljohn, 2008). Efforts to deal with skin toxicities via pre-emptive approach (i.e. applying emollients, hydrating and photoprotective creams, topical steroids and oral doxycyclin) in the STEPP ("Skin Toxicity Evaluation Protocol With Panitumumab") comparative clinical trial resulted in decreasing the frequency of Grade II or more severe forms already present from 62% to 28%. Quality of life improved significantly whereas the clinical

Fig. 1. 66-year-old male patient's acneiform rash after 2nd cycle (4th week) of pmab therapy

Fig. 2. Similar but more pronounced symptoms are visible on the back of the above patient.

for CRC with hepatic and pulmonary metastases.

efficacy of panitumumab treatment was unaffected. (Lacouture et al., 2010)

Fig. 3. Clearly visible inflammatory signs (pustules) differentiate EGFR-inhibitor therapy related rash from classical acne.

Fig. 4. Nail lesions (paronychia and overgrowth) developed on 6th week of pmab therapy. The disease did not resolve on conservative therapy, surgical treatment (exploration and drainage) was necessary.

It is important to modify or discontinue pmab administration according to the stage of rash. If the adverse events to (U. S. Department of Health And Human Services, U. S. National Institutes of Health, National Cancer Institute – Common Terminology Criteria for Adverse Events NCI-CTCAE) Grade 3 skin lesions emerge Vectibix should be suspended until the lesions resolve to Grade 2 or lower. In this case the product can be used by a 50% dose reduction; the dose can then be increased to the original in 25% increments every two weeks. If the rash persists or the symptoms recur in spite of dose reduction, pmab should be definitively discontinued (Alberta Health Services, 2010; EMA, 2011b; Pikó, 2009; Potthoff et

Panitumumab for the Treatment of Metastatic Colorectal Cancer 389

Fig. 5. Chest CT taken before starting planned pmab therapy of a mCRC patient who received therapy earlier in another institution. As the scan revealed pulmonary infiltration

2008; Pikó, 2009;, U. S. Department of Health And Human Services et al., 2009).

This is also a common side effect of EGFR inhibitors and indicates an injury of the intestinal mucosa similar to what is seen in dermatologic toxicities. Its frequency is not high; about 2% in patients with wild-type K-ras would develop diarrhoea. Its significance and its effect on

Symptoms are caused by the renal effects of EGFR inhibitors. Pronounced EGFR expression can be detected in the renal parenchyma (primarily in the ascending limb of loop of Henle, where magnesium and calcium are absorbed). Inhibition of EGFR in the renal tissue causes a decrease in the serum magnesium and calcium concentration. Following the recognition of these phenomenon patients involved in pmab studies have had their serum magnesium levels assessed. In 39% of cases the result proved to be abnormal, most often indicating mild hypomagnesaemia. The "Summary of Product Characteristics" requires regular assessments of serum magnesium and calcium levels before the treatment starts and for at least 8 weeks thereafter. Appropriate substitution is necessary for patients with mild-moderate disturbances, but the treatment may be discontinued in those who do not respond to substitution or present with severe clinical signs. Other electrolyte changes, such as hypokalaemia, have been detected as well. In such cases appropriate electrolyte substitution must be the primary step (Eaby, n. d.; EMA, 2011b; Pérez-Soler et al., 2005; Peeters et al.,

we did not administer Vectibix.

**3.3.5 Diarrhoea** 

**3.3.4 Hypomagnesaemia and hypocalcaemia** 

al., 2011; Widakowich et al., 2007). Nevertheless, skin and nail lesions are usually considered as positive predictive markers of efficacy and clinical response (Amado et al., 2008; Berardi et al., 2010; Busam et al., 2001; Eaby, n. d.; EMA 2011b; Grothey, 2006, 2007; Keating, 2010; Malik et al., 2005; Martinelli et al., 2007; Saif & Cohenuram, 2006; Siena et al., 2009; Widakowich, 2007).

#### **3.3.2 Ophthalmologic complications**

Since marketing authorization rare cases of keratitis and ulcerative keratitis has been reported, both representing a consequence of general mechanism of action of EGFR inhibitors (EMA, 2009; Burtness et al., 2009; Specenier et al., 2007; Thomas & Grandis, 2004; Xu et al., 2009). Retrospective analyses have shown that these complications were not severe in clinical trials, i.e. they did not reach Grade 2-4 (U. S. Department of Health And Human Services et al., 2009), and their incidence was between 0.2% and 0.7%. In clinical use as monotherapy, another case of severe keratitis and three cases of severe ulcerative keratitis have been reported (EMA 2011b). Care must be taken when the patients has a record of keratitis or ulcerative keratitis in his/her medical history. Consultation with an ophthalmologist is necessary in any instances the following symptoms are presented: inflammation of the eye, increased lacrimation, sensitivity to light, blurred vision, pain or redness of the eyes. The diagnosis of keratitis allows the oncologist to weigh the risk/benefit ratio of continuing or stopping Vectibix therapy, in cases of ulcerative keratitis however pmab treatment should be discontinued or suspended (EMA, 2011b; ManageCRC.com. 2011).

#### **3.3.3 Pulmonary complications**

Lung toxicity is a widely known complication of EGFR inhibitor therapies (interstitial lung disease ILD, interstitial pneumonitis, fibrosis) (Eaby, n. d.; Cohenuram & Saif, 2007; Gandara et al., 2006; Grothey, 2006; Inoue et al., 2003 ; Nagaria et al., 2005 ; Pikó, 2009 ; Saif & Cohenuram, 2006 ; Yoneda et al., 2007).

As patients suffering from the above lung diseases were excluded from pmab clinical trials before randomization, there are no available data on lung complications in these patients during pmab therapy (EMA, 2011b). If patients experience chest symptoms (dyspnea, dry cough, clinical or ECG signs of hypoxia, abnormalities of pulmonary function tests), at least simple (posterior-anterior) chest radiography or a more appropriate chest CT should be performed. If these examinations are indicative of an interstitial pulmonary disease, Vectibix should be discontinued. Depending on the severity of symptoms, symptomatic treatment with corticosteroids or diuretics (NCI-CTCAE Grade 2), oxygen supplementation (Grade 3), or intubation, tracheostomy or assisted respiration (Grade 4) may be necessary (Alberta Health Services, 2010; Peeters et al., 2008; U. S. Department of Health And Human Services et al., 2009).

It is important to differentiate pulmonary changes due to pmab therapy from signs of an underlying malignancy (e.g. well-defined metastases, carcinomatous lymphangiosis). Besides scrutinizing radio-morphologic features, other helpful measures, e.g. obtaining earlier radiographs, considering the dynamics of the process and sharing exact data with the radiologist (about the disease, signs, physical examination results, applied therapy) and further personal consultations may be appropriate as well and would underline the necessity of multidisciplinary oncological team-work.

al., 2011; Widakowich et al., 2007). Nevertheless, skin and nail lesions are usually considered as positive predictive markers of efficacy and clinical response (Amado et al., 2008; Berardi et al., 2010; Busam et al., 2001; Eaby, n. d.; EMA 2011b; Grothey, 2006, 2007; Keating, 2010; Malik et al., 2005; Martinelli et al., 2007; Saif & Cohenuram, 2006; Siena et al., 2009;

Since marketing authorization rare cases of keratitis and ulcerative keratitis has been reported, both representing a consequence of general mechanism of action of EGFR inhibitors (EMA, 2009; Burtness et al., 2009; Specenier et al., 2007; Thomas & Grandis, 2004; Xu et al., 2009). Retrospective analyses have shown that these complications were not severe in clinical trials, i.e. they did not reach Grade 2-4 (U. S. Department of Health And Human Services et al., 2009), and their incidence was between 0.2% and 0.7%. In clinical use as monotherapy, another case of severe keratitis and three cases of severe ulcerative keratitis have been reported (EMA 2011b). Care must be taken when the patients has a record of keratitis or ulcerative keratitis in his/her medical history. Consultation with an ophthalmologist is necessary in any instances the following symptoms are presented: inflammation of the eye, increased lacrimation, sensitivity to light, blurred vision, pain or redness of the eyes. The diagnosis of keratitis allows the oncologist to weigh the risk/benefit ratio of continuing or stopping Vectibix therapy, in cases of ulcerative keratitis however pmab treatment should be discontinued or suspended (EMA, 2011b; ManageCRC.com.

Lung toxicity is a widely known complication of EGFR inhibitor therapies (interstitial lung disease ILD, interstitial pneumonitis, fibrosis) (Eaby, n. d.; Cohenuram & Saif, 2007; Gandara et al., 2006; Grothey, 2006; Inoue et al., 2003 ; Nagaria et al., 2005 ; Pikó, 2009 ; Saif

As patients suffering from the above lung diseases were excluded from pmab clinical trials before randomization, there are no available data on lung complications in these patients during pmab therapy (EMA, 2011b). If patients experience chest symptoms (dyspnea, dry cough, clinical or ECG signs of hypoxia, abnormalities of pulmonary function tests), at least simple (posterior-anterior) chest radiography or a more appropriate chest CT should be performed. If these examinations are indicative of an interstitial pulmonary disease, Vectibix should be discontinued. Depending on the severity of symptoms, symptomatic treatment with corticosteroids or diuretics (NCI-CTCAE Grade 2), oxygen supplementation (Grade 3), or intubation, tracheostomy or assisted respiration (Grade 4) may be necessary (Alberta Health Services, 2010; Peeters et al., 2008; U. S. Department of Health And Human Services

It is important to differentiate pulmonary changes due to pmab therapy from signs of an underlying malignancy (e.g. well-defined metastases, carcinomatous lymphangiosis). Besides scrutinizing radio-morphologic features, other helpful measures, e.g. obtaining earlier radiographs, considering the dynamics of the process and sharing exact data with the radiologist (about the disease, signs, physical examination results, applied therapy) and further personal consultations may be appropriate as well and would underline the

Widakowich, 2007).

2011).

et al., 2009).

**3.3.2 Ophthalmologic complications** 

**3.3.3 Pulmonary complications** 

& Cohenuram, 2006 ; Yoneda et al., 2007).

necessity of multidisciplinary oncological team-work.

Fig. 5. Chest CT taken before starting planned pmab therapy of a mCRC patient who received therapy earlier in another institution. As the scan revealed pulmonary infiltration we did not administer Vectibix.

#### **3.3.4 Hypomagnesaemia and hypocalcaemia**

Symptoms are caused by the renal effects of EGFR inhibitors. Pronounced EGFR expression can be detected in the renal parenchyma (primarily in the ascending limb of loop of Henle, where magnesium and calcium are absorbed). Inhibition of EGFR in the renal tissue causes a decrease in the serum magnesium and calcium concentration. Following the recognition of these phenomenon patients involved in pmab studies have had their serum magnesium levels assessed. In 39% of cases the result proved to be abnormal, most often indicating mild hypomagnesaemia. The "Summary of Product Characteristics" requires regular assessments of serum magnesium and calcium levels before the treatment starts and for at least 8 weeks thereafter. Appropriate substitution is necessary for patients with mild-moderate disturbances, but the treatment may be discontinued in those who do not respond to substitution or present with severe clinical signs. Other electrolyte changes, such as hypokalaemia, have been detected as well. In such cases appropriate electrolyte substitution must be the primary step (Eaby, n. d.; EMA, 2011b; Pérez-Soler et al., 2005; Peeters et al., 2008; Pikó, 2009;, U. S. Department of Health And Human Services et al., 2009).

#### **3.3.5 Diarrhoea**

This is also a common side effect of EGFR inhibitors and indicates an injury of the intestinal mucosa similar to what is seen in dermatologic toxicities. Its frequency is not high; about 2% in patients with wild-type K-ras would develop diarrhoea. Its significance and its effect on

Panitumumab for the Treatment of Metastatic Colorectal Cancer 391

found that the most common side effects were dermatological symptoms (rash), already known in case of EGFR inhibitors. Presentation and severity of these symptoms were dose dependent and closely correlated with treatment results, while low haemoglobin and high alkaline phosphatase levels had a negative predictive value. No antibodies against ABX-

Based on the analysis of early Phase 1 study results subsequent studies with pmab were

In 2004 and 2005 results of a phase 2 study with pmab monotherapy, involving CRC patients relapsing following a subsequent irinotecan and oxaliplatin therapy, were published (Hecht et al., 2004; Malik et al., 2005). Data of 148 patients were evaluable in the analysis. Median progression-free survival (PFS) was 3.4 (2.0-4.0) months and overall survival (OS) was 9.4

Berlin and co-workers (Berlin et al., 2004) and Hecht and co-workers (Hecht et al., 2006b) used pmab with combinations containing irinotecan (IFL or FOLFIRI) in Phase 2 studies. The main adverse effects were dermatological symptoms and diarrhoea. In the IFL arm partial remission fulfilling the "Response Evaluation Criteria in Solid Tumors" (RECIST) was seen in 48% of patients and stable disease could be reached in 26% that equated to a tumour control in 74% of cases (Jaffe, 2006; Padhan & Ollivier, 2001; Therasse et al., 2000). Median PFS and OS were 5.6 and 17 months, respectively. When pmab was used in combination with FOLFIRI rates of remission, stable disease and total tumour control were 33%, 46%, and 79%, respectively. Progression-free survival was 10.9 months, but overall survival results could not have been calculated (overlapping results of the study had been

Patients were later divided into groups with negative or "low" (1 to 10%) (Hecht et al., 2006a), and "high" (above 10%) EGFR-expression (Berlin et al., 2006). No significant differences were found: at low EGFR levels 48% response rate and the rate 7.9 weeks median PFS were detected, while in patients with high EGFR levels 42% tumour response rate and 12-14 weeks PFS was seen. The adverse effect profile was similar. Grade 3/4 adverse events were presented in 19-24% (dermatological symptoms prevailed), and the rate

Van Cutsem et co-workers were the first to publish a comparison of Vectibix and "best supportive care" (BSC): they treated a total of 463 patients with EGFR expressing mCRC, after failure of irinotecan- and oxaliplatin-containing therapies (van Cutsem et al., 2007). Patients were given either pmab (6 mg/kg every two weeks, without premedication) in combination (with symptomatic treatment) or BSC alone, in 1:1 ratio. Patients in the BSC group could have been switched to the active arm in case of progression. Thirty-five percent of patients had been on adjuvant chemotherapy earlier, and all of them had had at least two treatment options due to metastatic disease. Thirty-seven percent of the patients had a disease progression after the third line of drug therapy. Treatment efficacy was assessed after week 8, 12, 16, 24, 32, and 40, and every 3 months thereafter according to the RECIST

The following chart represents the results of this study and shows the benefits of Vectibix

months (6.0-10.6). Results did not differ in EGFR positive or negative patients.

EGF have been detected in this study.

published by other authors in various forums).

of hypomagnesaemia was similar (8 and 12%).

**4.3 Phase 3 study and analysis of further results** 

(Jaffe, 2006; Padhan & Ollivier, 2001; Therasse et al., 2000).

compared to supportive care:

**4.2 Phase 2 studies** 

conducted in mCRC patients.

the continuability of pmab therapy depend on the severity of symptoms. Apart from lifestyle advices and loperamide administration, one should bear in mind that parenteral fluid replacement and normalization of electrolyte levels is essential in NCI-CTCAE Grade 3 diarrhoea (defecation more than 7 times per day or fecal incontinence, or necessity of hospitalization due to symptoms). If one fails to do so, calcium and magnesium electrolyte disturbances may increase in severity and acute renal failure may also develop (Berlin et al., 2007; Eaby, n. d.; EMA 2011b; Moy & Goss, 2007; Peeters et al., 2008; Pikó, 2009; Tuma, 2006; Widakowich et al., 2007).

#### **3.3.6 General symptoms and infusion complications**

Generally speaking, this term actually stands for adverse events (fever, chills and suffocation) which develop when a "foreign" protein is administered. Infusion complications emerge within 24 hours after administration. In most cases, premedication is needed to prevent general symptoms and infusion complications if human-animal chimeric or humanized monoclonal antibodies are used. As pmab is fully human, this is unnecessary when applying Vectibix. Nevertheless, infusion reactions might emerge during administration of fully human amino acid sequences despite using adequate protein filters to avoid complications. Several authors have reported however, that treatment with pmab may still be a viable and beneficial option for patients who suffered infusion reaction while being treated with the "chimeric" monoclonal antibody cetuximab (Cartwright & Genther, 2008; Chung, 2008; EMA, 2011b; Grothey, 2006; Helbling & Borner, 2007; Heun & Holen, 2007; Langerak et al., 2009; Lenz, 2007; Nielsen et al., 2009; O'Neil et al., 2007; Power et al., 2010; Saif et al., 2008).

Across all clinical studies, infusion-related reactions were reported in 3% of Vectibix-treated patients; of which < 1% were severe (NCI-CTC grade 3 or 4), i.e. required acute hospitalization or prolongation of hospitalization or was life-threatening. In the postmarketing setting serious infusion reactions have been reported, including rare reports of fatal outcome. If a severe or life-threatening reaction occurs during an infusion or at any time post-infusion, Vectibix should be permanently discontinued (U. S. Department of Health And Human Services et al., 2009).

#### **4. Results of clinical studies with panitumumab**

#### **4.1 Phase 1 studies**

At the Annual Meeting of the American Society of Clinical Oncology (ASCO) in 2002 Figlin and co-workers demonstrated the effect of a newly developed monoclonal antibody (called "ABX-EGF" in the presentation) on different tumors they evaluated in a phase 1 study (Figlin et al., 2002). The applied doses ranged from 0.01 mg/kg to 2.5 mg/kg. They found that the above therapy resulted in significantly long survival in certain cases. One patient with oesophageal cancer had stable disease for 7 months and minor response was reached in a patient with prostate cancer. No antibodies produced against ABX-EGF were detected, and its main side effect was rash.

In 2004, Rowinsky and co-workers published their results from a Phase 1 study with ABX-EGF (the agent later named pmab) used in renal cell carcinoma (Rowinsky et al., 2004). The highest dose used in this study was 2.5 mg per week. Although this dose could produce the highest rate of objective tumour response, the relationship between time to progression (median values were between 53 and 165 days) and the applied dose was unclear. It was found that the most common side effects were dermatological symptoms (rash), already known in case of EGFR inhibitors. Presentation and severity of these symptoms were dose dependent and closely correlated with treatment results, while low haemoglobin and high alkaline phosphatase levels had a negative predictive value. No antibodies against ABX-EGF have been detected in this study.

#### **4.2 Phase 2 studies**

390 Colorectal Cancer – From Prevention to Patient Care

the continuability of pmab therapy depend on the severity of symptoms. Apart from lifestyle advices and loperamide administration, one should bear in mind that parenteral fluid replacement and normalization of electrolyte levels is essential in NCI-CTCAE Grade 3 diarrhoea (defecation more than 7 times per day or fecal incontinence, or necessity of hospitalization due to symptoms). If one fails to do so, calcium and magnesium electrolyte disturbances may increase in severity and acute renal failure may also develop (Berlin et al., 2007; Eaby, n. d.; EMA 2011b; Moy & Goss, 2007; Peeters et al., 2008; Pikó, 2009; Tuma, 2006;

Generally speaking, this term actually stands for adverse events (fever, chills and suffocation) which develop when a "foreign" protein is administered. Infusion complications emerge within 24 hours after administration. In most cases, premedication is needed to prevent general symptoms and infusion complications if human-animal chimeric or humanized monoclonal antibodies are used. As pmab is fully human, this is unnecessary when applying Vectibix. Nevertheless, infusion reactions might emerge during administration of fully human amino acid sequences despite using adequate protein filters to avoid complications. Several authors have reported however, that treatment with pmab may still be a viable and beneficial option for patients who suffered infusion reaction while being treated with the "chimeric" monoclonal antibody cetuximab (Cartwright & Genther, 2008; Chung, 2008; EMA, 2011b; Grothey, 2006; Helbling & Borner, 2007; Heun & Holen, 2007; Langerak et al., 2009; Lenz, 2007; Nielsen et al., 2009; O'Neil et al., 2007; Power et al.,

Across all clinical studies, infusion-related reactions were reported in 3% of Vectibix-treated patients; of which < 1% were severe (NCI-CTC grade 3 or 4), i.e. required acute hospitalization or prolongation of hospitalization or was life-threatening. In the postmarketing setting serious infusion reactions have been reported, including rare reports of fatal outcome. If a severe or life-threatening reaction occurs during an infusion or at any time post-infusion, Vectibix should be permanently discontinued (U. S. Department of

At the Annual Meeting of the American Society of Clinical Oncology (ASCO) in 2002 Figlin and co-workers demonstrated the effect of a newly developed monoclonal antibody (called "ABX-EGF" in the presentation) on different tumors they evaluated in a phase 1 study (Figlin et al., 2002). The applied doses ranged from 0.01 mg/kg to 2.5 mg/kg. They found that the above therapy resulted in significantly long survival in certain cases. One patient with oesophageal cancer had stable disease for 7 months and minor response was reached in a patient with prostate cancer. No antibodies produced against ABX-EGF were detected,

In 2004, Rowinsky and co-workers published their results from a Phase 1 study with ABX-EGF (the agent later named pmab) used in renal cell carcinoma (Rowinsky et al., 2004). The highest dose used in this study was 2.5 mg per week. Although this dose could produce the highest rate of objective tumour response, the relationship between time to progression (median values were between 53 and 165 days) and the applied dose was unclear. It was

Widakowich et al., 2007).

2010; Saif et al., 2008).

**4.1 Phase 1 studies** 

and its main side effect was rash.

Health And Human Services et al., 2009).

**4. Results of clinical studies with panitumumab** 

**3.3.6 General symptoms and infusion complications** 

Based on the analysis of early Phase 1 study results subsequent studies with pmab were conducted in mCRC patients.

In 2004 and 2005 results of a phase 2 study with pmab monotherapy, involving CRC patients relapsing following a subsequent irinotecan and oxaliplatin therapy, were published (Hecht et al., 2004; Malik et al., 2005). Data of 148 patients were evaluable in the analysis. Median progression-free survival (PFS) was 3.4 (2.0-4.0) months and overall survival (OS) was 9.4 months (6.0-10.6). Results did not differ in EGFR positive or negative patients.

Berlin and co-workers (Berlin et al., 2004) and Hecht and co-workers (Hecht et al., 2006b) used pmab with combinations containing irinotecan (IFL or FOLFIRI) in Phase 2 studies. The main adverse effects were dermatological symptoms and diarrhoea. In the IFL arm partial remission fulfilling the "Response Evaluation Criteria in Solid Tumors" (RECIST) was seen in 48% of patients and stable disease could be reached in 26% that equated to a tumour control in 74% of cases (Jaffe, 2006; Padhan & Ollivier, 2001; Therasse et al., 2000). Median PFS and OS were 5.6 and 17 months, respectively. When pmab was used in combination with FOLFIRI rates of remission, stable disease and total tumour control were 33%, 46%, and 79%, respectively. Progression-free survival was 10.9 months, but overall survival results could not have been calculated (overlapping results of the study had been published by other authors in various forums).

Patients were later divided into groups with negative or "low" (1 to 10%) (Hecht et al., 2006a), and "high" (above 10%) EGFR-expression (Berlin et al., 2006). No significant differences were found: at low EGFR levels 48% response rate and the rate 7.9 weeks median PFS were detected, while in patients with high EGFR levels 42% tumour response rate and 12-14 weeks PFS was seen. The adverse effect profile was similar. Grade 3/4 adverse events were presented in 19-24% (dermatological symptoms prevailed), and the rate of hypomagnesaemia was similar (8 and 12%).

#### **4.3 Phase 3 study and analysis of further results**

Van Cutsem et co-workers were the first to publish a comparison of Vectibix and "best supportive care" (BSC): they treated a total of 463 patients with EGFR expressing mCRC, after failure of irinotecan- and oxaliplatin-containing therapies (van Cutsem et al., 2007). Patients were given either pmab (6 mg/kg every two weeks, without premedication) in combination (with symptomatic treatment) or BSC alone, in 1:1 ratio. Patients in the BSC group could have been switched to the active arm in case of progression. Thirty-five percent of patients had been on adjuvant chemotherapy earlier, and all of them had had at least two treatment options due to metastatic disease. Thirty-seven percent of the patients had a disease progression after the third line of drug therapy. Treatment efficacy was assessed after week 8, 12, 16, 24, 32, and 40, and every 3 months thereafter according to the RECIST (Jaffe, 2006; Padhan & Ollivier, 2001; Therasse et al., 2000).

The following chart represents the results of this study and shows the benefits of Vectibix compared to supportive care:

Panitumumab for the Treatment of Metastatic Colorectal Cancer 393

Number of patients 124 84 119 100 Median PFS (weeks) 16 8 8 8 Median OS (months) 8,1 7,6 4,9 4,4 Table 2. Progression-free survival (PFS) and overall survival (OS) of pmab + BSC vs. BSC

Side effects were more frequent and severe in the K-ras mutant subgroup that, apart from inefficacy, may lead to a worse tolerability and possibly higher treatment risks. Considering both efficacy results and side effects, it was proven that pmab should only be used in

Based on the consideration that the approval had been based on a clinical trial including pre-treated EGFR positive patients whose treatment was shown to be effective only in those with the K-ras wild type, the European Medicines Agency (EMA) summarizes treatment criteria in all of the Summaries of Product Characteristics congruently. "Vectibix is indicated as monotherapy for the treatment of patients with EGFR expressing metastatic colorectal carcinoma with non-mutated (wild-type) *KRAS* after failure of fluoropyrimidine-, oxaliplatin-, and irinotecan-containing chemotherapy regimens" (EMA, 2011b). The U. S. Food and Drug Administration **(**FDA) defines the same criteria in more detail: "Vectibix is an epidermal growth factor receptor antagonist indicated as a single agent for the treatment of metastatic colorectal carcinoma with disease progression on or following fluoropyrimidine, oxaliplatin, and irinotecan chemotherapy regimens. Approval is based on progression-free survival; no data demonstrate an improvement in disease-related symptoms or increased survival with Vectibix. Retrospective subset analyses of metastatic colorectal cancer trials have not shown a treatment benefit for Vectibix in patients whose tumors had *KRAS* mutations in codon 12 or 13. Use of Vectibix is not recommended for the treatment of colorectal cancer with

The Summary of Product Characteristics of other anti-mCRC targeted biologic therapies states that these agents can be used either only in combination with "traditional" antitumour chemotherapies (e.g. beva), or both in combination and as a stand-alone therapy (e.g. cmab). In contrast, pmab could only have been used as a monotherapy and in patients who have already had a definite cytostatic pre-treatment. Supposing that such timing of treatments does not provide optimal circumstances for the efficacy of monoclonal antibodies, possible combinations of Vectibix and cytostatic agents have been evaluated in

Following completion of a study involving 1183 patients titled "Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer to determine

**4.4 Vectibix summaries of product characteristics: A reinforcement of treatment** 

ras

treated patients by K-ras status. (Source: Amado et al., 2008)

these mutations" (U. S. Food and Drug Administration, 2009).

**5. Combining panitumumab with cytostatic agents** 

**5.1 Combination of pmab and chemotherapy as a first-line treatment** 

wild type K-

patients with the wild type K-ras.

clinical studies.

**criteria and results of the clinical trials** 

pmab + BSC BSC

wild type Kras

mutant K-ras

mutant K-ras


Table 1. Results of progression-free survival (PFS), response rate, stable disease, overall response rate treated with pmab + BSC vs. BSC alone (adapted from: van Cutsem et al., 2007)

In terms of all parameters (age, sex, site of primary tumour, ECOG performance status, former lines of chemotherapy applied, number of organs with metastases and degree of EGFR positivity), subgroup analyses unanimously showed that the active treatment arm (pmab) was superior to BSC. Degree of risk reduction was 46%, which was statistically significant (p<0.000000001). It is remarkable that among the 174 patients who were crossed over from BSC arm to the active (pmab) arm due to progression partial response (PR) could be reached in 9% and SD in 32% of cases, in spite of a more progressed disease (Cohenuram & Saif, 2007).

This study once again proved the correlation between side effects and efficacy, i.e. assessment of the results showed that skin symptoms are of good predictive value. These findings underline the fact that rash is one of the most important predictive factors of efficacy.

In the study designed to compare pmab and BSC, Siena and co-workers re-assessed response and survival data, and divided the group of responders into subgroups of patients with remission and those with stable disease. Differences between each group were statistically confirmed (Siena et al., 2007). Curves demonstrating treatment efficacy were also different, survival curve of patients with disease progression and that of those with no progression after 8 weeks of pmab therapy (equivalent with 4 treatment cycles) were compared. Based on these data the authors presumed with good reason that there must be another factor apart from the detectable EGFR expression (an inclusion criterion for all patients) that has an impact on treatment results.

The presumed factor was later proved to be the K-ras mutation status. Differences in treatment results could be explained by the presence of "normal" (wild type) or "abnormal" (mutated) K-ras genes. Amado and co-workers determined the frequency of mutations in the already known patient population (Amado et al., 2008). Although not all, 427 samples of the 463 patients were suitable for subsequent central laboratory evaluations and were eventually analyzed. Analyses showed mutations in 184 patients and "wild type" K-ras in 243 patients. Data analyses showed that no correlation can be detected between K-ras mutation and EGFR status (the latter determined by immunohistochemistry), neither by expression nor by the intensity of membrane staining.

Analyses of clinical results showed that (in accordance with the biological role of K-ras described earlier) tumour progression in mutation carriers is independent from the regulation of stimuli reaching the EGFR. Consequently, in these patients the EGFR inhibitor pmab is less effective and does not provide better results than BSC.

17 weeks NA

Studied parameters pmab + BSC (232 patients) BSC alone (231 patients)

Table 1. Results of progression-free survival (PFS), response rate, stable disease, overall response rate treated with pmab + BSC vs. BSC alone (adapted from: van Cutsem et al.,

In terms of all parameters (age, sex, site of primary tumour, ECOG performance status, former lines of chemotherapy applied, number of organs with metastases and degree of EGFR positivity), subgroup analyses unanimously showed that the active treatment arm (pmab) was superior to BSC. Degree of risk reduction was 46%, which was statistically significant (p<0.000000001). It is remarkable that among the 174 patients who were crossed over from BSC arm to the active (pmab) arm due to progression partial response (PR) could be reached in 9% and SD in 32% of cases, in spite of a more progressed disease (Cohenuram

This study once again proved the correlation between side effects and efficacy, i.e. assessment of the results showed that skin symptoms are of good predictive value. These findings underline the fact that rash is one of the most important predictive factors of

In the study designed to compare pmab and BSC, Siena and co-workers re-assessed response and survival data, and divided the group of responders into subgroups of patients with remission and those with stable disease. Differences between each group were statistically confirmed (Siena et al., 2007). Curves demonstrating treatment efficacy were also different, survival curve of patients with disease progression and that of those with no progression after 8 weeks of pmab therapy (equivalent with 4 treatment cycles) were compared. Based on these data the authors presumed with good reason that there must be another factor apart from the detectable EGFR expression (an inclusion criterion for all

The presumed factor was later proved to be the K-ras mutation status. Differences in treatment results could be explained by the presence of "normal" (wild type) or "abnormal" (mutated) K-ras genes. Amado and co-workers determined the frequency of mutations in the already known patient population (Amado et al., 2008). Although not all, 427 samples of the 463 patients were suitable for subsequent central laboratory evaluations and were eventually analyzed. Analyses showed mutations in 184 patients and "wild type" K-ras in 243 patients. Data analyses showed that no correlation can be detected between K-ras mutation and EGFR status (the latter determined by immunohistochemistry), neither by

Analyses of clinical results showed that (in accordance with the biological role of K-ras described earlier) tumour progression in mutation carriers is independent from the regulation of stimuli reaching the EGFR. Consequently, in these patients the EGFR inhibitor

PFS rate at week 24 18% 5% PFS rate at week 32 10% 34% Response rate (RR) 8% 0% Stable disease (SD) 28% 10% Overall response rate (ORR) 36% 10%

Median duration of

response

& Saif, 2007).

patients) that has an impact on treatment results.

expression nor by the intensity of membrane staining.

pmab is less effective and does not provide better results than BSC.

efficacy.

2007)


Table 2. Progression-free survival (PFS) and overall survival (OS) of pmab + BSC vs. BSC treated patients by K-ras status. (Source: Amado et al., 2008)

Side effects were more frequent and severe in the K-ras mutant subgroup that, apart from inefficacy, may lead to a worse tolerability and possibly higher treatment risks. Considering both efficacy results and side effects, it was proven that pmab should only be used in patients with the wild type K-ras.

#### **4.4 Vectibix summaries of product characteristics: A reinforcement of treatment criteria and results of the clinical trials**

Based on the consideration that the approval had been based on a clinical trial including pre-treated EGFR positive patients whose treatment was shown to be effective only in those with the K-ras wild type, the European Medicines Agency (EMA) summarizes treatment criteria in all of the Summaries of Product Characteristics congruently. "Vectibix is indicated as monotherapy for the treatment of patients with EGFR expressing metastatic colorectal carcinoma with non-mutated (wild-type) *KRAS* after failure of fluoropyrimidine-, oxaliplatin-, and irinotecan-containing chemotherapy regimens" (EMA, 2011b). The U. S. Food and Drug Administration **(**FDA) defines the same criteria in more detail: "Vectibix is an epidermal growth factor receptor antagonist indicated as a single agent for the treatment of metastatic colorectal carcinoma with disease progression on or following fluoropyrimidine, oxaliplatin, and irinotecan chemotherapy regimens. Approval is based on progression-free survival; no data demonstrate an improvement in disease-related symptoms or increased survival with Vectibix. Retrospective subset analyses of metastatic colorectal cancer trials have not shown a treatment benefit for Vectibix in patients whose tumors had *KRAS* mutations in codon 12 or 13. Use of Vectibix is not recommended for the treatment of colorectal cancer with these mutations" (U. S. Food and Drug Administration, 2009).

#### **5. Combining panitumumab with cytostatic agents**

The Summary of Product Characteristics of other anti-mCRC targeted biologic therapies states that these agents can be used either only in combination with "traditional" antitumour chemotherapies (e.g. beva), or both in combination and as a stand-alone therapy (e.g. cmab). In contrast, pmab could only have been used as a monotherapy and in patients who have already had a definite cytostatic pre-treatment. Supposing that such timing of treatments does not provide optimal circumstances for the efficacy of monoclonal antibodies, possible combinations of Vectibix and cytostatic agents have been evaluated in clinical studies.

#### **5.1 Combination of pmab and chemotherapy as a first-line treatment**

Following completion of a study involving 1183 patients titled "Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer to determine

Panitumumab for the Treatment of Metastatic Colorectal Cancer 395

An important aspect, also relevant for routine clinical practice, was investigated by Siena and co-workers in their subgroup analysis of the above study detailed in ASCO Annual Meeting 2011 (Siena et al., 2011). Patients with good performance status (ECOG 0-1) obviously profited from the addition of pmab to FOLFOX4 as PFS increased in these cases from 8.0 (7.5-9.3) to 10.4 months (9.3-11.3), OS from 20.7 (18.2-23.2) to 25.8 months (21.7-not estimable); whereas at ECOG2 (ambulatory and capable of all self-care but unable to carry out any work activities up and about more than 50% of waking hours) patients the addition a pmab decreased PFS from 7.6 (5.3-11.1) to 4.8 months (2.7-5.3), OS from 11.7 (8.0-15.7) to 7.0 (4.6-11.7) months. An adequate determination of performance status may serve as a simple and statistically convincing tool to predict the value of the addition of pmab to

Notably, besides performance status, quality of life may be a further parameter worth evaluating when analysing treatment results. Primary results from a phase II study involving 142 patients evaluating the combination a pmab and FOLFIRI as a first line chemotherapy in mCRC were published by Kohne and co-workers in 2010 (Kohne et al., 2010). Results showed 48% response rate (RR) for wild type and 29% RR for mutant K-ras patients, with no differences in side effects. Results of a secondary analysis of initial quality of life measures were published during ASCO Annual Meeting 2011 (Karthaus et al., 2011). The results demonstrated that those patients with better quality of life had better tumour responses as well by week 8 and 24 of the combination therapy. It does not seem to be an overstatement that the combination of pmab with cytostatics in the first line treatment of CRC is a promising option for patients in better clinical (performance and quality of life)

Peeters and co-workers compared the efficacy of pmab and FOLFIRI to FOLFIRI alone as the second-line treatment of mCRC patients in a phase 3, equally randomized trial (Peeters et al., 2010). The study was originally designed to compare the therapeutic effect in the entire population, but due to convincing external data it was modified before the efficacy assessments so that prospective assessments would be carried out as per the K-ras status of

A total number of 1186 patients were treated after randomization. Five hundred-ninety-two (50%) patients were given pmab and FOLFIRI, and 595 (50%) were given FOLFIRI alone. The K-ras status of 1083 patients (91%) was known (based on central laboratory tests): 597 patients (55%) had wild-type K-ras tumour and 486 (45%) had K-ras mutant metastatic colon

The eligible patients were older than 18 years and their ECOG performance status was 0, 1 or 2. Only one earlier chemotherapeutic scheme, i.e. first-line fluoropyrimidine-based chemotherapy was allowed for the treatment of mCRC. A radiologically verified progression by RECIST was required during the course of treatment or within 6 months. Known EGFR expression or K-ras status were not required for enrolment. Patients previously treated with irinotecan or anti-EGFR therapy were excluded from the study

Pmab (at 6 mg/kg) was administered over 60 minutes by infusion before chemotherapy; if patients tolerated the first dose, the following infusions were administered over 30 minutes. Every patient was given FOLFIRI: 180 mg/m2 irinotecan and 400 mg/m2 raceme leucovorin

**5.2 Combination of pmab and chemotherapy as a second-line treatment** 

(Jaffe, 2006), (Padhan & Ollivier, 2001), (Therasse et al., 2000).

FOLFOX4 in the first line treatment of mCRC.

status.

the tumour.

cancer.

Efficacy" (PRIME), Douillard and co-workers presented results of the application of pmab with FOLFOX4 (5-fluorouracil, leucovorin and oxaliplatin) versus FOLFOX4 alone as the first-line treatment in mCRC patients in open label, randomized, multicenter, Phase 3 trial (Douillard et al., 2010). Eligible patients were individuals older than 18 years who did not receive chemotherapy for their metastatic disease. 5-fluorouracil was allowed in adjuvant chemotherapy in case the disease recurred within 6 months after discontinuing the adjuvant therapy, but oxaliplatin was not allowed under any circumstances.

Pmab was administered every two weeks in a dose of 6 mg/kg by intravenous infusion over one hour on the day before FOLFOX4 chemotherapy was scheduled. If patients tolerated the first pmab infusion, the consecutive doses could have been administered over 30 minutes. FOLFOX4 was administered every two weeks: on day 1 oxaliplatin was administered at 85 mg/m2 and leucovorin at 200 mg/m2 (or equivalent dose) via infusion. On days 1 and 2 this was followed by fluorouracil at 400 mg/m2 by intravenous bolus and fluorouracil at 600 mg/m2 by a continuous 22-hour infusion. This treatment was continued until disease progression (adjudicated by an independent committee) or the occurrence of unacceptable side effects.

In terms of evaluation the study had four arms, as groups of K-ras mutant and wild-type patients were distinguished following previous laboratory assessment both in the FOLFOX4 alone and the pmab + FOLFOX4 arm.

The administration of the monoclonal antibody Vectibix to patients with wild-type K-ras increased PFS significantly from 8.0 to 9.6 months, while increase in overall survival (23.9 months as compared to 19.7 months) was clinically considerable and relevant nevertheless statistically non-significant, compared to FOLFOX4 alone arm. In K-ras mutated cases however, Vectibix with FOLFOX4 versus FOLFOX4 alone decreased the median PFS (7.3 vs. 8.8 months) and OS (15.5 vs. 19.3 months).

By a glance on the table summarizing side effects one can realize that apart from typical side effects of EGFR inhibitors in the Vectibix group no significant differences were revealed.

Antibodies against pmab were found in blood samples of 3.0% of patients (samples were drawn during treatment). After discontinuation, neutralizing antibodies were found in another 0.4% of patients.

A forest plot subgroup analysis with overlapping confidence intervals showed that pmab addition was generally beneficial in terms of improving progression-free survival. Treatment without pmab showed a tendency to be more beneficial in those with bad performance status (ECOG 2). Pmab seemed to be more beneficial in those with hepatic metastases, however in patients with dissemination in multiple organs and in cases presenting exclusively hepatic metastases no significant differences between the arms were shown. Subgroup analyses of overall survival revealed similar results, notably, poorer general condition (ECOG 2) seemed to be again more disadvantageous for Vectibix treated patients, age and gender showed marked but somewhat weaker interference than is PFS.

The authors claimed that adding pmab to FOLFOX4 increased PFS significantly in previously untreated mCRC patients with wild-type K-ras. Another clinically important feature of pmab is that severe infusion reactions are rare, and the standard 2-week protocol of Vectibix enables treating physicians to synchronize administration with chemotherapy schedules and decreases the number of visits to the minimum. As no premedication is required and no observation is necessary following treatment, the short outpatient therapy is advantageous for patients and caregivers as well.

Efficacy" (PRIME), Douillard and co-workers presented results of the application of pmab with FOLFOX4 (5-fluorouracil, leucovorin and oxaliplatin) versus FOLFOX4 alone as the first-line treatment in mCRC patients in open label, randomized, multicenter, Phase 3 trial (Douillard et al., 2010). Eligible patients were individuals older than 18 years who did not receive chemotherapy for their metastatic disease. 5-fluorouracil was allowed in adjuvant chemotherapy in case the disease recurred within 6 months after discontinuing the adjuvant

Pmab was administered every two weeks in a dose of 6 mg/kg by intravenous infusion over one hour on the day before FOLFOX4 chemotherapy was scheduled. If patients tolerated the first pmab infusion, the consecutive doses could have been administered over 30 minutes. FOLFOX4 was administered every two weeks: on day 1 oxaliplatin was administered at 85 mg/m2 and leucovorin at 200 mg/m2 (or equivalent dose) via infusion. On days 1 and 2 this was followed by fluorouracil at 400 mg/m2 by intravenous bolus and fluorouracil at 600 mg/m2 by a continuous 22-hour infusion. This treatment was continued until disease progression (adjudicated by an independent committee) or the occurrence of unacceptable

In terms of evaluation the study had four arms, as groups of K-ras mutant and wild-type patients were distinguished following previous laboratory assessment both in the FOLFOX4

The administration of the monoclonal antibody Vectibix to patients with wild-type K-ras increased PFS significantly from 8.0 to 9.6 months, while increase in overall survival (23.9 months as compared to 19.7 months) was clinically considerable and relevant nevertheless statistically non-significant, compared to FOLFOX4 alone arm. In K-ras mutated cases however, Vectibix with FOLFOX4 versus FOLFOX4 alone decreased the median PFS (7.3 vs.

By a glance on the table summarizing side effects one can realize that apart from typical side effects of EGFR inhibitors in the Vectibix group no significant differences were revealed. Antibodies against pmab were found in blood samples of 3.0% of patients (samples were drawn during treatment). After discontinuation, neutralizing antibodies were found in

A forest plot subgroup analysis with overlapping confidence intervals showed that pmab addition was generally beneficial in terms of improving progression-free survival. Treatment without pmab showed a tendency to be more beneficial in those with bad performance status (ECOG 2). Pmab seemed to be more beneficial in those with hepatic metastases, however in patients with dissemination in multiple organs and in cases presenting exclusively hepatic metastases no significant differences between the arms were shown. Subgroup analyses of overall survival revealed similar results, notably, poorer general condition (ECOG 2) seemed to be again more disadvantageous for Vectibix treated patients, age and gender showed marked but somewhat weaker interference than is PFS. The authors claimed that adding pmab to FOLFOX4 increased PFS significantly in previously untreated mCRC patients with wild-type K-ras. Another clinically important feature of pmab is that severe infusion reactions are rare, and the standard 2-week protocol of Vectibix enables treating physicians to synchronize administration with chemotherapy schedules and decreases the number of visits to the minimum. As no premedication is required and no observation is necessary following treatment, the short outpatient therapy

therapy, but oxaliplatin was not allowed under any circumstances.

side effects.

alone and the pmab + FOLFOX4 arm.

8.8 months) and OS (15.5 vs. 19.3 months).

is advantageous for patients and caregivers as well.

another 0.4% of patients.

An important aspect, also relevant for routine clinical practice, was investigated by Siena and co-workers in their subgroup analysis of the above study detailed in ASCO Annual Meeting 2011 (Siena et al., 2011). Patients with good performance status (ECOG 0-1) obviously profited from the addition of pmab to FOLFOX4 as PFS increased in these cases from 8.0 (7.5-9.3) to 10.4 months (9.3-11.3), OS from 20.7 (18.2-23.2) to 25.8 months (21.7-not estimable); whereas at ECOG2 (ambulatory and capable of all self-care but unable to carry out any work activities up and about more than 50% of waking hours) patients the addition a pmab decreased PFS from 7.6 (5.3-11.1) to 4.8 months (2.7-5.3), OS from 11.7 (8.0-15.7) to 7.0 (4.6-11.7) months. An adequate determination of performance status may serve as a simple and statistically convincing tool to predict the value of the addition of pmab to FOLFOX4 in the first line treatment of mCRC.

Notably, besides performance status, quality of life may be a further parameter worth evaluating when analysing treatment results. Primary results from a phase II study involving 142 patients evaluating the combination a pmab and FOLFIRI as a first line chemotherapy in mCRC were published by Kohne and co-workers in 2010 (Kohne et al., 2010). Results showed 48% response rate (RR) for wild type and 29% RR for mutant K-ras patients, with no differences in side effects. Results of a secondary analysis of initial quality of life measures were published during ASCO Annual Meeting 2011 (Karthaus et al., 2011). The results demonstrated that those patients with better quality of life had better tumour responses as well by week 8 and 24 of the combination therapy. It does not seem to be an overstatement that the combination of pmab with cytostatics in the first line treatment of CRC is a promising option for patients in better clinical (performance and quality of life) status.

#### **5.2 Combination of pmab and chemotherapy as a second-line treatment**

Peeters and co-workers compared the efficacy of pmab and FOLFIRI to FOLFIRI alone as the second-line treatment of mCRC patients in a phase 3, equally randomized trial (Peeters et al., 2010). The study was originally designed to compare the therapeutic effect in the entire population, but due to convincing external data it was modified before the efficacy assessments so that prospective assessments would be carried out as per the K-ras status of the tumour.

A total number of 1186 patients were treated after randomization. Five hundred-ninety-two (50%) patients were given pmab and FOLFIRI, and 595 (50%) were given FOLFIRI alone. The K-ras status of 1083 patients (91%) was known (based on central laboratory tests): 597 patients (55%) had wild-type K-ras tumour and 486 (45%) had K-ras mutant metastatic colon cancer.

The eligible patients were older than 18 years and their ECOG performance status was 0, 1 or 2. Only one earlier chemotherapeutic scheme, i.e. first-line fluoropyrimidine-based chemotherapy was allowed for the treatment of mCRC. A radiologically verified progression by RECIST was required during the course of treatment or within 6 months. Known EGFR expression or K-ras status were not required for enrolment. Patients previously treated with irinotecan or anti-EGFR therapy were excluded from the study (Jaffe, 2006), (Padhan & Ollivier, 2001), (Therasse et al., 2000).

Pmab (at 6 mg/kg) was administered over 60 minutes by infusion before chemotherapy; if patients tolerated the first dose, the following infusions were administered over 30 minutes. Every patient was given FOLFIRI: 180 mg/m2 irinotecan and 400 mg/m2 raceme leucovorin

Panitumumab for the Treatment of Metastatic Colorectal Cancer 397

al., 2011a, b). As the results of clinical studies with pmab concern distant metastatic diseases only, there is no significant difference between the two compilations. Like other agents affecting biological targets pmab is not allowed in any adjuvant indication except for clinical trials. Pmab is recommended in monotherapy or in combination with FOLFIRI in diseases with distant metastases whether or not resection of the primary malignancy was performed. It is considered reasonable to remove the primary malignancy (which has not been removed earlier) and the distant metastases in one or more surgeries following a 2- to 3-month treatment. (It is strongly highlighted in the recommendation, that K-ras evaluation must be performed and that the product should be administered only in patients with the wild type K-ras.) In non-resectable synchronous or metachronous distant metastases FOLFIRI pmab is an alternative of FOLFIRI bevacizumab or cetuximab as a first-line therapy at least 12

In patients eligible for intensive treatment, pmab FOLFOX is considered as the first-line therapy of metastatic diseases (among other combinations), while pmab FOLFIRI acts as a second-line therapy. Monoclonal antibody panitumumab is indicated as monotherapy in case the patient has decreased chemotherapy tolerance. Biological targeted agents such as pmab (depending on the previously administered agents) can be administered following a new progression (i.e. as a third treatment possibility), mostly in patients who do not tolerate irinotecan. Vectibix is recommended as a monotherapy by NCCN in patients who are

The European Society for Medical Oncology (ESMO) released guidelines in 2010. Pmab is not mentioned in the publications referring to the diagnosis, adjuvant therapy and followup of CRC (Labianca et al., 2010). This is compliant with the European Summary of Product Characteristics, which limits treatment possibilities much rigorously than the guidelines in the United States do. In guidelines detailing the treatment of advanced disease authors state (van Cutsem et al., 2010) that anti-EGFR antibodies pmab and cmab are effective as monotherapy for patients with chemorefractory mCRC, and wild-type state of K-ras is necessary to reach therapeutic effect. In comparison with BSC, pmab is considered beneficial in terms of PFS; this effect is not reflected in terms of overall survival (OS) due to the "crossover" design of trials. Pmab and polychemotherapy (FOLFOX4 as a first-line, and FOLFIRI as a second-line therapy) but the absence of significant improvement in OS is emphasized.

Being a fully human monoclonal antibody not requiring a special pre-treatment or saturation dosage, pmab belongs to a new group of biological targeted agents used in the treatment of metastatic colon or rectal cancer. Pmab binds to EGF receptors, and the poststudy pathologic evaluation of monotherapy registration trial samples provided convincing evidence of the crucial role K-ras status played in clinical efficacy: median progression-free survival was 16 weeks in the wild-type (vs. 8 weeks with best supportive care) patients group. Although pmab was practically ineffective in patients with mutant K-ras, side effects were more frequent and severe. According to effective Summaries of Product Characteristics the product can be applied in Europe as monotherapy in EGFR positive and K-ras wild-type

months after the administration of adjuvant FOLFOX.

Evidence level of all recommendations for pmab therapy is IB.

ineligible for intensive therapy.

**7. Summary** 

**6.2 Pmab in european guidelines** 

(or 200 mg/m2 l-leucovorin) by intravenous infusion on day 1 and 400 mg/m2 fluorouracil by intravenous bolus on day 1, followed by 2400 mg/m2 by continuous infusion on days 1 and 2. Patients were given chemotherapy with pmab or without pmab until the onset of progression or intolerance as per RECIST (confirmed by independent investigators) (Jaffe, 2006), (Padhan & Ollivier, 2001), (Therasse et al., 2000).

In terms of evaluation the study had four arms, groups of K-ras mutant and wild-type patients (as previously assessed) were distinguished both in the FOLFIRI (alone) and the pmab + FOLFIRI arm.

PFS improved significantly in the subgroup of wild-type K-ras patients if pmab was added to chemotherapy; the median PFS was 5.9 and 3.9 months in the pmab + FOLFIRI and the FOLFIRI alone group, respectively. A non-significant increase in OS was also observed, median OS was 14.5 and 12.5 months, the response rate improved from 10% to 35% with added pmab. Theoretical assumptions and earlier clinical experiences were confirmed by the fact that no difference was seen in terms of efficacy in patients with K-ras mutant tumors compared to chemotherapy alone.

Antibodies produced against pmab following therapy were found (by central laboratory) in less than 1% (4 out of 501) of patients. None of these antibodies had a neutralizing effect.

Subgroup analysis suggests that pmab was advantageous in every subgroup in terms of improved PFS with a similar age and gender tendency as seen in the "PRIME" study. In terms of OS, combination arm seemed equivocal with chemotherapy alone in patients previously treated with oxaliplatin, beside those overlapping confidence intervals and summary measures favouring panitumumab reinforced a positive tendency of improving OS.

The authors claimed that the study confirmed the efficacy of pmab with FOLFIRI in K-ras wild-type mCRC patients who were treated previously. PFS improved in a statistically significant manner in this group, which underlines the fact that K-ras status of the tumour can be considered as a predictive biomarker. With a Q2W administration, pmab was comfortably combined with FOLFIRI given at a similar dosing frequency. The toxicity profile was not different from that of EGFR inhibitors and chemotherapy combinations, toxicities could have been managed well.

Considering that, in Hungary bevacizumab is reimbursed only as a first line treatment by the state health fund - even though its use is not confined to a given line in mCRC by the effective Summaries of Product Characteristics (EMA, 2011a). Peeters and co-workers published data of critical relevance in ASCO Annual Meeting 2010 in this aspect (Peeters et al., 2010). The authors evaluated K-ras wild type patients from the above study previously treated with bevacizumab. According to the results, PFS was not different in bev pre-treated patients compared to the overall K-ras wild type study population (5.8 and 3.7 months vs. 5.9 and 3.9 months for pmab + FOLFIRI and FOLFIRI arms). In striking contrast OS improved when bev treatment preceded the pmab + FOLFIRI combination in second line from 14.5 months to 15.7 months.

#### **6. Panitumumab in current therapeutic guidelines**

#### **6.1 Pmab in U. S. guidelines**

From among clinical recommendations issued in the United States the first to review is the National Comprehensive Cancer Network's (NCCN) guidelines referring to the diagnosis and treatment of colon (Version 3.2011) and rectal carcinoma (Version 4.2011) (Engstrom et

(or 200 mg/m2 l-leucovorin) by intravenous infusion on day 1 and 400 mg/m2 fluorouracil by intravenous bolus on day 1, followed by 2400 mg/m2 by continuous infusion on days 1 and 2. Patients were given chemotherapy with pmab or without pmab until the onset of progression or intolerance as per RECIST (confirmed by independent investigators) (Jaffe,

In terms of evaluation the study had four arms, groups of K-ras mutant and wild-type patients (as previously assessed) were distinguished both in the FOLFIRI (alone) and the

PFS improved significantly in the subgroup of wild-type K-ras patients if pmab was added to chemotherapy; the median PFS was 5.9 and 3.9 months in the pmab + FOLFIRI and the FOLFIRI alone group, respectively. A non-significant increase in OS was also observed, median OS was 14.5 and 12.5 months, the response rate improved from 10% to 35% with added pmab. Theoretical assumptions and earlier clinical experiences were confirmed by the fact that no difference was seen in terms of efficacy in patients with K-ras mutant tumors

Antibodies produced against pmab following therapy were found (by central laboratory) in less than 1% (4 out of 501) of patients. None of these antibodies had a neutralizing effect. Subgroup analysis suggests that pmab was advantageous in every subgroup in terms of improved PFS with a similar age and gender tendency as seen in the "PRIME" study. In terms of OS, combination arm seemed equivocal with chemotherapy alone in patients previously treated with oxaliplatin, beside those overlapping confidence intervals and summary measures favouring panitumumab reinforced a positive tendency of

The authors claimed that the study confirmed the efficacy of pmab with FOLFIRI in K-ras wild-type mCRC patients who were treated previously. PFS improved in a statistically significant manner in this group, which underlines the fact that K-ras status of the tumour can be considered as a predictive biomarker. With a Q2W administration, pmab was comfortably combined with FOLFIRI given at a similar dosing frequency. The toxicity profile was not different from that of EGFR inhibitors and chemotherapy combinations,

Considering that, in Hungary bevacizumab is reimbursed only as a first line treatment by the state health fund - even though its use is not confined to a given line in mCRC by the effective Summaries of Product Characteristics (EMA, 2011a). Peeters and co-workers published data of critical relevance in ASCO Annual Meeting 2010 in this aspect (Peeters et al., 2010). The authors evaluated K-ras wild type patients from the above study previously treated with bevacizumab. According to the results, PFS was not different in bev pre-treated patients compared to the overall K-ras wild type study population (5.8 and 3.7 months vs. 5.9 and 3.9 months for pmab + FOLFIRI and FOLFIRI arms). In striking contrast OS improved when bev treatment preceded the pmab + FOLFIRI combination in second line

From among clinical recommendations issued in the United States the first to review is the National Comprehensive Cancer Network's (NCCN) guidelines referring to the diagnosis and treatment of colon (Version 3.2011) and rectal carcinoma (Version 4.2011) (Engstrom et

2006), (Padhan & Ollivier, 2001), (Therasse et al., 2000).

pmab + FOLFIRI arm.

improving OS.

compared to chemotherapy alone.

toxicities could have been managed well.

from 14.5 months to 15.7 months.

**6.1 Pmab in U. S. guidelines** 

**6. Panitumumab in current therapeutic guidelines** 

al., 2011a, b). As the results of clinical studies with pmab concern distant metastatic diseases only, there is no significant difference between the two compilations. Like other agents affecting biological targets pmab is not allowed in any adjuvant indication except for clinical trials. Pmab is recommended in monotherapy or in combination with FOLFIRI in diseases with distant metastases whether or not resection of the primary malignancy was performed. It is considered reasonable to remove the primary malignancy (which has not been removed earlier) and the distant metastases in one or more surgeries following a 2- to 3-month treatment. (It is strongly highlighted in the recommendation, that K-ras evaluation must be performed and that the product should be administered only in patients with the wild type K-ras.) In non-resectable synchronous or metachronous distant metastases FOLFIRI pmab is an alternative of FOLFIRI bevacizumab or cetuximab as a first-line therapy at least 12 months after the administration of adjuvant FOLFOX.

In patients eligible for intensive treatment, pmab FOLFOX is considered as the first-line therapy of metastatic diseases (among other combinations), while pmab FOLFIRI acts as a second-line therapy. Monoclonal antibody panitumumab is indicated as monotherapy in case the patient has decreased chemotherapy tolerance. Biological targeted agents such as pmab (depending on the previously administered agents) can be administered following a new progression (i.e. as a third treatment possibility), mostly in patients who do not tolerate irinotecan. Vectibix is recommended as a monotherapy by NCCN in patients who are ineligible for intensive therapy.

#### **6.2 Pmab in european guidelines**

The European Society for Medical Oncology (ESMO) released guidelines in 2010. Pmab is not mentioned in the publications referring to the diagnosis, adjuvant therapy and followup of CRC (Labianca et al., 2010). This is compliant with the European Summary of Product Characteristics, which limits treatment possibilities much rigorously than the guidelines in the United States do. In guidelines detailing the treatment of advanced disease authors state (van Cutsem et al., 2010) that anti-EGFR antibodies pmab and cmab are effective as monotherapy for patients with chemorefractory mCRC, and wild-type state of K-ras is necessary to reach therapeutic effect. In comparison with BSC, pmab is considered beneficial in terms of PFS; this effect is not reflected in terms of overall survival (OS) due to the "crossover" design of trials. Pmab and polychemotherapy (FOLFOX4 as a first-line, and FOLFIRI as a second-line therapy) but the absence of significant improvement in OS is emphasized. Evidence level of all recommendations for pmab therapy is IB.

#### **7. Summary**

Being a fully human monoclonal antibody not requiring a special pre-treatment or saturation dosage, pmab belongs to a new group of biological targeted agents used in the treatment of metastatic colon or rectal cancer. Pmab binds to EGF receptors, and the poststudy pathologic evaluation of monotherapy registration trial samples provided convincing evidence of the crucial role K-ras status played in clinical efficacy: median progression-free survival was 16 weeks in the wild-type (vs. 8 weeks with best supportive care) patients group. Although pmab was practically ineffective in patients with mutant K-ras, side effects were more frequent and severe. According to effective Summaries of Product Characteristics the product can be applied in Europe as monotherapy in EGFR positive and K-ras wild-type

Panitumumab for the Treatment of Metastatic Colorectal Cancer 399

Bardelli, A. & Sien, S. (2010). Molecular Mechanisms of Resistance to Cetuximab and

BC Cancer Agency Cancer Management Guidelines (2006). Guidelines for the management

6DE017C9B4C4/19258/Managementforbevacizumabsideeffects\_1Dec06.pdf Benson, AB.; Catalano, PJ.; Meropol, NJ., O'Dwyer PJ. & Giantonio, BJ. (2003). Bevacizumab (anti-

Benvenuti, S.; Sartore-Bianchi, A.; Di Nicolantonio, F.; Zanon, C.; Moroni, M., et al. (2007).

Berardi, R.; Onofri, A.; Pistelli, M.; Maccaroni, E.; Scartozzi, M., et al. (2010). Panitumumab:

Berlin, J.; Neubauer, M.; Swanson, P.; Harker, WG.; Burris, H.**,** et al: (2006). Panitumumab

Berlin, J.; Posey, J.; Tchekmedyian, S.; Hu, E.; Chan, D., et al. (2007). Panitumumab with

Blazer, M. (2010). Advances in the Treatment of Colorectal Cancer. (*The Oncology Nurse –* 

Borgstrom, P.; Gold, DP.; Hillan, KJ. & Ferrara, N. (1999). Importance of VEGF for breast

Burtness, B.; Anadkat, M.; Basti, S.; Hughes, M.; Lacouture, ME., et al. (2009). NCCN Task

*Society of Clinical Oncology* Vol.22: 2003 (abstr 975), ISSN: 0736-7589

March 2010), pp. 1254-1261*.*, ISSN: 02773732

6D39414F-EC1A-4BE2-9ACB-

ISSN 0923-7534

Online: 1938-0674

treatment-colorectal-cancer

4214, Print ISSN: 0250-7005

Panitumumab in Colorectal Cancer. *Journal of Clinical Oncology,* Vol.28, No.7 (1

of side effects of bevacizumab (Avastin®) in patients with colorectal cancer. (*BC Cancer Agency)*. Available from http://www.bccancer.bc.ca/NR/rdonlyres/

VEGF) plus FOLFOX4 in previously treated advanced colorectal cancer (advCRC): An interim toxicity analysis of the Eastern Cooperative Oncology Group (ECOG) study E3200. *2003 ASCO Annual Meeting*, Citation: *Proceeding of the Annual Meeting – American* 

Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. *Cancer Research* Vol.67, Nr.6, (15 March 2007), pp. 2643-2648, ISSN: 0008-5472.

the evidence for its use in the treatment of metastatic colorectal cancer. *Core Evidence*, Vol.5, Issue1, (24 September 2010), pp. 61–76, ISSN Online: 1555-175X Bergers, G. & Benjamin, LE. (2003). Tumorigenesis and the angiogenic switch. *Nature Reviews Cancer*, Vol.3, Nr.6, (June 2003), pp. 401-410, ISSN: 1474-175X Berlin, J.; Malik, I. & Picus, J. (2004). Panitumumab therapy with irinotecan, 5-fluorouracil,

and leukovorin (IFL) in metastatic colorectal patients. *Proceeding European Society of Medical Oncology 2004*, Citation: *Annals of Oncology,* Vol.15 suppl. 3. (October 2004),

antitumor activity in patients (pts) with metastatic colorectal cancer (mCRC) expressing ≥10% epidermal growth factor receptor (EGFr)*. 2006 ASCO Annual Meeting,* Citation*: Journal of Clinical Oncology, 2006 ASCO Annual Meeting Proceedings,* Part I. Vol.24, No.18S (June 20 Supplement), 2006: 3548, ISSN: 02773732

irinotecan/leucovorin/5-fluorouracil for first-line treatment of metastatic colorectal cancer. *Clinical Colorectal Cancer,* Vol.6, Nr.6, (March 2007), pp. 427-32, 2007, ISSN

*APN/PNA*), Available from http://www.theoncologynurse.com/article/advances-

cancer angiogenesis in vivo: Implications from intravital microscopy of combination treatments with an anti-VEGF neutralizing monoclonal antibody and doxorubicin. *Anticancer Research* Vol.19, Nr.5B, (September-October 1999), pp. 4203-

Force Report: Management of Dermatologic and Other Toxicities Associated With EGFR Inhibition in Patients With Cancer. *Journal of the National Comprehensive Cancer Network*, Vol.7, Suppl.1, (May 2009), pp. S5-S21, ISSN (electronic): 1540-1413.

mCRC patients after fluoropyrimidine, oxaliplatin and irinotecan-based chemotherapeutic protocols had failed.

Based on clinical study results published in 2011, the addition of panitumumab to FOLFOX4 polychemotherapy as a first-line treatment in wild-type K-ras resulted in a significant increase in progression-free survival (PFS) (8.0 to 9.6 months), while increase in overall survival (OS) (19.7 to 23.9 vs. FOLFOX4 alone) was clinically considerable but nonsignificant. In K-ras mutant cases however, Vectibix with FOLFOX4 versus FOLFOX4 alone decreased the median PFS (8.8 to 7.3 months) and OS (19.3 to 15.5 months). PFS improved significantly in the group of wild-type K-ras patients if pmab was added to the FOLFIRI protocol as a second-line treatment; median PFS was 5.9 and 3.9 months in the pmab + FOLFIRI and the FOLFIRI alone groups, respectively. A non-significant increase in OS was also observed; median OS was 14.5 and 12.5 months, and response rate significantly improved from 10% to 35% with added pmab. In mutant K-ras, PFS was 5.0 months with added monoclonal antibodies and 4.9 months with FOLFIRI alone, while OS was 11.8 and 11.1 months, respectively, i.e. no difference could have been statistically confirmed. Following a positive EMA's Committee for Medicinal Products for Human Use (CHMP) opinion in the middle of 2011, both the FOLFOX4 (1st line) and the FOLFIRI (2nd line) combinations will be likely authorized in the EU for the treatment of mCRC.

The side effect profile matches other EGFR inhibitors (the spectrum as a whole being utterly different from that of conventional cytostatics), with dermatologic symptoms (rash), nail diseases, lung infiltration, diarrhoea and electrolyte disturbances of renal origin may develop. Infusion complications are not common. Panitumumab therapy is safe in cases where followed-up carefully, this may mean temporary suspension of treatment, dose reduction or therapy discontinuation if justified by above detailed side effect related signs and symptoms.

#### **8. Acknowledgments**

I would like to express my gratitude to all those whom I got the chance to learn from, my coauthors who are my co-workers as well, my wife, my daughter, and all my colleagues who did their best to provide a quiet environment for this work. I am grateful to my patients for the experiences I gained during their treatment encouraged me to write this chapter. Last but not least, I thank Amgen Inc. for the invaluable technical support.

#### **9. References**


mCRC patients after fluoropyrimidine, oxaliplatin and irinotecan-based chemotherapeutic

Based on clinical study results published in 2011, the addition of panitumumab to FOLFOX4 polychemotherapy as a first-line treatment in wild-type K-ras resulted in a significant increase in progression-free survival (PFS) (8.0 to 9.6 months), while increase in overall survival (OS) (19.7 to 23.9 vs. FOLFOX4 alone) was clinically considerable but nonsignificant. In K-ras mutant cases however, Vectibix with FOLFOX4 versus FOLFOX4 alone decreased the median PFS (8.8 to 7.3 months) and OS (19.3 to 15.5 months). PFS improved significantly in the group of wild-type K-ras patients if pmab was added to the FOLFIRI protocol as a second-line treatment; median PFS was 5.9 and 3.9 months in the pmab + FOLFIRI and the FOLFIRI alone groups, respectively. A non-significant increase in OS was also observed; median OS was 14.5 and 12.5 months, and response rate significantly improved from 10% to 35% with added pmab. In mutant K-ras, PFS was 5.0 months with added monoclonal antibodies and 4.9 months with FOLFIRI alone, while OS was 11.8 and 11.1 months, respectively, i.e. no difference could have been statistically confirmed. Following a positive EMA's Committee for Medicinal Products for Human Use (CHMP) opinion in the middle of 2011, both the FOLFOX4 (1st line) and the FOLFIRI (2nd line)

combinations will be likely authorized in the EU for the treatment of mCRC.

but not least, I thank Amgen Inc. for the invaluable technical support.

hp-cancer-guide-panitumumab.pdf

The side effect profile matches other EGFR inhibitors (the spectrum as a whole being utterly different from that of conventional cytostatics), with dermatologic symptoms (rash), nail diseases, lung infiltration, diarrhoea and electrolyte disturbances of renal origin may develop. Infusion complications are not common. Panitumumab therapy is safe in cases where followed-up carefully, this may mean temporary suspension of treatment, dose reduction or therapy discontinuation if justified by above detailed side effect related signs

I would like to express my gratitude to all those whom I got the chance to learn from, my coauthors who are my co-workers as well, my wife, my daughter, and all my colleagues who did their best to provide a quiet environment for this work. I am grateful to my patients for the experiences I gained during their treatment encouraged me to write this chapter. Last

Alberta Health Services. (2010). Panitumumab (Vectibix®) Administration Guidelines. In:

Allen, JA.**;** Adlakha, A. & Bergethon, PR**.** (2006). Reversible posterior leukoencephalopathy

*Alberta Health Services*, Available from http://www.albertahealthservices.ca/hp/if-

syndrome after bevacizumab/FOLFIRI regimen for metastatic colon cancer. *Archives of Neurology*, Vol.63, Nr.10, (October 2006), pp. 1475-1478, ISSN: 0003-9942 Amado, RG.; Wolf, M.; Peeters, M.; van Cutsem, E.; Siena, S., et al*.* (2008). Wild-Type *KRAS* 

Is Required for Panitumumab Efficacy in Patients With Metastatic Colorectal Cancer. *Journal of Clinical Oncology,* Vol.26, Nr.10. *(*April 2008). pp. 1626-1634*,* ISSN:

protocols had failed.

and symptoms.

**9. References** 

02773732

**8. Acknowledgments** 


6DE017C9B4C4/19258/Managementforbevacizumabsideeffects\_1Dec06.pdf


Panitumumab for the Treatment of Metastatic Colorectal Cancer 401

Engstrom, PF.; Arnoletti, JP.; Benson, AB.; Chan E., Chen YJ., et al. (2011). Rectal cancer,

European Medicines Agency. (2011). Avastin. EPAR – Product information. 27/04/2011. In:

jsp&murl=menus/medicines/medicines.jsp&mid=WC0b01ac058001d125 European Medicines Agency. (2011). Vectibix. EPAR – Product information. 16/02/2011. In:

Fakih, MG. & Lombardo, JC. (2006). Bevacizumab-induced Nasal Septum Perforation. *The Oncologist*, Vol.11, No.1, (January 2006), pp. 85-86, ISSN Online 1549-490X Figlin, RA.; Belldegrun AS.; Crawford, J.; Lohner, M.; Roskos, L., et al. (2002). ABX-EGF, a

Folkman, J. (1971). Tumor angiogenesis: therapeutic implications. *The New England Journal of Medicine,* Vol. 285, Nr.21, (November 18, 1971) pp. 1182-1186, ISSN: 00284793 Gandara, DR.; Yoneda, K.; Shelton, D.; Beckett, LA.; Ramies, DA., et al. (2006). Independent

Garcia-Sáenz, JA.; Sastre, J. & Diaz-Rubio Garcia, E. (2009). Biomarkers and anti-EGFR

Giantonio, BJ.; Chen, HX.; Catalano, PJ.; Meropol, NJ.; O'Dwyer, PJ., et al. (2004). Bowel

Goldberg, RM.; Rothenberg ML.; van Cutsem, EB.; Benson, AB.; Blanke. CD., et al (2007).

Vol.22, No.14S (15 July Supplement), 2004: 3017, ISSN: 02773732

*Oncology,* Vol.11, Nr.11, (November 2009), pp. 737-747, ISSN: 1699048X Gaudi, I. & Kásler, M. (2002). The course of cancer mortality in Hungary between 1975 and

*Oncology* Vol.21: 2002 (abstr 35), ISSN: 0736-7589

291–295, ISSN Online 2060-0399

1549-490X

http://www.nccn.org/professionals/physician\_gls/pdf/rectal.pdf Esteller M.; Gonzalez, S.; Risques, RA.; Marcuello, E.; Mangues, R., et al (2001). K-ras and

*Oncology,* Vol.19, No.2 (15 January 2001), pp. 299-304*,* ISSN: 02773732 European Medicines Agency. (2009). Erbitux. EPAR – Product information. 14/07/2009. In:

WC500029119.pdf

500047710.pdf

Version 4.2011, In: *NCCN Clinical Practice Guidelines in Oncology*), Available from

p16 aberrations confer poor prognosis in human colorectal cancer. *Journal of Clinical* 

*European Medicines Agency.* Available from http://www.ema.europa.eu/ docs/en\_GB/document\_library/EPAR\_-\_Product\_Information/human/000558/

*European Medicines Agency*. Available from http://www.ema.europa.eu/ema/ index.jsp?curl=pages/medicines/human/medicines/000582/human\_med\_000663.

*European Medicines Agency*. Available from http://www.ema.europa.eu/docs/ en\_GB/document\_library/EPAR\_-\_Product\_Information/human/000741/WC

fully human anti-epidermal growth factor receptor (EGFR) monoclonal antibody (mAb) in patients with advanced cancer: phase 1 clinical results. *2002 ASCO Annual Meeting*. Citation: *Proceeding of the Annual Meeting – American Society of Clinical* 

review of fatal interstitial lung disease (ILD) in TRIBUTE: paclitaxel + carboplatin ± erlotinib in advanced non-small cell lung cancer (NSCLC). *2006 ASCO Annual Meeting*, Citation: *Journal of Clinical Oncology, 2006 ASCO Annual Meeting Proceedings* Part I. Vol. 24, No. 18S (20 June Supplement), 2006: 7071, ISSN: 02773732

therapies for KRAS wild-type metastatic colorectal cancer. *Clinical and Translational* 

2001. (Article in Hungarian) *Hungarian Oncology,* Vol.46, Nr.4. (December 2002), pp.

perforation and fistula formation in colorectal cancer patients treated on Eastern Cooperative Oncology Group (ECOG) studies E2200 and E3200, *2004 ASCO Annual Meeting Proceedings*, Citation: *Journal of Clinical Oncology*, *Post-Meeting Edition,*

The Continuum of Care: A Paradigm for the Management of Metastatic Colorectal Cancer. *The Oncologist*, Vol. 12, No. 1, Online (January 2007), pp. 38-50, ISSN Online


Busam, KJ.; Capodieci, P.; Motzer, R.; Kiehn, T.; Det P., et al. (2001). Cutaneous side-effects

Cartwright TH. & Genther R. (2008). Successful Administration of Panitumumab Alone

Chau, I. & Cunningham, D. (2009). Treatment in advanced colorectal cancer: what, when

Chung, CH. (2008). Managing Premedications and the Risk for Reactions to Infusional

Cohenuram, M. & Saif, MW. (2007). Panitumumab the first fully human monoclonal

Coutinho, AK. & Rocha Lima, CM. (2003). Metastatic colorectal cancer: systemic treatment

Dahabreh, J.; Terasawa, T.; Castaldi, PJ. &. Trikalinos, TA. (2011). Systematic review: Anti–

de Bono, JS. & Rowinsky, EK. (2002). Therapeutics targeting signal transduction for patients

De Roock, W.; Claes, B.; Bernasconi, D.; De Schutter, J.; Biesmans, B., et al. (2010). Effects of

Douillard, JY.; Siena, S.; Cassidy, J.; Tabernero, J.; Burkes, R., et al. (2010). Randomized, Phase

Eng, C. (2010). The Evolving Role of Monoclonal Antibodies in Colorectal Cancer: Early

Engstrom, PF.; Arnoletti, JP.; Benson, AB.; Chan E., Chen YJ., et al. (2011). Colon cancer,

http://www.nccn.org/professionals/physician\_gls/pdf/colon.pdf

Online 1365-2133

1719, ISSN 0007-0920

ISSN: 1073-2748

732, ISSN Online 1549-490X

pp. 227-254, ISSN: 00071420

2010), pp.753-762, ISSN: 1470-2045

article.cfm?c=16&s=59&ss=224&id=961

Nr.1 (January 2010), pp.73–84, ISSN Online 1549-490X

2007), pp. 7–15, ISSN (printed): 0959-4973

(January 2011), pp. 37-49**,** ISSN: 0003-4819

1938-0674

in cancer patients treated with the antiepidermal growth factor receptor antibody C225. *British Journal of Dermatology*, Vol.144, Nr.6, (June 2001), pp.1169-1176, ISSN

After Severe Infusion Reaction to Cetuximab in a Patient with Metastatic Colorectal Cancer. *Clinical Colorectal Cancer*, Vol.7, Nr.3, (May 2008), pp. 202-203, ISSN Online:

and how? (Review). *British Journal of Cancer,* Vol.100, Nr.11, (June 2009), pp: 1704–

Monoclonal Antibody Therapy. *The Oncologist,* Vol.13, Nr.6 (June 2008), pp. 725–

antibody: from the bench to the clinic. *Anti-Cancer Drugs,* Vol.18, Nr.1. (January

in the new millennium. *Cancer Control,* Vol.10, Nr. 3. (May/June 2003), pp. 224-238,

Epidermal Growth Factor Receptor treatment effect modification by KRAS Mutations in advanced colorectal cancer. *Annals of Internal Medicine*, Vol. 154, Nr.1

with colorectal carcinoma. *British Medical Bulletin,* Vol.64, Issue1, (December 2002)

KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis. *The Lancet Oncology*, Vol.11, Nr.8, (Augustus

III Trial of Panitumumab with Infusional Fluorouracil, Leucovorin, and Oxaliplatin (FOLFOX4) versus FOLFOX4 alone as First-Line Treatment in Patients With Previously Untreated Metastatic Colorectal Cancer: The PRIME Study. *Journal of Clinical Oncology,* Vol.28, No.31 (1 November 2010), pp. 4697-4705*.*, ISSN: 02773732 Eaby, B. (n. d.). Nursing Management of Patients Receiving EGFR Inhibitors. In: *OncoLink* 

*Cancer Resources > Nurses' Notes > Nursing Continuing Education > Continuing Education Modules*. Available from http://www.oncolink.org/resources/

Presumptions and Impact on Clinical Trial Development. *The Oncologist,* Vol.15,

Version 3.2011 In: *NCCN Clinical Practice Guidelines in Oncology*, Available from


Panitumumab for the Treatment of Metastatic Colorectal Cancer 403

Hochster, HS. (2006). Use of Monoclonal Antibodies in Treatment of Advanced Colorectal

Inoue, A.; Saijo, Y.; Maemondo, M.; Gomi, K.; Tokue, Y. et al. (2003). Severe acute interstitial

Jaffe, CC. (2006). Measures of Response: RECIST, WHO, and New Alternatives. *Journal of Clinical Oncology*, Vol.24, No.20 (July 10 2006), pp. 3245-3251, ISSN: 02773732 Karthaus, M.; Thaler, J.; Hofheinz, R.; Mineur, L.; Letocha, H., et al. (2011). The relationship

Keating, GM. (2010). Panitumumab. A Review of its Use in Metastatic Colorectal Cancer.

Khan, NF.; Ward, A.; Watson, E.; Austoker, J. & Rose, PW. (2008). Long-term survivors of

*Journal of Cancer***,** Vol. 44, Nr. 2, (January 2008), pp 195-204, ISSN**:** 0959-8049 Khan, NF.; Mant, D. & Rose, PW. (2010). Quality of Care for Chronic Diseases in a British

Kilickap, S.; Abali, H. & Celik, I. (2003). Bevacizumab, bleeding, thrombosis, and warfarin.

Kim, KJ.; Li, B.; Winer, J.; Armanini, M.; Gillett, N., et al. (1993). Inhibition of vascular

*Nature*, Vol.362, Nr. 6423, (29 April 1993), pp. 841-844, 1993, ISSN: 0028-0836 Knudson, D. (2007). Nursing considerations when administering panitumumab. *Community* 

Kohne, C.; Mineur, L.; Greil R.; Letocha, H.; Thaler, J., et al. (2010). Primary analysis of a

Labianca, R.; Nordlinger., B; Beretta, GD.; Brouquet, A. & Cervantes, A. (2010). Primary

*Drugs,* Vol.70, Nr.8, (May 2010), pp. 1059-1078, ISSN 0012-6667

(September/October 2010), pp. 418-424, ISSN (printed): 1544-1709

*Oncology,* Vol.4, Nr.3, (March, 2007), pp. 125-126, ISSN: 1548-5315

(January 2009), pp: 22-28, ISSN Online 1549-490X.

139, ISSN: 01406736

3542 , ISSN: 02773732

72&abstractID=1456

02773732

Cancer: An Expert Interview With Dr. Howard Hochster. (*Medscape Hematology-Oncology),* Available from http://www.medscape.org/viewarticle/550028\_print Hoda, D.; Simon GR. & Garrett CR. (2008). Targeting colorectal cancer with anti-epidermal

growth factor receptor antibodies: focus on panitumumab. *Therapeutics and Clinical Risk Management*, Vol.4, Nr.6, (December 2008), pp. 1221–1227, ISSN: 1178-203X Hurwitz, HL.; Jing, Y.; Ince, W.; Novotny, WF. & Rosen, O. (2009). The Clinical Benefit of

Bevacizumab in Metastatic Colorectal Cancer Is Independent of K-ras Mutation Status: Analysis of a Phase III Study of Bevacizumab with Chemotherapy in Previously Untreated Metastatic Colorectal Cancer. *The Oncologist,* Vol.4. Nr.1

pneumonia and gefitinib. *The Lancet*, Vol.361, Issue9352, (11 January 2003), pp.137-

between quality of life (QoL) and tumor response in patients (pts) with metastatic colorectal cancer (mCRC) receiving panitumumab (pmab) plus FOLFIRI as firstline therapy: An analysis of study 314*. 2011. ASCO Annual Meeting*, Abstract No. 3634, Citation: *Journal of Clinical Oncology*, Vol.29, Suppl. abstr. 3534. 2011, ISSN:

adult cancers and uptake of primary health services: A systematic review. *European* 

Cohort of Long-Term Cancer Survivors. *Annals of Family Medicine,* Vol.8, No.5,

(Correspondence) *Journal of Clinical Oncology*, Vol.21. No.18 (September 2003), pp.

endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo.

phase II study (20060314) combining first-line panitumumab (pmab) with FOLFIRI in the treatment of patients (pts) with metastatic colorectal cancer (mCRC). *2010 Gastrointestinal Cancers Symposium*, Abstract No. 414. Available from http:// www.asco.org/ascov2/Meetings/Abstracts?&vmview=abst\_detail\_view&confID=

colon cancer: ESMO Clinical Practice Guidelines for diagnosis, adjuvant treatment


Grothey, A.; Sargent, D.; Goldberg, RM. & Schmoll, H-J. (2004*).* Survival of Patients With

*Clinical Oncology*, Vol. 22, No. 7 (April 2004), pp. 1209-1214, ISSN: 02773732 Grothey, A. (2006). Recognizing and Managing Toxicities of Molecular Targeted Therapies

Grothey, A. (2007). Rational treatment planning for metastatic disease. *Community Oncology,*

Haller, DG. (2007). Current approaches and management of advanced colorectal cancer. *John* 

Hamilton, SR. (2008). Targeted therapy of cancer: new roles for pathologists in colorectal

Harari, PM. (2004). Epidermal growth factor receptor inhibition strategies in oncology.

Hecht, J.; Patnaik, A.; Malik, I.; Venook, A.; Berlin, J., et al. (2004). ABX-EGF monotherapy in

Hecht, J.; Mitchell, E.; Barada, J.; Malik I.; Richards, D., et al: (2006). Panitumumab antitumor

Hecht, J.; Posey, J.; Tchekmedyian, S., Hu, E.; Chan, D., et al. (2006). Panitumumab in

Helbling, D. & Borner, M. (2007). Successful challenge with the fully human EGFR antibody

Heun, J. & Holen, K. (2007). Treatment with Panitumumab After a Severe Infusion Reaction to

*Colorectal Cancer,* Vol.6, Nr.7, (May 2007), pp. 529-531, ISSN Online: 1938-0674

Abstracts?&vmview=abst\_detail\_view&confID=41&abstractID=179 Heinemann, V.; Stintzing, S.; Kirchner, T.; Boeck, S. & Jung, A. (2009). Clinical relevance of

cancernetwork.com/display/article/10165/60418

1558-0334

0088

2004: 3511, ISSN: 02773732

Supplement), 2006: 3547, ISSN: 02773732

2009), pp. 262–271, ISSN: 0305-7372

Issue 5, (March 2002), pp. 1593–1611, ISSN: 0008-543X

Vol.4, Suppl.3 (June 2007), pp. 24-29, ISSN: 1548-5315

Advanced Colorectal Cancer Improves With the Availability of Fluorouracil-Leucovorin, Irinotecan, and Oxaliplatin in the Course of Treatment. *Journal of* 

for Colorectal Cancer. In: *Cancernetwork,* Available from http://www.

*Hopkins Advanced Studies in Medicine,* Vol.7, Nr.2. (February 2007), pp. 39-44, ISSN-

cancer. *Modern Pathology*, Nr.21, Suppl.2, (May 2008), pp. S23–S30, ISSN: 0893-3952

*Endocrine-Related Cancer,* Vol.11, Nr.4, (December 2004), pp. 689–708, ISSN: 1351-

patients (pts) with metastatic colorectal cancer (mCRC): An updated analysis. *2004 ASCO Annual Meeting*. Citation: *Journal of Clinical Oncology, 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition),* Vol.22, No,14S (July 15 Supplement),

activity in patients with metastatic colorectal cancer expressing low (1–9%) or negative levels of EGFR *2006 ASCO Annual Meeting,* Citation: *Journal of Clinical Oncology, 2006 ASCO Annual Meeting Proceedings,* Part I. Vol. 24, No.18S (June 20

combination with irinotecan, 5-fluorouracil, and leukovorin (IFL) or FOLFIRI for first-line treatment of metastatic colorectal cancer. *2006 Gastrointestinal Cancers Symposium*; (abstr. 237), Available from http://www.asco.org/ascov2/Meetings/

EGFR- and KRAS-status in colorectal cancer patients treated with monoclonal antibodies directed against the EGFR. *Cancer Treatment Reviews*, Vol.35, Nr.3 (May

panitumumab following an infusion reaction with the chimeric EGFR antibody cetuximab. *Annals of Oncology*, Vol.18, Nr.5, (May 2007), pp. 963-964, ISSN 0923-7534 Herbst, RS. & Shin, DM. (2002). Monoclonal Antibodies to Target Epidermal Growth Factor

Receptor–Positive Tumors. A New Paradigm for Cancer Therapy. *Cancer,* Vol.94,

Cetuximab in a Patients with Metastatic Colorectal Cancer: A Case Report. *Clinical* 


Panitumumab for the Treatment of Metastatic Colorectal Cancer 405

Moy, B, & Goss, PE. (2007). Lapatinib-Associated Toxicity and Practical Management

Nagaria, NC.; Cogswell, J.; Choe, JK. & Kasimis, B. (2005) Side Effects and Good Effects from

Nagasaka, T.; Sasamoto, H.; Notohara, K.; Cullings, HM.; Takeda, M., et al. (2004).

*Oncology*, Vol.22, No.22 (15 November 2004), pp. 4584-4594, ISSN: 02773732 National Cancer Institute. (n. d.) Stat Fact Sheets: Colon and Rectum - Survival & Stage.

Nielsen, DL.; Pfeiffer, P. & Jensen, BV. (2009). Six cases of treatment with panitumumab in

Normanno, N.; Tejpar, S.; Morgillo, F.; De Luca, A.; van Cutsem, E., et al. (2009).

O'Neil, BH.; Allen, R.; Spigel, DR.; Stinchcombe., TE; Moore, DT., et al. (2007). High

Ottó, S & Kásler, M. (2005). Trends in cancer mortality and morbidity in Hungarian and

Padhani, R. & Ollivier, L. (2001). The RECIST criteria: implications for diagnostic

Peeters, M; Balfour, J & Arnold, D. (2008) Review article: panitumumab - a fully human anti-

Peeters, M.; Price, TJ.; Cervantes, A.; Sobrero, AF.; Ducreux, M. et al. (2010). Randomized

Pereg, D. & Lishner, M. (2008). Bevacizumab treatment for cancer patients with

Pérez-Soler, R.; Delord, JP.; Halpern, A.; Kelly, K.; Krueger J.**,** et al. (2005). HER1/EGFR

*Oncologist,* Vol.10, Nr.5, (May 2005), pp. 345–356, ISSN Online 1549-490X

Vol.25, No.24, (20 August 2007), pp. 3644-3648, ISSN: 02773732

seer.cancer.gov/statfacts/html/colorect.html#survival

Vol.20, Nr.4. (2 March 2009), p. 789, ISSN 0923-7534

(March 2005), pp. 99–107, ISSN Online 2060-0399

November 2010), pp. 4706-4713, ISSN: 02773732

9, (October 2008), pp. 2325-2326, ISSN: 0195668X

1549-490X

1743-4254

986, ISSN**:** 1748880X

02692813

2005), pp. 2423-2428, ISSN: 02773732

Recommendations, *The Oncologist*, Vol.12, Nr.7 (July 2007), pp.756-765, ISSN Online

New Chemotherapeutic Agents*. Journal of Clinical Oncology*, Vol.23, No.10, (1 April

Colorectal Cancer With Mutation in BRAF, KRAS, and Wild-Type With Respect to Both Oncogenes Showing Different Patterns of DNA Methylation*. Journal of Clinical* 

(*SEER (Surveillance Epidemiology and End Results*), Available from: http://

patients with severe hypersensitivity reactions to cetuximab. *Annals of Oncology*,

Implications for KRAS status and EGFR-targeted therapies in metastatic CRC. *Nature Reviews Clinical Oncology,* Vol.6, Nr.9, (September 2009), pp. 519-527, ISSN:

Incidence of Cetuximab-Related Infusion Reactions in Tennessee and North Carolina and the Association With Atopic History. *Journal of Clinical Oncology,* 

international statistics. Characteristics and potential outcome of public health screening programmes. (Article in Hungarian). *Hungarian Oncology.* Vol.49, Nr. 2,

radiologists. *The British Journal of Radiology*, Vol.74, No.887 (November 2001), 983–

EGFR monoclonal antibody for treatment of metastatic colorectal cancer, *Alimentary Pharmacology & Therapeutics*, Vol.28, Issue 3, (16 April 2008), pp. 269–281, ISSN:

Phase III Study of Panitumumab with Fluorouracil, Leucovorin, and Irinotecan (FOLFIRI) Compared with FOLFIRI alone as Second-Line Treatment in Patients with Metastatic Colorectal Cancer. *Journal of Clinical Oncology,* Vol.28, No.31, (1

cardiovascular disease: a double edged sword? *European Heart Journal*, Vol.29, Nr.

Inhibitor-Associated Rash: Future Directions for Management and Investigation Outcomes from the HER1/EGFR Inhibitor Rash Management Forum. *The* 

and follow-up. *Annals of Oncology,* Vol. 21, Suppl. 5, (May 2010), pp. v70–v77, ISSN 0923-7534


Lacouture, ME. (2009). The Growing Importance of Skin Toxicity in EGFR Inhibitor

Lacouture, ME.; Mitchell, EP.; Piperdi, B.; Pillai, MV.; Sheare, H., et al. (2010). Skin Toxicity

Langerak, A.; River, G.; Mitchell, E.; Cheema P. & Shing, M. (2009). Panitumumab

Lenz, HJ. (2007). Management and Preparedness for Infusion and Hypersensitivity Reactions. *The Oncologist*, Vol.12, No.5, (May 2007), pp. 601-609, ISSN Online 1549-490X LoBuglio, AF.; Wheeler, RH.; Trang J.; Haynes, A.; Rogers, K.,el al. (1989). Mouse/human

Malik I.; Hecht, JR.; Patnaik, A.; Venook, A.; Berlin, J., et al. (2005). Safety and efficacy of

Malumbres, M. & Barbacid, M. (2003). RAS oncogenes: The first 30 years, *Nature Reviews.* 

ManageCRC.com. (2011), Ocular Changes Secondary to Chemotherapy. In:

Martel, CL.; Presant, CA.; Ebrahimi, B.; Upadhyaya, G.; Vakil, M., et al. (2006).

Vol.2, Issue2, (30 November 2007), pp. 81–88, ISSN Online: 1555-175X Mayer, RJ. (2009). Targeted Therapy for Advanced Colorectal Cancer —More Is Not Always

Melosky, B.; Burkes, R.; Rayson, D.; Alcindor, T.; Shear, N.; et al. (2009). Management of skin

Michael, M. & Zalcberg, JR**.** (2000). Chemotherapy for advanced colorectal cancer. It can

*Journal,* Vol.321, Nr.7260, (September 2000), p. 521, ISSN: 0959-8138

*Cancer,* Vol.3, Nr.6. (June 2003) pp. 459-465, ISSN: 1474-175X

ArticleReader.aspx?article=295&page=1

2009). pp. 623-625, ISSN: 00284793

2009), pp. 18-28, ISSN: 1198-0052

Nr.1, (January 2009), pp. 49-54, ISSN Online: 1938-0674

0923-7534

0890-9091

02773732

ISSN: 0027-8424

and follow-up. *Annals of Oncology,* Vol. 21, Suppl. 5, (May 2010), pp. v70–v77, ISSN

Therapy. *Oncology (Williston Park)*, Vol.23, Nr.2, (February 2009), pp.194-196, ISSN:

Evaluation Protocol With Panitumumab (STEPP), a Phase II, Open-Label, Randomized Trial Evaluating the Impact of a Pre-Emptive Skin Treatment Regimen on Skin Toxicities and Quality of Life in Patients With Metastatic Colorectal Cancer. *Journal of Clinical Oncology,* Vol.28, No.8, (10 March 2010), pp. 1351-1357, ISSN:

Monotherapy in Patients with Metastatic Colorectal Cancer and Cetuximab Infusion Reactions: A Series of Four Case Reports. *Clinical Colorectal Cancer*, Vol.8,

chimeric monoclonal antibody in man: Kinetics and immune response. *Proceeding of the National Academy of Sciences (U. S*), Vol.86, Nr.11, (June 1989) pp. 4220-4224,

panitumumab monotherapy in patients with metastatic colorectal cancer (mCRC)**.**  *2005 ASCO Annual Meeting*. Citation: *Journal of Clinical Oncology*, *2005 ASCO Annual Meeting Proceedings,* Vol.23, No 16S (1 June Supplement) 2005: 3520, ISSN: 02773732

*Manage.CRC.com*, Available from: http://www.managecrc.com/Articles/

Bevacizumab-related toxicities: association of hypertension and proteinuria. *Community Oncology,* Vol.3, Nr. 2. (February 2006), pp.90-93, ISSN: 1548-5315 Martinelli, E.; Morgillo, F.; Troiani, T.; Tortora, G. & Ciardiello, F. (2007). Panitumumab: the

evidence for its use in the treatment of metastatic colorectal cancer. *Core Evidence*,

Better. (Editorial)**.** *The New England Journal of Medicine,* Vol.360, Nr.6 (5 February

rash during EGFR targeted monoclonal antibody treatment for gastrointestinal malignancies: Canadian recommendations. *Current Oncology*, Vol.16, Nr.1, (January

improve quality of care and offer modest increases in survival**.** *British Medical* 


Panitumumab for the Treatment of Metastatic Colorectal Cancer 407

Specenier, P.; Koppen, C- & Vermorken, JB. (2007). Diffuse punctate keratitis in a patient

Sudoyo, AW. (n.d.). Chemotherapy and Targeting therapy in Colon Cancer (n.d.). (*Scribd*),

Takimoto, CH. & Calvo, E. (2005). Principles of oncologic pharmacotherapy. In: *Cancer* 

Therasse, P.; Arbuck, SG.; Eisenhauer, EA.; Wanders, J.; Kaplan, RS., et al. (2000). New

Thomas, SF. & Grandis, JR. (2004). Pharmacokinetic and pharmacodynamic properties of

Traina, TF.; Norton, L.; Drucker, K & Singh, B. (2006). Nasal Septum Perforation in a

Tuma, RS. (2006). Panitumumab safe and effective as alternative to cetuximab in colorectal

U. S. Department Of Health And Human Services, U. S. National Institutes of Health,

U. S. Food and Drug Administration. (2009). Avastin. Drug Details. 7/31/2009. *U. S. Food* 

van Cutsem, E.; Peeters, M.; Siena, S.; Humblet, Y.; Hendlisz, A., et al. (2007). Open-Label

van Cutsem, E.; Kohne, CH.; Hitre, E.; , Zaluski, J.; Chien, CRC., et al. (2009). Cetuximab and

No.10, (November 2006), pp. 1070-1071, ISSN Online 1549-490X

http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE\_4.03\_2010-06-

cder/drugsatfda/index.cfm?fuseaction=Search.DrugDetails

Healthcare Media, ISBN: 1-891483-35-8, Washington, USA

Suppl. abstr. 3567. 2011, ISSN: 02773732

2007), pp. 961-962, ISSN 0923-7534

Targeting-therapy-in-Colon-Cancer#

(May 2004), pp. 255–268, ISSN: 0305-7372

14\_QuickReference\_5x7.pdf

4713, ISSN: 02773732

00284793

00278874

Eastern Cooperative Oncology Group (ECOG) performance status (PS). *2011 ASCO Annual Meeting*. Abstract No: 3567. Citation: *Journal of Clinical Oncology,* Vol.29,

treated with cetuximab as monotherapy. *Annals of Oncology*, Vol.18, Nr.5, (May

Available from http://www.scribd.com/doc/43602320/Chemotherapy-and-

*Management: A Multidisciplinary Approach. Medical, Surgical & Radiation Oncology*. Pazdur, R.; Coia, LR.; Hoskins WJ & Wagman, LD. (Eds.), pp. 23-43, CMP

Guidelines to Evaluate the Response to Treatment in Solid Tumors. *Journal of the National Cancer Institute*, Vol.92, No.3, (February 2, 2000), pp. 205-216, ISSN:

EGFR inhibitors under clinical investigation. *Cancer Treatment Reviews*, Vol.30, Nr.3,

Bevacizumab-Treated Patient with Metastatic Breast Cancer*. The Oncologist,* Vol.11,

cancer. *Oncology Times (UK Edition)*, Vol.3, Nr.3 (March 2006), p.20, ISSN: 07262234

National Cancer Institute. (2009) Skin and subcutaneous tissue disorders**.** In: *Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0*., Available from

*and Drug Administration*, Available from http://www.accessdata.fda.gov/scripts/ cder/drugsatfda/index.cfm?fuseaction=Search.Label\_ApprovalHistory#apphist U. S. Food and Drug Administration. (2009). Vectibix. Drug Details. 7/17/2009**.** *U. S. Food* 

*and Drug Administration*, Available from http://www.accessdata.fda.gov/scripts/

Phase III Trial of Panitumumab Plus Best Supportive Care Compared With Best Supportive Care Alone in Patients with Chemotherapy-Refractory Metastatic Colorectal Cancer. *Journal of Clinical Oncology,* Vol.25, No. 1, (1 May 2007), pp. 4706-

Chemotherapy as Initial Treatment for Metastatic Colorectal Cancer. *The New England Journal of Medicine,* Vol.360, Nr.14, (April 2, 2009) pp. 1408-1417, ISSN:


Phillips, JL. & Currow, DC. (2010). Cancer as a chronic disease. *Collegian*, Vol.17, Nr.2, (July

Pikó, B. (2009). Panitumumab-treatment of metastatic colorectal cancer. (Article in

Potthoff, K; Hofheinz, R; Hassel, JC; Volkenandt M; Lordick, F, et al. (2011). Interdisciplinary

*Annals of Oncology*, Vol.22, Nr.(3) (March 2011), pp. 524-35, ISSN 0923-7534 Power DG.; Manish, AS.; Asmis, TR.; Garcia, JJ. & Kemeny, NE. (2010). Safety and efficacy of

Rakkar, AN. (2007). A welcome addition to the therapeutic armamentarium against

Ritter, CA. & Arteaga, CL. (2003).The epidermal growth factor receptor-tyrosine kinase: a

Rowinsky, EK.; Schwartz, GH.; Gollob, JA., et al. (2004.) Safety, pharmacokinetics, and

*Oncology,* Vol.22, Nr.15. (Augustus 2004), pp. 3003-3015, ISSN: 02773732 Saif, MW. & Cohenuram, M. (2006). Role of Panitumumab in the Management of Metastatic

Saif, MW.; Peccerillo, J. & Potter, V. (2008). Successful re-challenge with panitumumab in

Scappaticci, FA.; Skillings, JR.; Holden, SN.; Gerber, HP.; Miller, K., et al. (2007). Arterial

Segaert, S. & van Cutsem, E. (2005). Clinical signs, pathophysiology and management of

Siena, S.; Sartore-Bianchi, A.; Di Nicolantonio, F.; Balfour, J. & Bardelli, A. (2009).

*Institute*, Vol.101, Issue.19, (7 October 2009), pp. 1308–1324, ISSN: 00278874 Siena, S.; Cassidy, J.; Tabernero, J.; Burkes, KM.; Barugel, ME., et al. (2011). Randomized

Issue 16 (15 August 2007), pp. 1232-1239, ISSN: 00278874

Nr. 11. (December 2007), pp. 1469-1474, ISSN 0007-0920

Hungarian). *Hungarian Oncology,* Vol.53, Nr.2, (June 2009), pp. 135-142, ISSN

management of EGFR-inhibitor induced skin reactions: a German expert opinion*.* 

panitumumab following cetuximab: retrospective review of the Memorial Sloan-Kettering experience. *Investigational New Drugs*, Vol. 28, Nr.3, (June 2010), pp. 353-

colorectal cancer. *Community Oncology,* Vol.4, Nr. 3, (March, 2007), pp. 123-125,

promising therapeutic target in solid tumors. *Seminars in Oncology,* Vol. 30. Suppl 1.

activity of ABX-EGF, a fully human anti-epidermal growth factor receptor monoclonal antibody in patients with metastatic renal cell cancer. *Journal of Clinical* 

Colorectal Cancer. *Clinical Colorectal Cancer,* Vol.6, No.2, (July 2006), pp. 118-124,

patients who developed hypersensitivity reactions to cetuximab: report of three cases and review of literature. *Cancer Chemotherapy and Pharmacology*, Vol.63, Nr. 6,

Thromboembolic Events in Patients with Metastatic Carcinoma Treated with Chemotherapy and Bevacizumab. *Journal of the National Cancer Institute,* Vol.99,

skin toxicity during therapy with epidermal growth factor receptor inhibitors. *Annals of Oncology,* Vol.16, Nr.9 (September 2005), pp. 1425–1433, ISSN 0923-7534 Siena, S.; Peeters, M.; van Cutsem E.; Humblet Y.; Conte, P., et al. (2007). Association of

progression-free survival with patient-reported outcomes and survival: results from a randomised phase 3 trial of panitumumab. *British Journal of Cancer*, Vol.9,

Biomarkers Predicting Clinical Outcome of Epidermal Growth Factor Receptor – Targeted Therapy in Metastatic Colorectal Cancer. *Journal of the National Cancer* 

phase III study of panitumumab (pmab) with FOLFOX4 compared with FOLFOX4 alone as first-line treatment (tx) for metastatic colorectal cancer (mCRC): Results by

2010), pp. 47-50, ISSN: 1322-7696

(February 2003), pp. 3-11, ISSN: 0093-7754

Online 2060-0399

60 ISSN: 0167-6997

ISSN: 1548-5315

ISSN Online: 1938-0674

pp. 1017-1022, ISSN: 0344-5704

Eastern Cooperative Oncology Group (ECOG) performance status (PS). *2011 ASCO Annual Meeting*. Abstract No: 3567. Citation: *Journal of Clinical Oncology,* Vol.29, Suppl. abstr. 3567. 2011, ISSN: 02773732


**21** 

*Bulgaria* 

**Resection for Colorectal Liver Metastases** 

*Department of Surgery, Naval Hospital of Varna, and Division of Surgery,* 

Colorectal cancer is the third most frequent cancer in the Western world. About half of the patients develop synchronous or metachronous metastases. The liver is the most common site of such metastases and thus hepatic metastatic disease is a significant socio-medical problem. If it is not treated, the median patient survival is only some months. Surgical resection is the treatment of choice for patients with isolated colorectal liver metastases when feasible. For patients with four or fewer isolated hepatic lesions, five year relapse-free survival rates range from 24 to 58 percent and ten year survival rates vary between 17 and 33 percent. There is a convincing socio-epidemiological evidence of the dramatic unfavourable influence on population wealth of untimely diagnosis and inadequate treatment of the patients with advanced and metastatic colorectal cancer worldwide (Hata et

The purposes of the present paper are to define the variety of liver resections as an important component of the modern treatment of colorectal liver metastases, to describe their operative techniques and postoperative results, to illustrate some peculiar resection patterns from our own patients' contingent and, based on our own experience with the complex preoperative diagnostic algorithm and the individualized indications and contraindications for surgery and multimodal therapy, to outline the advantages of different types of hepatic resections in properly selected cases with colorectal liver metastases as

Patients' contingent included a total of 158 patients who have undergone liver resections for colorectal liver metastases in the Department of Surgery, Naval Hospital of Varna and in the Division of Surgery, Marko Markov Interregional Dispensary and Hospital of Oncological Diseases of Varna, Bulgaria, during a 10-year period (January 1, 2000 - December 31, 2010). Results concerning 108 patients dynamically followed-up for at least one year after the operation were illustrated in this comprehensive retrospective study. Demographic characteristics, preoperative clinical, laboratory and functional diagnosis of the patients,

al., 2010; Kostov & Kobakov, 2006a; Stillwell et al., 2011; Tsoulfas et al., 2011).

manifested by improved patient's quality of life and survival.

**1. Introduction** 

**2. Purposes of the study** 

**3. Material** 

Daniel Kostov and Georgi Kobakov

*Marko Markov Interregional Dispensary and Hospital of Oncological Diseases of Varna, Varna,* 


### **Resection for Colorectal Liver Metastases**

#### Daniel Kostov and Georgi Kobakov

*Department of Surgery, Naval Hospital of Varna, and Division of Surgery, Marko Markov Interregional Dispensary and Hospital of Oncological Diseases of Varna, Varna, Bulgaria* 

#### **1. Introduction**

408 Colorectal Cancer – From Prevention to Patient Care

van Cutsem, E.; Nordlinger, B. & Cervantes, A. (2010). Advanced colorectal cancer: ESMO

van der Velden, LFJ.; Francke, AL.; Hingstman, L & Willems, DL. Dying from cancer or

van Heeckeren, WJ.; Ortiz, J.; Cooney, MM. & Remick, SC. (2007). Hypertension,

Widakowich, C.; Castro, G.; Azambuja, E.; Dinh, P. & Awada, A. (2007). Review: Side Effects

Willett, CG.; Duda, DG.; Czito, BG.; Bendell JC.; Clark JW., et al. (2007). Targeted Therapy in

Winkeljohn DL. (2008). Review of Panitumumab: A Targeted Therapy. *Clinical Journal of Oncology Nursing,* Vol.12, Nr.1, (February 2008), pp. 30-32, ISSN Online 1538-067X World Health Organization – International Agency for Research on Cancer. (2008).

Xu, KP,; Li, Y,; Ljubimov, AV. & Yu, FSX. (2009). High Glucose Suppresses Epidermal

Yang, XD.; Jia, XC.; Corvalan, JFR.; Wang, P. & Davis, CG. (2001.) Development of ABX-

Yoneda, KY.; Shelton, DK.; Beckett, LA. & Gandara, DR. (2007). Independent review of

No.12, (December 2007), pp. 1443-1455, ISSN Online 1549-490X

*Oncology*, Vol.25, No. 21 (20 July 2007), pp. 2993-2995, ISSN: 02773732 van Kleffens, T.; van Baarsen, B. & van Leeuwen, E. (2004). The medical practice of patient

5. (May 2010), pp. v93–v97, ISSN 0923-7534

9536

passim. ISSN: 0890-9091

http://globocan.iarc.fr/

0737-9587

Online: 1556-1380

2009) ISSN Online 1939-327X.

http://www.biomedcentral.com/1472-684X/8/4

Clinical Practice Guidelines for treatment. *Annals of Oncology,* Vol. 21, Supplement

other chronic diseases in the Netherlands: ten-year trends derived from death certificate data. In: *BioMed Central Palliative Care,* 4.02.2009, Available from:

Proteinuria, and Antagonism of Vascular Endothelial Growth Factor Signaling: Clinical Toxicity, Therapeutic Target, or Novel Biomarker? *Journal of Clinical* 

autonomy and cancer treatment refusals: a patients' and physicians' perspective. *Social Science and Medicin*e, Vol.58, Nr.11. (June 2004), pp. 2325-2336, ISSN: 0277-

of Approved Molecular Targeted Therapies in Solid Cancer. *The Oncologist*, Vol.12,

Rectal Cancer. *Oncology (Williston Park),* Vol.21., Nr.9, (August 2007), pp. 1056-

*GLOBOCAN 2008 (Cancer Incidence and Mortality Worldwide in 2008)*. Available from

Growth Factor Receptor/Phosphatidylinositol 3-Kinase/Akt Signaling Pathway and Attenuates Corneal Epithelial Wound Healing. *Diabetes*, Vol.58, No5, (May

EGF, a fully human anti-EGF receptor monoclonal antibody, for cancer therapy. *Critical Reviews in Oncology/Hematology,* Vol.38, Nr.1. (April 2001), pp. 17–23, ISSN:

interstitial lung disease associated with death in TRIBUTE (paclitaxel and carboplatin with or without concurrent erlotinib) in advanced non-small cell lung cancer. *Journal of Thoracic Oncology*, Vol.2, Nr. 6, (June 2007), pp. 537-543, ISSN Colorectal cancer is the third most frequent cancer in the Western world. About half of the patients develop synchronous or metachronous metastases. The liver is the most common site of such metastases and thus hepatic metastatic disease is a significant socio-medical problem. If it is not treated, the median patient survival is only some months. Surgical resection is the treatment of choice for patients with isolated colorectal liver metastases when feasible. For patients with four or fewer isolated hepatic lesions, five year relapse-free survival rates range from 24 to 58 percent and ten year survival rates vary between 17 and 33 percent. There is a convincing socio-epidemiological evidence of the dramatic unfavourable influence on population wealth of untimely diagnosis and inadequate treatment of the patients with advanced and metastatic colorectal cancer worldwide (Hata et al., 2010; Kostov & Kobakov, 2006a; Stillwell et al., 2011; Tsoulfas et al., 2011).

#### **2. Purposes of the study**

The purposes of the present paper are to define the variety of liver resections as an important component of the modern treatment of colorectal liver metastases, to describe their operative techniques and postoperative results, to illustrate some peculiar resection patterns from our own patients' contingent and, based on our own experience with the complex preoperative diagnostic algorithm and the individualized indications and contraindications for surgery and multimodal therapy, to outline the advantages of different types of hepatic resections in properly selected cases with colorectal liver metastases as manifested by improved patient's quality of life and survival.

#### **3. Material**

Patients' contingent included a total of 158 patients who have undergone liver resections for colorectal liver metastases in the Department of Surgery, Naval Hospital of Varna and in the Division of Surgery, Marko Markov Interregional Dispensary and Hospital of Oncological Diseases of Varna, Bulgaria, during a 10-year period (January 1, 2000 - December 31, 2010). Results concerning 108 patients dynamically followed-up for at least one year after the operation were illustrated in this comprehensive retrospective study. Demographic characteristics, preoperative clinical, laboratory and functional diagnosis of the patients,

Resection for Colorectal Liver Metastases 411

Volume of resection Type of resection n=108

Sg1 3 Sg2 2 Sg3 3 Sg4 5 Sg5 2 Sg6 3 Sg7 2 Sg8 2

Sg6,7 4 Sg5,8 3 Sg2,3 4 Sg5,6 3 Sg4b,5 2 Sg1,4 2 Sg7,8 2 Sg4a,8 2 Sg3,4b 1

Sg4 + parts of Sg 1,2,3,5 1

Sg4 + part of Sg2,3 + metastasectomy of Sg8 1

Sg4,5,8 3 Sg1,4,5,8 2 Sg1,4b,5,6 1 Sg5,6,7 1 Sg3,5,6,7 1 Sg3,4b,5 2

Sg4b,6,7 1 Sg6,7,8 1

Sg3+ parts of Sg4,6,8 1

Left hemihepatectomy (Sg2,3,4±1) 13 Right hemihepatectomy (Sg5,6,7,8±1) 32 Right trisectionectomy (Sg1,4,5,6,7,8) 2 Left trisectionectomy (Sg1,2,3,4,5,8) 1

Monosegmentectomy

Bisegmentectomy n=24 (22%)

Multisegmentectomy

Table 2. Type and volume of liver resections

n=62 (57%)

n=22 (20%)

types of surgical interventions and conservative therapy as well as metastatic tumour localization, volume, number and staging were systematized. Only some of our data could be presented in the present paper.

Number and mean age of male and female patients can be seen on Table 1.


Table 1. Patients' distribution according to gender and mean age

#### **4. Methods**

The algorithm for contemporary diagnostic evaluation comprised total colonoscopy, conventional chest radiography, conventional blood tests, serum levels of some tumour markers such as carcinoembryonic antigen (CEA), carbohydrate antigen CA 19.9 and carbohydrate antigen CA 242, abdominal preoperative and intraoperative ultrasonography (for estimation of the type and volume of liver resection), intraoperative cholangiography (for pre- and postoperative bile drainage evaluation in remnant hepatic parenchyma), methyene-blue injection through the portal vein (for assessment of afferent and efferent blood flow in remnant hepatic parenchyma), contrast-enhanced and spiral computed abdominal tomography, MRI in case of contradictory computer tomographic data and histopathology of enlarged hilar lymph nodes. The volume of the liver resection was determined not only by the number, size and localization of the metastases but also by the degree of compensatory hypertrophy of the intact hepatic volume.

The presentation of all the types of surgical interventions included the following:

i. types of surgical access, liver mobilization, hilar dissection, and hepatic-vein control,


Patients' distribution according to the volume of liver resection is demonstrated on Table 2.

During the last 8 years, a total of 14 patients underwent repeated liver resections. A third resection was done in three of these patients, and a fourth resection was done in two of these patients.

Table 3 indicates the consecutive number and volume of primary and repeated liver resections.

Some essential parameters of colorectal liver metastases in these repeated resections of different consecutive number are summarized on Table 4.

Additionally, multimodal treatment of all the patients with colorectal liver metastases included a variety of chemotherapeutic protocols as aneoadjuvant and/or adjuvant chemotherapy along with radiofrequent ablation, portal vein embolization, and two-stage resection of bilobar colorectal liver metastases (Kostov & Kobakov, 2006). The effect of neoadjuvant chemotherapy was assessed according to the Response Evaluation Criteria in Solid Tumours (Eisenhauer et al., 2009).

types of surgical interventions and conservative therapy as well as metastatic tumour localization, volume, number and staging were systematized. Only some of our data could

Males 68 63 58 36-81 Females 40 37 54 32-79 Total 108 100 59 32-81

The algorithm for contemporary diagnostic evaluation comprised total colonoscopy, conventional chest radiography, conventional blood tests, serum levels of some tumour markers such as carcinoembryonic antigen (CEA), carbohydrate antigen CA 19.9 and carbohydrate antigen CA 242, abdominal preoperative and intraoperative ultrasonography (for estimation of the type and volume of liver resection), intraoperative cholangiography (for pre- and postoperative bile drainage evaluation in remnant hepatic parenchyma), methyene-blue injection through the portal vein (for assessment of afferent and efferent blood flow in remnant hepatic parenchyma), contrast-enhanced and spiral computed abdominal tomography, MRI in case of contradictory computer tomographic data and histopathology of enlarged hilar lymph nodes. The volume of the liver resection was determined not only by the number, size and localization of the metastases but also by the

Patients Mean age n % years range

Number and mean age of male and female patients can be seen on Table 1.

Table 1. Patients' distribution according to gender and mean age

degree of compensatory hypertrophy of the intact hepatic volume.

different consecutive number are summarized on Table 4.

Solid Tumours (Eisenhauer et al., 2009).

The presentation of all the types of surgical interventions included the following:

i. types of surgical access, liver mobilization, hilar dissection, and hepatic-vein control,

Patients' distribution according to the volume of liver resection is demonstrated on Table 2. During the last 8 years, a total of 14 patients underwent repeated liver resections. A third resection was done in three of these patients, and a fourth resection was done in two of these

Table 3 indicates the consecutive number and volume of primary and repeated liver

Some essential parameters of colorectal liver metastases in these repeated resections of

Additionally, multimodal treatment of all the patients with colorectal liver metastases included a variety of chemotherapeutic protocols as aneoadjuvant and/or adjuvant chemotherapy along with radiofrequent ablation, portal vein embolization, and two-stage resection of bilobar colorectal liver metastases (Kostov & Kobakov, 2006). The effect of neoadjuvant chemotherapy was assessed according to the Response Evaluation Criteria in

be presented in the present paper.

Gender

**4. Methods** 

ii. operative approaches, and iii. operative volumes.

patients.

resections.


Table 2. Type and volume of liver resections

Resection for Colorectal Liver Metastases 413

One- and three-year survival data were retrospectively recorded up to December, 2010. Memorial Sloan-Kettering Cancer Center Clinical Risk Score (MSKCC-CRS) was used to

Kaplan-Meier estimates outlined differences with Kaplan-Meier curves. Comparisons of sex and age between segmentectomy and major hepatectomy patients applied chi-square and *t*test. The *t*-test compared mean blood loss, diameter of colorectal liver metastases, duration of surgery, length of hospital stay, and resection margins. The postoperative complications were compared by means of Fisher's exact test. Patients' homogeneity was comparatively

The following types of surgical access for liver mobilization, hilar dissection, and hepatic-

Upper medial laparotomy with transversal enlargement to the right until 9th intercostal space along with Makuushi incision is most commonly performed to access right or left hemiliver while Mercedes-Benz incision is suitable to access both left and right hemilivers. Complete liver mobilization passes through five stages: i) interruption of *lig. teres hepatis* between two ligatures, ii) cutting of *lig. falciforme hepatis* up to the subdiaphragmatic part of *vena cava inferior*, iii) search for an accessory left hepatic artery as a branch of *a. gastrica sinistra* when cutting *lig. hepatogastricum*, iv) cutting to the left of both *lig. triangularе sinistrum* and *lig. coronarium hepatis* to left hepatic vein trunk and v) cutting to the right of

Hilar dissection aims at dividing the vessels designed for the left and right hemiliver that enables the application of hemi-Pringle maneuver. Right hepatic vein extrahepatic part can be reached by interruption of Makuushi ligament. Right hepatic vein is lifted on rubber

both *lig. triangularе dextrum* and *lig. coronarium hepatis* to right hepatic vein trunk.

Fig. 1. Right hepatic vein mobilization and short retrohepatic veins

evaluate the postoperative prognosis of the patients (Arru et al., 2008).

assessed by means of the log rank and Wilcoxon tests.

**5. Operative techniques of liver resections** 

vein control can be used (Kostov & Kobakov, 2010):

**5.1 Types of surgical access** 

holder (Fig. 1).


Table 3. Consecutive number and volume of resections


Table 4. Characteristics of colorectal liver metastases in repeated liver resections

One Sg Two Sg Three Sg Trisection-

First 14 2 6 2 2 2 - Second 14 - - - - - 14 Third 3 - - - - - 3 Fourth 2 - - - - - 2

CEA > 200 ng/mL 9 9 3 2 CEA ≤ 200 ng/mL 5 5 - synchronous 5 - - metachronous 9 14 3 2 after < 12 months 6 - - after ≥ 12 months 8 - - -

One 2 8 2 two-three 7 6 1 - ≥ three 5 - - 2 unilobar 8 12 3 2 bilobar 6 2 - diameter < 20 mm 2 7 2 diameter of 20-50 mm 8 5 1 1 diameter ≥ 50 mm 4 1 - 1

*duodenale* - - - 2 positive margins - - - 2 negative margins 14 14 3 -

0-2 factors 8 8 1 - 3-5 factors 6 6 2 2

Table 4. Characteristics of colorectal liver metastases in repeated liver resections

Volume of resection

ectomy

resection Third resection Fourth

Hemihepatectomy

Wedge resection

resection

Number of resection

total number

nodes in *lig. hepato-*

MSKCC-CRS

Number of patients

Table 3. Consecutive number and volume of resections

Parameter First resection Second

One- and three-year survival data were retrospectively recorded up to December, 2010. Memorial Sloan-Kettering Cancer Center Clinical Risk Score (MSKCC-CRS) was used to evaluate the postoperative prognosis of the patients (Arru et al., 2008).

Kaplan-Meier estimates outlined differences with Kaplan-Meier curves. Comparisons of sex and age between segmentectomy and major hepatectomy patients applied chi-square and *t*test. The *t*-test compared mean blood loss, diameter of colorectal liver metastases, duration of surgery, length of hospital stay, and resection margins. The postoperative complications were compared by means of Fisher's exact test. Patients' homogeneity was comparatively assessed by means of the log rank and Wilcoxon tests.

#### **5. Operative techniques of liver resections**

#### **5.1 Types of surgical access**

The following types of surgical access for liver mobilization, hilar dissection, and hepaticvein control can be used (Kostov & Kobakov, 2010):

Upper medial laparotomy with transversal enlargement to the right until 9th intercostal space along with Makuushi incision is most commonly performed to access right or left hemiliver while Mercedes-Benz incision is suitable to access both left and right hemilivers.

Complete liver mobilization passes through five stages: i) interruption of *lig. teres hepatis* between two ligatures, ii) cutting of *lig. falciforme hepatis* up to the subdiaphragmatic part of *vena cava inferior*, iii) search for an accessory left hepatic artery as a branch of *a. gastrica sinistra* when cutting *lig. hepatogastricum*, iv) cutting to the left of both *lig. triangularе sinistrum* and *lig. coronarium hepatis* to left hepatic vein trunk and v) cutting to the right of both *lig. triangularе dextrum* and *lig. coronarium hepatis* to right hepatic vein trunk.

Hilar dissection aims at dividing the vessels designed for the left and right hemiliver that enables the application of hemi-Pringle maneuver. Right hepatic vein extrahepatic part can be reached by interruption of Makuushi ligament. Right hepatic vein is lifted on rubber holder (Fig. 1).

Fig. 1. Right hepatic vein mobilization and short retrohepatic veins

Resection for Colorectal Liver Metastases 415

Clamping the three hepatic veins enables a complete vascular exclusion with blood flow preservation through *vena cava inferior*. With single 'hanging'-maneuver, a rubber tape passes cranially between the right and middle hepatic veins but caudally - between hilar

This method is applied in right/left hemihepatectomy or right segmentectomy. With double 'hanging'-maneuver, a second rubber tape is additionally used which passes cranially between the middle and left hepatic veins but caudally - between hilar vessels for the right and left hemiliver (Fig. 4). This method is applied in mesohepatectomy or proximal segmentectomy. With complete vascular exclusion and blood flow interruption through *vena cava inferior* the latter is clamped over the three hepatic veins and over the inflow of renal veins. For that purpose, *vena cava inferior* is mobilized at two sites - below the diaphragm and over the inflow of renal veins ( Fig. 5). Right suprarenal vein is obligatorily

vessels for the right and left hemiliver (Fig. 3).

Fig. 4. Double 'hanging'-maneuver

interrupted.

Usually, both left and middle hepatic veins present with a common trunk as their bifurcation is intraparenchymally located (Fig. 2).

Fig. 2. Extrahepatic mobilization of three hepatic veins enables a complete vascular exclusion of the liver and preserves blood flow through *vena cava inferior* 

Fig. 3. Single 'hаnging'-maneuver

Usually, both left and middle hepatic veins present with a common trunk as their

Fig. 2. Extrahepatic mobilization of three hepatic veins enables a complete vascular

exclusion of the liver and preserves blood flow through *vena cava inferior* 

Fig. 3. Single 'hаnging'-maneuver

bifurcation is intraparenchymally located (Fig. 2).

Clamping the three hepatic veins enables a complete vascular exclusion with blood flow preservation through *vena cava inferior*. With single 'hanging'-maneuver, a rubber tape passes cranially between the right and middle hepatic veins but caudally - between hilar vessels for the right and left hemiliver (Fig. 3).

This method is applied in right/left hemihepatectomy or right segmentectomy. With double 'hanging'-maneuver, a second rubber tape is additionally used which passes cranially between the middle and left hepatic veins but caudally - between hilar vessels for the right and left hemiliver (Fig. 4). This method is applied in mesohepatectomy or proximal segmentectomy. With complete vascular exclusion and blood flow interruption through *vena cava inferior* the latter is clamped over the three hepatic veins and over the inflow of renal veins. For that purpose, *vena cava inferior* is mobilized at two sites - below the diaphragm and over the inflow of renal veins ( Fig. 5). Right suprarenal vein is obligatorily interrupted.

Fig. 4. Double 'hanging'-maneuver

Resection for Colorectal Liver Metastases 417

The extrahepatic approach requires interruption of the afferent and efferent blood supply to the hepatic part outside the liver which is subject to removal. Among monosegmentectomies, only Sg1 devascularization can be entirely done through such an approach. With isolated segmentectomy 1, the line of parenchymal transsection passing

Fig. 6. Line of parenchymal transsection (2) when removing Sg 1. Line of dividing the liver into left and right hemiliver (1); left (A), middle (B), and right veins (Liau et al., 2004)

Fig. 7. Removed Sg1 - view from the left. LHV - left hepatic vein; MHV - middle hepatic

vein; RHV- right hepatic vein

behind the three hepatic veins is of interest (Fig. 6 through Fig. 8).

Fig. 5. Preparation for complete vascular exclusion and blood flow interruption through vena cava inferior. Cranial rubber tape passes circularly over the three hepatic veins but the caudal one does over renal veins. Right suprarenal vein is interrupted

#### **5.2 Operative approaches**

The following operative approaches can be made use of (Kostov & Kobakov, 2010):


#### **5.3 Operative volumes**

According to the localization and expansion of the pathologic process, one of the following operative volumes should be selected by liver surgeons (Kostov & Kobakov, 2010):

#### **5.3.1 Segmentectomies**

Stages of the following monosegmentectomies - segmentectomy 1 ( Sg 1), Sg 2, Sg 3, Sg 4, Sg 5, Sg 6, Sg 7, Sg 8, and wedge resection:

Segmentectomy 1 passes through five stages: i) devascularization of *proc. caudatus*, ii) devascularization of Spiegel's lobe, iii) interruption of short retrohepatic veins which enter directly *vena cava inferior*, iv) mobilization of right, middle and left hepatic veins and v) parenchymal transection at the borderline between Sg1 and Sg4. Resection to the right ends at the borderline to Sg7.

Fig. 5. Preparation for complete vascular exclusion and blood flow interruption through vena cava inferior. Cranial rubber tape passes circularly over the three hepatic veins but the

The following operative approaches can be made use of (Kostov & Kobakov, 2010):

i. extrahepatic approach to the hepatic inflow pedicles for ligation of a portal triad to the Sg 1 and 4, right anterior section, right posterior section, left hemiliver and right

ii. intrahepatic anterior approach to the hepatic inflow pedicles for ligation of a portal

iii. intrahepatic posterior approach to the hepatic pedicles by using Glissonian sheaths, and iv. combined extrahepatic and intrahepatic approaches for ligation of a portal triad were used in some bisegmentectomies, right trisectionectomies, and left trisectionectomies.

According to the localization and expansion of the pathologic process, one of the following

Stages of the following monosegmentectomies - segmentectomy 1 ( Sg 1), Sg 2, Sg 3, Sg 4, Sg

Segmentectomy 1 passes through five stages: i) devascularization of *proc. caudatus*, ii) devascularization of Spiegel's lobe, iii) interruption of short retrohepatic veins which enter directly *vena cava inferior*, iv) mobilization of right, middle and left hepatic veins and v) parenchymal transection at the borderline between Sg1 and Sg4. Resection to the right ends

operative volumes should be selected by liver surgeons (Kostov & Kobakov, 2010):

caudal one does over renal veins. Right suprarenal vein is interrupted

triad to an individual Sg (2, 3, 5, 6, 7 and 8),

**5.2 Operative approaches** 

hemiliver,

**5.3 Operative volumes** 

**5.3.1 Segmentectomies** 

at the borderline to Sg7.

5, Sg 6, Sg 7, Sg 8, and wedge resection:

The extrahepatic approach requires interruption of the afferent and efferent blood supply to the hepatic part outside the liver which is subject to removal. Among monosegmentectomies, only Sg1 devascularization can be entirely done through such an approach. With isolated segmentectomy 1, the line of parenchymal transsection passing behind the three hepatic veins is of interest (Fig. 6 through Fig. 8).

Fig. 6. Line of parenchymal transsection (2) when removing Sg 1. Line of dividing the liver into left and right hemiliver (1); left (A), middle (B), and right veins (Liau et al., 2004)

Fig. 7. Removed Sg1 - view from the left. LHV - left hepatic vein; MHV - middle hepatic vein; RHV- right hepatic vein

Resection for Colorectal Liver Metastases 419

portal veins through extrahepatic access, iii) interruption of descendent portal veins during parenchymal transection along *lig. falciforme hepatis* (Fig. 11), iv) opening and interruption of bile ducts for Sg4 in Rex recessus and v) parenchymal transection along Rex-Cantlie line as

Segmentectomy 5 passes through three stages: i) definition of resection borderlines of Sg5. Clamping the vessels for Sg5,8 causes their ischaemic demarcation and visualizes the left and right resection borderlines. The complete Sg5 volume is visualized after injection into segmental portal vein of 5 mL of methylene blue under echographic control, ii) parenchymal

middle hepatic vein can be either interrupted, or preserved.

Fig. 10. Anterior intrahepatic access to left hepatic vein

Fig. 11. Ischaemic demarcation of Sg4

Fig. 8. Removed Sg1 - view from the right.

Both segmentectomy 2 and segmentectomy 3 pass through three stages each: i) definition of borderlines of Sg2 and Sg3 ii) parenchymal transection with intraparenchymal interruption of portal triad vessels for Sg2 and Sg3 (Fig. 9) and iii) interruption of branches of left hepatic vein for Sg2 and Sg3 through anterior intrahepatic access (Fig. 10).

Fig. 9. Sites for interruption of portal triad vessels for Sg2 and Sg3

Segmentectomy 4 passes through five stages: i) extrahepatic interruption of the artery for *lobus quadratus* which, normally, is left hepatic artery branch, ii) ligation of some ascendent

Both segmentectomy 2 and segmentectomy 3 pass through three stages each: i) definition of borderlines of Sg2 and Sg3 ii) parenchymal transection with intraparenchymal interruption of portal triad vessels for Sg2 and Sg3 (Fig. 9) and iii) interruption of branches of left hepatic vein

Segmentectomy 4 passes through five stages: i) extrahepatic interruption of the artery for *lobus quadratus* which, normally, is left hepatic artery branch, ii) ligation of some ascendent

Fig. 8. Removed Sg1 - view from the right.

for Sg2 and Sg3 through anterior intrahepatic access (Fig. 10).

Fig. 9. Sites for interruption of portal triad vessels for Sg2 and Sg3

portal veins through extrahepatic access, iii) interruption of descendent portal veins during parenchymal transection along *lig. falciforme hepatis* (Fig. 11), iv) opening and interruption of bile ducts for Sg4 in Rex recessus and v) parenchymal transection along Rex-Cantlie line as middle hepatic vein can be either interrupted, or preserved.

Fig. 10. Anterior intrahepatic access to left hepatic vein

Fig. 11. Ischaemic demarcation of Sg4

Segmentectomy 5 passes through three stages: i) definition of resection borderlines of Sg5. Clamping the vessels for Sg5,8 causes their ischaemic demarcation and visualizes the left and right resection borderlines. The complete Sg5 volume is visualized after injection into segmental portal vein of 5 mL of methylene blue under echographic control, ii) parenchymal

Resection for Colorectal Liver Metastases 421

vessels for Sg5 and middle hepatic vein preservation as portal triad vessels for Sg8 is caudally identified and interrupted. Parenchymal transection continues cranially to the borderline with middle hepatic vein where the vein for Sg8 is interrupted. Sg8 devascularization visualizes its borderlines with Sg7 and Sg5 and iii) parenchymal

Direct access consists in immediate intervention on Sg8. Parenchymal transection passes through three stages: i) definition of resection borderlines of Sg8 visualized after injection into PV for Sg8 of 5 mL of methylene blue under echographic control (Fig. 13). Clamping the artery and vein for Sg5,8 causes ischaemic demarcation of left and right resection borderlines. Right hepatic vein is clamped, if necessary, ii) parenchymal transection at the borderline between Sg4a and Sg8 as, caudally, the vein draining blood from Sg8 into middle hepatic vein and segmental portal triad vessels are consecutively interrupted. Sg8 demarcation visualizes its borderlines with Sg5 and Sg7 and iii) parenchymal transection

Fig. 13. Borderlines of Sg8 visualized after injection into PV for Sg8 of methylene blue under

Stages of the following bisegmentecomies - bisegmentectomy 2,3; 6,7; 5,8; 3,4b; 1,4; 4b,5; 5,6;

Bisegmentectomy 2,3 passes through three stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare sinistrum* and *lig. coronarium hepatis sinistrum*, ii) parenchymal transection along the left edge of *lig. falciforme hepatis* and caudal interruption of portal triad vessels for Sg2,3 and iii) left hepatic vein interruption either through extrahepatic access, or through anterior intrahepatic access at the end of parenchymal

Bisegmentectomy 6,7 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Extrahepatic

echographic control

7,8, and 4a,8:

transection.

**5.3.2 Bisegmentectomies** 

transection along these bordelines in order to preserve right hepatic vein.

along these borderlines and obligatory preservation of right hepatic vein. Wedge resection consists in removal of some part of a given liver segment only.

transection along Rex-Cantlie line with intraparenchymal interruption of the vessels for Sg5 and middle hepatic vein preservation. Sg5 devascularization induces ischaemic demarcation of its borderlines and iii) parenchymal transection at the borderline with Sg8 and Sg6. Resection line should pass over Ganz furrow in which the vessels for Sg6 are located.

Segmentectomy 6 passes through three stages: i) definition of resection borderlines of Sg6. Clamping the artery and portal vein for Sg6,7 causes their ischaemic demarcation and visualizes the borderline to Sg5,8. The resection borderlines are visualized after injection into portal vein for Sg6,7 or into segmental portal vein for Sg6 of 5 mL of methylene blue under echographic control (Fig. 12). (ii) parenchymal transsection at the borderline between Sg5 and Sg6 with consecutive interruption of the vein and portal triad for Sg6. Sg6 ischaemia allows visualization of its borderline to Sg7 and iii) parenchymal transection along this bordeline.

Fig. 12. Sg6 resection borderlines after metthylene-blue injection into segmental PV under echographic control

Segmentectomy 7 passes through three stages: i) definition of resection borderlines of Sg7. Clamping the artery and portal vein for Sg6,7 causes their ischaemic demarcation and visualizes the borderline to Sg5,8. The resection borderlines are visualized after injection into portal vein for Sg6,7 or into segmental portal vein for Sg7 of 5 mL of methylene blue under echographic control. Right hepatic vein mobilization prevents bleeding during resection, (ii) parenchymal transection at the borderline between Sg7 and Sg8 with consecutive interruption of right hepatic vein right branch and portal triad vessels for Sg7. Sg7 ischaemic demarcation visualizes its borderline to Sg6 and iii) parenchymal transection along this borderline.

Segmentectomy 8 can be performed through indirect and direct access.

Indirect access - segmentectomy 8 passes through three stages: i) definition of resection borderlines of Sg8. Clamping the artery and vein for Sg5,8 causes their ischaemic demarcation. Right hepatic vein mobilization prevents bleeding during resection, (ii) parenchymal transection along Rex-Cantlie line with intraparenchymal interruption of the

transection along Rex-Cantlie line with intraparenchymal interruption of the vessels for Sg5 and middle hepatic vein preservation. Sg5 devascularization induces ischaemic demarcation of its borderlines and iii) parenchymal transection at the borderline with Sg8 and Sg6.

Segmentectomy 6 passes through three stages: i) definition of resection borderlines of Sg6. Clamping the artery and portal vein for Sg6,7 causes their ischaemic demarcation and visualizes the borderline to Sg5,8. The resection borderlines are visualized after injection into portal vein for Sg6,7 or into segmental portal vein for Sg6 of 5 mL of methylene blue under echographic control (Fig. 12). (ii) parenchymal transsection at the borderline between Sg5 and Sg6 with consecutive interruption of the vein and portal triad for Sg6. Sg6 ischaemia allows visualization of its borderline to Sg7 and iii) parenchymal transection along this bordeline.

Fig. 12. Sg6 resection borderlines after metthylene-blue injection into segmental PV under

Segmentectomy 8 can be performed through indirect and direct access.

Segmentectomy 7 passes through three stages: i) definition of resection borderlines of Sg7. Clamping the artery and portal vein for Sg6,7 causes their ischaemic demarcation and visualizes the borderline to Sg5,8. The resection borderlines are visualized after injection into portal vein for Sg6,7 or into segmental portal vein for Sg7 of 5 mL of methylene blue under echographic control. Right hepatic vein mobilization prevents bleeding during resection, (ii) parenchymal transection at the borderline between Sg7 and Sg8 with consecutive interruption of right hepatic vein right branch and portal triad vessels for Sg7. Sg7 ischaemic demarcation visualizes its borderline to Sg6 and iii) parenchymal transection

Indirect access - segmentectomy 8 passes through three stages: i) definition of resection borderlines of Sg8. Clamping the artery and vein for Sg5,8 causes their ischaemic demarcation. Right hepatic vein mobilization prevents bleeding during resection, (ii) parenchymal transection along Rex-Cantlie line with intraparenchymal interruption of the

echographic control

along this borderline.

Resection line should pass over Ganz furrow in which the vessels for Sg6 are located.

vessels for Sg5 and middle hepatic vein preservation as portal triad vessels for Sg8 is caudally identified and interrupted. Parenchymal transection continues cranially to the borderline with middle hepatic vein where the vein for Sg8 is interrupted. Sg8 devascularization visualizes its borderlines with Sg7 and Sg5 and iii) parenchymal transection along these bordelines in order to preserve right hepatic vein.

Direct access consists in immediate intervention on Sg8. Parenchymal transection passes through three stages: i) definition of resection borderlines of Sg8 visualized after injection into PV for Sg8 of 5 mL of methylene blue under echographic control (Fig. 13). Clamping the artery and vein for Sg5,8 causes ischaemic demarcation of left and right resection borderlines. Right hepatic vein is clamped, if necessary, ii) parenchymal transection at the borderline between Sg4a and Sg8 as, caudally, the vein draining blood from Sg8 into middle hepatic vein and segmental portal triad vessels are consecutively interrupted. Sg8 demarcation visualizes its borderlines with Sg5 and Sg7 and iii) parenchymal transection along these borderlines and obligatory preservation of right hepatic vein.

Wedge resection consists in removal of some part of a given liver segment only.

Fig. 13. Borderlines of Sg8 visualized after injection into PV for Sg8 of methylene blue under echographic control

#### **5.3.2 Bisegmentectomies**

Stages of the following bisegmentecomies - bisegmentectomy 2,3; 6,7; 5,8; 3,4b; 1,4; 4b,5; 5,6; 7,8, and 4a,8:

Bisegmentectomy 2,3 passes through three stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare sinistrum* and *lig. coronarium hepatis sinistrum*, ii) parenchymal transection along the left edge of *lig. falciforme hepatis* and caudal interruption of portal triad vessels for Sg2,3 and iii) left hepatic vein interruption either through extrahepatic access, or through anterior intrahepatic access at the end of parenchymal transection.

Bisegmentectomy 6,7 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Extrahepatic

Resection for Colorectal Liver Metastases 423

Fig. 15. Mobilization and preparation for interruption of portal vein for Sg5,8. RPV - right

Fig. 16. Ischaemic demarcation of Sg5,8 after their devascularization

Bisegmentectomy 1,4 passes through five stages: i) mobilization of left and right hemiliver and definition of Sg4 resection borderlines. To the left, parenchymal transection passes along *lig. falciforme hepatis* but to the right it follows middle hepatic vein course as defined by means of intraoperative echography, ii) Sg4 devascularization by consecutive interruption of the artery and ascendant portal veins for Sg4 through extrahepatic access, iii) parenchymal transection along *lig. falciforme hepatis* and, caudally, interruption of descendent portal veins for Sg4. After interruption of portal veins for *lobus quadratus*, Rex recessus is reached where the bile duct for Sg4 is identified and interrupted, iv) Sg1

portal vein

portion of right hepatic vein is liberated and lifted on a rubber tape, ii) definition of resection borderline after interruption of the artery and portal vein for Sg6,7 or by injection of 5 mL of methylene blue into portal vein for Sg6,7, iii) parenchymal transection at the borderline between Sg6,7 and Sg5,8. Caudally, both the vein for Sg6 draining blood into right hepatic vein anterior branch and portal triad vessels for Sg6,7 under it are interrupted and iv) at the end of parenchymal transection, right hepatic vein posterior branch draining blood from Sg7 is interrupted as its anterior branch is preserved.

Bisegmentectomy 5,8 passes through four stages: i) mobilization of right hemiliver through interruption of lig. triangulare dextrum and lig. coronarium hepatis dextrum. Extrahepatic portion of right hepatic vein is liberated and lifted on a rubber tape, ii) definition of resection borderlines of Sg5,8. Extrahepatic interruption of the artery (Fig. 14) and portal vein (Fig. 15) for Sg5,8 causes their ischaemic demarcation (Fig. 16). Sg 5,8 visualization after injection of 5 mL of methylene blue into portal vein for Sg5,8, iii) parenchymal transection at the borderline to lobus quadratus through consecutive interruption of the vein for Sg5 draining blood into middle hepatic vein, of the portal triad vessels for Sg5,8 and the vein draining blood from Sg8 into middle hepatic vein. Sg5,8 devascularization results in ischaemic demarcation line at the borderline to Sg6,7 and iv) parenchymal transection along this borderline includes interruprion of the vein draining blood from Sg5 into right hepatic vein as both middle and right hepatic veins are obligatorily preserved.

Bisegmentectomy 3,4b passes through three stages: i) definition of resection borderlines by means of intraoperative echography. To the left, parenchymal transection passes along left hepatic vein but to the right it does along Rex-Cantlie line. Both left and middle hepatic veins are preserved, ii) parenchymal transection along *lig. falciforme hepatis* reaching caudally to the vessels for left segments. Only portal triad vessels for Sg3 and bile ducts for Sg4b are interrupted. Sg3 ischaemia causes demarcation of its borderline to Sg2 along which parenchymal transection is performed and iii) parenchymal transection along Rex-Cantlie line at the borderline between Sg4b and Sg5.

Fig. 14. Mobilization and preparation for interruption of the artery for Sg5,8. RHA - right hepatic artery

portion of right hepatic vein is liberated and lifted on a rubber tape, ii) definition of resection borderline after interruption of the artery and portal vein for Sg6,7 or by injection of 5 mL of methylene blue into portal vein for Sg6,7, iii) parenchymal transection at the borderline between Sg6,7 and Sg5,8. Caudally, both the vein for Sg6 draining blood into right hepatic vein anterior branch and portal triad vessels for Sg6,7 under it are interrupted and iv) at the end of parenchymal transection, right hepatic vein posterior branch draining

Bisegmentectomy 5,8 passes through four stages: i) mobilization of right hemiliver through interruption of lig. triangulare dextrum and lig. coronarium hepatis dextrum. Extrahepatic portion of right hepatic vein is liberated and lifted on a rubber tape, ii) definition of resection borderlines of Sg5,8. Extrahepatic interruption of the artery (Fig. 14) and portal vein (Fig. 15) for Sg5,8 causes their ischaemic demarcation (Fig. 16). Sg 5,8 visualization after injection of 5 mL of methylene blue into portal vein for Sg5,8, iii) parenchymal transection at the borderline to lobus quadratus through consecutive interruption of the vein for Sg5 draining blood into middle hepatic vein, of the portal triad vessels for Sg5,8 and the vein draining blood from Sg8 into middle hepatic vein. Sg5,8 devascularization results in ischaemic demarcation line at the borderline to Sg6,7 and iv) parenchymal transection along this borderline includes interruprion of the vein draining blood from Sg5 into right hepatic

Bisegmentectomy 3,4b passes through three stages: i) definition of resection borderlines by means of intraoperative echography. To the left, parenchymal transection passes along left hepatic vein but to the right it does along Rex-Cantlie line. Both left and middle hepatic veins are preserved, ii) parenchymal transection along *lig. falciforme hepatis* reaching caudally to the vessels for left segments. Only portal triad vessels for Sg3 and bile ducts for Sg4b are interrupted. Sg3 ischaemia causes demarcation of its borderline to Sg2 along which parenchymal transection is performed and iii) parenchymal transection along Rex-Cantlie

Fig. 14. Mobilization and preparation for interruption of the artery for Sg5,8. RHA - right

blood from Sg7 is interrupted as its anterior branch is preserved.

vein as both middle and right hepatic veins are obligatorily preserved.

line at the borderline between Sg4b and Sg5.

hepatic artery

Fig. 15. Mobilization and preparation for interruption of portal vein for Sg5,8. RPV - right portal vein

Fig. 16. Ischaemic demarcation of Sg5,8 after their devascularization

Bisegmentectomy 1,4 passes through five stages: i) mobilization of left and right hemiliver and definition of Sg4 resection borderlines. To the left, parenchymal transection passes along *lig. falciforme hepatis* but to the right it follows middle hepatic vein course as defined by means of intraoperative echography, ii) Sg4 devascularization by consecutive interruption of the artery and ascendant portal veins for Sg4 through extrahepatic access, iii) parenchymal transection along *lig. falciforme hepatis* and, caudally, interruption of descendent portal veins for Sg4. After interruption of portal veins for *lobus quadratus*, Rex recessus is reached where the bile duct for Sg4 is identified and interrupted, iv) Sg1

Resection for Colorectal Liver Metastases 425

one hand, and between left and right hemiliver, on the other hand. Transversal resection borderline between Sg8 and Sg5 is established by means of intraoperative echography while the left borderline passes along *lig. falciforme hepatis*, iii) parenchymal transection along Rex-Cantlie line and consecutive interruption of the veins draining blood from Sg5,8 into middle hepatic vein and portal triad vessels for Sg8. Sg8 devascularization results in ischaemic demarcation line at the borderline to Sg5 and iv) parenchymal transection along this line continuing in a sagittal plane between Sg7 and Sg8 and reaching cranially up to the borderline between the right and the middle hepatic vein. Next follows parenchymal transection along *lig. falciforme hepatis* in order to liberate Sg4a ending, cranially, at the borderline between the left and the middle hepatic vein. Finally, both Sg8 and Sg4a are

i. Stages of the following multisegmentectomies - mesohepatectomy with preservation of Sg1; mesohepatectomy together with Sg1; resection of Sg4b,5,6 and Spiegel's lobe; resection of Sg3,4b,5; resection of Sg3,5-7; resection of parts of Sg3,4b,5,6,8; resection of

Mesohepatectomy (Sg4,5,8) consists in removal of three segments both of which (Sg5 and Sg8) belong anatomically to the right hemiliver while Sg4 belongs to the left hemiliver. This operation is applied in centrally located liver metastases enabling R0 (Fig. 17). Mesohepatectomy with preservation of Sg1 passes through six stages: i) mobilization of left and right hemiliver as double 'hanging'-maneuver lifting on a holder of the right and left hepatic veins facilitates hepatic resection, ii) definition of right resection borderline by consecutive interruption of the artery and portal vein for Sg5,8 (Fig. 18). Parenchymal transection visualizes right resection borderline between Sg5,8 and Sg6,7, iii) interruption of the artery and ascendant portal veins for Sg4 through extrahepatic access. To the left, resection line passes along *lig. falciforme hepatis.* Sg4,5,8 devascularization enables mesohepatectomy at minimal blood loss, iv) parenchymal transection along *lig. falciforme* 

entirely mobilized around middle hepatic vein which is interrupted at its basis.

**5.3.3 Multisegmentectomies** 

Sg4b,6,7, and resection of Sg6-8:

Fig. 17. CT image of liver metastasis in Sg4,5,8

devascularization through extrahepatic access and v) parenchymal transection along Rex-Cantlie line. Cranially, middle hepatic vein is interrupted, if necessary. After bisegmentectomy 1,4, a large parenchymal defect is formed at which bottom the retrohepatic portion of *vena cava inferior* is visible.

Bisegmentectomy 4b,5 passes through four stages: i) definition of resection borderlines. To the left, parenchymal transection passes along *lig. falciforme hepatis* but to the right it does along Rex-Cantlie line. Hilar dissection with division of vessels for right and left hemiliver enables selective clamping the artery and portal vein for Sg2-8, if necessary, ii) parenchymal transection along *lig. falciforme hepatis* as the artery for Sg4 is provisorily clamped as well as ascendant portal veins and bile duct for Sg4b are interrupted, iii) parenchymal transection in a transversal plane at the borderline between Sg4b and Sg5, on the one hand, and between Sg4a and Sg8, on the other hand. Resection line is defined by means of intraoperative echography. Consecutively, distal portion of middle hepatic vein and portal triad vessels for Sg5 are interrupted and iv) Sg5 ischaemia results in demarcation line at the borderline to Sg6 along which parenchymal transection is performed. Caudally, the vein draining blood from Sg5 into right hepatic vein anterior branch is interrupted and portal triad vessels for Sg6-8 are obligatorily preserved.

Bisegmentectomy 5,6 passes through three stages: i) hilar dissection and isolation of the vessels for right hemiliver. Their clamping visualizes the parenchymal transection line between Sg5 and Sg6. Resection borderlines in bisegmentectomy 5,6 are defined by means of intraoperative echography, too, ii) parenchymal transection at the borderline between Sg4 and Sg5 and interruption of the vein draining blood from Sg5 into middle hepatic vein. Caudally, isolation of portal triad vessels for right hemiliver and consecutive interruption of blood supply for Sg5,6. Their ischaemia results in demarcation of their borderline to Sg7,8 and iii) parenchymal transection along this borderline. Caudally, right hepatic vein anterior branch draining blood from Sg5,6 is interrupted and portal triad vessels for Sg7,8 are obligatorily preserved.

Bisegmentectomy 7,8 is possible only in the presence of inferior right hepatic vein draining blood from Sg6. Bisegmentectomy 7,8 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Right hemiliver is luxated to the left and the retrohepatic portion of *vena cava inferior* is liberated. Identification of inferior right hepatic vein enables technical performance of bisegmentectomy 7,8, ii) definition of resection borderlines. Hilar clamping the vessels for right hemiliver results in ischaemic demarcation line at the borderline between Sg5,8 and Sg4. Parenchymal transection along this line between Sg8 and Sg4. 'Hanging'-maneuver facilitates hepatic resection. Resection borderline between proximal (Sg7,8) and transversal (Sg5,6) segments is defined by means of intraoperative echography, iii) parenchymal transection along resection borderlines with Sg7,8 devascularization starting at the borderline between Sg6 and Sg7. Initially, right hepatic vein anterior branch is identified and interrupted. Then, in a transversal plane, portal triad vessels for Sg7,8 are reached and interrupted and iv) parenchymal transection in a sagittal plane at the borderline between Sg8 and Sg4. Caudally, the vein draining blood from Sg8 into middle hepatic vein is ligated. Finally, right hepatic vein branch is interrupted.

Bisegmentectomy 4a,8 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Right hemiliver is luxated to the left and then right hepatic vein branch is extrahepatically mobilized, ii) definition of resection borderlines. Hilar dissection enables the isolation of the artery and portal vein for Sg5,8*.* Their clamping visualizes the borderline between Sg8 and Sg6, on the

devascularization through extrahepatic access and v) parenchymal transection along Rex-Cantlie line. Cranially, middle hepatic vein is interrupted, if necessary. After bisegmentectomy 1,4, a large parenchymal defect is formed at which bottom the

Bisegmentectomy 4b,5 passes through four stages: i) definition of resection borderlines. To the left, parenchymal transection passes along *lig. falciforme hepatis* but to the right it does along Rex-Cantlie line. Hilar dissection with division of vessels for right and left hemiliver enables selective clamping the artery and portal vein for Sg2-8, if necessary, ii) parenchymal transection along *lig. falciforme hepatis* as the artery for Sg4 is provisorily clamped as well as ascendant portal veins and bile duct for Sg4b are interrupted, iii) parenchymal transection in a transversal plane at the borderline between Sg4b and Sg5, on the one hand, and between Sg4a and Sg8, on the other hand. Resection line is defined by means of intraoperative echography. Consecutively, distal portion of middle hepatic vein and portal triad vessels for Sg5 are interrupted and iv) Sg5 ischaemia results in demarcation line at the borderline to Sg6 along which parenchymal transection is performed. Caudally, the vein draining blood from Sg5 into right hepatic vein anterior branch is interrupted and portal triad vessels for

Bisegmentectomy 5,6 passes through three stages: i) hilar dissection and isolation of the vessels for right hemiliver. Their clamping visualizes the parenchymal transection line between Sg5 and Sg6. Resection borderlines in bisegmentectomy 5,6 are defined by means of intraoperative echography, too, ii) parenchymal transection at the borderline between Sg4 and Sg5 and interruption of the vein draining blood from Sg5 into middle hepatic vein. Caudally, isolation of portal triad vessels for right hemiliver and consecutive interruption of blood supply for Sg5,6. Their ischaemia results in demarcation of their borderline to Sg7,8 and iii) parenchymal transection along this borderline. Caudally, right hepatic vein anterior branch draining blood from Sg5,6 is interrupted and portal triad vessels for Sg7,8 are

Bisegmentectomy 7,8 is possible only in the presence of inferior right hepatic vein draining blood from Sg6. Bisegmentectomy 7,8 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Right hemiliver is luxated to the left and the retrohepatic portion of *vena cava inferior* is liberated. Identification of inferior right hepatic vein enables technical performance of bisegmentectomy 7,8, ii) definition of resection borderlines. Hilar clamping the vessels for right hemiliver results in ischaemic demarcation line at the borderline between Sg5,8 and Sg4. Parenchymal transection along this line between Sg8 and Sg4. 'Hanging'-maneuver facilitates hepatic resection. Resection borderline between proximal (Sg7,8) and transversal (Sg5,6) segments is defined by means of intraoperative echography, iii) parenchymal transection along resection borderlines with Sg7,8 devascularization starting at the borderline between Sg6 and Sg7. Initially, right hepatic vein anterior branch is identified and interrupted. Then, in a transversal plane, portal triad vessels for Sg7,8 are reached and interrupted and iv) parenchymal transection in a sagittal plane at the borderline between Sg8 and Sg4. Caudally, the vein draining blood from Sg8 into middle hepatic vein is ligated.

Bisegmentectomy 4a,8 passes through four stages: i) mobilization of right hemiliver through interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum.* Right hemiliver is luxated to the left and then right hepatic vein branch is extrahepatically mobilized, ii) definition of resection borderlines. Hilar dissection enables the isolation of the artery and portal vein for Sg5,8*.* Their clamping visualizes the borderline between Sg8 and Sg6, on the

retrohepatic portion of *vena cava inferior* is visible.

Sg6-8 are obligatorily preserved.

obligatorily preserved.

Finally, right hepatic vein branch is interrupted.

one hand, and between left and right hemiliver, on the other hand. Transversal resection borderline between Sg8 and Sg5 is established by means of intraoperative echography while the left borderline passes along *lig. falciforme hepatis*, iii) parenchymal transection along Rex-Cantlie line and consecutive interruption of the veins draining blood from Sg5,8 into middle hepatic vein and portal triad vessels for Sg8. Sg8 devascularization results in ischaemic demarcation line at the borderline to Sg5 and iv) parenchymal transection along this line continuing in a sagittal plane between Sg7 and Sg8 and reaching cranially up to the borderline between the right and the middle hepatic vein. Next follows parenchymal transection along *lig. falciforme hepatis* in order to liberate Sg4a ending, cranially, at the borderline between the left and the middle hepatic vein. Finally, both Sg8 and Sg4a are entirely mobilized around middle hepatic vein which is interrupted at its basis.

#### **5.3.3 Multisegmentectomies**

i. Stages of the following multisegmentectomies - mesohepatectomy with preservation of Sg1; mesohepatectomy together with Sg1; resection of Sg4b,5,6 and Spiegel's lobe; resection of Sg3,4b,5; resection of Sg3,5-7; resection of parts of Sg3,4b,5,6,8; resection of Sg4b,6,7, and resection of Sg6-8:

Mesohepatectomy (Sg4,5,8) consists in removal of three segments both of which (Sg5 and Sg8) belong anatomically to the right hemiliver while Sg4 belongs to the left hemiliver. This operation is applied in centrally located liver metastases enabling R0 (Fig. 17). Mesohepatectomy with preservation of Sg1 passes through six stages: i) mobilization of left and right hemiliver as double 'hanging'-maneuver lifting on a holder of the right and left hepatic veins facilitates hepatic resection, ii) definition of right resection borderline by consecutive interruption of the artery and portal vein for Sg5,8 (Fig. 18). Parenchymal transection visualizes right resection borderline between Sg5,8 and Sg6,7, iii) interruption of the artery and ascendant portal veins for Sg4 through extrahepatic access. To the left, resection line passes along *lig. falciforme hepatis.* Sg4,5,8 devascularization enables mesohepatectomy at minimal blood loss, iv) parenchymal transection along *lig. falciforme* 

Fig. 17. CT image of liver metastasis in Sg4,5,8

Resection for Colorectal Liver Metastases 427

Mesohepatectomy with segmentectomy 1 is indicated in tumours of central location and passes through seven stages already described in detail in single segmentectomy chapters: i) mobilization of left and right hemiliver by means of double 'hanging'-maneuver, ii) Sg5,8 devascularization through extrahepatic access, iii) Sg4 devascularization through extrahepatic access for the artery and ascendant portal veins, iv) Sg1 devascularization with liberation of the retrohepatic part of *vena cava inferior*, v) parenchymal transection along *lig. falciforme hepatis* ending at the borderline between the middle and the left hepatic vein. Caudally, identification and interruption of descendent portal veins for Sg4 enabling the opening of Rex recessus. Interruption of bile ducts for Sg4 located in Rex recessus, vi) parenchymal transection at the borderline between Sg5,8 and Sg6,7 and vii) middle hepatic vein interruption at the end of parenchymal transection and obligatory intraoperative cholangiography after resection of Sg1,4,5,8 for control of bile drainage from the remnant

Resection of Sg4b,5,6 and Spiegel's lobe combines the already described stages in resections

Resection of Sg3,4b,5 combines the already described stages in resections of Sg3,4b,5. Portal

Resection of Sg3,5-7 combines the already described stages in resections of Sg3,5-7. Portal

Resection of parts of Sg3,4b,5,6,8 combines the already described stages in resections of

Resection of Sg6-8 combines the already described stages in resections of Sg6-8. Portal triad

ii. Stages of the following hemihepatectomies - left hemihepatectomy (Sg2-4), and right

Resection of Sg4b,6,7 combines the already described stages in resections of Sg4b,6,7.

Fig. 20. Residual liver volume after mesohepatectomy (Sg4,5,8)

of Sg1,4-6. Portal triad vessels for Sg7,8 are obligatorily preserved.

triad vessels for Sg6,8 are obligatorily preserved.

triad vessels for Sg8 are obligatorily preserved.

vessels for Sg5 are obligatorily preserved.

hemihepatectomy (Sg5-8):

liver parenchyma.

Sg3,4b,5,6,8.

*hepatis* ending at the borderline between the middle and the left hepatic vein. Consecutive interruption of descendent portal veins, the bile duct for Sg4 and the vein draining blood from *lobus quadratus* in left hepatic vein, v) parenchymal transection at the borderline between Sg5,8 and Sg6,7 and consecutive interruption of the vein draining blood from Sg5 into right hepatic vein anterior branch and portal triad vessels for Sg5,8. To the right, parenchymal transection ends at the borderline between the right and the middle hepatic vein and vi) interruption of middle hepatic vein around which these already liberated Sg5,8 and Sg4 are located (Fig. 19). Residual liver volume after removal of Sg4,5,8 is shown on Fig. 20.

Fig. 18. Liver resection volume in mesohepatectomy (Sg4,5,8)

Fig. 19. The site for interruption of the middle hepatic vein is indicated by a circle. MHV middle hepatic vein; RHV - right hepatic vein

*hepatis* ending at the borderline between the middle and the left hepatic vein. Consecutive interruption of descendent portal veins, the bile duct for Sg4 and the vein draining blood from *lobus quadratus* in left hepatic vein, v) parenchymal transection at the borderline between Sg5,8 and Sg6,7 and consecutive interruption of the vein draining blood from Sg5 into right hepatic vein anterior branch and portal triad vessels for Sg5,8. To the right, parenchymal transection ends at the borderline between the right and the middle hepatic vein and vi) interruption of middle hepatic vein around which these already liberated Sg5,8 and Sg4 are located (Fig. 19).

Residual liver volume after removal of Sg4,5,8 is shown on Fig. 20.

Fig. 18. Liver resection volume in mesohepatectomy (Sg4,5,8)

middle hepatic vein; RHV - right hepatic vein

Fig. 19. The site for interruption of the middle hepatic vein is indicated by a circle. MHV -

Fig. 20. Residual liver volume after mesohepatectomy (Sg4,5,8)

Mesohepatectomy with segmentectomy 1 is indicated in tumours of central location and passes through seven stages already described in detail in single segmentectomy chapters: i) mobilization of left and right hemiliver by means of double 'hanging'-maneuver, ii) Sg5,8 devascularization through extrahepatic access, iii) Sg4 devascularization through extrahepatic access for the artery and ascendant portal veins, iv) Sg1 devascularization with liberation of the retrohepatic part of *vena cava inferior*, v) parenchymal transection along *lig. falciforme hepatis* ending at the borderline between the middle and the left hepatic vein. Caudally, identification and interruption of descendent portal veins for Sg4 enabling the opening of Rex recessus. Interruption of bile ducts for Sg4 located in Rex recessus, vi) parenchymal transection at the borderline between Sg5,8 and Sg6,7 and vii) middle hepatic vein interruption at the end of parenchymal transection and obligatory intraoperative cholangiography after resection of Sg1,4,5,8 for control of bile drainage from the remnant liver parenchyma.

Resection of Sg4b,5,6 and Spiegel's lobe combines the already described stages in resections of Sg1,4-6. Portal triad vessels for Sg7,8 are obligatorily preserved.

Resection of Sg3,4b,5 combines the already described stages in resections of Sg3,4b,5. Portal triad vessels for Sg6,8 are obligatorily preserved.

Resection of Sg3,5-7 combines the already described stages in resections of Sg3,5-7. Portal triad vessels for Sg8 are obligatorily preserved.

Resection of parts of Sg3,4b,5,6,8 combines the already described stages in resections of Sg3,4b,5,6,8.

Resection of Sg4b,6,7 combines the already described stages in resections of Sg4b,6,7.

Resection of Sg6-8 combines the already described stages in resections of Sg6-8. Portal triad vessels for Sg5 are obligatorily preserved.

ii. Stages of the following hemihepatectomies - left hemihepatectomy (Sg2-4), and right hemihepatectomy (Sg5-8):

Resection for Colorectal Liver Metastases 429

Fig. 21. Right and middle hepatic arteries as well as portal vein right branch are interrupted.

Fig. 22. Sites for interruption of the veins for Sg4 through a combined access.

ligation of the middle and the right hepatic veins (Fig. 26).

The interruption of the right hepatic duct is presented on Fig. 23. A preserved left hepatic vein (single 'hanging'-maneuver) is indicated on Fig. 24, iv) parenchymal transection along *lig. falciforme hepatis* and consecutive interruption of descendent portal veins for Sg 4. Entering Rex *recessus* with interruption of the bile ducts for Sg4 (Fig. 25), and v) cranial

LHA – left hepatic artery; RHA - right hepatic artery

Left hemihepatectomy passes through three stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis sinistrum.*  Devascularization of left hemiliver with ligation of left hepatic artery and portal vein left branch. If possible, left hepatic duct is liberated without its interruption. Extrahepatically, the trunk of the left and the middle hepatic vein is mobilized enabling the application of 'hanging'-maneuver, ii) parenchymal transection along the ischaemic demarcation line between left and right hemiliver. Resection borderline is defined after injection of 10 mL of methylene blue through portal vein left branch and iii) finally, consecutive interruption of left hepatic duct and left hepatic vein-middle hepatic vein branch. Middle hepatic vein is preserved, if indicated.

Right hemihepatectomy passes through three stages: i) mobilization of right hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum*, ii) interruption of the right hepatic and the portal vein right branch. If possible, right hepatic duct is liberated without its interruption. Right hepatic vein mobilization by means of 'hanging'-maneuver facilitates liver resection. Right hemiliver devascularization results in ischaemic demarcation line at the borderline to left hemiliver, and iii) parenchymal transection at the borderline between left and right hemiliver and consecutive ligation of the veins draining blood from Sg5,8 into middle hepatic vein. Finally, interruption of right hepatic duct and right hepatic vein as well as of middle hepatic vein, if indicated.

iii. Stages of the following trisectionectomies - left trisectionectomy (Sg2-5,8), and right trisectionectomy (Sg4-8):

Left trisectionectomy is used in tumours affecting left hemiliver and Sg5,8. No preoperative embolization of portal vein left branch is needed as preserved Sg7 and Sg 6 amount to 30- 35% of standard liver volume. Left trisectionectomy passes through five stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare sinistrum*  and *lig. coronarium hepatis sinistrum*. 'Hanging'-maneuver application facilitates resection, ii) devascularization of left hemiliver with ligation of left hepatic artery and portal vein left branch, iii) Sg5,8 devascularization through extrahepatic access . Sg5,8 ischaemia causes demarcation line at the borderline to Sg6,7 along which parenchymal transection is performed, iv) consecutive interruption of the vein draining blood from Sg5 into middle hepatic vein and intraparenchymal portal triad vessels for Sg5,8. Parenchymal transection ends at the borderline between the right and the middle hepatic vein. Right hepatic vein is lifted on a rubber holder in order to prevent its injury and v) finally, the trunk of the middle and the left hepatic vein is interrupted. Portal triad vessels for Sg6,7 are obligatorily preserved.

Right trisectionectomy is used in tumours affecting right hemiliver and Sg4. In most cases, preoperative embolization of portal vein right branch is needed in order to achieve hypertrophy of the left hemiliver and, in particular, of Sg2 and Sg3. Right trisectionectomy passes through five stages: i) mobilization of right hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum* and short retrohepatic veins; ii) devascularization of right hemiliver with ligation of the right and middle hepatic arteries and the right branch of the portal vein (Fig. 21); iii) devascularization of *lobus quadratus* by means of interruption of the artery and ascendent portal veins for Sg4 through extrahepatic access (Fig. 22).

Left hemihepatectomy passes through three stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis sinistrum.*  Devascularization of left hemiliver with ligation of left hepatic artery and portal vein left branch. If possible, left hepatic duct is liberated without its interruption. Extrahepatically, the trunk of the left and the middle hepatic vein is mobilized enabling the application of 'hanging'-maneuver, ii) parenchymal transection along the ischaemic demarcation line between left and right hemiliver. Resection borderline is defined after injection of 10 mL of methylene blue through portal vein left branch and iii) finally, consecutive interruption of left hepatic duct and left hepatic vein-middle hepatic vein branch. Middle hepatic vein is

Right hemihepatectomy passes through three stages: i) mobilization of right hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum*, ii) interruption of the right hepatic and the portal vein right branch. If possible, right hepatic duct is liberated without its interruption. Right hepatic vein mobilization by means of 'hanging'-maneuver facilitates liver resection. Right hemiliver devascularization results in ischaemic demarcation line at the borderline to left hemiliver, and iii) parenchymal transection at the borderline between left and right hemiliver and consecutive ligation of the veins draining blood from Sg5,8 into middle hepatic vein. Finally, interruption of right hepatic duct and right hepatic vein as well as of middle hepatic vein, if

iii. Stages of the following trisectionectomies - left trisectionectomy (Sg2-5,8), and right

Left trisectionectomy is used in tumours affecting left hemiliver and Sg5,8. No preoperative embolization of portal vein left branch is needed as preserved Sg7 and Sg 6 amount to 30- 35% of standard liver volume. Left trisectionectomy passes through five stages: i) mobilization of left hemiliver through consecutive interruption of *lig. triangulare sinistrum*  and *lig. coronarium hepatis sinistrum*. 'Hanging'-maneuver application facilitates resection, ii) devascularization of left hemiliver with ligation of left hepatic artery and portal vein left branch, iii) Sg5,8 devascularization through extrahepatic access . Sg5,8 ischaemia causes demarcation line at the borderline to Sg6,7 along which parenchymal transection is performed, iv) consecutive interruption of the vein draining blood from Sg5 into middle hepatic vein and intraparenchymal portal triad vessels for Sg5,8. Parenchymal transection ends at the borderline between the right and the middle hepatic vein. Right hepatic vein is lifted on a rubber holder in order to prevent its injury and v) finally, the trunk of the middle and the left hepatic vein is interrupted. Portal triad vessels for Sg6,7 are obligatorily

Right trisectionectomy is used in tumours affecting right hemiliver and Sg4. In most cases, preoperative embolization of portal vein right branch is needed in order to achieve hypertrophy of the left hemiliver and, in particular, of Sg2 and Sg3. Right trisectionectomy passes through five stages: i) mobilization of right hemiliver through consecutive interruption of *lig. triangulare dextrum* and *lig. coronarium hepatis dextrum* and short retrohepatic veins; ii) devascularization of right hemiliver with ligation of the right and middle hepatic arteries and the right branch of the portal vein (Fig. 21); iii) devascularization of *lobus quadratus* by means of interruption of the artery and ascendent portal veins for Sg4

preserved, if indicated.

trisectionectomy (Sg4-8):

through extrahepatic access (Fig. 22).

indicated.

preserved.

Fig. 21. Right and middle hepatic arteries as well as portal vein right branch are interrupted. LHA – left hepatic artery; RHA - right hepatic artery

Fig. 22. Sites for interruption of the veins for Sg4 through a combined access.

The interruption of the right hepatic duct is presented on Fig. 23. A preserved left hepatic vein (single 'hanging'-maneuver) is indicated on Fig. 24, iv) parenchymal transection along *lig. falciforme hepatis* and consecutive interruption of descendent portal veins for Sg 4. Entering Rex *recessus* with interruption of the bile ducts for Sg4 (Fig. 25), and v) cranial ligation of the middle and the right hepatic veins (Fig. 26).

Resection for Colorectal Liver Metastases 431

Fig. 25. Entering Rex *recessus* with interruption of the bile ducts for Sg4

In case of damaged blood supply to common hepatic duct, these vessels should be removed with subsequent biliodigestive anastomosis between left hepatic duct and intestinal loop isolated after Roux. Fig. 27 shows residual liver volume following right trisectionectomy.

Fig. 26. Interruption of the middle and the right hepatic veins. MHV – middle hepatic vein;

RHV – right hepatic vein, VCI - *vena cava inferior*

Fig. 23. Interruption of right hepatic duct

Fig. 24. Preservation of the left hepatic vein through a single 'hanging'-maneuver

Fig. 23. Interruption of right hepatic duct

Fig. 24. Preservation of the left hepatic vein through a single 'hanging'-maneuver

Fig. 25. Entering Rex *recessus* with interruption of the bile ducts for Sg4

In case of damaged blood supply to common hepatic duct, these vessels should be removed with subsequent biliodigestive anastomosis between left hepatic duct and intestinal loop isolated after Roux. Fig. 27 shows residual liver volume following right trisectionectomy.

Fig. 26. Interruption of the middle and the right hepatic veins. MHV – middle hepatic vein; RHV – right hepatic vein, VCI - *vena cava inferior*

Resection for Colorectal Liver Metastases 433

number and percentage of patients with complications

surgical liver damage 8 (7.4%) 19 (18%) 0.009

respiratory tract damage 6 (5.5%) 11 (10%) 0.034

other complications 3 (2.7%) 8 (7.4%) 0.060

complications of general nature 8 (7.4%) 15 (13.8%) 0.014

Parameters Number of removed segments p one or two ≥ three positive resection area 2 (1.8%) 4 (3.7%) 0.299 duration of surgery (min) 224±19 211±21 0.368 total blood loss (mL) 480±52 682±48 < 0.001 necessity of haemotransfusion (patients) 16 (14.8%) 45 (41.6%) < 0.001 necessity of Pringle-maneuver (patients) 20 (18.5%) 52 (48.1%) < 0.001 stay in reanimation ward (days) 2±1.2 1.5±0.5 0.459 hospital stay (days) 14.7±1.4 13.5±1.6 0.269 repeated operation 2 (1.8%) 9 (8.3%) < 0.001

Three-year patients' survival assessed by means of the variables of 22 prognostic criteria is

37 (34%) 11 (10%) 26 (24%) 0.092

Complications Number of removed segments p one or two ≥ three

hemorrhage 2 (1.8%) 4 (3.7%) liver failure 2 (1.8%) 14 (13%) bilirrhagia from an opened bile duct 3 (2.7%) 9 (8.3%) extrahepatic biliary tree necrosis - 1 (1%) purulent perihepatic collection 2 (1.8%) 4 (3.7%) cholangitis 2 (1.8%) 3 (2.7%) mechanical jaundice - 2 (1.8%) peritonitis - 2 (1.8%) ascites 2 (1.8%) 14 (13%)

pneumothorax 2 (1.8%) 2 (1.8%) pulmonary thromboembolism - 1 (1%) respiratory failure 2 (1.8%) 9 (8.3%) pleural effusion > 200 mL 4 (3.7%) 9 (8.3%)

drug-resistant renal failure 2 (1.8%) 7 (6.5%) drug-resistant heart failure 2 (1.8%) 7 (6.5%) sepsis 2 (1.8%) 8 (7.4%) deep vein thrombosis 1 (1%) 1 (1%)

operative wound suppuration 8 (7.4%) 12 (11%) operative wound dehiscence 3 (2.7%) 6 (5.5%) postoperative herniation 8 (7.4%) 15 (13.8%)

Table 5. Complications after liver resections

Table 6. Surgical patterns of patients with liver resections

presented on Table 7.

Fig. 27. Residual liver volume after right trisectionectomy

#### **6. Results**

The results concerning various clinical and laboratory characteristics of the patients having undergone different types of segmentectomies and major liver resections were comparatively demonstrated. The main attention was paid to the following basic parameters: number, diameter, and localization of colorectal liver metastases; postoperative mortality rate; complications; blood loss and required blood transfusions; operative duration; length of hospital stay; resection margins, one-, two and three-year disease-free and overall survival rates.

Some of them are shown in the present paper.

Postoperative complications following monosegmentectomies and bisegmentectomies, on the one hand, and multisegmentectomies, on the other hand, are comparatively presented on Table 5.

It is evident that, as a whole, liver damage caused by the surgical intervention itself occurs statistically significantly more commonly in the patients who have undergone multisegmentectomies.

Some surgical characteristics of monosegmentectomies and bisegmentectomies, on the one hand, and multisegmentectomies, on the other hand, are comparatively presented on Table 6.

Obviously, several surgical patterns of undoubted medical and socio-economic importance such as total blood loss, necessity of blood transfusions and of application of Pringlemaneuver are statistically significantly more unfavourable in the patients who have undergone multisegmentectomies.

Besides, these patients require statistically significantly more often the performance of repeated surgical interventions on the occasion of colorectal liver metastases than those who have undergone mono- or bisegmentectomies.

The results concerning various clinical and laboratory characteristics of the patients having undergone different types of segmentectomies and major liver resections were comparatively demonstrated. The main attention was paid to the following basic parameters: number, diameter, and localization of colorectal liver metastases; postoperative mortality rate; complications; blood loss and required blood transfusions; operative duration; length of hospital stay; resection margins, one-, two and three-year disease-free

Postoperative complications following monosegmentectomies and bisegmentectomies, on the one hand, and multisegmentectomies, on the other hand, are comparatively presented

It is evident that, as a whole, liver damage caused by the surgical intervention itself occurs statistically significantly more commonly in the patients who have undergone

Some surgical characteristics of monosegmentectomies and bisegmentectomies, on the one hand, and multisegmentectomies, on the other hand, are comparatively presented on Table 6. Obviously, several surgical patterns of undoubted medical and socio-economic importance such as total blood loss, necessity of blood transfusions and of application of Pringlemaneuver are statistically significantly more unfavourable in the patients who have

Besides, these patients require statistically significantly more often the performance of repeated surgical interventions on the occasion of colorectal liver metastases than those who

Fig. 27. Residual liver volume after right trisectionectomy

**6. Results** 

on Table 5.

and overall survival rates.

multisegmentectomies.

undergone multisegmentectomies.

have undergone mono- or bisegmentectomies.

Some of them are shown in the present paper.


Table 5. Complications after liver resections


Table 6. Surgical patterns of patients with liver resections

Three-year patients' survival assessed by means of the variables of 22 prognostic criteria is presented on Table 7.

Resection for Colorectal Liver Metastases 435

We identify a small number of prognostic criteria which could be considered statistically significant in the patients with colorectal liver metastases. Here belong the increased levels of CEA, the higher number of colorectal liver metastases (more than three), the negative resection areas, the presence of negative lymph nodes in *lig. hepatogastroduodenale,* the implementation of multisegmentectomy as a less sparing surgical intervention, the presence of at least 3 factors of MSKCC-CRS and the absence of extrahepatic dissemination of the

Thus our investigations should be enlarged in future in order to more comprehensively explain the dynamic interactions between the single risk factors for the relatively poor

Our own results demonstrate the substantial advantages of segmental resection for colorectal liver metastases over major liver resection (Kobakov & Kostov, 2006; Kostov & Kobakov, 2006b; Kostov & Kobakov, 2009). They are the following: conservation of a sufficient liver volume, achievement of lower perioperative morbidity and mortality rates as well as warranting the similar disease-free and overall survival rates. Liver conservation is essential in normal and damaged liver. It reduces the risk of postoperative liver insufficiency from a small liver remnant and in the patients at advanced age or with

The following prognostic factors exert a statistically significant effect on short- and longterm survival rates after liver resections for colorectal liver metastases: CEA level, presence of metastatic nodes along *lig. hepatoduodenale*, number of metastases, extension of liver resection, resection volume, number of prognostic factors according to MSKCC, and

The following therapeutic strategy should be recommended: i) liver resection for resectable colorectal metastases (at stages IVA and IVB); ii) neoadjuvant chemotherapy for primarily non-resectable colorectal metastases (at stage IVC) when downstaging is feasible to allow radical surgery, and iii) only chemotherapy for colorectal metastases in stage IVD patients. Recent literature data convincingly indicate the uninterrupted progress in the interdisciplinary field of oncologic liver surgery. Along with original investigations, a lot of review papers, meta-analyses, multicentre reports and randomized controlled trials are

In this respect, multimodal therapy deserves a special attention. It increases the number of resections and improves long-term survival rate (currently more than 40% at 5 years) (Neumann et al., 2010). Advances in staging, surgical technique, perioperative care and systemic chemotherapy contribute to improvement in oncologic outcomes of stage IV colorectal cancer patients (Abdalla, 2011). The limits of resection expand to include cases with more, larger and bilateral colorectal liver metastases as 5-year overall survival exceeds 50% following resection. Tailored, patient-centered treatment includes a variety of liver resections, liver volumetry, and portal vein embolization for preoperative enhancement of

Multimodality approach of laparoscopic liver resection is feasible and safe in selected patients. It is associated with a low complications rate (Isoniemi et al., 2011, Lai et al., 2011). Intraoperative ablation extends the limits of hepatectomy in the patients not amenable to complete resection (Brown et al., 2011; Govindarajan et al., 2011; Hammill et al., 2011;

the volume and function of the planned future remnant liver (Abdalla, 2011).

pathological process.

**7. Discussion** 

cirrhosis.

prognosis of this contingent of patients.

extrahepatic dissemination of primary colorectal cancer.

currently published by authors from all over the world.


Table 7. Prognostic criteria for three-year survival

We identify a small number of prognostic criteria which could be considered statistically significant in the patients with colorectal liver metastases. Here belong the increased levels of CEA, the higher number of colorectal liver metastases (more than three), the negative resection areas, the presence of negative lymph nodes in *lig. hepatogastroduodenale,* the implementation of multisegmentectomy as a less sparing surgical intervention, the presence of at least 3 factors of MSKCC-CRS and the absence of extrahepatic dissemination of the pathological process.

Thus our investigations should be enlarged in future in order to more comprehensively explain the dynamic interactions between the single risk factors for the relatively poor prognosis of this contingent of patients.

#### **7. Discussion**

434 Colorectal Cancer – From Prevention to Patient Care

Males females age < 65 years age ≥ 65 years T2-T3 category T4 category

G1-G2 tumour differentiation G3 tumour differentiation

colonic primary tumour rectal primary tumour CEA ≤ 200 ng/mL CEA > 200 ng/mL synchronous metastases metachronous metastases after < 12 months after ≥ 12 months diameter < 50 mm diameter ≥ 50 mm ≤ 3 metastases > 3 metastases unilobar metastases bilobar metastases positive resection areas negative resection areas

resection distance ≥ 10 mm resection distance of 5-10 mm resection distance ≤ 5 mm

application of Pringle-maneuver no application of Pringle-maneuver postoperative complications no postoperative complications 0-2 factors of MSKCC-CRS 3-5 factors of MSKCC-CRS extrahepatic dissemination no extrahepatic dissemination neoadjuvant chemotherapy no neoadjuvant chemotherapy

multisegmentectomy blood loss > 500 mL blood loss ≤ 500 mL

negative lymph nodes during colorectal cancer surgery positive lymph nodes during colorectal cancer surgery

positive lymph nodes in *lig. hepatogastroduodenale* negative lymph nodes in *lig. hepatogastroduodenale*

monosegmentectomy and bisegmentectomy

Table 7. Prognostic criteria for three-year survival

Variables of prognostic criteria n % p

0.582

0.743 0.944 0.934

0.677

0.877

0.983

< 0.001

0.834

0.221

0.712

< 0.001

0.069

< 0.001

< 0.001 0.700 0.790

< 0.001

0.644

0.736

0.743

< 0.001

< 0.001

0.628

Our own results demonstrate the substantial advantages of segmental resection for colorectal liver metastases over major liver resection (Kobakov & Kostov, 2006; Kostov & Kobakov, 2006b; Kostov & Kobakov, 2009). They are the following: conservation of a sufficient liver volume, achievement of lower perioperative morbidity and mortality rates as well as warranting the similar disease-free and overall survival rates. Liver conservation is essential in normal and damaged liver. It reduces the risk of postoperative liver insufficiency from a small liver remnant and in the patients at advanced age or with cirrhosis.

The following prognostic factors exert a statistically significant effect on short- and longterm survival rates after liver resections for colorectal liver metastases: CEA level, presence of metastatic nodes along *lig. hepatoduodenale*, number of metastases, extension of liver resection, resection volume, number of prognostic factors according to MSKCC, and extrahepatic dissemination of primary colorectal cancer.

The following therapeutic strategy should be recommended: i) liver resection for resectable colorectal metastases (at stages IVA and IVB); ii) neoadjuvant chemotherapy for primarily non-resectable colorectal metastases (at stage IVC) when downstaging is feasible to allow radical surgery, and iii) only chemotherapy for colorectal metastases in stage IVD patients.

Recent literature data convincingly indicate the uninterrupted progress in the interdisciplinary field of oncologic liver surgery. Along with original investigations, a lot of review papers, meta-analyses, multicentre reports and randomized controlled trials are currently published by authors from all over the world.

In this respect, multimodal therapy deserves a special attention. It increases the number of resections and improves long-term survival rate (currently more than 40% at 5 years) (Neumann et al., 2010). Advances in staging, surgical technique, perioperative care and systemic chemotherapy contribute to improvement in oncologic outcomes of stage IV colorectal cancer patients (Abdalla, 2011). The limits of resection expand to include cases with more, larger and bilateral colorectal liver metastases as 5-year overall survival exceeds 50% following resection. Tailored, patient-centered treatment includes a variety of liver resections, liver volumetry, and portal vein embolization for preoperative enhancement of the volume and function of the planned future remnant liver (Abdalla, 2011).

Multimodality approach of laparoscopic liver resection is feasible and safe in selected patients. It is associated with a low complications rate (Isoniemi et al., 2011, Lai et al., 2011). Intraoperative ablation extends the limits of hepatectomy in the patients not amenable to complete resection (Brown et al., 2011; Govindarajan et al., 2011; Hammill et al., 2011;

Resection for Colorectal Liver Metastases 437

timely and individualized diagnostic and complex therapeutic approach by an interdisciplinary physician's team. Medical staff's behaviour should be maximally sparing,

New advances in image diagnostic modalities such as positron emission tomography/computer-aided tomography, steadily improved surgical and microsurgical techniques such as laparoscopic resections along with emerging opportunities for costeffective chemotherapy and multimodal management promise better perspectives in this

Abdalla, E.K. (2011). Resection of colorectal liver metastases. *Journal of Gastrointestinal* 

Arru, M.; Aldrighetti, L.; Castoldi, R.; Di Palo, S.; Orsenigo, E.; Stella, M.; Pulitanò, C.;

Brown, R.E.; Martin, R.C. 2nd & Scoggins, C.R. (2011). Ablative therapies for colorectal liver

Chen, J.; Li, Q.; Wang, C.; Zhu, H.; Shi, Y. & Zhao, G. (2011). Simultaneous vs. staged

Coimbra, F.J.; Pires, T.C.; Costa Junior, W.L.; Diniz, A.L. & Ribeiro, H.S. (2011). Advances in

de Haas, R.J.; Wicherts, D.A.; Andreani, P.; Pascal, G.; Saliba, F.; Ichai, P.; Adam, R.;

*Annals of Surgery*, Vol.253, No.6, (June 2011), pp. 1069-1079, ISSN 0003-4932 Eisenhauer, E.A.; Therasse, P., Bogaerts, J.; Schwartz, L.H.; Sargent, D.; Ford, R.; Dancey, J.;

Ferrero, A.; Russolillo, N.; Viganò, L.; Lo Tesoriere, R.; Muratore, A. & Capussotti, L. (2010).

Gomez, D. & Cameron, I.C. (2010). Prognostic scores for colorectal liver metastasis: clinically

Govindarajan, A.; Arnaoutakis, D.; D'Angelica, M.; Allen, P.J.; Dematteo, R.P.; Blumgart,

*Brasileira*, Vol.57, No.2, (April 2011), pp. 220-227, ISSN 0104-4230

Vol.45, No.2, (January 2009), pp. 228-247, ISSN 0959-8049

Gavazzi, F.; Ferla, G.; Di Carlo, V. & Staudacher, C. (2008). Analysis of prognostic factors influencing long-term survival after hepatic resection for metastatic colorectal cancer. *World Journal of Surgery*, Vol.32, No.1, (January 2008), pp. 93-103,

metastases. *Surgical Oncology Clinics of North America*, Vol.20, No.2, (April 2011), pp.

resection for synchronous colorectal liver metastases: a metaanalysis. *International Journal of Colorectal Diseases*, Vol.26, No.2, (February 2011), pp. 191-1999, ISSN 0179-

surgical treatment of colorectal liver metastases. *Revista da Associacao Medica* 

Castaing, D. & Azoulay, D. (2011). Impact of expanding criteria for resectability of colorectal metastases on short-and long-term outcomes after hepatic resection.

Arbuck, S.; Gwyther, S.; Mooney, M.; Rubinstein, L.; Shankar, L.; Dodd, L.; Kaplan, R.; Lacombe, D. & Verweij, J. (2009). New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). *European Journal of Cancer*,

Does Pringle maneuver affect survival in patients with colorectal liver metastases? *World Journal of Surgery*, Vol.34, No.10, (October 2010), pp. 2418-2425, ISSN 0364-

important or an academic exercise? *HPB* (*Oxford*), Vol.12, No.4, (May 2010), pp.

L.H.; Jarnagin, W.R. & Fong, Y. (2011). Use of intraoperative ablation as an adjunct

*Surgery*, Vol.15, No.4 (March 2011), pp. 416-419, ISSN 1091-255X

when possible.

**9. References** 

field of permanently rising social significance.

259-271, vii, ISSN 1055-3207

ISSN 1091-255X

1958

2131

227-238, ISSN 1365-182X

Hompes et al., 2011). Portal vein embolization, radiofrequency ablation, two-stage hepatectomy, conversion therapy and reverse treatment strategy along with hepatectomy are used in the presence of extrahepatic disease (Coimbra et al. 2011; Narita et al., 2011, Tsim et al., 2011). Resection of advanced colorectal liver metastases after a second-line chemotherapy regimen is safe and promising in certain cases. The addition of neoadjuvant chemotherapy should, however, be cost-effective.

Positron emission tomography/computed tomography have a higher accuracy for detection of extra-hepatic and colorectal liver metastases than computed tomography alone (Patel et al., 2011). In patients treated with neoadjuvant chemotherapy, magnetic resonance imaging measurements of steatosis show the highest correlation coefficient and the best diagnostic accuracy, as compared to computed tomography ones (Marsman et al., 2011). Intraoperative ultrasound and preoperative imaging significantly increase the diagnostic accuracy of patients undergoing liver resection for colorectal liver metastases (Lordan et al., 2011).

Metachronous resections have a better outcome than synchronous. Iterative resection is very encouraging and justifies an aggressive surgical approach (Tonelli et al., 2010). Simultaneous resection is safe and efficient in the treatment of patients with synchronous colorectal liver metastases while avoiding a second major operation (Chen et al., 2011). In patients with bilobar synchronous colorectal liver metastases who are candidates for twostage hepatectomy, combined resection of the primary tumour and first-stage he patectomy reduces the number of procedures, optimizes chemotherapy administration and may improve outcome (Karoui et al., 2010). The two-stage strategy for colorectal liver metastases can be performed with acceptable morbidity and mortality. The second stage is not feasible in 20-25% of patients. Patients completing the two-stage approach may have long-term survival comparable to those treated with a planned single-stage hepatectomy (Tsai et al., 2010). Concomitant extrahepatic disease in a patient with colorectal liver metastases should not be a contraindication to their resection.

As there is no significant difference in morbidity, mortality, recurrence rate, or survival in anatomical and nonanatomical liver resections, the latter can be used as a save procedure to preserve liver parenchyma (Lalmahomed et al., 2011). The Pringle maneuver does not seem to affect the survival of patients with liver metastases (Ferrero et al., 2010). Ultrasoundguided finger compression of sectional portal pedicle feeding the right posterior section is a feasible, safe, and effective method for performing anatomical right posterior sectionectomy (Torzilli et al., 2011).

Prognostic factors and score systems occupy an important place in oncologic liver surgery (de Haas et al., 2011; Peng et al., 2011; Pulitanò et al., 2011). Although twelve prognostic scoring systems have been identified from 1996 to 2009, there is no 'ideal' system for the clinical management of patients with colorectal liver metastases (Gomez et al., 2010). A predicted positive surgical margin (R1 resection) is not any absolute contraindication to surgery for aggressive or advanced colorectal liver metastases (Tanaka et al., 2011). Liver resection has superior long-term survival which is, however, significantly reduced by the occurrence of post-surgical complications (Schepers et al., 2010). Superior overall healthrelated quality of life merits an aggressive surgical approach and intensive follow-up to detect recurrence early (Wiering et al., 2011).

#### **8. Conclusion**

Based on our own results and reliable scientific evidence available worldwide up-to-date, it can be concluded that the patient presenting with colorectal liver metastases deserves a timely and individualized diagnostic and complex therapeutic approach by an interdisciplinary physician's team. Medical staff's behaviour should be maximally sparing, when possible.

New advances in image diagnostic modalities such as positron emission tomography/computer-aided tomography, steadily improved surgical and microsurgical techniques such as laparoscopic resections along with emerging opportunities for costeffective chemotherapy and multimodal management promise better perspectives in this field of permanently rising social significance.

#### **9. References**

436 Colorectal Cancer – From Prevention to Patient Care

Hompes et al., 2011). Portal vein embolization, radiofrequency ablation, two-stage hepatectomy, conversion therapy and reverse treatment strategy along with hepatectomy are used in the presence of extrahepatic disease (Coimbra et al. 2011; Narita et al., 2011, Tsim et al., 2011). Resection of advanced colorectal liver metastases after a second-line chemotherapy regimen is safe and promising in certain cases. The addition of neoadjuvant

Positron emission tomography/computed tomography have a higher accuracy for detection of extra-hepatic and colorectal liver metastases than computed tomography alone (Patel et al., 2011). In patients treated with neoadjuvant chemotherapy, magnetic resonance imaging measurements of steatosis show the highest correlation coefficient and the best diagnostic accuracy, as compared to computed tomography ones (Marsman et al., 2011). Intraoperative ultrasound and preoperative imaging significantly increase the diagnostic accuracy of patients undergoing liver resection for colorectal liver metastases (Lordan et al., 2011). Metachronous resections have a better outcome than synchronous. Iterative resection is very encouraging and justifies an aggressive surgical approach (Tonelli et al., 2010). Simultaneous resection is safe and efficient in the treatment of patients with synchronous colorectal liver metastases while avoiding a second major operation (Chen et al., 2011). In patients with bilobar synchronous colorectal liver metastases who are candidates for twostage hepatectomy, combined resection of the primary tumour and first-stage he patectomy reduces the number of procedures, optimizes chemotherapy administration and may improve outcome (Karoui et al., 2010). The two-stage strategy for colorectal liver metastases can be performed with acceptable morbidity and mortality. The second stage is not feasible in 20-25% of patients. Patients completing the two-stage approach may have long-term survival comparable to those treated with a planned single-stage hepatectomy (Tsai et al., 2010). Concomitant extrahepatic disease in a patient with colorectal liver metastases should

As there is no significant difference in morbidity, mortality, recurrence rate, or survival in anatomical and nonanatomical liver resections, the latter can be used as a save procedure to preserve liver parenchyma (Lalmahomed et al., 2011). The Pringle maneuver does not seem to affect the survival of patients with liver metastases (Ferrero et al., 2010). Ultrasoundguided finger compression of sectional portal pedicle feeding the right posterior section is a feasible, safe, and effective method for performing anatomical right posterior sectionectomy

Prognostic factors and score systems occupy an important place in oncologic liver surgery (de Haas et al., 2011; Peng et al., 2011; Pulitanò et al., 2011). Although twelve prognostic scoring systems have been identified from 1996 to 2009, there is no 'ideal' system for the clinical management of patients with colorectal liver metastases (Gomez et al., 2010). A predicted positive surgical margin (R1 resection) is not any absolute contraindication to surgery for aggressive or advanced colorectal liver metastases (Tanaka et al., 2011). Liver resection has superior long-term survival which is, however, significantly reduced by the occurrence of post-surgical complications (Schepers et al., 2010). Superior overall healthrelated quality of life merits an aggressive surgical approach and intensive follow-up to

Based on our own results and reliable scientific evidence available worldwide up-to-date, it can be concluded that the patient presenting with colorectal liver metastases deserves a

chemotherapy should, however, be cost-effective.

not be a contraindication to their resection.

detect recurrence early (Wiering et al., 2011).

(Torzilli et al., 2011).

**8. Conclusion** 


Resection for Colorectal Liver Metastases 439

Liau, K.H.; Blumgart, L.H. & DeMatteo, R.P. (2004). Segment-oriented approach to liver

Lordan, J.T.; Stenson, K.M. & Karanjia, N.D. (2011). The value of intraoperative ultrasound

Marsman, H.A.; van der Pool, A.E.; Verheij, J.; Padmos, J.; Ten Kate, F.J.; Dwarkasing, R.S.;

Narita, M.; Oussoultzoglou, E.; Bachellier, P.; Rosso, E.; Pessaux, P. & Jaeck, D. (2011). Two-

Neumann, U.P.; Seehofer, D. & Neuhaus, P. (2010). The surgical treatment of hepatic

Patel, S.; McCall, M.; Ohinmaa, A.; Bigam, D. & Dryden, D.M. (2011). Positron emission

Peng, P.D.; van Vledder, M.G., Tsai, S.; de Jong, M.C.; Makary, M.; Ng, J.; Edil, B.H.;

Pulitanò, C.; Bodingbauer, M.; Aldrighetti, L.; de Jong, M.C.; Castillo, F.; Schulick, R.D.;

Schepers, A.; Mieog, S.; van de Burg, B.B.; van Schaik, J.; Liefers, G.J. & Marang-van de

Stillwell, A.P.; Ho, Y.H. & Veitch, C. (2011). Systematic review of prognostic factors related

Tanaka, K.; Nojiri, K.; Kumamoto, T.; Takeda, K. & Endo, I. (2011). R1 resection for

*Surgery*, Vol.253, No.4, (April 2011), pp. 666-671, ISSN 0003-4932

Vol.18, No.5, (July 2011), pp. 1380-1388, ISSN 1068-9265

Vol.37, No.4, (April 2011), pp. 336-343, ISSN 0748-7983

(November 2010), pp. e91-97, ISSN 0022-4804

Vol.93, No.3, (April 2011), pp. 246-249, ISSN 0035-8843

No.19, (May 2010), pp. 335-342 ISSN 1866-0452

ISSN 0039-6109

ISSN 0022-4790

pp. 121-126, ISSN 0253-4886

439-446, ISSN 1365-182X

0364-2131

resection. *Surgical Clinics of North America*, Vol.84, No.2 (April 2004), pp. 543-561,

and preoperative imaging, individually and in combination, in liver resection for metastatic colorectal cancer. Annals *of the Royal College of Surgeons of England*,

van Gulik, T.M.; Ijzermans, J.N. & Verhoef, C. (2011). Hepatic steatosis assessment with CT or MRI in patients with colorectal liver metastases after neoadjuvant chemotherapy. *Journal of Surgical Oncology*, Vol. 104, No.1, (July 2011), pp. 10-16,

stage hepatectomy procedure to treat initially unresectable multiple bilobar colorectal liver metastases: technical aspects. *Digestive Surgery*, Vol.28, No.2, (2011),

metastases in colorectal carcinoma. *Deutsches Ärzteblatt International*, Vol.107,

tomography/computed tomographic scans compared to computed tomographic scans for detecting colorectal liver metastases: a systematic review. *Annals of* 

Wolfgang, C.L.; Schulick, R.D.; Choti, M.A.; Kamel, I. & Pawlik, T.M. (2011). Sarcopenia negatively impacts short-term outcomes in patients undergoing hepatic resection for colorectal liver metastasis. *HPB* (*Oxford*), Vol.13, No.7, (July 2011), pp.

Parks, R.W.; Choti, M.A.; Wigmore, S.J.; Gruenberger, T. & Pawlik, T.M. (2011). Liver resection for colorectal metastasis in presence of extrahepatic disease: results from an international multi-institutional analysis. *Annals of Surgical Oncology*,

Mheen, P.J. (2010). Impact of complications after surgery for colorectal liver metastasis on patient survival. *The Journal of Surgical Research*, Vol.164, No.1,

to overall survival in patients with stage IV colorectal cancer and unresectable metastases. *World Journal of Surgery*, Vol.35, No.3, (March 2011), pp. 684-692, ISSN

aggressive or advanced colorectal liver metastases is justified in combination with effective prehepatectomy chemotherapy. *European Journal of Surgical Oncology*,

to surgical resection in the treatment of recurrent colorectal liver metastases. *Journal of Gastrointestinal Surgery*, Vol.15, No.7, (July 2011), pp. 1168-1172, ISSN 1091-255X


Hata, M.; Sakamoto, K.; Doneza, J.; Sumida, K.; Sugimoto, K.; Ishiyama, S.; Takahashi, M.;

Hompes, D.; Prevoo, W. & Ruers, T. (2011). Radiofrequency ablation as a treatment tool for

Isoniemi, H. & Osterlund, P. (2011). Surgery combined with oncological treatments in liver

Karoui, M.; Vigano, L.; Goyer, P.; Ferrero, A.; Luciani, A.; Aglietta, M.; Delbaldo, C.; Cirillo,

Kobakov, G. & Kostov, D. (2006). Colorectal liver metastases - general characteristics and

Kostov, D. & Kobakov, G. Eds. (2006a). *Colorectal Liver Metastases - Diagnosis and* 

Kostov, D. & Kobakov, G. (2006b). Surgical treatment of liver metastases, In: *Colorectal Liver* 

Kostov, D. & Kobakov, G. (2010) *A Colour Atlas of Liver Resections,* Marko Markov

Lai, E.C.; Tang, C.N.; Yang, G.P. & Li, M.K. (2011). Multimodality laparoscopic liver

Lalmahomed, Z.S.; Ayez, N.; van der Pool, A.E.; Verheij, J.; IJzermans, J.N. & Verhoef, C.

Vol.35, No.3, (March 2011), pp. 656-661, ISSN 0364-2131

Medical University of Varna, ISBN 954-9685-33-0, Varna: (in Bulgarian). Kostov, D. V. & Kobakov, G. L. (2009). Segmental liver resection for colorectal metastases.

1947-1954, ISSN 1068-9265

(2011), pp. 35-41 ISSN 1457-4969

9625

ISSN 1740-5025

1362, ISSN 0007-1323

453, ISSN 1841-8724

954-92432-1-5, Varna, Bulgaria

(2011), pp. 324-328 ISSN 1743-9191

Bulgarian)

33-0, Varna, (in Bulgarian)

to surgical resection in the treatment of recurrent colorectal liver metastases. *Journal of Gastrointestinal Surgery*, Vol.15, No.7, (July 2011), pp. 1168-1172, ISSN 1091-255X Hammill, C.W.; Billingsley, K.G.; Cassera, M.A.; Wolf, R.F.; Ujiki, M.B. & Hansen, P.D.

(2011). Outcome after laparoscopic radiofrequency ablation of technically resectable colorectal liver metastases. *Annals of Surgical Oncology*, Vol.18, No.7, (July 2011), pp.

Kojima, Y.; Tomiki, Y. & Machi, J. (2010)Improvement of long-term survival of colorectal cancer in Japanese-Americans of Hawaii from 1990 to 2001. *International Journal of Clinical Oncology*, Vol.15, No.6, (December 2010), pp. 559-564, ISSN 1341-

liver metastases of colorectal origin. *Cancer Imaging*, Vol.11, (March 2011), pp. 23-30

metastases from colorectal cancer. *Scandinavian Journal of Surgery*, Vol.100, No.1,

S.; Capussotti, L. & Cherqui, D. (2010). Combined first-stage hepatectomy and colorectal resection in a two-stage hepatectomy strategy for bilobar synchronous liver metastases. *British Journal of Surgery*, Vol.97, No.9, (September 2010), pp. 1354-

therapeutic opportunities, In: *Colorectal Liver Metastases - Diagnosis and Management,* D. Kostov & G. Kobakov (Eds.), 55-62, Medical University of Varna, ISBN 954-9685-

*Management,* Medical University of Varna, ISBN 954-9685-33-0, Varna, (in

*Metastases - Diagnosis and Management,* D. Kostov & G. Kobakov (Eds.), 63-87,

*Journal of Gastrointestinal and Liver Diseases,* Vol.18, No.4, (December 2009), pp. 447-

Interregional Dispensary and Hospital of Oncological Diseases of Varna, ISBN 978-

resection for hepatic malignancy - from conventional total laparoscopic approach to robot-assisted laparoscopic approach. *International Journal of Surgery*, Vol.9, No.4,

(2011). Anatomical versus nonanatomical resection of colorectal liver metastases: is there a difference in surgical and oncological outcome? *World Journal of Surgery*,


**22** 

*1Germany 2Syria* 

**Experimental Colorectal Cancer Liver Metastasis** 

*1Toxicology and Chemotherapy Unit, Heidelberg,* 

Rania B. Georges1,2, Hassan Adwan1 and Martin R. Berger1

*2Toxicology and Pharmacology Unit, Faculty of Pharmacy, Al-Baath University, Homs,* 

With estimated 1 080 000 diagnosed cases each year, which account for 1.1% of all deaths, colorectal carcinoma (CRC) ranks fourth in cancer-related deaths in both sexes worldwide (WHOSIS, 2008). In Europe, CRC is the third most lethal malignancy after lung and stomach

CRC progression is characterized by increased growth of the primary carcinoma as well as lymphatic and haematogenic spread. The liver is often the first vascular bed in which disseminating colorectal cancer cells are trapped and therefore is affected in up to 10-20% of CRC patients at the time of presentation (Berney, et al., 1998). Another 40-50% of patients will eventually develop liver metastasis during the course of their illness, which is commonly the cause of death (Bentrem, et al., 2005, Stangl, et al., 1994, Sugarbaker, 1990). At present, liver resection is considered the treatment of choice for suited patients with colorectal liver metastases, offering a five-year survival rate of 25-44% (Choti, et al., 2002, Garden, et al., 2006, Zacharias, et al., 2004) to those 20-25% of patients with isolated liver metastasis (Adson, et al., 1984, Bismuth, et al., 1996, Fong, et al., 1999). Unfortunately, this procedure is feasible only in patients with no signs of irresectable extra-hepatic disease, whereas the median survival is only 9–19 months for patients with unresectable disease who receive systemic chemotherapy (de Gramont, et al., 2000, Giacchetti, et al., 2000, Meyerhardt

However, the fact that CRC malignancy develops over a long period and can only be efficiently controlled if detected early provokes many efforts to better understand the neoplastic progression of this cancer. It is well known, that there is a continuous shedding of tumor cells from a primary CRC (Chambers, et al., 2002), but not all disseminated CRC cells develop into macrometastases. It was hypothesized that sub-populations of malignant cells evolve a genetic advantage to become "highly metastatic". These clones are skilled to dissociate from the primary cancer, to intravasate into nearby blood and lymphatic vessels, to travel through the lymphatic and hematogenous systems, to survive the immune surveillance, to extravasate into distant tissues forming micrometastases, and to eventually

In this cascade, the epithelial-mesenchymal transition (EMT), characterized by the loss of cell-to-cell adhesion and cell polarity (Thiery, 2003), plays a crucial role in different stages;

cancers in men, and it ranks second after breast cancer in women (WHOSIS, 2008).

**1. Introduction** 

and Mayer, 2005, Saltz, et al., 2000).

colonize the target organ.


### **Experimental Colorectal Cancer Liver Metastasis**

Rania B. Georges1,2, Hassan Adwan1 and Martin R. Berger1 *1Toxicology and Chemotherapy Unit, Heidelberg, 2Toxicology and Pharmacology Unit, Faculty of Pharmacy, Al-Baath University, Homs, 1Germany 2Syria* 

#### **1. Introduction**

440 Colorectal Cancer – From Prevention to Patient Care

Terminology Committee of the International Hepato-Pancreato-Biliary Association;

Tonelli, F.; Leo, F.; Nobili, S.; Mini, E. & Batignani, G. (2010). Prognostic factors in primary

Torzilli, G.; Procopio, F.; Donadon, M.; Palmisano, A.; Fabbro, D.D. & Montorsi, M. (2011).

Tsai, S.; Marques, H.P.; de Jong, M.C.; Mira, P.; Ribeiro, V.; Choti, M.A.; Schulick, R.D.;

Tsim, N.; Healey, A.J.; Frampton, A.E.; Habib, N.A.; Bansi, D.S.; Wasan, H.; Cleator, S.J.;

*Surgical Oncology*, Vol.18, No.7, (July 2011), pp. 1936-1946, ISSN 1068-9265 Tsoulfas, G.; Pramateftakis, M.G. & Kanellos, I. (2011). Surgical treatment of hepatic

Wiering, B.; Oyen, W.J.; Adang, E.M.; van der Sijp, J.R.; Roumen, R.M.; de Jong, K.P.; Ruers,

No.5, (October 2010), pp. 358-363, ISSN 1120-009X

No.1, (January 2011), pp. 1-9, ISSN 1948-5204

333-339, ISSN 1365-182X

95 ISSN 2038-131X

262-269, ISSN 1365-182X

565-571, ISSN 0007-1323

Strasberg, S.M.; Belghiti, J.; Claven, P.-A. et al. (2000). The Brisbane 2000 Terminology of Liver Anatomy and Resections. *HPB (Oxford)*, Vol.2, *(*2000), pp.

and iterative surgery of colorectal liver metastases. *Journal of Chemotherapy*, Vol.22,

Anatomical right posterior sectionectomy: a further expansion of the ultrasoundguided compression technique. *Updates in Surgery*, Vol.63, No.2, (June 2011), pp. 91-

Barroso, E. & Pawlik, T.M. (2010). Two-stage strategy for patients with extensive bilateral colorectal liver metastases. *HPB (Oxford)*, Vol.12, No.4, (May 2010), pp.

Stebbing, J.; Lowdell, C.P.; Jackson, J.E.; Tait, P. & Jiao, L.R. (2011). Two-stage resection for bilobar colorectal liver metastases: r0 resection is the key. *Annals of* 

metastases from colorectal cancer. *World Journal of Gastrointestinal Oncology*, Vol.3,

T.J. & Krabbe, P.F. (2011). Long-term global quality of life in patients treated for colorectal liver metastases. *British Journal of Surgery*, Vol.98, No.4, (April 2011), pp. With estimated 1 080 000 diagnosed cases each year, which account for 1.1% of all deaths, colorectal carcinoma (CRC) ranks fourth in cancer-related deaths in both sexes worldwide (WHOSIS, 2008). In Europe, CRC is the third most lethal malignancy after lung and stomach cancers in men, and it ranks second after breast cancer in women (WHOSIS, 2008).

CRC progression is characterized by increased growth of the primary carcinoma as well as lymphatic and haematogenic spread. The liver is often the first vascular bed in which disseminating colorectal cancer cells are trapped and therefore is affected in up to 10-20% of CRC patients at the time of presentation (Berney, et al., 1998). Another 40-50% of patients will eventually develop liver metastasis during the course of their illness, which is commonly the cause of death (Bentrem, et al., 2005, Stangl, et al., 1994, Sugarbaker, 1990). At present, liver resection is considered the treatment of choice for suited patients with colorectal liver metastases, offering a five-year survival rate of 25-44% (Choti, et al., 2002, Garden, et al., 2006, Zacharias, et al., 2004) to those 20-25% of patients with isolated liver metastasis (Adson, et al., 1984, Bismuth, et al., 1996, Fong, et al., 1999). Unfortunately, this procedure is feasible only in patients with no signs of irresectable extra-hepatic disease, whereas the median survival is only 9–19 months for patients with unresectable disease who receive systemic chemotherapy (de Gramont, et al., 2000, Giacchetti, et al., 2000, Meyerhardt and Mayer, 2005, Saltz, et al., 2000).

However, the fact that CRC malignancy develops over a long period and can only be efficiently controlled if detected early provokes many efforts to better understand the neoplastic progression of this cancer. It is well known, that there is a continuous shedding of tumor cells from a primary CRC (Chambers, et al., 2002), but not all disseminated CRC cells develop into macrometastases. It was hypothesized that sub-populations of malignant cells evolve a genetic advantage to become "highly metastatic". These clones are skilled to dissociate from the primary cancer, to intravasate into nearby blood and lymphatic vessels, to travel through the lymphatic and hematogenous systems, to survive the immune surveillance, to extravasate into distant tissues forming micrometastases, and to eventually colonize the target organ.

In this cascade, the epithelial-mesenchymal transition (EMT), characterized by the loss of cell-to-cell adhesion and cell polarity (Thiery, 2003), plays a crucial role in different stages;

Experimental Colorectal Cancer Liver Metastasis 443

Yeatman and Chambers, 2003). In colorectal cancer, gene profiling studies have identified a positive correlation between advanced or metastatic colon tumors and abundant OPN expression (Wai and Kuo, 2004). Increased OPN expression is associated with tumor invasion or metastasis in cancers of the breast (Tuck, et al., 1999, Tuck, et al., 1998, Tuck, et al., 1997), stomach (Ue, et al., 1998), lung (Chambers, et al., 1996, Shijubo, et al., 1999), prostate (Thalmann, et al., 1999), liver (Gotoh, et al., 2002, Pan, et al., 2003), and colon (Agrawal, et al., 2002, Yeatman and Chambers, 2003). Analysis of the OPN promoter has uncovered multiple

The main aim of our experimental series was first to generate a model suited to be used for investigating the efficacy of new drugs (Eyol, et al., 2008, Seelig, et al., 2004, Wittmer, et al., 1999). One of the few well-characterized animal models for hepatic colorectal cancer makes use of the rat CC531 cell line. Following topical injection of CC531 cells, liver metastases develop and their growth has been frequently used for studying effects of various anticancer treatments (Marinelli, et al., 1991, Oldenburg, et al., 1994, Veenhuizen, et al., 1996). A second aim was to identify metastasis-related changes in gene expression in tumor cells, which differ from those in the primary tumor and probably play a crucial role in metastasis formation. Therefore, temporal changes in gene expression of CRC cells homing to the liver have been investigated using the above *in vivo* model, which is characterized by a defined onset of metastatic proliferation in rat liver following intraportal inoculation of CC531 tumor cells. This, in turn, permits a close following of the time-dependent modulation of gene expression, as the tumor cells home into the liver and then grow to a lethal size. The technique of re-isolating these tumor cells from rat liver permitted to monitor for the first time the expression of several candidate genes in a time-dependent manner (Georges, et

Initially, the CC531 cell line was induced by treatment of WAG rats with 1,2 dimethylhydrazine (DMH). Forty weeks after 6 weekly injections of 30 mg/kg DMH, a carcinoma originated in the ascending colon of injected WAG rats. After serial implantation to male rats, the resulting transplantable tumor was described to be moderately

The majority of animal liver metastasis models available at that time was based on the subcutaneous or intraperitoneal injection of tumor cells (Venditti, et al., 1984). To imitate as closely as possible the physiological metastatic spread of colon cancer, different orthotopic models were developed. These included intraportal (Griffini, et al., 1997, Griffini, et al., 1996, Thomas, et al., 1993) and spleen injections (Fukumura, et al., 1997) resulting in a diffuse outgrowth of tumor cells in the liver, as well as implantation of tumor tissue fragments or cells under the Glisson's liver capsule giving a local, limited, and nodal growth pattern (Aguiar, et al., 1987, Bartkowski, et al., 1986, Kamphorst, et al., 1999). Quantification of tumor growth in the latter models is often done by measuring tumor diameters (Aguiar, et al., 1987, Bartkowski, et al., 1986, Kamphorst, et al., 1999), whereas the diffuse models couldn't be quantified easily. To encounter this obstacle, many efforts have been made, including the 3-D reconstruction of metastases by consecutive serial sections (Griffini, et al., 1997), counting tumor nodules in the liver (Thomas, et al., 1993), or the use of tumor specific

consensus binding sites for known transcription factors (Hijiya, et al., 1994).

al., 2011).

**2. Description of the CC531 rat model** 

differentiated on histological examination (Marquet, et al., 1984).

**2.1 Generation of the model** 

namely the dissemination of tumor cells as well as their intra- and extra-vasation (Gupta and Massague, 2006). Several known transcription factors, such as Snail, Slug, and Twist, were found to induce EMT on one hand and were implicated with tumor progression and metastasis on the other hand. In line with this, some downstream genes of these regulatory factors are responsible for cell-to-cell adhesion, specially E-cadherin and claudins. For example, up-regulation of Snail induces EMT and down-regulates the transcription of different tight junctions proteins (TJs), such as claudins and occludin (Findley and Koval, 2009). Claudins (CLDNs) form the structural backbone of TJs, and comprise at least 27 members of integral transmembrane proteins (Mineta, et al., 2011) ranging in size between 20-27 kDa (Tsukita, et al., 2001). Recently, the altered expression of various claudins has been implicated in the progression of several human cancers (Cheung, et al., 2005, Hough, et al., 2000, Johnson, et al., 2005, Kominsky, et al., 2003, Long, et al., 2001, Michl, et al., 2001, Morin, 2005, Sanada, et al., 2006, Swisshelm, et al., 2005). In contrast to the published notion that claudin expression would decrease from tumorigenesis as tight junctions are lost during cellular transformation, the claudin status seems to change in a tissue-specific manner. For example, over-expression of *Cldn2* has been correlated to colorectal cancer (Kinugasa, et al., 2007), whereas decreased *Cldn7* expression has been reported in head and neck cancer (Usami, et al., 2006), invasive ductal breast carcinoma (Kominsky, et al., 2003), and metastatic breast cancer (Sauer, et al., 2005). In addition, *Cldn3* and *Cldn4* have been found repeatedly elevated in a variety of cancers including pancreatic ductal adenocarcinoma (Michl, et al., 2003) as well as ovarian, uterine, prostate, and breast cancers (Rangel, et al., 2003). In partial contrast, reduced expression of *Cldn4* and *Cldn5* was detected in hepatocellular and renal carcinomas (Soini, 2005). In CRC, both, up- and down-regulation of claudin4 expression have been described (de Oliveira, et al., 2005, Ueda, et al., 2007), as well as aberrant expression of *Cldn1.* 

Another type of cell connection has been named cell-to-extracellular matrix (ECM) contacts. On their disruption, they are presumably also implicated in tumor initiation. It is well known from each stage of malignant progression that tumor cells communicate with their microenvironment and thereby elicit responses from it. This microenvironment is mainly composed of tumor cells, extracellular matrix (ECM), stromal cells, immune cells and microvessels (Farrow, et al., 2008, Jung, et al., 2002). The ECM is a scaffold of extracellular proteins that maintain tissue shape and provide the cellular compartment with structural support (Bosman and Stamenkovic, 2003). However, by influencing cell adhesion, migration, differentiation, proliferation and survival, the ECM is a remodeling network that contributes substantially to tumor progression and metastasis (Engbring and Kleinman, 2003, Ioachim, et al., 2002). Remodeling and deposition of the ECM is mostly regulated by a functional family of extracellular proteins known as matricellular proteins, which contribute to the structural integrity and composition of the ECM (Bornstein and Sage, 2002). One of the most important characteristics of matricellular proteins is their ability to manipulate the integration and turn-over of ECM (Bornstein and Sage, 2002, Kyriakides, et al., 2001). Furthermore, by playing a linker role between the ECM and the cell surface, matricellular proteins can also direct cell fate, survival, adhesion and motility (Bornstein and Sage, 2002, Brekken and Sage, 2000, Kyriakides, et al., 2001).

Osteopontin (OPN) is an acidic extracellular matrix phosphoprotein of 298-amino acids secreted by a wide variety of cancers, which functionally favours tumor progression (Gao, et al., 2003, Weber, 2001). The secreted phosphoprotein binds to the integrins (e.g. ITG-αvβ3 or ITG-αvβ5) and CD44 families of receptors to propagate cellular signals (Agrawal, et al., 2002,

namely the dissemination of tumor cells as well as their intra- and extra-vasation (Gupta and Massague, 2006). Several known transcription factors, such as Snail, Slug, and Twist, were found to induce EMT on one hand and were implicated with tumor progression and metastasis on the other hand. In line with this, some downstream genes of these regulatory factors are responsible for cell-to-cell adhesion, specially E-cadherin and claudins. For example, up-regulation of Snail induces EMT and down-regulates the transcription of different tight junctions proteins (TJs), such as claudins and occludin (Findley and Koval, 2009). Claudins (CLDNs) form the structural backbone of TJs, and comprise at least 27 members of integral transmembrane proteins (Mineta, et al., 2011) ranging in size between 20-27 kDa (Tsukita, et al., 2001). Recently, the altered expression of various claudins has been implicated in the progression of several human cancers (Cheung, et al., 2005, Hough, et al., 2000, Johnson, et al., 2005, Kominsky, et al., 2003, Long, et al., 2001, Michl, et al., 2001, Morin, 2005, Sanada, et al., 2006, Swisshelm, et al., 2005). In contrast to the published notion that claudin expression would decrease from tumorigenesis as tight junctions are lost during cellular transformation, the claudin status seems to change in a tissue-specific manner. For example, over-expression of *Cldn2* has been correlated to colorectal cancer (Kinugasa, et al., 2007), whereas decreased *Cldn7* expression has been reported in head and neck cancer (Usami, et al., 2006), invasive ductal breast carcinoma (Kominsky, et al., 2003), and metastatic breast cancer (Sauer, et al., 2005). In addition, *Cldn3* and *Cldn4* have been found repeatedly elevated in a variety of cancers including pancreatic ductal adenocarcinoma (Michl, et al., 2003) as well as ovarian, uterine, prostate, and breast cancers (Rangel, et al., 2003). In partial contrast, reduced expression of *Cldn4* and *Cldn5* was detected in hepatocellular and renal carcinomas (Soini, 2005). In CRC, both, up- and down-regulation of claudin4 expression have been described (de Oliveira, et al., 2005, Ueda, et al., 2007), as

Another type of cell connection has been named cell-to-extracellular matrix (ECM) contacts. On their disruption, they are presumably also implicated in tumor initiation. It is well known from each stage of malignant progression that tumor cells communicate with their microenvironment and thereby elicit responses from it. This microenvironment is mainly composed of tumor cells, extracellular matrix (ECM), stromal cells, immune cells and microvessels (Farrow, et al., 2008, Jung, et al., 2002). The ECM is a scaffold of extracellular proteins that maintain tissue shape and provide the cellular compartment with structural support (Bosman and Stamenkovic, 2003). However, by influencing cell adhesion, migration, differentiation, proliferation and survival, the ECM is a remodeling network that contributes substantially to tumor progression and metastasis (Engbring and Kleinman, 2003, Ioachim, et al., 2002). Remodeling and deposition of the ECM is mostly regulated by a functional family of extracellular proteins known as matricellular proteins, which contribute to the structural integrity and composition of the ECM (Bornstein and Sage, 2002). One of the most important characteristics of matricellular proteins is their ability to manipulate the integration and turn-over of ECM (Bornstein and Sage, 2002, Kyriakides, et al., 2001). Furthermore, by playing a linker role between the ECM and the cell surface, matricellular proteins can also direct cell fate, survival, adhesion and motility (Bornstein and Sage, 2002,

Osteopontin (OPN) is an acidic extracellular matrix phosphoprotein of 298-amino acids secreted by a wide variety of cancers, which functionally favours tumor progression (Gao, et al., 2003, Weber, 2001). The secreted phosphoprotein binds to the integrins (e.g. ITG-αvβ3 or ITG-αvβ5) and CD44 families of receptors to propagate cellular signals (Agrawal, et al., 2002,

well as aberrant expression of *Cldn1.* 

Brekken and Sage, 2000, Kyriakides, et al., 2001).

Yeatman and Chambers, 2003). In colorectal cancer, gene profiling studies have identified a positive correlation between advanced or metastatic colon tumors and abundant OPN expression (Wai and Kuo, 2004). Increased OPN expression is associated with tumor invasion or metastasis in cancers of the breast (Tuck, et al., 1999, Tuck, et al., 1998, Tuck, et al., 1997), stomach (Ue, et al., 1998), lung (Chambers, et al., 1996, Shijubo, et al., 1999), prostate (Thalmann, et al., 1999), liver (Gotoh, et al., 2002, Pan, et al., 2003), and colon (Agrawal, et al., 2002, Yeatman and Chambers, 2003). Analysis of the OPN promoter has uncovered multiple consensus binding sites for known transcription factors (Hijiya, et al., 1994).

The main aim of our experimental series was first to generate a model suited to be used for investigating the efficacy of new drugs (Eyol, et al., 2008, Seelig, et al., 2004, Wittmer, et al., 1999). One of the few well-characterized animal models for hepatic colorectal cancer makes use of the rat CC531 cell line. Following topical injection of CC531 cells, liver metastases develop and their growth has been frequently used for studying effects of various anticancer treatments (Marinelli, et al., 1991, Oldenburg, et al., 1994, Veenhuizen, et al., 1996).

A second aim was to identify metastasis-related changes in gene expression in tumor cells, which differ from those in the primary tumor and probably play a crucial role in metastasis formation. Therefore, temporal changes in gene expression of CRC cells homing to the liver have been investigated using the above *in vivo* model, which is characterized by a defined onset of metastatic proliferation in rat liver following intraportal inoculation of CC531 tumor cells. This, in turn, permits a close following of the time-dependent modulation of gene expression, as the tumor cells home into the liver and then grow to a lethal size. The technique of re-isolating these tumor cells from rat liver permitted to monitor for the first time the expression of several candidate genes in a time-dependent manner (Georges, et al., 2011).

#### **2. Description of the CC531 rat model**

#### **2.1 Generation of the model**

Initially, the CC531 cell line was induced by treatment of WAG rats with 1,2 dimethylhydrazine (DMH). Forty weeks after 6 weekly injections of 30 mg/kg DMH, a carcinoma originated in the ascending colon of injected WAG rats. After serial implantation to male rats, the resulting transplantable tumor was described to be moderately differentiated on histological examination (Marquet, et al., 1984).

The majority of animal liver metastasis models available at that time was based on the subcutaneous or intraperitoneal injection of tumor cells (Venditti, et al., 1984). To imitate as closely as possible the physiological metastatic spread of colon cancer, different orthotopic models were developed. These included intraportal (Griffini, et al., 1997, Griffini, et al., 1996, Thomas, et al., 1993) and spleen injections (Fukumura, et al., 1997) resulting in a diffuse outgrowth of tumor cells in the liver, as well as implantation of tumor tissue fragments or cells under the Glisson's liver capsule giving a local, limited, and nodal growth pattern (Aguiar, et al., 1987, Bartkowski, et al., 1986, Kamphorst, et al., 1999). Quantification of tumor growth in the latter models is often done by measuring tumor diameters (Aguiar, et al., 1987, Bartkowski, et al., 1986, Kamphorst, et al., 1999), whereas the diffuse models couldn't be quantified easily. To encounter this obstacle, many efforts have been made, including the 3-D reconstruction of metastases by consecutive serial sections (Griffini, et al., 1997), counting tumor nodules in the liver (Thomas, et al., 1993), or the use of tumor specific

Experimental Colorectal Cancer Liver Metastasis 445

Fig. 2. Comparison of therapeutic efficacy: The mean tumor cell number of treated groups is

Thus, the comparison of HACE with 5-FU or Gem showed that the efficacy of Gem in reducing the hepatic tumor cell load was significantly higher and its therapeutic ratio was

In a subsequent experiment, the effect of HACE with irinotecan was compared vs. 5-FU as a standard agent in rat liver metastasis (Saenger, et al., 2004). Briefly, 4 × 106 CC531-lac-Z cells were intraportally injected into male Wag/Rij rats. Irinotecan (10, 30 and 60 mg/kg) and 5- FU (40, 60 and 90 mg/kg) were administered concomitantly with DSM (30 mg/kg) for temporary embolization. The tumor cell load was determined quantitatively using the

HACE with irinotecan induced a complete remission in 44% of the animals and the highest dose reduced the mean tumor cell load by 66% (P<0.001). In contrast, the highest dose of 5-FU caused a reduction of only 18% (P = 0.026) and altogether 23% complete remissions were observed in response to 5-FU (Table 1 and Fig. 3). Collectively, HACE with irinotecan had a greater effect

than that of HACE with 5-FU, setting the basis for further investigation in clinical trials.

Table 1. Results of hepatic arterial chemoembolization on tumor cell reduction

given in percent of the respective control group (T/C\* 100)

chemoluminescence assay mentioned above.

greater than that of 5-FU.

antibodies for immunohistological examination (Thomas, et al., 1993). These assays are protracted and allow only a gross grading of tumor mass or cytostatic-induced loss of tumor load in the liver. The use of reporter genes such as green fluorescent protein (GFP), luciferase, or *β*-galactosidase (GLB1) for tracing tumor cells has greatly facilitated both, quantification and localization at the single cell level (Chishima, et al., 1997, Dooley, et al., 1993, Zhang, et al., 1994). In view of that, we aimed in our study to develop an orthotopic, diffusely growing liver metastasis model that can be used for diagnostic and therapeutic studies. For this purpose the CC531 rat colorectal cancer cell line with its natural homing into the liver was transfected by the *Glb1* gene.

The stable transfection of these cells with the *Glb1* gene (Fig. 1) allowed quantitation of the tumor cell load at any time after implantation and hence a quick evaluation of the efficacy of therapy that can be used for diagnostic and therapeutic studies (Wittmer, et al., 1999).

Fig. 1. CC531 cells growing in RPMI-1640 medium were stained by the activity of GLB1 converting X-gal, a chromogenic substrate for GLB1. Parental cells **(a)** are compared with transfected **(b)** and subcloned **(c)** CC531 cells. The magnification is identical for all three photographs (see scale bar)

#### **2.2 Chemosensitivity of the model**

#### **2.2.1 Chemoembolization of rat liver metastasis with microspheres and gemcitabine or irinotecan followed by evaluation of tumor cell load by chemiluminescence**

These experiments (Seelig, et al., 2004) were performed to determine the potential of hepatic artery chemoembolization (HACE) for reducing the tumor cell load. Seven days after the intraportal injection of CC531-lac-Z cells to male WAG/Rij rats, tumor positive animals were treated by intra-hepatic artery injection with solvent (n=17), degradable starch microspheres (DSM, 30 mg/kg; n=16), DSM plus 5-fluorouracil (5-FU; 90, 60, and 40 mg/kg; ntotal=43) or DSM plus gemcitabine (Gem; 100, 80, 50, and 10 mg/kg; ntotal=46). After 3 more weeks the experiment was terminated, the livers were weighed and the number of CC531 lac-Z cells per liver was determined. Injection of DSM reduced the tumor cell load by 21% (T/C%=79), whereas the combination with 5-FU reduced tumor cell number more intensively at 60 mg/kg (T/C%=86), and 90 mg/kg (T/C%=19). None of these effects was significantly different from controls. The combination of DSM plus Gem was well tolerated and significantly (p<0.05) effective at 80, 50 and 10 mg/kg (T/C%= 16, 9 and 26, respectively; Fig. 2).

antibodies for immunohistological examination (Thomas, et al., 1993). These assays are protracted and allow only a gross grading of tumor mass or cytostatic-induced loss of tumor load in the liver. The use of reporter genes such as green fluorescent protein (GFP), luciferase, or *β*-galactosidase (GLB1) for tracing tumor cells has greatly facilitated both, quantification and localization at the single cell level (Chishima, et al., 1997, Dooley, et al., 1993, Zhang, et al., 1994). In view of that, we aimed in our study to develop an orthotopic, diffusely growing liver metastasis model that can be used for diagnostic and therapeutic studies. For this purpose the CC531 rat colorectal cancer cell line with its natural homing

The stable transfection of these cells with the *Glb1* gene (Fig. 1) allowed quantitation of the tumor cell load at any time after implantation and hence a quick evaluation of the efficacy of therapy that can be used for diagnostic and therapeutic studies (Wittmer, et al., 1999).

Fig. 1. CC531 cells growing in RPMI-1640 medium were stained by the activity of GLB1 converting X-gal, a chromogenic substrate for GLB1. Parental cells **(a)** are compared with transfected **(b)** and subcloned **(c)** CC531 cells. The magnification is identical for all three

**2.2.1 Chemoembolization of rat liver metastasis with microspheres and gemcitabine or irinotecan followed by evaluation of tumor cell load by chemiluminescence** 

These experiments (Seelig, et al., 2004) were performed to determine the potential of hepatic artery chemoembolization (HACE) for reducing the tumor cell load. Seven days after the intraportal injection of CC531-lac-Z cells to male WAG/Rij rats, tumor positive animals were treated by intra-hepatic artery injection with solvent (n=17), degradable starch microspheres (DSM, 30 mg/kg; n=16), DSM plus 5-fluorouracil (5-FU; 90, 60, and 40 mg/kg; ntotal=43) or DSM plus gemcitabine (Gem; 100, 80, 50, and 10 mg/kg; ntotal=46). After 3 more weeks the experiment was terminated, the livers were weighed and the number of CC531 lac-Z cells per liver was determined. Injection of DSM reduced the tumor cell load by 21% (T/C%=79), whereas the combination with 5-FU reduced tumor cell number more intensively at 60 mg/kg (T/C%=86), and 90 mg/kg (T/C%=19). None of these effects was significantly different from controls. The combination of DSM plus Gem was well tolerated and significantly (p<0.05) effective at 80, 50 and 10 mg/kg (T/C%= 16, 9 and 26,

into the liver was transfected by the *Glb1* gene.

photographs (see scale bar)

respectively; Fig. 2).

**2.2 Chemosensitivity of the model** 

Thus, the comparison of HACE with 5-FU or Gem showed that the efficacy of Gem in reducing the hepatic tumor cell load was significantly higher and its therapeutic ratio was greater than that of 5-FU.

In a subsequent experiment, the effect of HACE with irinotecan was compared vs. 5-FU as a standard agent in rat liver metastasis (Saenger, et al., 2004). Briefly, 4 × 106 CC531-lac-Z cells were intraportally injected into male Wag/Rij rats. Irinotecan (10, 30 and 60 mg/kg) and 5- FU (40, 60 and 90 mg/kg) were administered concomitantly with DSM (30 mg/kg) for temporary embolization. The tumor cell load was determined quantitatively using the chemoluminescence assay mentioned above.

HACE with irinotecan induced a complete remission in 44% of the animals and the highest dose reduced the mean tumor cell load by 66% (P<0.001). In contrast, the highest dose of 5-FU caused a reduction of only 18% (P = 0.026) and altogether 23% complete remissions were observed in response to 5-FU (Table 1 and Fig. 3). Collectively, HACE with irinotecan had a greater effect than that of HACE with 5-FU, setting the basis for further investigation in clinical trials.


Table 1. Results of hepatic arterial chemoembolization on tumor cell reduction

Experimental Colorectal Cancer Liver Metastasis 447

Fig. 4. Summary of treatment effects; columns denote the respective therapeutic efficacy as quotient of treated and control values (T/C%) for **(a)** tumor cell number determined by βgalactosidase assay and **(b)** wet liver weight. White columns indicate the expected

method and were fractionated to produce an average size of 75 µm. The DEBs were loaded with the desired concentration of either doxorubicin hydrochloride or irinotecan hydrochloride prior to administration by immersion in the drug solution, yielding basically

Fig. 5. Aspect of loaded and unloaded beads **(a)** Unloaded Beads; **(b)** Irinotecan DEB;

combination effect. Bars symbolize standard error of the mean

100% loading efficiency (Fig. 5).

**(c)** Doxorubicin DEB

Fig. 3. The columns give the percentage of complete remissions (defined as chemoluminescence signal below that of a healthy control liver) in relation to the respective treatment group

#### **2.2.2 Combination treatment of CC531-lac-Z rat liver metastases by chemoembolization with pemetrexed disodium and gemcitabine**

The aim here was to evaluate the combination effect of pemetrexed disodium (MTA; Alimta; LY 231514) and gemcitabine (GEM) administered by hepatic artery and portal vein chemoembolization (HACE and PVCE) in our rat liver metastasis model (Rodenbach, et al., 2005). After implantation of CC531 cells, MTA (30, 60 and 90 mg/kg) was administered locoregionally by PVCE and compared with repeated systemic intravenous injection. GEM (50 mg/kg) was also given locoregionally by HACE as well as systemically. All routes of administration were examined alone as well as in combination. Efficacy of treatment in terms of liver metastases burden was determined with the chemoluminescence assay. Locoregional administration by HACE with GEM was significantly more effective than systemic intravenous bolus treatment (P=0.03). Repeated systemic treatment with MTA yielded a slight reduction in tumor cell load that was significant *vs.* control at the medium and high doses (60 mg/kg, P=0.009; 90 mg/kg, P=0.046) but not *vs.* PVCE. The combination treatment of systemic (60 and 90 mg/kg) or locoregional (60 mg/kg) MTA with HACE using GEM (50 mg/kg) resulted in >80% tumor growth inhibition; this antineoplastic combination effect was maximally additive (Fig. 4). HACE with GEM was superior to systemic intravenous bolus treatment, while PVCE with MTA was ineffective. The optimal *in vivo* regimen of MTA (intravenous or PVCE) preceding GEM (HACE) resulted in a maximally additive tumor growth inhibition indicating that MTA and GEM can successfully be combined and favor further evaluation in patients.

#### **2.2.3 Chemoembolisation of rat colorectal liver metastases with drug eluting beads loaded with irinotecan or doxorubicin**

Chemoembolisation with drug eluting beads (DEBs) designed to deliver drug at the target over a prolonged period was tested as a new strategy to reduce the tumor burden of liver metastases (Eyol, et al., 2008). Accordingly, DEBs possessing anionic groups capable of ionically complexing with cationic drugs were synthesized by a suspension polymerization

Fig. 3. The columns give the percentage of complete remissions (defined as

**2.2.2 Combination treatment of CC531-lac-Z rat liver metastases by chemoembolization with pemetrexed disodium and gemcitabine** 

be combined and favor further evaluation in patients.

**loaded with irinotecan or doxorubicin** 

treatment group

chemoluminescence signal below that of a healthy control liver) in relation to the respective

The aim here was to evaluate the combination effect of pemetrexed disodium (MTA; Alimta; LY 231514) and gemcitabine (GEM) administered by hepatic artery and portal vein chemoembolization (HACE and PVCE) in our rat liver metastasis model (Rodenbach, et al., 2005). After implantation of CC531 cells, MTA (30, 60 and 90 mg/kg) was administered locoregionally by PVCE and compared with repeated systemic intravenous injection. GEM (50 mg/kg) was also given locoregionally by HACE as well as systemically. All routes of administration were examined alone as well as in combination. Efficacy of treatment in terms of liver metastases burden was determined with the chemoluminescence assay. Locoregional administration by HACE with GEM was significantly more effective than systemic intravenous bolus treatment (P=0.03). Repeated systemic treatment with MTA yielded a slight reduction in tumor cell load that was significant *vs.* control at the medium and high doses (60 mg/kg, P=0.009; 90 mg/kg, P=0.046) but not *vs.* PVCE. The combination treatment of systemic (60 and 90 mg/kg) or locoregional (60 mg/kg) MTA with HACE using GEM (50 mg/kg) resulted in >80% tumor growth inhibition; this antineoplastic combination effect was maximally additive (Fig. 4). HACE with GEM was superior to systemic intravenous bolus treatment, while PVCE with MTA was ineffective. The optimal *in vivo* regimen of MTA (intravenous or PVCE) preceding GEM (HACE) resulted in a maximally additive tumor growth inhibition indicating that MTA and GEM can successfully

**2.2.3 Chemoembolisation of rat colorectal liver metastases with drug eluting beads** 

Chemoembolisation with drug eluting beads (DEBs) designed to deliver drug at the target over a prolonged period was tested as a new strategy to reduce the tumor burden of liver metastases (Eyol, et al., 2008). Accordingly, DEBs possessing anionic groups capable of ionically complexing with cationic drugs were synthesized by a suspension polymerization

Fig. 4. Summary of treatment effects; columns denote the respective therapeutic efficacy as quotient of treated and control values (T/C%) for **(a)** tumor cell number determined by βgalactosidase assay and **(b)** wet liver weight. White columns indicate the expected combination effect. Bars symbolize standard error of the mean

method and were fractionated to produce an average size of 75 µm. The DEBs were loaded with the desired concentration of either doxorubicin hydrochloride or irinotecan hydrochloride prior to administration by immersion in the drug solution, yielding basically 100% loading efficiency (Fig. 5).

Fig. 5. Aspect of loaded and unloaded beads **(a)** Unloaded Beads; **(b)** Irinotecan DEB; **(c)** Doxorubicin DEB

Experimental Colorectal Cancer Liver Metastasis 449

Fig. 7. **(A-E).** Photographs of rat liver taken at 3, 6, 9, 14 and 21 days after inoculation of

As mentioned before, the rats were kept for 3, 6, 9, 13 and 21 days after tumor cell implantation (Georges, et al., 2011). Then, the abdominal cavity was opened and a 22 G cannula was inserted into the portal vein, through which the liver was perfused with HBSS medium (20 ml/min, 37° C for 10 min). This medium was replaced with pre-warmed perfusion medium [125 ml HBSS containing CaCl2 1M, 0.1% pronase, 100 mg collagenase Type IV (Serva, Heidelberg, Germany), 37 ° C, for the following 10 min] to digest connective tissues. After getting the cells in suspension, they were filtered through a sterile filter (Cell strainer, 70 µm Nylon, BD, Germany) and centrifuged. The resulting cell suspension of liver and tumor cells was transferred into 50ml-tubes and layered carefully onto a Ficoll gradient

Fig. 8. The density plot illustrates the CC531 population which was obtained by FACS sorting. The marker protein RFP was used for isolating CC531 cells without contaminating liver cells

**3.1 The re-isolation technique of tumor cells, hepatocytes and Kupffer cells** 

CC531 cells, before re-isolation of the metastatic tumor cells

medium (Amersham pharmacia Biotech AB, Uppsala, Sweden).

After injection of CC531 cells as previously mentioned, DEBs loaded with irinotecan or doxorubicin were administered by single injection into the hepatic artery. The resulting reduction in liver tumor burden and the corresponding reduction in liver weight indicated significant anticancer activity (Fig. 6).

Comparing the two agents, irinotecan appeared more advantageous because of its significant activity and excellent tolerability following administration at 2 dosages of either 20 or 30 mg/kg. Doxorubicin showed a narrower activity window, being effective at 4 mg/kg but ineffective at the lower dose of 2 mg/kg. Therefore, HACE with DEBs with either agent may have potential for treating patients with colorectal liver metastasis.

Fig. 6. Gross pathological aspects of implanted control and treated livers using irinotecan DEB (top line) and doxorubicin DEB (bottom line)

#### **3. Search for genes that are involved in colorectal cancer liver metastasis**

To identify genes that are involved in the metastatic phenotype of CC531 cells, cDNA microarrays were used to analyze mRNA expression profiles of these cells for changes related to their homing into the liver. Briefly, CC531 cells were intraportally implanted into the liver of Wag-Rij rats and re-isolated after 3, 6, 9, 14 and 21 days (Fig. 7 (A-E)). For the reisolation purposes, the CC531 cells had been marked with stains for viable cells *i.e.* eGFP and RFP markers.

After injection of CC531 cells as previously mentioned, DEBs loaded with irinotecan or doxorubicin were administered by single injection into the hepatic artery. The resulting reduction in liver tumor burden and the corresponding reduction in liver weight indicated

Comparing the two agents, irinotecan appeared more advantageous because of its significant activity and excellent tolerability following administration at 2 dosages of either 20 or 30 mg/kg. Doxorubicin showed a narrower activity window, being effective at 4 mg/kg but ineffective at the lower dose of 2 mg/kg. Therefore, HACE with DEBs with

either agent may have potential for treating patients with colorectal liver metastasis.

Fig. 6. Gross pathological aspects of implanted control and treated livers using irinotecan

**3. Search for genes that are involved in colorectal cancer liver metastasis** 

To identify genes that are involved in the metastatic phenotype of CC531 cells, cDNA microarrays were used to analyze mRNA expression profiles of these cells for changes related to their homing into the liver. Briefly, CC531 cells were intraportally implanted into the liver of Wag-Rij rats and re-isolated after 3, 6, 9, 14 and 21 days (Fig. 7 (A-E)). For the reisolation purposes, the CC531 cells had been marked with stains for viable cells *i.e.* eGFP

DEB (top line) and doxorubicin DEB (bottom line)

and RFP markers.

significant anticancer activity (Fig. 6).

Fig. 7. **(A-E).** Photographs of rat liver taken at 3, 6, 9, 14 and 21 days after inoculation of CC531 cells, before re-isolation of the metastatic tumor cells

#### **3.1 The re-isolation technique of tumor cells, hepatocytes and Kupffer cells**

As mentioned before, the rats were kept for 3, 6, 9, 13 and 21 days after tumor cell implantation (Georges, et al., 2011). Then, the abdominal cavity was opened and a 22 G cannula was inserted into the portal vein, through which the liver was perfused with HBSS medium (20 ml/min, 37° C for 10 min). This medium was replaced with pre-warmed perfusion medium [125 ml HBSS containing CaCl2 1M, 0.1% pronase, 100 mg collagenase Type IV (Serva, Heidelberg, Germany), 37 ° C, for the following 10 min] to digest connective tissues. After getting the cells in suspension, they were filtered through a sterile filter (Cell strainer, 70 µm Nylon, BD, Germany) and centrifuged. The resulting cell suspension of liver and tumor cells was transferred into 50ml-tubes and layered carefully onto a Ficoll gradient medium (Amersham pharmacia Biotech AB, Uppsala, Sweden).

Fig. 8. The density plot illustrates the CC531 population which was obtained by FACS sorting. The marker protein RFP was used for isolating CC531 cells without contaminating liver cells

Experimental Colorectal Cancer Liver Metastasis 451

The cDNA microarray results showed that compared to control CC531 cells, claudin1 and

Interestingly, both genes were at first down-regulated with a nadir (8 or 11 fold downregulation) on day 6, followed by gradual up-regulation within the observation period. These results were confirmed with RT-PCR (maximum down-regulation of 80% on day 6 for both genes; Fig. 9B) and Western blot (specific bands below detection limit, >90% inhibition on day 6; Fig. 9C). It is noteworthy that the transcription repressor gene *Snail* showed an inverse modulation: an increased expression during the first week (up to 3.8 fold) with the peak of its expression corresponding to the nadir of *Cldn1* and *Cldn4* down-regulation

Next, two experiments were performed to explain the initial down-regulation of *Cldn1* and *Cldn4*; these included co-culture of CC531 cells with isolated rat hepatocytes or Kupffer cells

Briefly, this model is based on a two-compartment system in which hepatocytes or Kupffer cells, plated in the lower compartment, are co-cultured with CC531 tumor cells growing in the upper compartment, with the two cell types being separated by a porous membrane (0.4 µm pore size). This system, preventing a direct contact between the two compartments, allows the cells to be only indirectly influenced by molecules secreted from the cells in the

No down-regulation effect on claudin expression was noticed, whereas both genes were up-

Fig. 10. Expression of claudins (1, 4) in CC531 cells co-cultured for 24 to 72 h with Kupffer cells (KCs) and hepatocytes (HCs) in comparison to the housekeeping gene ү-tubulin as

2x106 CC531 cells were seeded in 25 cm2 flat-bottom flasks or into round 50 ml glass bottles (Steiner GmbH, Siegen Eiserfeld, Germany), which were rotated on a roller (Stovall Life Science Incorporated, Greensboro, NC USA) at a speed of 1rpm, preventing the cells from adhesion to each other and onto the flask bottom. After 24 h, the cells in flat flasks and half

claudin4 were among the ≥8-fold initially down-regulated genes (Fig. 9A).

(Fig. 9A).

and the physical forces' effect experiment:

**3.2 Co-culture/two compartment model** 

regulated after co-culture with hepatocytes (Fig. 10).

other layer, respectively.

shown by RT-PCR

**3.3 Physical forces' effect experiment** 

After centrifugation, the tumor cells were obtained from the top of the interface and resuspended in RPMI medium. To obtain a high purity of isolated tumor cells, CC531 cells were subsequently isolated by fluorescence-activated cell sorting technology using red fluorescent protein (RFP) as marker (Fig. 8).

Afterwards, the pure cells were pelleted and snap frozen at -80 °C. An aliquot of the cells, which were isolated on day 21, was used for re-culturing CC531 cells *in vitro*. These cells were propagated every 3 days, but two time points (14 and 22 days after tumor cell explantation) were chosen for subsequent microarray analysis, PCR, and Western blot.

For the isolation of rat hepatocytes (HCs) and Kupffer cells (KCs), the same perfusion method was performed as described above. However, to separate parenchymal (PCs) from non-parenchymal cells (NPCs), cell suspensions were gently pelleted and the resulting pellet, containing mainly hepatocytes, was taken up in Maintenance-Medium without FCS. Trypan blue exclusion (1 part trypan blue: 2 parts cell suspension) was used for cell counting and assessing their viability. 4×107 hepatocytes with 95% viability were usually obtained from one rat liver. Afterwards, the 25%/50% two-step Percoll gradient was used to isolate Kupffer cells as pure as possible.

Fig. 9. Down-regulation of *Cldn1* and *Cldn4* in CC531 cells homing into the liver. **(A)** Expression profile of claudins (1, 4) and Snail in CC531 cells as shown by microarray analysis. The values represent the gene expression in isolated metastasizing cells in comparison to the expression in cells growing *in vitro*. **(B)** The diagram represents the mRNA or protein expression levels in re-isolated CC531 cells in % of the expression detected in control cells (100%). Values were calculated using the pixel density of each PCR/or Western blot band normalized to the corresponding value of γ-tubulin (*Tubg1*) or ERK2, respectively. **(C)** Expression of claudins (1, 4) in CC531 cells as shown by RT-PCR. Lane 1: control CC531 cells, lanes 2-6: CC531 cells isolated from the liver after 3, 6, 9, 14 and 21 days, respectively, lanes 7, 8: CC531 cells re-isolated after 21 days and cultured *in vitro* for 14 and 22 days, respectively. **(D)** Expression of CLDNs (1, 4) in CC531 cells as shown by Western blot. Lanes 1-5: CC531 cells isolated from the liver after 3, 6, 9, 14 and 21 days, respectively, lanes 6, 7: CC531 cells isolated after 21 days and cultured *in vitro* for 14 and 22 days, respectively, lane 8: control CC531 cells

After centrifugation, the tumor cells were obtained from the top of the interface and resuspended in RPMI medium. To obtain a high purity of isolated tumor cells, CC531 cells were subsequently isolated by fluorescence-activated cell sorting technology using red

Afterwards, the pure cells were pelleted and snap frozen at -80 °C. An aliquot of the cells, which were isolated on day 21, was used for re-culturing CC531 cells *in vitro*. These cells were propagated every 3 days, but two time points (14 and 22 days after tumor cell explantation) were chosen for subsequent microarray analysis, PCR, and Western blot. For the isolation of rat hepatocytes (HCs) and Kupffer cells (KCs), the same perfusion method was performed as described above. However, to separate parenchymal (PCs) from non-parenchymal cells (NPCs), cell suspensions were gently pelleted and the resulting pellet, containing mainly hepatocytes, was taken up in Maintenance-Medium without FCS. Trypan blue exclusion (1 part trypan blue: 2 parts cell suspension) was used for cell counting and assessing their viability. 4×107 hepatocytes with 95% viability were usually obtained from one rat liver. Afterwards, the 25%/50% two-step Percoll gradient was used to

Fig. 9. Down-regulation of *Cldn1* and *Cldn4* in CC531 cells homing into the liver.

**(A)** Expression profile of claudins (1, 4) and Snail in CC531 cells as shown by microarray analysis. The values represent the gene expression in isolated metastasizing cells in comparison to the expression in cells growing *in vitro*. **(B)** The diagram represents the mRNA or protein expression levels in re-isolated CC531 cells in % of the expression detected in control cells (100%). Values were calculated using the pixel density of each PCR/or Western blot band normalized to the corresponding value of γ-tubulin (*Tubg1*) or ERK2, respectively. **(C)** Expression of claudins (1, 4) in CC531 cells as shown by RT-PCR. Lane 1: control CC531 cells, lanes 2-6: CC531 cells isolated from the liver after 3, 6, 9, 14 and 21 days, respectively, lanes 7, 8: CC531 cells re-isolated after 21 days and cultured *in vitro* for 14 and 22 days, respectively. **(D)** Expression of CLDNs (1, 4) in CC531 cells as shown by Western blot. Lanes 1-5: CC531 cells isolated from the liver after 3, 6, 9, 14 and 21 days, respectively, lanes 6, 7: CC531 cells isolated after 21 days and cultured *in vitro* for 14 and 22

fluorescent protein (RFP) as marker (Fig. 8).

isolate Kupffer cells as pure as possible.

days, respectively, lane 8: control CC531 cells

The cDNA microarray results showed that compared to control CC531 cells, claudin1 and claudin4 were among the ≥8-fold initially down-regulated genes (Fig. 9A).

Interestingly, both genes were at first down-regulated with a nadir (8 or 11 fold downregulation) on day 6, followed by gradual up-regulation within the observation period. These results were confirmed with RT-PCR (maximum down-regulation of 80% on day 6 for both genes; Fig. 9B) and Western blot (specific bands below detection limit, >90% inhibition on day 6; Fig. 9C). It is noteworthy that the transcription repressor gene *Snail* showed an inverse modulation: an increased expression during the first week (up to 3.8 fold) with the peak of its expression corresponding to the nadir of *Cldn1* and *Cldn4* down-regulation (Fig. 9A).

Next, two experiments were performed to explain the initial down-regulation of *Cldn1* and *Cldn4*; these included co-culture of CC531 cells with isolated rat hepatocytes or Kupffer cells and the physical forces' effect experiment:

#### **3.2 Co-culture/two compartment model**

Briefly, this model is based on a two-compartment system in which hepatocytes or Kupffer cells, plated in the lower compartment, are co-cultured with CC531 tumor cells growing in the upper compartment, with the two cell types being separated by a porous membrane (0.4 µm pore size). This system, preventing a direct contact between the two compartments, allows the cells to be only indirectly influenced by molecules secreted from the cells in the other layer, respectively.

No down-regulation effect on claudin expression was noticed, whereas both genes were upregulated after co-culture with hepatocytes (Fig. 10).

Fig. 10. Expression of claudins (1, 4) in CC531 cells co-cultured for 24 to 72 h with Kupffer cells (KCs) and hepatocytes (HCs) in comparison to the housekeeping gene ү-tubulin as shown by RT-PCR

#### **3.3 Physical forces' effect experiment**

2x106 CC531 cells were seeded in 25 cm2 flat-bottom flasks or into round 50 ml glass bottles (Steiner GmbH, Siegen Eiserfeld, Germany), which were rotated on a roller (Stovall Life Science Incorporated, Greensboro, NC USA) at a speed of 1rpm, preventing the cells from adhesion to each other and onto the flask bottom. After 24 h, the cells in flat flasks and half

Experimental Colorectal Cancer Liver Metastasis 453

Furthermore, transferring tumor cells from a non-adhesive state in round bottles to growing in flat bottom flasks for 24 h caused >2.5-fold increased expression of *Cldn4*, whereas no effect on *Cldn1* expression was noticed. Accordingly, the physical conditions and the adhesion status of the cells affected differently the expression of *Cldn1* and *Cldn4*,

CC531 cells cultured in 6-well-plates were transfected with 100 nM siRNA or negative control using Lipofectamine 2000 (Invitrogen) following the manufacturer's instructions. The cells were harvested at 24, 48 and 72 h after treatment. As shown by RT-PCR in Fig. 13A, exposure to siRNA species directed against *Cldn1* and *Cldn4* caused reduced

To further investigate a possible interdependence of these two genes, the expression of *Cldn4* and *Cldn1* was investigated in CC531si.*Cldn1* cells and CC531si.*Cldn4* cells, respectively. Expression of *Cldn4* was down-regulated by 50% in tumor cells transfected with siRNA against *Cldn1* (Fig. 13B), whereas inhibition of *Cldn4* did not exert the same effect on *Cldn1*

Fig. 13. **(A)** Down-regulation of claudins (1, 4) in CC531 cells after siRNA transfection as shown by RT-PCR. **(B)** Down-regulation of *Cldn4* after 24-72 h in CC531si.*Cldn1* cells

**3.5 Expression of CLDN1 and CLDN4 in neoplastic human CRC tissues** 

The effect of *Cldn1* and *Cldn4* knockdown on cell growth (MTT assay), cell migration and colony formation (Adwan, et al., 2004, Georges, et al., 2011) was investigated as well. These *in vitro* experiments showed significantly increased migration and decreased clonogenic growth of tumor cells (p<0.05), but no effect on cell proliferation was noticed (Fig. 14).

For the immunohistochemical (IHC) analyses of CLDN1 and CLDN4, 32 primary CRC tissue specimens with adjacent non-neoplastic tissue and 8 liver metastases were obtained

The patients had a median age of 65 years and were classified into UICC stages ll (n=24) and IV (n=8) and graded as G2 (n=25) and G3 (n=7). The histopathological analysis revealed that the expression of CLDN1 was high in 91% (n=30) and that of CLDN4 in 85% (n=28) of all tumor specimens. Comparing the CLDN expression related to UICC stages, CLDN1 and

suggesting a direct relationship with the latter, but not with the former gene.

**3.4 Small interfering RNA (siRNA) knockdown experiments** 

expression of mRNA to 24% and 15%, respectively.

(compared to CC531nonsense cells) as shown by RT-PCR

from the Institute of Pathology, University of Heidelberg.

expression (data not shown).

of the cells in round bottles were harvested for PCR analysis; the other half of cells was seeded in flat flasks till the next day to investigate the influence of adhesion status on claudin expression and then harvested after determining their viability under the microscope. This procedure was done daily for 3 days after seeding the cells (Fig. 11).

Fig. 11. Scheme indicating the experimental procedure for assessing the physical force's effects

As shown in Fig. 12, no change in *Cldn1* expression was noticed either in CC531 cells growing continuously in flat flasks or in round bottles. On the contrary, *Cldn4* mRNA expression was 2- and 1.8- fold higher in CC531 cells growing in flat flasks than their counterparts growing in round bottles at 48 and 72 h after seeding the cells, respectively.

Fig. 12. Expression of claudins (1, 4) in CC531 cells harvested from round and flat flasks as shown by RT-PCR. Lanes 1, 3, 6: CC531 cells harvested from round bottles after 24, 48 and 72 h, respectively. Lanes 2, 5, 8, 10: CC531 cells harvested from flat flasks after 24, 48, 72 and 96 h, respectively. Lanes 4, 7, 9: CC531 cells harvested from flat flasks after being transferred from round bottles after 24, 48 and 72 h, respectively

of the cells in round bottles were harvested for PCR analysis; the other half of cells was seeded in flat flasks till the next day to investigate the influence of adhesion status on claudin expression and then harvested after determining their viability under the microscope. This procedure was done daily for 3 days after seeding the cells (Fig. 11).

Fig. 11. Scheme indicating the experimental procedure for assessing the physical force's

As shown in Fig. 12, no change in *Cldn1* expression was noticed either in CC531 cells growing continuously in flat flasks or in round bottles. On the contrary, *Cldn4* mRNA expression was 2- and 1.8- fold higher in CC531 cells growing in flat flasks than their counterparts growing in round bottles at 48 and 72 h after seeding the cells, respectively.

Fig. 12. Expression of claudins (1, 4) in CC531 cells harvested from round and flat flasks as shown by RT-PCR. Lanes 1, 3, 6: CC531 cells harvested from round bottles after 24, 48 and 72 h, respectively. Lanes 2, 5, 8, 10: CC531 cells harvested from flat flasks after 24, 48, 72 and 96 h, respectively. Lanes 4, 7, 9: CC531 cells harvested from flat flasks after being transferred

from round bottles after 24, 48 and 72 h, respectively

effects

Furthermore, transferring tumor cells from a non-adhesive state in round bottles to growing in flat bottom flasks for 24 h caused >2.5-fold increased expression of *Cldn4*, whereas no effect on *Cldn1* expression was noticed. Accordingly, the physical conditions and the adhesion status of the cells affected differently the expression of *Cldn1* and *Cldn4*, suggesting a direct relationship with the latter, but not with the former gene.

#### **3.4 Small interfering RNA (siRNA) knockdown experiments**

CC531 cells cultured in 6-well-plates were transfected with 100 nM siRNA or negative control using Lipofectamine 2000 (Invitrogen) following the manufacturer's instructions. The cells were harvested at 24, 48 and 72 h after treatment. As shown by RT-PCR in Fig. 13A, exposure to siRNA species directed against *Cldn1* and *Cldn4* caused reduced expression of mRNA to 24% and 15%, respectively.

To further investigate a possible interdependence of these two genes, the expression of *Cldn4* and *Cldn1* was investigated in CC531si.*Cldn1* cells and CC531si.*Cldn4* cells, respectively. Expression of *Cldn4* was down-regulated by 50% in tumor cells transfected with siRNA against *Cldn1* (Fig. 13B), whereas inhibition of *Cldn4* did not exert the same effect on *Cldn1* expression (data not shown).

Fig. 13. **(A)** Down-regulation of claudins (1, 4) in CC531 cells after siRNA transfection as shown by RT-PCR. **(B)** Down-regulation of *Cldn4* after 24-72 h in CC531si.*Cldn1* cells (compared to CC531nonsense cells) as shown by RT-PCR

The effect of *Cldn1* and *Cldn4* knockdown on cell growth (MTT assay), cell migration and colony formation (Adwan, et al., 2004, Georges, et al., 2011) was investigated as well. These *in vitro* experiments showed significantly increased migration and decreased clonogenic growth of tumor cells (p<0.05), but no effect on cell proliferation was noticed (Fig. 14).

#### **3.5 Expression of CLDN1 and CLDN4 in neoplastic human CRC tissues**

For the immunohistochemical (IHC) analyses of CLDN1 and CLDN4, 32 primary CRC tissue specimens with adjacent non-neoplastic tissue and 8 liver metastases were obtained from the Institute of Pathology, University of Heidelberg.

The patients had a median age of 65 years and were classified into UICC stages ll (n=24) and IV (n=8) and graded as G2 (n=25) and G3 (n=7). The histopathological analysis revealed that the expression of CLDN1 was high in 91% (n=30) and that of CLDN4 in 85% (n=28) of all tumor specimens. Comparing the CLDN expression related to UICC stages, CLDN1 and

Experimental Colorectal Cancer Liver Metastasis 455

Fig. 15. Expression of CLDN1 and CLDN4 proteins in human CRC and liver metastasis tissues compared to normal mucosa as shown by immunohistochemistry. **(A)**, **(C)** and **(E)** Expression of CLDN1 in normal mucosa, cancerous tissue and liver metastasis, respectively.

The 67 CRC patients (42 men and 25 women) had an average age of 67 years and were classified into 4 UICC stages (I, n=11; II, n=25; III, n=20; IV, n=11). The expression levels of *CLDN1* and *CLDN4* were significantly correlated (P<0.05; Fig. 16A). No correlation between *CLDN1* expression and age (p=0.19), tumor stage (p=0.88), or overall survival (p=0.2) was seen. With respect to *CLDN4*, also no correlation with age (p=0.69) or tumor stage (p=0.38) was noticed. However, the overall survival of CRC patients with high or low risk in relation to the median split of *CLDN4* expression levels differed significantly according to the logrank test (p=0.018; Fig. 16B). Similarly, using a Cox model, there was not significant difference (p=0.07) between the low and high risk groups taking all stages together, whereas in patients with tumor stages l-lll, an elevated *CLDN4* level was clearly associated with a

**(B)**, **(D)** and **(F)** expression of CLDN4 in normal mucosa, cancerous tissue and liver

metastasis, respectively. Magnification x64

less favorable prognosis (p=0.05; Fig. 16C).

CLDN4 had significantly lower expression in stage IV than in stage ll (p=0.01 and p=0.05, respectively). In line with this, liver metastases showed lower expression of CLDN1 and CLDN4 than in the corresponding primary carcinomas (Fig. 15). This difference was significant for CLDN1 (p<0.05) but not for CLDN4.

Fig. 14. Knockdown effects of *Cldn1* and *Cldn4* on cellular functions of CC531 cells . **(A)** Proliferation of CC531 cells in response to si.*Cldn1* or si.*Cldn4*. **(B)** Increased migration of CC531 cells in response to si.*Cldn1* or si.*Cldn4*. **(C)** Inhibition of colony formation of CC531 cells in response to siRNA down-regulation of claudins (1, 4). **(D)** Inhibition of large colony formation of CC531 cells in response to siRNA down-regulation of *Cldn1* or *Cldn4*. Data (n=3) are shown as means ± S.D. in % of nonsense-transfected cells, an asterisk denotes a significant difference to control cells (p<0.05)

#### **3.6 Correlation of CLDN4 or CLDN1 expression with prognosis in CRC patients**

For real-time PCR analysis, 67 sporadic CRC patients, who were admitted and underwent surgery in the time between (01/98 - 07/01) at the Municipal Hospital in Nürnberg (Department of Abdominal-, Thorax-, and Endocrine Surgery) were selected.

The samples included in this study (for IHC and real-time PCR) were used based on the patients informed consent and approved by the Ethics Committee of the Universities of Heidelberg and Erlangen.

CLDN4 had significantly lower expression in stage IV than in stage ll (p=0.01 and p=0.05, respectively). In line with this, liver metastases showed lower expression of CLDN1 and CLDN4 than in the corresponding primary carcinomas (Fig. 15). This difference was

Fig. 14. Knockdown effects of *Cldn1* and *Cldn4* on cellular functions of CC531 cells .

**3.6 Correlation of CLDN4 or CLDN1 expression with prognosis in CRC patients** 

(Department of Abdominal-, Thorax-, and Endocrine Surgery) were selected.

For real-time PCR analysis, 67 sporadic CRC patients, who were admitted and underwent surgery in the time between (01/98 - 07/01) at the Municipal Hospital in Nürnberg

The samples included in this study (for IHC and real-time PCR) were used based on the patients informed consent and approved by the Ethics Committee of the Universities of

significant difference to control cells (p<0.05)

Heidelberg and Erlangen.

**(A)** Proliferation of CC531 cells in response to si.*Cldn1* or si.*Cldn4*. **(B)** Increased migration of CC531 cells in response to si.*Cldn1* or si.*Cldn4*. **(C)** Inhibition of colony formation of CC531 cells in response to siRNA down-regulation of claudins (1, 4). **(D)** Inhibition of large colony formation of CC531 cells in response to siRNA down-regulation of *Cldn1* or *Cldn4*. Data (n=3) are shown as means ± S.D. in % of nonsense-transfected cells, an asterisk denotes a

significant for CLDN1 (p<0.05) but not for CLDN4.

Fig. 15. Expression of CLDN1 and CLDN4 proteins in human CRC and liver metastasis tissues compared to normal mucosa as shown by immunohistochemistry. **(A)**, **(C)** and **(E)** Expression of CLDN1 in normal mucosa, cancerous tissue and liver metastasis, respectively. **(B)**, **(D)** and **(F)** expression of CLDN4 in normal mucosa, cancerous tissue and liver metastasis, respectively. Magnification x64

The 67 CRC patients (42 men and 25 women) had an average age of 67 years and were classified into 4 UICC stages (I, n=11; II, n=25; III, n=20; IV, n=11). The expression levels of *CLDN1* and *CLDN4* were significantly correlated (P<0.05; Fig. 16A). No correlation between *CLDN1* expression and age (p=0.19), tumor stage (p=0.88), or overall survival (p=0.2) was seen. With respect to *CLDN4*, also no correlation with age (p=0.69) or tumor stage (p=0.38) was noticed. However, the overall survival of CRC patients with high or low risk in relation to the median split of *CLDN4* expression levels differed significantly according to the logrank test (p=0.018; Fig. 16B). Similarly, using a Cox model, there was not significant difference (p=0.07) between the low and high risk groups taking all stages together, whereas in patients with tumor stages l-lll, an elevated *CLDN4* level was clearly associated with a less favorable prognosis (p=0.05; Fig. 16C).

Experimental Colorectal Cancer Liver Metastasis 457

Fig. 17. Expression of OPN on mRNA (left; RT-PCR) and protein (right; Western blot) levels. Lane 1: CC531 cells explanted from the tumor, lanes (2-6): CC531 cells after 11, 13, 15, 18,

In conclusion, the model described in this chapter evolved in two main steps. First, for quantitation of tumor cell load, the CC531 cells had been marked with GLB1. This marker proved useful for the purposes of therapeutic studies with various anti-cancer agents. Using the GLB1-based chemoluminescence assay, it could be shown that the efficacy of hepatic artery chemoembolization with gemcitabine-microspheres was significantly higher than

Also it was shown, that HACE with irinotecan-microspheres had a greater effect than that of

The same assay allowed us to evaluate the combination effect of pemetrexed disodium and gemcitabine administered by hepatic artery and portal vein chemoembolization in our rat liver metastasis model. These experiments showed that HACE/gemcitabine was superior to systemic intravenous bolus treatment, while PVCE/pemetrexed disodium was ineffective. Interestingly, a maximum additive tumor growth inhibition *in vivo* was noticed using a regimen of (intravenous or PVCE)/pemetrexed disodium preceding HACE/gemcitabine indicating that these two agents can successfully be combined and favor further evaluation

The third set of therapeutic experiments, using HACE with drug eluting beads (DEBs) loaded with either doxorubicin hydrochloride or irinotecan hydrochloride showed that irinotecan is more advantageous because of its significant activity and excellent tolerability. In addition, HACE with DEBs with either agent may have potential for treating patients

The CC531 tumor cells marked with GLB1 allowed a determination of the tumor cell load only after termination of the experiment. To further develop the model for determining the tumor cell load in living animals, other markers had to be introduced. These were eGFP and RFP, which enabled the next series of experiments that were related to liver metastasis genes. Realizing the fact that the most common cause of CRC-related death is the development of metastasis, especially into the liver, denotes the importance of identifying the metastasisrelated changes in tumor cells, which probably differ from those related to the primary tumor. Therefore, temporal changes in gene expression of CRC cells homing into the liver have been investigated using our *in vivo* model, which had been improved by eGFP and RFP markers. The intraportal inoculation of CC531 cells defines the onset of metastatic proliferation in rat liver. This, in turn, permits a close following of the time-dependent modulation of gene expression, as the tumor cells home into the liver and then grow to a lethal size. The technique of re-isolating these tumor cells from rat liver permits monitoring for the first time the expression of several candidate genes in a time-dependent manner. The

HACE with 5-FU, setting the basis for further investigations in clinical trials.

and 21 days; respectively

with 5-FU in reducing the tumor cell load.

**4. Conclusion** 

in patients.

with colorectal liver metastasis.

Fig. 16. Correlation of *CLDN1* or *CLDN4* expression levels with prognosis in 67 CRC patients and with each other. **(A)** Scatterplot of the correlation between *CLDN1* and *CLDN4* expression levels assessed with the non-parametric correlation coefficient from Spearman **(B)** The Kaplan-Meier plot represents the overall survival probability of CRC patients with high or low risk according to *CLDN4* median split for overall survival. The log-rank test shows a significant difference (p=0.018) between the two groups. **(C)** The Kaplan-Meier plot demonstrates the overall survival probability of CRC patients with high or low *CLDN4* expression, dichotomized into high and low risk groups by the *CLDN4* median split for overall survival and after separating the stages into I-III and IV stages. The Cox model shows a significant association between *CLDN4* elevated levels and less overall survival (p=0.05)

#### **3.7 OPN expression profile in CC531 cells ex-vivo**

After explanting the liver, a piece of liver containing tumor cells was resected and used for re-culturing CC531 cells (Georges, et al., 2010). These cells were cultured for three weeks and within this period passaged five times corresponding to 11, 13, 15, 18 and 21 days; respectively (passages 1-5). At these intervals the cells were investigated for their mRNAand protein expression levels of OPN.

At the mRNA level, *Opn* was up-regulated in CC531 cells explanted from the tumor. Over time, this *Opn* mRNA level gradually decreased till it disappeared after the third passage; i.e. after two weeks (corresponding to a reduction by 88%; Fig. 17, left).

The OPN protein level, as shown by Western blot (Fig. 17, right), was first highly expressed but then down-regulated to minimally 5% within the next 15 days. A slight increase to 20% of the initial level was seen at the final passages 4 and 5.

Fig. 17. Expression of OPN on mRNA (left; RT-PCR) and protein (right; Western blot) levels. Lane 1: CC531 cells explanted from the tumor, lanes (2-6): CC531 cells after 11, 13, 15, 18, and 21 days; respectively

#### **4. Conclusion**

456 Colorectal Cancer – From Prevention to Patient Care

Fig. 16. Correlation of *CLDN1* or *CLDN4* expression levels with prognosis in 67 CRC patients and with each other. **(A)** Scatterplot of the correlation between *CLDN1* and *CLDN4* expression levels assessed with the non-parametric correlation coefficient from Spearman **(B)** The Kaplan-Meier plot represents the overall survival probability of CRC patients with high or low risk according to *CLDN4* median split for overall survival. The log-rank test shows a significant difference (p=0.018) between the two groups. **(C)** The Kaplan-Meier plot demonstrates the

dichotomized into high and low risk groups by the *CLDN4* median split for overall survival and after separating the stages into I-III and IV stages. The Cox model shows a significant

After explanting the liver, a piece of liver containing tumor cells was resected and used for re-culturing CC531 cells (Georges, et al., 2010). These cells were cultured for three weeks and within this period passaged five times corresponding to 11, 13, 15, 18 and 21 days; respectively (passages 1-5). At these intervals the cells were investigated for their mRNA-

At the mRNA level, *Opn* was up-regulated in CC531 cells explanted from the tumor. Over time, this *Opn* mRNA level gradually decreased till it disappeared after the third passage;

The OPN protein level, as shown by Western blot (Fig. 17, right), was first highly expressed but then down-regulated to minimally 5% within the next 15 days. A slight increase to 20%

overall survival probability of CRC patients with high or low *CLDN4* expression,

association between *CLDN4* elevated levels and less overall survival (p=0.05)

i.e. after two weeks (corresponding to a reduction by 88%; Fig. 17, left).

**3.7 OPN expression profile in CC531 cells ex-vivo** 

of the initial level was seen at the final passages 4 and 5.

and protein expression levels of OPN.

In conclusion, the model described in this chapter evolved in two main steps. First, for quantitation of tumor cell load, the CC531 cells had been marked with GLB1. This marker proved useful for the purposes of therapeutic studies with various anti-cancer agents. Using the GLB1-based chemoluminescence assay, it could be shown that the efficacy of hepatic artery chemoembolization with gemcitabine-microspheres was significantly higher than with 5-FU in reducing the tumor cell load.

Also it was shown, that HACE with irinotecan-microspheres had a greater effect than that of HACE with 5-FU, setting the basis for further investigations in clinical trials.

The same assay allowed us to evaluate the combination effect of pemetrexed disodium and gemcitabine administered by hepatic artery and portal vein chemoembolization in our rat liver metastasis model. These experiments showed that HACE/gemcitabine was superior to systemic intravenous bolus treatment, while PVCE/pemetrexed disodium was ineffective. Interestingly, a maximum additive tumor growth inhibition *in vivo* was noticed using a regimen of (intravenous or PVCE)/pemetrexed disodium preceding HACE/gemcitabine indicating that these two agents can successfully be combined and favor further evaluation in patients.

The third set of therapeutic experiments, using HACE with drug eluting beads (DEBs) loaded with either doxorubicin hydrochloride or irinotecan hydrochloride showed that irinotecan is more advantageous because of its significant activity and excellent tolerability. In addition, HACE with DEBs with either agent may have potential for treating patients with colorectal liver metastasis.

The CC531 tumor cells marked with GLB1 allowed a determination of the tumor cell load only after termination of the experiment. To further develop the model for determining the tumor cell load in living animals, other markers had to be introduced. These were eGFP and RFP, which enabled the next series of experiments that were related to liver metastasis genes.

Realizing the fact that the most common cause of CRC-related death is the development of metastasis, especially into the liver, denotes the importance of identifying the metastasisrelated changes in tumor cells, which probably differ from those related to the primary tumor. Therefore, temporal changes in gene expression of CRC cells homing into the liver have been investigated using our *in vivo* model, which had been improved by eGFP and RFP markers. The intraportal inoculation of CC531 cells defines the onset of metastatic proliferation in rat liver. This, in turn, permits a close following of the time-dependent modulation of gene expression, as the tumor cells home into the liver and then grow to a lethal size. The technique of re-isolating these tumor cells from rat liver permits monitoring for the first time the expression of several candidate genes in a time-dependent manner. The

Experimental Colorectal Cancer Liver Metastasis 459

Brekken, R. A. & Sage, E. H. (2000). SPARC, a matricellular protein: at the crossroads of cell-

Chambers, A. F., et al. (2002). Dissemination and growth of cancer cells in metastatic sites.

Chambers, A. F., et al. (1996). Osteopontin expression in lung cancer. *Lung Cancer*, Vol.15,

Cheung, S. T., et al. (2005). Claudin-10 expression level is associated with recurrence of

Chishima, T., et al. (1997). Cancer invasion and micrometastasis visualized in live tissue by

Choti, M. A., et al. (2002). Trends in long-term survival following liver resection for hepatic colorectal metastases. *Ann Surg*, Vol.235, No.6, (2002), pp.(759-66), 0003-4932 de Gramont, A., et al. (2000). Leucovorin and fluorouracil with or without oxaliplatin as

de Oliveira, S. S., et al. (2005). Claudins upregulation in human colorectal cancer. *FEBS Lett*,

Dooley, T. P., et al. (1993). Evaluation of a nude mouse tumor model using beta-

Engbring, J. A. & Kleinman, H. K. (2003). The basement membrane matrix in malignancy. *J* 

Eyol, E., et al. (2008). Chemoembolisation of rat colorectal liver metastases with drug eluting

Farrow, B., et al. (2008). The role of the tumor microenvironment in the progression of pancreatic cancer. *J Surg Res*, Vol.149, No.2, (2008), pp.(319-28), 1095-8673 Findley, M. K. & Koval, M. (2009). Regulation and roles for claudin-family tight junction

Fong, Y., et al. (1999). Clinical score for predicting recurrence after hepatic resection for

Fukumura, D., et al. (1997). Effect of host microenvironment on the microcirculation of human colon adenocarcinoma. *Am J Pathol*, Vol.151, No.3, (1997), pp.(679-88), 0002-9440 Gao, C., et al. (2003). Osteopontin-dependent CD44v6 expression and cell adhesion in HepG2 cells. *Carcinogenesis*, Vol.24, No.12, (2003), pp.(1871-8), 0143-3334 Garden, O. J., et al. (2006). Guidelines for resection of colorectal cancer liver metastases. *Gut*,

Georges, R., et al. (2010). Regulation of osteopontin and related proteins in rat CC531 colorectal cancer cells. *Int J Oncol*, Vol.37, No.2, (2010), pp.(249-56), 1791-2423 Georges, R., et al. (2011). Sequential biphasic changes in claudin1 and claudin4 expression

Georges, R. B., et al. (2011). The insulin-like growth factor binding proteins 3 and 7 are

proteins. *IUBMB Life*, Vol.61, No.4, (2009), pp.(431-7), 1521-6551

No.3, (1999), pp.(309-18; discussion 318-21), 0003-4932

Vol.55 Suppl 3, (2006), pp.(iii1-8), 0017-5749

primary hepatocellular carcinoma. *Clin Cancer Res*, Vol.11, No.2 Pt 1, (2005),

green fluorescent protein expression. *Cancer Res*, Vol.57, No.10, (1997), pp.(2042-7),

first-line treatment in advanced colorectal cancer. *J Clin Oncol*, Vol.18, No.16, (2000),

galactosidase-expressing melanoma cells. *Lab Anim Sci*, Vol.43, No.1, (1993), pp.(48-

beads loaded with irinotecan or doxorubicin. *Clin Exp Metastasis*, Vol.25, No.3,

metastatic colorectal cancer: analysis of 1001 consecutive cases. *Ann Surg*, Vol.230,

are correlated to colorectal cancer progression and liver metastasis. *J Cell Mol Med*,

associated with colorectal cancer and liver metastasis. *Cancer Biol Ther*, Vol.12,

matrix. *Matrix Biol*, Vol.19, No.7, (2000), pp.(569-580), 0945-053X

*Nat Rev Cancer*, Vol.2, No.8, (2002), pp.(563-72), 1474-175X

Vol.579, No.27, (2005), pp.(6179-6185), 0014-5793

*Pathol*, Vol.200, No.4, (2003), pp.(465-70), 0022-3417

No.3, (1996), pp.(311-23), 0169-5002

pp.(551-556), 1078-0432

pp.(2938-47), 0732-183X

(2008), pp.(273-82), 0262-0898

0008-5472

57), 0023-6764

(2011), 1582-4934

No.1, (2011), 1555-8576

following cDNA microarray analysis showed that the initial phase of rat CRC cells homing into the liver involves a transient down-regulation of *Cldn1* and in particular of *Cldn4*. The transcription repressor Snail, which regulates both claudins, was concomitantly upregulated during the early stages of metastasis before returning to normal expression levels. Silencing of *Cldn1* and *Cldn4* by siRNA increased migration and reduced colony formation, with these phenotypes consistent with metastatic homing. These experimental results were paralleled in human CRC tumor samples, which show increased *CLDN1* and *CLDN4* expression in UICC stages I–III, and significantly reduced expression in stage IV and in liver metastasis. The results obtained with human specimens give first evidence of a modulated claudin expression similar to those in the rat model. However, a prospective study is needed to corroborate these results, taking into account separately the entities, colonic and rectal cancers. That research could be driven by our hypothesis that primary CRC tumors have an initial growth advantage from increased claudin expression, whereas metastasizing cells require a transient reduction in claudin expression to be liberated from the primary tumor and to then initiate metastatic growth in the liver.

Based on these experiments we conclude that the CC531 colorectal cancer liver metastasis model in rats is attractive for translational research: it is suited for therapeutic studies as well as for identifying genes involved in colorectal cancer liver metastasis.

#### **5. References**


following cDNA microarray analysis showed that the initial phase of rat CRC cells homing into the liver involves a transient down-regulation of *Cldn1* and in particular of *Cldn4*. The transcription repressor Snail, which regulates both claudins, was concomitantly upregulated during the early stages of metastasis before returning to normal expression levels. Silencing of *Cldn1* and *Cldn4* by siRNA increased migration and reduced colony formation, with these phenotypes consistent with metastatic homing. These experimental results were paralleled in human CRC tumor samples, which show increased *CLDN1* and *CLDN4* expression in UICC stages I–III, and significantly reduced expression in stage IV and in liver metastasis. The results obtained with human specimens give first evidence of a modulated claudin expression similar to those in the rat model. However, a prospective study is needed to corroborate these results, taking into account separately the entities, colonic and rectal cancers. That research could be driven by our hypothesis that primary CRC tumors have an initial growth advantage from increased claudin expression, whereas metastasizing cells require a transient reduction in claudin expression to be liberated from the primary tumor

Based on these experiments we conclude that the CC531 colorectal cancer liver metastasis model in rats is attractive for translational research: it is suited for therapeutic studies as

Adson, M. A., et al. (1984). Resection of hepatic metastases from colorectal cancer. *Arch Surg*,

Adwan, H., et al. (2004). Downregulation of osteopontin and bone sialoprotein II is related

Aguiar, J. L., et al. (1987). Feasible model for locoregional and systemic longterm

Bartkowski, R., et al. (1986). Experiments on the efficacy and toxicity of locoregional

Bentrem, D. J., et al. (2005). Surgical therapy for metastatic disease to the liver. *Annu Rev* 

Berney, T., et al. (1998). Results of surgical resection of liver metastases from non-colorectal

Bismuth, H., et al. (1996). Resection of nonresectable liver metastases from colorectal cancer

Bornstein, P. & Sage, E. H. (2002). Matricellular proteins: extracellular modulators of cell function. *Curr Opin Cell Biol*, Vol.14, No.5, (2002), pp.(608-16), 0955-0674 Bosman, F. T. & Stamenkovic, I. (2003). Functional structure and composition of the extracellular matrix. *J Pathol*, Vol.200, No.4, (2003), pp.(423-8), 0022-3417

*Cancer Res Clin Oncol*, Vol.113, No.1, (1987), pp.(27-30), 0171-5216

primaries. *Br J Surg*, Vol.85, No.10, (1998), pp.(1423-7), 0007-1323

to reduced colony formation and metastasis formation of MDA-MB-231 human breast cancer cells. *Cancer Gene Ther*, Vol.11, No.2, (2004), pp.(109-120), 0929-1903 Agrawal, D., et al. (2002). Osteopontin identified as lead marker of colon cancer progression,

using pooled sample expression profiling. *J Natl Cancer Inst*, Vol.94, No.7, (2002),

administration of drugs and concomitant blood sampling in Sprague-Dawley rats. *J* 

chemotherapy of liver tumors with 5-fluoro-2'-deoxyuridine (FUDR) and 5 fluorouracil (5-FU) in an animal model. *J Cancer Res Clin Oncol*, Vol.111, No.1,

after neoadjuvant chemotherapy. *Ann Surg*, Vol.224, No.4, (1996), pp.(509-20;

well as for identifying genes involved in colorectal cancer liver metastasis.

Vol.119, No.6, (1984), pp.(647-51), 0004-0010

and to then initiate metastatic growth in the liver.

pp.(513-21), 0027-8874

(1986), pp.(42-6), 0171-5216

discussion 520-2), 0003-4932

*Med*, Vol.56, (2005), pp.(139-56), 0066-4219

**5. References** 


Experimental Colorectal Cancer Liver Metastasis 461

Meyerhardt, J. A. & Mayer, R. J. (2005). Systemic therapy for colorectal cancer. *N Engl J Med*,

Michl, P., et al. (2003). Claudin-4 expression decreases invasiveness and metastatic potential of pancreatic cancer. *Cancer Res*, Vol.63, No.19, (2003), pp.(6265-71), 0008-5472 Michl, P., et al. (2001). Claudin-4: a new target for pancreatic cancer treatment using

Mineta, K., et al. (2011). Predicted expansion of the claudin multigene family. *FEBS Lett*, Vol.585, No.4, (2011), pp.(606-12), 1873-3468 (Electronic), 0014-5793 (Linking) Morin, P. J. (2005). Claudin proteins in human cancer: promising new targets for diagnosis and therapy. *Cancer Res*, Vol.65, No.21, (2005), pp.(9603-9606), 0008-5472 Oldenburg, J., et al. (1994). Characterization of resistance mechanisms to cis-

cell line in vitro. *Cancer Res*, Vol.54, No.2, (1994), pp.(487-93), 0008-5472 Pan, H. W., et al. (2003). Overexpression of osteopontin is associated with intrahepatic

Clostridium perfringens enterotoxin. *Gastroenterology*, Vol.121, No.3, (2001),

diamminedichloroplatinum(II) in three sublines of the CC531 colon adenocarcinoma

metastasis, early recurrence, and poorer prognosis of surgically resected hepatocellular carcinoma. *Cancer*, Vol.98, No.1, (2003), pp.(119-27), 0008-543X Rangel, L. B., et al. (2003). Tight junction proteins claudin-3 and claudin-4 are frequently

overexpressed in ovarian cancer but not in ovarian cystadenomas. *Clin Cancer Res*,

chemoembolization with pemetrexed disodium and gemcitabine. *J Cancer Res Clin* 

quantification of tumor cell reduction. *J Cancer Res Clin Oncol*, Vol.130, No.4, (2004),

cancer. Irinotecan Study Group. *N Engl J Med*, Vol.343, No.13, (2000), pp.(905-14),

analysis of gene expression data analysis, in gastric cancer with an intestinal

carcinomas correlate with grading and metastatic disease. *Cytopathology*, Vol.16,

gemcitabine followed by evaluation of tumor cell load by chemiluminescence.

adenocarcinoma. *Am J Respir Crit Care Med*, Vol.160, No.4, (1999), pp.(1269-73),

Rodenbach, M., et al. (2005). Combination treatment of CC531-lac-Z rat liver metastases by

Saenger, J., et al. (2004). Chemoembolization of rat liver metastasis with irinotecan and

Saltz, L. B., et al. (2000). Irinotecan plus fluorouracil and leucovorin for metastatic colorectal

Sanada, Y., et al. (2006). Down-regulation of the claudin-18 gene, identified through serial

Sauer, T., et al. (2005). Reduced expression of Claudin-7 in fine needle aspirates from breast

Seelig, M. H., et al. (2004). Chemoembolization of rat liver metastasis with microspheres and

Shijubo, N., et al. (1999). Vascular endothelial growth factor and osteopontin in stage I lung

Soini, Y. (2005). Expression of claudins 1, 2, 3, 4, 5 and 7 in various types of tumours.

Stangl, R., et al. (1994). Factors influencing the natural history of colorectal liver metastases.

Sugarbaker, P. H. (1990). Surgical decision making for large bowel cancer metastatic to the

phenotype. *J Pathol*, Vol.208, No.5, (2006), pp.(633-642), 0022-3417

*Oncol Rep*, Vol.11, No.5, (2004), pp.(1107-13), 1021-335X

*Histopathology*, Vol.46, No.5, (2005), pp.(551-60), 0309-0167

*Lancet*, Vol.343, No.8910, (1994), pp.(1405-10), 0140-6736

liver. *Radiology*, Vol.174, No.3 Pt 1, (1990), pp.(621-6), 0033-8419

Vol.352, No.5, (2005), pp.(476-87), 1533-4406

Vol.9, No.7, (2003), pp.(2567-75), 1078-0432

*Oncol*, Vol.131, No.5, (2005), pp.(289-99), 0171-5216

pp.(678-684), 0016-5085

pp.(203-10), 0171-5216

No.4, (2005), pp.(193-8), 0956-5507

0028-4793

1073-449X


Giacchetti, S., et al. (2000). Phase III multicenter randomized trial of oxaliplatin added to

colorectal cancer. *J Clin Oncol*, Vol.18, No.1, (2000), pp.(136-47), 0732-183X Gotoh, M., et al. (2002). Overexpression of osteopontin in hepatocellular carcinoma. *Pathol* 

Griffini, P., et al. (1997). Three-dimensional reconstruction of colon carcinoma metastases in

Griffini, P., et al. (1996). Kupffer cells and pit cells are not effective in the defense against

Gupta, G. P. & Massague, J. (2006). Cancer metastasis: building a framework. *Cell*, Vol.127,

Hijiya, N., et al. (1994). Cloning and characterization of the human osteopontin gene and its promoter. *Biochem J*, Vol.303 ( Pt 1), (1994), pp.(255-62), 0264-6021 (Print) Hough, C. D., et al. (2000). Large-scale serial analysis of gene expression reveals genes

Ioachim, E., et al. (2002). Immunohistochemical expression of extracellular matrix

Johnson, A. H., et al. (2005). Expression of tight-junction protein claudin-7 is an early event in gastric tumorigenesis. *Am J Pathol*, Vol.167, No.2, (2005), pp.(577-84), 0002-9440 Jung, Y. D., et al. (2002). The role of the microenvironment and intercellular cross-talk in tumor angiogenesis. *Semin Cancer Biol*, Vol.12, No.2, (2002), pp.(105-12), 1044-579X Kamphorst, E. J., et al. (1999). New technique for superselective arterial (chemo-) embolization of the rat liver. *Lab Anim Sci*, Vol.49, No.2, (1999), pp.(216-9), 0023-6764 Kinugasa, T., et al. (2007). Selective up-regulation of claudin-1 and claudin-2 in colorectal cancer. *Anticancer Res*, Vol.27, No.6A, (2007), pp.(3729-3734), 0250-7005 Kominsky, S. L., et al. (2003). Loss of the tight junction protein claudin-7 correlates with

liver. *J Microsc*, Vol.187, No.Pt 1, (1997), pp.(12-21), 0022-2720

*Int*, Vol.52, No.1, (2002), pp.(19-24), 1320-5463

Vol.14, No.4, (1996), pp.(367-80), 0262-0898

Vol.38, No.18, (2002), pp.(2362-70), 0959-8049

No.4, (2006), pp.(679-95), 0092-8674

pp.(6281-6287), 0008-5472

pp.(842-9), 1525-0016

pp.(7878-7881), 0008-5472

pp.(689-92), 0020-7136

chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic

experimentally induced colon carcinoma metastasis in rat liver. *Clin Exp Metastasis*,

differentially expressed in ovarian cancer. *Cancer Res*, Vol.60, No.22, (2000),

components tenascin, fibronectin, collagen type IV and laminin in breast cancer: their prognostic value and role in tumour invasion and progression. *Eur J Cancer*,

histological grade in both ductal carcinoma in situ and invasive ductal carcinoma

histological grade in both ductal carcinoma in situ and invasive ductal carcinoma

localized, matrix-mediated antisense gene delivery. *Mol Ther*, Vol.3, No.6, (2001),

and claudin-4 in prostate cancer epithelium. *Cancer Res*, Vol.61, No.21, (2001),

experimental rat liver tumours: comparison of isolated liver perfusion and hepatic

adenocarcinoma: importance of tumor site. *Int J Cancer*, Vol.33, No.5, (1984),

of the breast. *Oncogene*, Vol.22, No.13, (2003), pp.(2021-2033), 0950-9232 Kominsky, S. L., et al. (2003). Loss of the tight junction protein claudin-7 correlates with

of the breast. *Oncogene*, Vol.22, No.13, (2003), pp.(2021-33), 0950-9232 Kyriakides, T. R., et al. (2001). Regulation of angiogenesis and matrix remodeling by

Long, H., et al. (2001). Expression of Clostridium perfringens enterotoxin receptors claudin-3

Marinelli, A., et al. (1991). Increasing the effective concentration of melphalan in

artery infusion. *Br J Cancer*, Vol.64, No.6, (1991), pp.(1069-75), 0007-0920 Marquet, R. L., et al. (1984). Interferon treatment of a transplantable rat colon


**Part 6** 

**Study Reports** 


**Part 6** 

**Study Reports** 

462 Colorectal Cancer – From Prevention to Patient Care

Swisshelm, K., et al. (2005). Role of claudins in tumorigenesis. *Adv Drug Deliv Rev*, Vol.57,

Thalmann, G. N., et al. (1999). Osteopontin: possible role in prostate cancer progression. *Clin* 

Thiery, J. P. (2003). Epithelial-mesenchymal transitions in development and pathologies.

Thomas, C., et al. (1993). Liver metastasis model of colon cancer in the rat:

Tsukita, S., et al. (2001). Multifunctional strands in tight junctions. *Nat Rev Mol Cell Biol*,

Tuck, A. B., et al. (1999). Osteopontin induces increased invasiveness and plasminogen

Tuck, A. B., et al. (1998). Osteopontin expression in a group of lymph node negative breast cancer patients. *Int J Cancer*, Vol.79, No.5, (1998), pp.(502-8), 0020-7136 Tuck, A. B., et al. (1997). Osteopontin and p53 expression are associated with tumor

Ue, T., et al. (1998). Co-expression of osteopontin and CD44v9 in gastric cancer. *Int J Cancer*,

Ueda, J., et al. (2007). Heterogeneous expression of claudin-4 in human colorectal cancer:

Usami, Y., et al. (2006). Reduced expression of claudin-7 correlates with invasion and

Veenhuizen, R. B., et al. (1996). Intraperitoneal photodynamic therapy of the rat CC531 adenocarcinoma. *Br J Cancer*, Vol.73, No.11, (1996), pp.(1387-92), 0007-0920 Venditti, J. M., et al. (1984). Current NCI preclinical antitumor screening in vivo: results of

Wai, P. Y. & Kuo, P. C. (2004). The role of Osteopontin in tumor metastasis. *J Surg Res*,

Weber, G. F. (2001). The metastasis gene osteopontin: a candidate target for cancer therapy.

WHOSIS. (2008). WHO Statistical Information System In: Available from: <http://www. who.int/healthinfo/global\_burden\_disease/GBD\_report\_2004update\_full.pdf> Wittmer, A., et al. (1999). Quantitative detection of lac-Z-transfected CC531 colon carcinoma

Yeatman, T. J. & Chambers, A. F. (2003). Osteopontin and colon cancer progression. *Clin Exp* 

Zacharias, T., et al. (2004). First and repeat resection of colorectal liver metastases in elderly

Zhang, L., et al. (1994). Luciferase activity as a marker of tumor burden and as an indicator

*Biochim Biophys Acta*, Vol.1552, No.2, (2001), pp.(61-85), 0006-3002

patients. *Ann Surg*, Vol.240, No.5, (2004), pp.(858-65), 0003-4932

*Arch Pathol Lab Med*, Vol.121, No.6, (1997), pp.(578-84), 0003-9985

metastasis. *Pathobiology*, Vol.74, No.1, (2007), pp.(32-41), 1015-2008

immunohistochemical characterization. *Invasion Metastasis*, Vol.13, No.2, (1993),

activator expression of human mammary epithelial cells. *Oncogene*, Vol.18, No.29,

progression in a case of synchronous, bilateral, invasive mammary carcinomas.

decreased claudin-4 expression at the invasive front correlates cancer invasion and

metastasis in squamous cell carcinoma of the esophagus. *Hum Pathol*, Vol.37, No.5,

tumor panel screening, 1976-1982, and future directions. *Adv Pharmacol Chemother*,

cells in an orthotopic rat liver metastasis model. *Clin Exp Metastasis*, Vol.17, No.5,

of tumor response to antineoplastic therapy in vivo. *Clin Exp Metastasis*, Vol.12,

No.6, (2005), pp.(919-928), 0169-409X

Vol.2, No.4, (2001), pp.(285-293), 1471-0072

Vol.79, No.2, (1998), pp.(127-32), 0020-7136

pp.(102-12), 0251-1789

(1999), pp.(4237-46), 0950-9232

(2006), pp.(569-77), 0046-8177

Vol.20, (1984), pp.(1-20), 0065-3144

(1999), pp.(369-76), 0262-0898

No.2, (1994), pp.(87-92), 0262-0898

Vol.121, No.2, (2004), pp.(228-41), 0022-4804

*Metastasis*, Vol.20, No.1, (2003), pp.(85-90), 0262-0898

*Cancer Res*, Vol.5, No.8, (1999), pp.(2271-7), 1078-0432

*Curr Opin Cell Biol*, Vol.15, No.6, (2003), pp.(740-6), 0955-0674

**23** 

*Japan*

**Risk Factors for Wound Infection** 

**A Matched Case – Control Study** 

*Department of Surgery, Kitasato University School of Medicine,* 

Elective surgery for colorectal cancer involves a semi-contaminated operation, with a 3% to 26% incidence of postoperative wound infection (1). Risk factors for postoperative wound infection include high body-mass index (BMI) (2, diabetes mellitus(3), body-weight loss(4), advanced age(5), smoking(6), blood transfusion(7), and high intraoperative blood loss (8). The development of wound infections can cause considerable discomfort and stress, as well as prolong the hospital stay, substantially increasing healthcare costs. Measures to prevent wound infections have been refined by adjusting the duration of antibiotic treatment and improving techniques for preoperative bowel preparation and drain placement. We performed a matched case-control study to clarify risk factors for perioperative wound infection in patients who underwent standard surgery for colorectal cancer, performed by the same operator at the same hospital. All patients received similar levels of perioperative

From January 2004 through December 2006, a total of 264 patients underwent surgery in our hospital for primary colorectal cancer with a solitary lesion. The same surgeon (T.N.) served as the operator or assistant. Preoperatively, all patients received mechanical bowel preparation. Two tablets of sennoside were given orally 2 days before surgery, and 1 packet of magnesium citrate and 1 bottle of sodium picosulfate were given the day before surgery. For wound closure, the peritoneum and fascia were sutured together with interrupted, polydioxanone absorbable sutures (0-PDS IITM, Johnson and Johnson Co. Ltd ). The same suture material was used to close the abdomen in patients who underwent laparoscopic

In patients who underwent high-pressure irrigation of their wounds, the muscle layer was sutured, and the subcutaneous tissue was washed with warm physiological saline solution applied under high pressure, using a 23-gauge ophthalmic washing catheter attached to a 20-mL syringe. The tip of the needle was positioned about 10 cm from the wound (Fig. 1). The applied volume of physiological saline solution was 400 mL for open surgery, and 200

**1. Introduction**

care.

**2. Subjects and methods** 

surgery and those who underwent open surgery.

Takatoshi Nakamura and Masahiko Watanabe

**After Surgery for Colorectal Cancer:** 

### **Risk Factors for Wound Infection After Surgery for Colorectal Cancer: A Matched Case – Control Study**

Takatoshi Nakamura and Masahiko Watanabe *Department of Surgery, Kitasato University School of Medicine, Japan*

#### **1. Introduction**

Elective surgery for colorectal cancer involves a semi-contaminated operation, with a 3% to 26% incidence of postoperative wound infection (1). Risk factors for postoperative wound infection include high body-mass index (BMI) (2, diabetes mellitus(3), body-weight loss(4), advanced age(5), smoking(6), blood transfusion(7), and high intraoperative blood loss (8). The development of wound infections can cause considerable discomfort and stress, as well as prolong the hospital stay, substantially increasing healthcare costs. Measures to prevent wound infections have been refined by adjusting the duration of antibiotic treatment and improving techniques for preoperative bowel preparation and drain placement. We performed a matched case-control study to clarify risk factors for perioperative wound infection in patients who underwent standard surgery for colorectal cancer, performed by the same operator at the same hospital. All patients received similar levels of perioperative care.

#### **2. Subjects and methods**

From January 2004 through December 2006, a total of 264 patients underwent surgery in our hospital for primary colorectal cancer with a solitary lesion. The same surgeon (T.N.) served as the operator or assistant. Preoperatively, all patients received mechanical bowel preparation. Two tablets of sennoside were given orally 2 days before surgery, and 1 packet of magnesium citrate and 1 bottle of sodium picosulfate were given the day before surgery. For wound closure, the peritoneum and fascia were sutured together with interrupted, polydioxanone absorbable sutures (0-PDS IITM, Johnson and Johnson Co. Ltd ). The same suture material was used to close the abdomen in patients who underwent laparoscopic surgery and those who underwent open surgery.

In patients who underwent high-pressure irrigation of their wounds, the muscle layer was sutured, and the subcutaneous tissue was washed with warm physiological saline solution applied under high pressure, using a 23-gauge ophthalmic washing catheter attached to a 20-mL syringe. The tip of the needle was positioned about 10 cm from the wound (Fig. 1). The applied volume of physiological saline solution was 400 mL for open surgery, and 200

Risk Factors for Wound Infection After Surgery

Table 1. Demographic characteristics of the patients

surgery was not switched to open surgery in any subject.

for Colorectal Cancer: A Matched Case – Control Study 467

Laparoscopic surgery was performed if tumors invaded the lamina propria (Tis), the submucosa (T1), or the muscularis propria (T2). Open surgery was performed if tumors directly invaded other organs or structures and/or perforated the visceral peritoneum (T4, AI). If tumors invaded through the muscularis propria into the subserosa or into nonperitonealized pericolic or perirectal tissues (T3, A), the surgical procedure was decided in a randomized control trial after obtaining informed consent from the patient. Laparoscopic

After skin closure, the wound was applied a polyurethane film dressing for 48 hours in all patients. Subsequently, the wound was uncovered and was not sterilized. To prevent infection during and after surgery, cefmetazole sodium (1 g/time) or flomoxef sodium (1 g/time) was given as a continuous intravenous infusion 1 hour before surgery and at 3-hour intervals thereafter. On the day after surgery, antibiotics were administered only one time. After the operator confirmed the wound, wound infection was evaluated according to the 1999 Guidelines for the Prevention of Surgical Site Infection (SSI)(9 ). These guidelines do not require the results of culture studies to assess wound infection. In our study, however, a wound culture was performed if fluid or discharge was exuded from the wound. The median postoperative follow-up period was 15 months (range, 7 to 30) in the high-pressure irrigation group and 30 months (range, 21 to 39) in the non-high-pressure irrigation group. During follow-up, patients visited the outpatient clinic at 2- to 4-week intervals. Follow-up

mL for laparoscopic surgery. After irrigation, the skin was closed with polydioxanone absorbable sutures (4-0 PDS II™, Johnson and Johnson Co., Ltd.).

Fig. 1. Before closure of the abdomen during surgery, high-pressure irrigation of the subcutaneous tissue after muscle layer suturing

In patients who did not undergo high-pressure irrigation (non-high-pressure irrigation group), the peritoneum and fascia were sutured together with interrupted, polydioxanone absorbable sutures (0-PDS II™, Johnson and Johnson Co., Ltd.) Then, In the non-highpressure lavage group, no syringe was used. The subcutaneous tissue was just washed with 500 mL of warm physiological saline solution.

The subcutaneous adipose tissue was sutured with polyglactin 910 absorbable sutures (3-0 Vicryl™, Johnson and Johnson Co., Ltd.), and the skin was closed with a skin stapler.

During the 19 months from January 2004 through July 2005, a total of 145 patients underwent surgery without high-pressure irrigation. During 16 months from August 2005 through December 2006, a total of 119 patients underwent surgery with high-pressure irrigation. The two groups were matched for the following 6 variables: sex, age ( 5 years), tumor location (right side of colon, transverse colon, left side of colon, rectum), surgical procedure (laparoscopic surgery, open surgery), tumor-node-metastasis (TNM) classification, and BMI ( 1). We studied a total of 100 patients: 50 in the high-pressure irrigation group and 50 in the non-high-pressure irrigation group (Table 1).

As for the demographic characteristics of the patients, the American Society of Anesthesiologists' physical status classification was class I in 37 patients (74%), class II in 10 (20%), and class III in 3 (6%) in the high-pressure irrigation group and class I in 37 patients (4%), class II in 11 (22%), and class III in 2 (4%) in the non-high-pressure irrigation group. The difference between the groups was not significant (p = 0.884). No patient had a preoperative hemoglobin level of ≤8.0 g/dL. Four patients received blood transfusions during surgery.

mL for laparoscopic surgery. After irrigation, the skin was closed with polydioxanone

Fig. 1. Before closure of the abdomen during surgery, high-pressure irrigation of the

In patients who did not undergo high-pressure irrigation (non-high-pressure irrigation group), the peritoneum and fascia were sutured together with interrupted, polydioxanone absorbable sutures (0-PDS II™, Johnson and Johnson Co., Ltd.) Then, In the non-highpressure lavage group, no syringe was used. The subcutaneous tissue was just washed with

The subcutaneous adipose tissue was sutured with polyglactin 910 absorbable sutures (3-0

During the 19 months from January 2004 through July 2005, a total of 145 patients underwent surgery without high-pressure irrigation. During 16 months from August 2005 through December 2006, a total of 119 patients underwent surgery with high-pressure irrigation. The two groups were matched for the following 6 variables: sex, age ( 5 years), tumor location (right side of colon, transverse colon, left side of colon, rectum), surgical procedure (laparoscopic surgery, open surgery), tumor-node-metastasis (TNM) classification, and BMI ( 1). We studied a total of 100 patients: 50 in the high-pressure

As for the demographic characteristics of the patients, the American Society of Anesthesiologists' physical status classification was class I in 37 patients (74%), class II in 10 (20%), and class III in 3 (6%) in the high-pressure irrigation group and class I in 37 patients (4%), class II in 11 (22%), and class III in 2 (4%) in the non-high-pressure irrigation group. The difference between the groups was not significant (p = 0.884). No patient had a preoperative hemoglobin level of ≤8.0 g/dL. Four patients received blood transfusions

Vicryl™, Johnson and Johnson Co., Ltd.), and the skin was closed with a skin stapler.

irrigation group and 50 in the non-high-pressure irrigation group (Table 1).

subcutaneous tissue after muscle layer suturing

500 mL of warm physiological saline solution.

during surgery.

absorbable sutures (4-0 PDS II™, Johnson and Johnson Co., Ltd.).


#### Table 1. Demographic characteristics of the patients

Laparoscopic surgery was performed if tumors invaded the lamina propria (Tis), the submucosa (T1), or the muscularis propria (T2). Open surgery was performed if tumors directly invaded other organs or structures and/or perforated the visceral peritoneum (T4, AI). If tumors invaded through the muscularis propria into the subserosa or into nonperitonealized pericolic or perirectal tissues (T3, A), the surgical procedure was decided in a randomized control trial after obtaining informed consent from the patient. Laparoscopic surgery was not switched to open surgery in any subject.

After skin closure, the wound was applied a polyurethane film dressing for 48 hours in all patients. Subsequently, the wound was uncovered and was not sterilized. To prevent infection during and after surgery, cefmetazole sodium (1 g/time) or flomoxef sodium (1 g/time) was given as a continuous intravenous infusion 1 hour before surgery and at 3-hour intervals thereafter. On the day after surgery, antibiotics were administered only one time.

After the operator confirmed the wound, wound infection was evaluated according to the 1999 Guidelines for the Prevention of Surgical Site Infection (SSI)(9 ). These guidelines do not require the results of culture studies to assess wound infection. In our study, however, a wound culture was performed if fluid or discharge was exuded from the wound. The median postoperative follow-up period was 15 months (range, 7 to 30) in the high-pressure irrigation group and 30 months (range, 21 to 39) in the non-high-pressure irrigation group. During follow-up, patients visited the outpatient clinic at 2- to 4-week intervals. Follow-up

Risk Factors for Wound Infection After Surgery

infection, wound dehiscence, or adhesive ileus.

**4. Discussion** 

management in the same hospital.

Table 3. Results of multivariate analysis of risk for wound infection

group. The difference between the groups was not significant (p = 0.884).

*aureus* infections were sensitive to the prophylactically administered antibiotics.

for Colorectal Cancer: A Matched Case – Control Study 469

stay after surgery was 8 days (range, 5 to 31) in patients without wound infection, as compared with 15 (range, 7 to 40) in those with wound infection. This difference was significant (p = 0.041). During observation after discharge, there was no flare-up of wound

The American Society of Anesthesiologists' Physical Status Classification was class I in 56 patients (76%), class II in 15 (20%), and class III in 3 (4%) in the laparoscopic surgery group and class I in 16 patients (67%), class II in 6 (23%), and class III in 2 (7%) in the open surgery

Pus samples from infected wounds were cultured in 9 of the 11 patients with wound infection (2 in the high-pressure irrigation group and 7 in the non-high-pressure irrigation group). Pus cultures were positive in all patients in both the high-pressure irrigation group and the nonhigh-pressure irrigation group. The most common pathogen was bacteroides in 6 patients, followed by *Staphylococcus aureus* in 2 and *Pseudomonas aeruginosa* in 1. Pathogens did not differ between the high-pressure irrigation group and the non-high-pressure irrigation group. Strains isolated from the 6 patients with bacteroides infections and the 2 with *Staphylococcus* 

Our study showed that that non-high-pressure irrigation and open surgery were independent risk factors for postoperative wound infection in patients with colorectal cancer. To minimize potential effects of confounding factors, our study was conducted under consistent conditions, i.e., the same operator performed surgery and perioperative

Carlos et al. performed a randomized control trial to evaluate whether syringe pressure irrigation of surgical wounds decreases wound infection after appendectomy (10). Patients were randomly assigned to receive prophylactic antibiotics alone before surgery or prophylactic antibiotics plus syringe pressure irrigation of their wounds. Irrigation was performed by applying 300 mL of physiological saline solution under high pressure with the use of a 20-mL syringe with a 19-gauge intravenous catheter. The tip of the catheter was placed about 2 cm from the wound. Among the 283 patients confirmed to have appendicitis, 95 (33.6%) had complications. Of these 95 patients, wound infection developed in 9 (16.3%) of the 55 patients who received prophylactic antibiotics plus wound irrigation, as compared with 29 (72.5%) of the 40 patients who received prophylactic antibiotics alone (p = 0.0006).

examinations included examination of the surgical wound, administration of adjuvant chemotherapy, and computed tomography (CT) of the chest and abdomen. The preoperative and postoperative status of each patient was retrospectively examined on the basis of medical records.

A total of 10 potential risk factors for wound infection were compared between the groups: the presence or absence of high-pressure irrigation before wound closure, sex, age (<65 years, ≥65 years), BMI (≤25 kg/m2, >25 kg/m2), tumor location (colon, rectum), surgical procedure (laparoscopic surgery, open surgery), operation time (<180 minutes, ≥180 minutes), bleeding volume (<100 mL, ≥100 mL), tumor stage (I or II, III or IV), and antibiotic treatment (cefmetazole, flomoxef), (Table 1).

Chi-square tests with Yates' correction and multivariate logistic regression analysis were performed to identify variables with p values of less than 0.1. P values of less than 0.05 were considered to indicate statistical significance. SPSS version 8.0J (SPSS Inc., Chicago, USA) software was used for analysis.

#### **3. Results**

Wound infections developed after colorectal cancer surgery in 11 (11%) of the 100 patients. On univariate analysis, wound infection was significantly associated with tumors located in the rectum (p = 0.011), open surgery (p = 0.032), and non-high-pressure irrigation of wounds (Table 2). On multivariate analysis, independent risk factors for wound infection were wound treatment (non-high-pressure irrigation) (p = 0.034; odds ratio, 5.968) and surgical procedure (open surgery) (p = 0.039; odds ratio, 4.266) (Table 3).


Table 2. Wound infection detected free of wound infection

Wound infection developed in 5 (7%) of the 74 patients who underwent laparoscopic surgery and 6 (23%) of the 26 patients who underwent open surgery. Wound infection occurred in 2 (4%) of the 50 patients in the high-pressure irrigation group and 9 (18%) of the 50 patients in the non-high-pressure irrigation group. The mean duration of the hospital stay after surgery was 8 days (range, 5 to 31) in patients without wound infection, as compared with 15 (range, 7 to 40) in those with wound infection. This difference was significant (p = 0.041). During observation after discharge, there was no flare-up of wound infection, wound dehiscence, or adhesive ileus.


468 Colorectal Cancer – From Prevention to Patient Care

examinations included examination of the surgical wound, administration of adjuvant chemotherapy, and computed tomography (CT) of the chest and abdomen. The preoperative and postoperative status of each patient was retrospectively examined on the

A total of 10 potential risk factors for wound infection were compared between the groups: the presence or absence of high-pressure irrigation before wound closure, sex, age (<65 years, ≥65 years), BMI (≤25 kg/m2, >25 kg/m2), tumor location (colon, rectum), surgical procedure (laparoscopic surgery, open surgery), operation time (<180 minutes, ≥180 minutes), bleeding volume (<100 mL, ≥100 mL), tumor stage (I or II, III or IV), and antibiotic

Chi-square tests with Yates' correction and multivariate logistic regression analysis were performed to identify variables with p values of less than 0.1. P values of less than 0.05 were considered to indicate statistical significance. SPSS version 8.0J (SPSS Inc., Chicago, USA)

Wound infections developed after colorectal cancer surgery in 11 (11%) of the 100 patients. On univariate analysis, wound infection was significantly associated with tumors located in the rectum (p = 0.011), open surgery (p = 0.032), and non-high-pressure irrigation of wounds (Table 2). On multivariate analysis, independent risk factors for wound infection were wound treatment (non-high-pressure irrigation) (p = 0.034; odds ratio, 5.968) and surgical

Wound infection developed in 5 (7%) of the 74 patients who underwent laparoscopic surgery and 6 (23%) of the 26 patients who underwent open surgery. Wound infection occurred in 2 (4%) of the 50 patients in the high-pressure irrigation group and 9 (18%) of the 50 patients in the non-high-pressure irrigation group. The mean duration of the hospital

procedure (open surgery) (p = 0.039; odds ratio, 4.266) (Table 3).

Table 2. Wound infection detected free of wound infection

basis of medical records.

treatment (cefmetazole, flomoxef), (Table 1).

software was used for analysis.

**3. Results** 

Table 3. Results of multivariate analysis of risk for wound infection

The American Society of Anesthesiologists' Physical Status Classification was class I in 56 patients (76%), class II in 15 (20%), and class III in 3 (4%) in the laparoscopic surgery group and class I in 16 patients (67%), class II in 6 (23%), and class III in 2 (7%) in the open surgery group. The difference between the groups was not significant (p = 0.884).

Pus samples from infected wounds were cultured in 9 of the 11 patients with wound infection (2 in the high-pressure irrigation group and 7 in the non-high-pressure irrigation group). Pus cultures were positive in all patients in both the high-pressure irrigation group and the nonhigh-pressure irrigation group. The most common pathogen was bacteroides in 6 patients, followed by *Staphylococcus aureus* in 2 and *Pseudomonas aeruginosa* in 1. Pathogens did not differ between the high-pressure irrigation group and the non-high-pressure irrigation group. Strains isolated from the 6 patients with bacteroides infections and the 2 with *Staphylococcus aureus* infections were sensitive to the prophylactically administered antibiotics.

#### **4. Discussion**

Our study showed that that non-high-pressure irrigation and open surgery were independent risk factors for postoperative wound infection in patients with colorectal cancer. To minimize potential effects of confounding factors, our study was conducted under consistent conditions, i.e., the same operator performed surgery and perioperative management in the same hospital.

Carlos et al. performed a randomized control trial to evaluate whether syringe pressure irrigation of surgical wounds decreases wound infection after appendectomy (10). Patients were randomly assigned to receive prophylactic antibiotics alone before surgery or prophylactic antibiotics plus syringe pressure irrigation of their wounds. Irrigation was performed by applying 300 mL of physiological saline solution under high pressure with the use of a 20-mL syringe with a 19-gauge intravenous catheter. The tip of the catheter was placed about 2 cm from the wound. Among the 283 patients confirmed to have appendicitis, 95 (33.6%) had complications. Of these 95 patients, wound infection developed in 9 (16.3%) of the 55 patients who received prophylactic antibiotics plus wound irrigation, as compared with 29 (72.5%) of the 40 patients who received prophylactic antibiotics alone (p = 0.0006).

Risk Factors for Wound Infection After Surgery

resection Ann Surg 2004; 239:599-607

randmised trial Lancet 2004; 359:224-229

Epudermiol 1956; 65:85-90

1956; 65:85-90

Hospital Infection 2006 ; 64 : 30-35.

open colectomy Surg Endosc 2002; 16:1420-1425

Control 2005; 33:353-359

**5. References** 

for Colorectal Cancer: A Matched Case – Control Study 471

Many surgical wound infections are caused by bacteroides and *Staphylococcus aureus*. Although these bacteria were sensitive to the antibiotics we administered prophylactically in our study, wound infection developed. The dose and duration of treatment with antibiotics should thus be reassessed. Further multicenter studies are needed to more clearly define risk

[1] Braga M, Vignali A, Gianotti L, et al Laparoscopic versus open colorectal surgery:a Randmized trial on short-Term Outcomes Ann Surg 2002; 236:759-767 [2] Smith RL, Bohl JK, Mcelearney ST, et al Wound infection after elective colorectal

[3] Talbot TR Diabetes mellitus and cardiothoracic surgical site infections Am J Infect

[4] Braga M, Gianotti L, Vinali A Preoperative oral arginine and n-3 fatty acid

[5] Heyland DK, Novak F, Drover JW Should immunonutrition become routine in critically ill patients? A systematic review of the evidence JAMA 2001; 286:944-953 [6] Nagachinta T, Stephens M, Reitz B, et al Risk factors for surgical-wound infections

[7] Lacy AM, Garcia-valdecasas JC, Devadora Salvadora, et al Laparoscopic assisted

[8] Hebert PC:A multicenter, randomized, contorolled clinical trial of transfusion

[9] Mangram AJ Guideline for prevention of surgical site Infection Infect Cont Host

[10] Cervantes-Sanchez CR, Gutierrez-Vega R, Vazquez-Carpizo JA Syringe pressure

[11] Johnson A, Young D, Reilly J Caesarean section surgical infection surveillance. J

[12] Elek SD Experimental staphyloccal infections in the skin of the man Ann NY Acad Sci

[13] Leung KL, Kwok SP, Lam SC Laparoscopic resection of rectosigmoid

[14] Abraham N.S, Young J.M, Solomon M.J Meta-analysis of short-term outcomes after laparoscopic resection for colorectal cancer Br J Surg 2004; 91:1111-1124 [15] Robson MC, Shaw RC, Heggers JP The reclosure of postoperative incisional abscesses based on bacterial quantification of the wound Ann Surg 1970; 171:279-282 [16] Winslow ER, Fleshman JW, Brinbaum EH Wound complications of laparoscopic vs

carcinoma:prospective randomized trial Lancet 2004; 363:1187-1192

canadian critical care trial group N Engl J Med 1999; 340:409-417

postoperative wound infection World J Surg 2000; 24:38-41

colorectal resection for cancer Surgery 2002; 132:805-814

following cardiac surgery J Infect Dis 1987; 156:967-973

supplementation improves the immunometabolic host response and outcome after

colectomy versus colectomy for treatment of non-metastatic colon cancer:a

requirements in critical Care:transfusion Requirements in critical care investigators

irrigation of subdermic tissue after appendectomy to decreasend the incidence of

factors and establish the most effective prophylactic treatment for wound infection.

Johnson et al. prospectively studied the incidence of SSI in 715 patients who underwent a cesarean section procedure (11).The use of subcuticular sutures for skin closure was associated with a significantly higher incidence of SSI than was the use of staples (7.9%, 20/252 vs. 13.0%, 59/459; p = 0.021).

To our knowledge, no previous study has evaluated high-pressure irrigation at the time of wound closure after muscle-layer suture in patients with colorectal cancer. This technique was originally developed in our hospital. A bacterial count of greater than 105 bacteria per gram of tissue is considered necessary for the development of wound infection(12). We believe that our method for high-pressure irrigation after muscle-layer suture effectively decreases the bacterial count in subcutaneous tissue.

Our study showed that the incidence of postoperative wound infection was significantly lower after laparoscopic surgery than after open surgery. Some previous studies have reported that the incidence of wound infection after laparoscopic surgery was similar to that after open surgery, whereas others have found that the incidence of wound infection was significantly lower after laparoscopic surgery(13). A meta-analysis performed by Abraham et al. showed that the incidence of postoperative wound infection was significantly lower after laparoscopic surgery (3.9%) than after open surgery (8.9%) (p<0.005). Our findings support the results of this meta-analysis(14).

The presence of suture material in a closed wound has been reported to increase the number of bacteria to 104 per gram of tissue (15). Open surgery requires a longer skin incision and more stitches than laparoscopic surgery, increasing the bacterial count and potentially increasing the risk of wound infection. One study reported that the incidence of wound infection after laparoscopic surgery was 2.7% at the trocar site and 10.8% at the site of colorectal removal (16). In our hospital, no patient has had wound infection at the trocar site, and the incidence of wound infection at the site of colorectal removal was only 4%, which was lower than that reported previously(17). In general, wound infection has been reported to occur at an incidence of about 20% after open surgery, which is consistent with the incidence of 23% in our study.

The relation between surgical wound infection and diabetes mellitus in patients who undergo surgery for colorectal cancer remains unclear. The risk of SSI after cardiac surgery has been reported to be 2 to 3 times higher in patients with diabetes mellitus than in those without diabetes mellitus, even after adjustment for other risk factors (3). The higher risk of SSI may be related to long-term metabolic and microcirculatory disorders induced by diabetes mellitus, perioperative hyperglycemia, and other risk factors associated with diabetes mellitus, such as obesity. Because we strictly control blood levels before and after surgery in our hospital, diabetes mellitus was not a risk factor for wound infection.

As for the relation between BMI and wound infection, Smith et al. reported that the incidence of wound infection increases in parallel to BMI (2). In their study, 53% of the patients had a BMI of ≥25 kg/m2 or higher. In contrast, only 21% of our patients had a BMI of ≥25 kg/m2. This lower proportion of patients with an increased BMI may have accounted for BMI not being a risk factor for wound infection.

In patients with a BMI of ≥25 kg/m2, we close the wound with subcuticular sutures after inserting a closed subcutaneous drain to maintain fluid drainage. However, firm evidence supporting the use of a subcutaneous drain after colorectal cancer surgery has yet to be obtained. The insertion of a drain may increase healthcare costs and negatively affect patients' ability to walk after surgery. Therefore, further studies are needed.

Many surgical wound infections are caused by bacteroides and *Staphylococcus aureus*. Although these bacteria were sensitive to the antibiotics we administered prophylactically in our study, wound infection developed. The dose and duration of treatment with antibiotics should thus be reassessed. Further multicenter studies are needed to more clearly define risk factors and establish the most effective prophylactic treatment for wound infection.

#### **5. References**

470 Colorectal Cancer – From Prevention to Patient Care

Johnson et al. prospectively studied the incidence of SSI in 715 patients who underwent a cesarean section procedure (11).The use of subcuticular sutures for skin closure was associated with a significantly higher incidence of SSI than was the use of staples (7.9%,

To our knowledge, no previous study has evaluated high-pressure irrigation at the time of wound closure after muscle-layer suture in patients with colorectal cancer. This technique was originally developed in our hospital. A bacterial count of greater than 105 bacteria per gram of tissue is considered necessary for the development of wound infection(12). We believe that our method for high-pressure irrigation after muscle-layer suture effectively

Our study showed that the incidence of postoperative wound infection was significantly lower after laparoscopic surgery than after open surgery. Some previous studies have reported that the incidence of wound infection after laparoscopic surgery was similar to that after open surgery, whereas others have found that the incidence of wound infection was significantly lower after laparoscopic surgery(13). A meta-analysis performed by Abraham et al. showed that the incidence of postoperative wound infection was significantly lower after laparoscopic surgery (3.9%) than after open surgery (8.9%) (p<0.005). Our findings

The presence of suture material in a closed wound has been reported to increase the number of bacteria to 104 per gram of tissue (15). Open surgery requires a longer skin incision and more stitches than laparoscopic surgery, increasing the bacterial count and potentially increasing the risk of wound infection. One study reported that the incidence of wound infection after laparoscopic surgery was 2.7% at the trocar site and 10.8% at the site of colorectal removal (16). In our hospital, no patient has had wound infection at the trocar site, and the incidence of wound infection at the site of colorectal removal was only 4%, which was lower than that reported previously(17). In general, wound infection has been reported to occur at an incidence of about 20% after open surgery, which is consistent with the

The relation between surgical wound infection and diabetes mellitus in patients who undergo surgery for colorectal cancer remains unclear. The risk of SSI after cardiac surgery has been reported to be 2 to 3 times higher in patients with diabetes mellitus than in those without diabetes mellitus, even after adjustment for other risk factors (3). The higher risk of SSI may be related to long-term metabolic and microcirculatory disorders induced by diabetes mellitus, perioperative hyperglycemia, and other risk factors associated with diabetes mellitus, such as obesity. Because we strictly control blood levels before and after

As for the relation between BMI and wound infection, Smith et al. reported that the incidence of wound infection increases in parallel to BMI (2). In their study, 53% of the patients had a BMI of ≥25 kg/m2 or higher. In contrast, only 21% of our patients had a BMI of ≥25 kg/m2. This lower proportion of patients with an increased BMI may have accounted

In patients with a BMI of ≥25 kg/m2, we close the wound with subcuticular sutures after inserting a closed subcutaneous drain to maintain fluid drainage. However, firm evidence supporting the use of a subcutaneous drain after colorectal cancer surgery has yet to be obtained. The insertion of a drain may increase healthcare costs and negatively affect

surgery in our hospital, diabetes mellitus was not a risk factor for wound infection.

patients' ability to walk after surgery. Therefore, further studies are needed.

20/252 vs. 13.0%, 59/459; p = 0.021).

decreases the bacterial count in subcutaneous tissue.

support the results of this meta-analysis(14).

for BMI not being a risk factor for wound infection.

incidence of 23% in our study.


**Modelling and Inference in Screening:** 

*The University of Texas Health Science Center at Houston, Austin,* 

*School of Public Health and Information Sciences, University of Louisville, Louisville,* 

Colorectal cancer (CRC) is the third most common form of cancer and the third leading cancer killer for both genders in the United States (American Cancer Society, 2011). In 2011, the American Cancer Society estimates 141,210 new cases in the U.S. and 49,380 deaths due to CRC (American Cancer Society, 2011). Similarly, the World Health Organization estimates over 940,000 new cases occurring annually worldwide (World Health Organization, 2003), and

Colorectal cancer is often found in people older than fifty, and its mortality rates are higher each year than HIV/AIDS and breast cancer combined (Colon Cancer Prevention Project, 2011). The age-specific colorectal cancer risk rises continuously with advancing age (National Cancer Institute, 2011). CRC is also one of the most preventable among cancers, studies show that regular screening could substantially reduce mortalities (Kronborg et al., 1996, Mandel et al., 1993, 1999). The reason is that colorectal cancer can take 5-15 years to develop, and screening exam, such as digital rectal exam (DRE), colonoscopy, flexible sigmoidoscopy (FSG) and faecal occult blood test (FOBT), can detect polyps before the cancer develops (Screen for Life: National Colorectal Cancer Action Campaign, 2009). The current guidelines in the U.S. for CRC include several options among men and women aged 50 and older at average risk (American Cancer Society, 2011). Many considerations are needed when following the guidelines, including risk factors, type of test, and test interval including flexible sigmoidoscopy, colonoscopy, double-contrast barium enema, computed tomographic colonography, faecal occult blood test and stool DNA test (American Cancer Society, 2011). However, the acceptance of people about CRC screening has been low in the U.S. In most areas of the U.S., less than half of the population is in compliance with recommended CRC guidelines (American Cancer Society, 2011), and only one-third colorectal cancers are being diagnosed at an early and treatable stage, due to lack of

The purpose of colorectal cancer screening is to detect early stage of the disease before clinical symptoms take place. CRC has a 90% treatable rate when detected early, therefore

nearly 610,000 died from CRC in 2008 (World Health Organization, 2011).

screening or lack of disease symptoms (ARUP Laboratories, 2010).

**1. Introduction** 

**Exemplification with the Faecal** 

**Occult Blood Test** 

*USA* 

Dongfeng Wu1 and Adriana Pérez2 *1Department of Bioinformatics and Biostatistics,* 

*2Division of Biostatistics, School of Public Health,* 

[17] Dalibon N, Moutafis M, Fischler M A comparison of laproscopically assisted and open colectomy for colon cancer N Engl J Med 2004; 350: 2050-2059 **24** 

### **Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test**

#### Dongfeng Wu1 and Adriana Pérez2

*1Department of Bioinformatics and Biostatistics, School of Public Health and Information Sciences, University of Louisville, Louisville, 2Division of Biostatistics, School of Public Health, The University of Texas Health Science Center at Houston, Austin, USA* 

#### **1. Introduction**

472 Colorectal Cancer – From Prevention to Patient Care

[17] Dalibon N, Moutafis M, Fischler M A comparison of laproscopically assisted and open

Colorectal cancer (CRC) is the third most common form of cancer and the third leading cancer killer for both genders in the United States (American Cancer Society, 2011). In 2011, the American Cancer Society estimates 141,210 new cases in the U.S. and 49,380 deaths due to CRC (American Cancer Society, 2011). Similarly, the World Health Organization estimates over 940,000 new cases occurring annually worldwide (World Health Organization, 2003), and nearly 610,000 died from CRC in 2008 (World Health Organization, 2011).

Colorectal cancer is often found in people older than fifty, and its mortality rates are higher each year than HIV/AIDS and breast cancer combined (Colon Cancer Prevention Project, 2011). The age-specific colorectal cancer risk rises continuously with advancing age (National Cancer Institute, 2011). CRC is also one of the most preventable among cancers, studies show that regular screening could substantially reduce mortalities (Kronborg et al., 1996, Mandel et al., 1993, 1999). The reason is that colorectal cancer can take 5-15 years to develop, and screening exam, such as digital rectal exam (DRE), colonoscopy, flexible sigmoidoscopy (FSG) and faecal occult blood test (FOBT), can detect polyps before the cancer develops (Screen for Life: National Colorectal Cancer Action Campaign, 2009).

The current guidelines in the U.S. for CRC include several options among men and women aged 50 and older at average risk (American Cancer Society, 2011). Many considerations are needed when following the guidelines, including risk factors, type of test, and test interval including flexible sigmoidoscopy, colonoscopy, double-contrast barium enema, computed tomographic colonography, faecal occult blood test and stool DNA test (American Cancer Society, 2011). However, the acceptance of people about CRC screening has been low in the U.S. In most areas of the U.S., less than half of the population is in compliance with recommended CRC guidelines (American Cancer Society, 2011), and only one-third colorectal cancers are being diagnosed at an early and treatable stage, due to lack of screening or lack of disease symptoms (ARUP Laboratories, 2010).

The purpose of colorectal cancer screening is to detect early stage of the disease before clinical symptoms take place. CRC has a 90% treatable rate when detected early, therefore

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 475

definition, an estimator of sensitivity should be *n nn* 11 11 21 /( ) . However, since the number *n*<sup>21</sup> is unknown, it cannot be obtained directly from data collected in the screening study.

Positive (+) True Positive ( *n*<sup>11</sup> ) False Positive ( *n*<sup>12</sup> )

Negative (-) False Negative ( *n*<sup>21</sup> ) True Negative ( *n*<sup>22</sup> )

Sojourn time refers to the time beginning when the disease first develops until the manifestation of clinical symptoms, which is the time length in the preclinical state. For individuals diagnosed with cancer by screening exams, they will be treated immediately; hence the onset of the clinical state *Sc* is not observable. For individuals diagnosed with cancer between screenings (the interval case), though the onset of the clinical state is available, the onset of the preclinical state is still unknown. Therefore, estimation of the sojourn time distribution is difficult from data collected in screening studies. However, the sojourn time duration can be estimated under model assumptions, the preclinical phase of colorectal cancer may last more than 5 years (American Cancer Society, 2011, Prevost et al.,

The transition probability into the preclinical stage is the probability density function (PDF) of making a transition from the disease-free state to the preclinical state. It is continuously changing with one's age (Wu et al., 2009a) on CRC, and is difficult to estimate without

These three parameters are the key parameters for the estimation of other important indicators in cancer screening, and they cannot be easily estimated from data. We will briefly review the age-dependent likelihood method that we used in estimating these three parameters, and provide the key result using the MCCCS data (Wu et al., 2005, 2009b).

Consider a cohort of initially asymptomatic individuals who enrolled in a screening program. The sensitivity is (t), where t is the individual's age at the screening exam. The probability density function (PDF) of making a transition from *S*<sup>0</sup> to *Sp* at age *t* is *w t*( ) . Let

Consider a cohort of men or women in the study of interest who are all aged 0*t* at study entry, and a protocol for *K* ordered screening exams occurring at ages 01 1 , *<sup>K</sup> tt t* where *<sup>i</sup>* <sup>0</sup> *tti* for annual screening exams. Define the i-th screening interval as the time interval between the i-th and the (i+1)-th screening exams <sup>1</sup> ( , ), *i i t t* i=1, 2,, K-1. We let <sup>1</sup> *t* 0. For each screening exam, let 0 *ni t*, be the total number of individuals in this cohort examined at the i-th screening, 0 *i t*, *s* is the number of diagnosed and confirmed cancer cases

*z Q z q x dx* 

which

*q x*( ) be the probability density function of the sojourn time in *Sp* . Let () ()

is the survivor function of the sojourn time in the preclinical state *Sp* .

Table 1. Illustration of different kinds of counts in a screening study

Disease Status Cancer Non-Cancer

Screening Result

1998).

proper modelling.

**2.1 Model and method** 

screening saves lives (Colon Cancer Prevention Project, 2011). Prevention efforts in the population requires reliable estimates of the sensitivity of the test, the sojourn time of the disease, the transition probabilities from the disease-free state to the preclinical state, the lead time of the disease and the indirect effects in the screening per se in the estimates of rates of the disease.

The aim of this chapter is to introduce the concept of probability modelling in colorectal cancer screening, and the statistical methods developed by the authors in this area (Wu et al., 2005, 2007, 2009a, 2009b). We will estimate these essential components from a population based perspective. In section 2, we provide the definition, model, methods and application of essential parameters needed when estimating indicators of cancer screening. In section 3, we provide the methods and application for estimating the distribution of the lead time in cancer screening. In section 4, we provide the definition, methods and application when evaluating the long term screening outcomes in CRC. Finally, conclusions and recommendations for future research are provided in section 5. We will focus on one particular test, the faecal occult blood test (FOBT). FOBT has been used as a sign of colon cancer, given that tumours tend to bleed and blood in the stool can be detected using this test. We will apply our methods to the Minnesota Colorectal Cancer Control Study (MCCCS) (Mandel et al., 1999), to inform the readers about the benefits of probability modelling in colorectal cancer screening using FOBT, as well as reached recommendations for the test.

The Minnesota Colorectal Cancer Control Study (MCCCS) was carried out between 1976 and 1982 in Minnesota, U.S.A. (Mandel et al., 1999). Approximately 46,000 subjects were randomized to receive either: five annual FOBT screenings, three biennial FOBT screenings or no screening (usual care at the time of the study). Each screening cycle consisted of six hemoccult slides (Hemoccult®, Beckman Coulter, Palo Alto, California); about 83% of slides were re-hydrated. If any of the slides was positive, then the screen was considered positive and a definitive follow-up exam was done, including colonoscopy (Mandel et al., 1999). Due to a lower than expected death rate among the usual care group, the investigators resumed screening between 1986 and 1992. We restricted this analysis to the annual group and to the original five screenings.

#### **2. Sensitivity, sojourn time and transition probability in colon cancer screening**

We assume that the disease develops by progressing through three states, denoted by *SSS* <sup>0</sup> *<sup>p</sup> <sup>c</sup>* . *S*<sup>0</sup> represents the disease-free state. *Sp* represents the preclinical disease state, in which an asymptomatic individual unknowingly has disease that the screening exam can detect. Similarly, *Sc* represents the clinical state when the disease manifests itself in clinical symptoms.

Sensitivity is the probability that the screening exam is positive given that the individual is in the preclinical state *Sp* . The sensitivity cannot be easily estimated from data collected during screening, but can be estimated using probability modelling (Wu et al., 2005, 2009b). We exemplify the rationale for this issue using Table 1. Let us assume the data in Table 1 is generated from a single screening study. It is neither cost effective nor ethical to obtain a biopsy from each individual with a negative screening result. Therefore, the numbers *n*<sup>21</sup> and *n*22 are not available; but *n*<sup>11</sup> , *n*12 and *n n* 21 22 are available information. Using the


definition, an estimator of sensitivity should be *n nn* 11 11 21 /( ) . However, since the number *n*<sup>21</sup> is unknown, it cannot be obtained directly from data collected in the screening study.

Table 1. Illustration of different kinds of counts in a screening study

Sojourn time refers to the time beginning when the disease first develops until the manifestation of clinical symptoms, which is the time length in the preclinical state. For individuals diagnosed with cancer by screening exams, they will be treated immediately; hence the onset of the clinical state *Sc* is not observable. For individuals diagnosed with cancer between screenings (the interval case), though the onset of the clinical state is available, the onset of the preclinical state is still unknown. Therefore, estimation of the sojourn time distribution is difficult from data collected in screening studies. However, the sojourn time duration can be estimated under model assumptions, the preclinical phase of colorectal cancer may last more than 5 years (American Cancer Society, 2011, Prevost et al., 1998).

The transition probability into the preclinical stage is the probability density function (PDF) of making a transition from the disease-free state to the preclinical state. It is continuously changing with one's age (Wu et al., 2009a) on CRC, and is difficult to estimate without proper modelling.

These three parameters are the key parameters for the estimation of other important indicators in cancer screening, and they cannot be easily estimated from data. We will briefly review the age-dependent likelihood method that we used in estimating these three parameters, and provide the key result using the MCCCS data (Wu et al., 2005, 2009b).

#### **2.1 Model and method**

474 Colorectal Cancer – From Prevention to Patient Care

screening saves lives (Colon Cancer Prevention Project, 2011). Prevention efforts in the population requires reliable estimates of the sensitivity of the test, the sojourn time of the disease, the transition probabilities from the disease-free state to the preclinical state, the lead time of the disease and the indirect effects in the screening per se in the estimates of

The aim of this chapter is to introduce the concept of probability modelling in colorectal cancer screening, and the statistical methods developed by the authors in this area (Wu et al., 2005, 2007, 2009a, 2009b). We will estimate these essential components from a population based perspective. In section 2, we provide the definition, model, methods and application of essential parameters needed when estimating indicators of cancer screening. In section 3, we provide the methods and application for estimating the distribution of the lead time in cancer screening. In section 4, we provide the definition, methods and application when evaluating the long term screening outcomes in CRC. Finally, conclusions and recommendations for future research are provided in section 5. We will focus on one particular test, the faecal occult blood test (FOBT). FOBT has been used as a sign of colon cancer, given that tumours tend to bleed and blood in the stool can be detected using this test. We will apply our methods to the Minnesota Colorectal Cancer Control Study (MCCCS) (Mandel et al., 1999), to inform the readers about the benefits of probability modelling in colorectal cancer screening using FOBT, as well as reached recommendations

The Minnesota Colorectal Cancer Control Study (MCCCS) was carried out between 1976 and 1982 in Minnesota, U.S.A. (Mandel et al., 1999). Approximately 46,000 subjects were randomized to receive either: five annual FOBT screenings, three biennial FOBT screenings or no screening (usual care at the time of the study). Each screening cycle consisted of six hemoccult slides (Hemoccult®, Beckman Coulter, Palo Alto, California); about 83% of slides were re-hydrated. If any of the slides was positive, then the screen was considered positive and a definitive follow-up exam was done, including colonoscopy (Mandel et al., 1999). Due to a lower than expected death rate among the usual care group, the investigators resumed screening between 1986 and 1992. We restricted this analysis to the annual group and to the

**2. Sensitivity, sojourn time and transition probability in colon cancer** 

We assume that the disease develops by progressing through three states, denoted by *SSS* <sup>0</sup> *<sup>p</sup> <sup>c</sup>* . *S*<sup>0</sup> represents the disease-free state. *Sp* represents the preclinical disease state, in which an asymptomatic individual unknowingly has disease that the screening exam can detect. Similarly, *Sc* represents the clinical state when the disease manifests itself in clinical

Sensitivity is the probability that the screening exam is positive given that the individual is in the preclinical state *Sp* . The sensitivity cannot be easily estimated from data collected during screening, but can be estimated using probability modelling (Wu et al., 2005, 2009b). We exemplify the rationale for this issue using Table 1. Let us assume the data in Table 1 is generated from a single screening study. It is neither cost effective nor ethical to obtain a biopsy from each individual with a negative screening result. Therefore, the numbers *n*<sup>21</sup> and *n*22 are not available; but *n*<sup>11</sup> , *n*12 and *n n* 21 22 are available information. Using the

rates of the disease.

for the test.

original five screenings.

**screening** 

symptoms.

Consider a cohort of initially asymptomatic individuals who enrolled in a screening program. The sensitivity is (t), where t is the individual's age at the screening exam. The probability density function (PDF) of making a transition from *S*<sup>0</sup> to *Sp* at age *t* is *w t*( ) . Let

*q x*( ) be the probability density function of the sojourn time in *Sp* . Let () () *z Q z q x dx* which

is the survivor function of the sojourn time in the preclinical state *Sp* .

Consider a cohort of men or women in the study of interest who are all aged 0*t* at study entry, and a protocol for *K* ordered screening exams occurring at ages 01 1 , *<sup>K</sup> tt t* where *<sup>i</sup>* <sup>0</sup> *tti* for annual screening exams. Define the i-th screening interval as the time interval between the i-th and the (i+1)-th screening exams <sup>1</sup> ( , ), *i i t t* i=1, 2,, K-1. We let <sup>1</sup> *t* 0. For each screening exam, let 0 *ni t*, be the total number of individuals in this cohort examined at the i-th screening, 0 *i t*, *s* is the number of diagnosed and confirmed cancer cases

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 477

2005-2007 was about 4.97% for females and 5.30% for males. Wu et al. 2005, 2009b shows the detailed justifications on how these age effect functions were chosen. Models in equation 1-5 were estimated using programs C/C++ (Silicon Graphics, I, 2003, Stroustrup, B, 2011) and we applied the likelihood separately for men and women in the MCCCS. Markov Chain Monte Carlo (MCMC) was used to generate random samples from the joint posterior distribution of the parameters in the likelihood for Bayesian inference (Wu et al., 2005, 2009b). The posterior distribution within the MCMC was partitioned into four sub-chains,

priors were used for all parameters (Wu et al., 2009b). Each MCMC was run for 20,000 steps; after a burn-in of 15,000 iterations, then posterior samples were collected every 20 steps, which finally provided 250 samples from each chain (Wu et al., 2009b). Because four overdispersed chains were simulated using MCMC, a pool of 1000 posterior samples were used for the analysis presented below. These Bayesian posterior samples are notated as \*

Bayes estimates of the highest posterior density (HPD) interval were also computed, which are similar to confidence interval from a frequentist perspective and also known as credible

The original FOBT screening data from MCCCS for each age group, male and female, are

the standard errors for each gender are listed in table 3 in Wu et al. 2009b. The age-

30 0.117 0.325 0.372 0.024 0.113 0.190 35 0.192 0.361 0.368 0.090 0.207 0.258 45 0.458 0.494 0.331 0.488 0.562 0.332 55 0.791 0.749 0.213 1.101 1.122 0.278 65 0.943 0.863 0.187 1.594 1.597 0.381 75 0.980 0.879 0.191 1.683 1.692 0.467 85 0.994 0.868 0.219 1.474 1.442 0.364

Median Mean S.E. Median Mean S.E.

published in table 1 and 2 in Wu et al. 2009b. The posterior estimates for parameters

The sensitivity appears to increase with age for both male and female. A Bayes hypothesis test of 0 1 *H b*: 0 versus 1 1 *H b*: 0 showed that for males, the posterior probability of a positive slope was <sup>1</sup> *P b Data* ( 0| ) 0.806 ; for females, this posterior probability of a positive

0 1 ( , ), , ,( , ) *b b* 

separately. Non-informative

and the transition density *w t*( ) for males

Transition probabilitya for *w t*( )

and *w t*( ) for male participants.

*<sup>j</sup>* .

and

e.g. sampling the posterior distribution for <sup>2</sup>

intervals from the Bayesian perspective.

dependent Bayesian estimates of the sensitivity

and females are listed in table 2 and 3 here.

Age Sensitivity

Table 2. Bayesian posterior estimates of

**2.2 Results** 

aThe unit is <sup>3</sup> 10 .

slope was 0.941.

at the i-th screening exam, and 0 *i t*, *r* is the number of cases diagnosed in the clinical state *Sc* within the interval <sup>1</sup> ( ,) *i i t t* , the interval cases.

The likelihood function for each gender cohort is:

$$L = \prod\_{t\_0} \prod\_{k=1}^{K} D\_{k,t\_0}^{s\_{k,t\_0}} I\_{k,t\_0}^{r\_{k,t\_0}} (1 - D\_{k,t\_0} - I\_{k,t\_0})^{n\_{k,t\_0} - s\_{k,t\_0} - r\_{k,t\_0}} \ . \tag{1}$$

To facilitate the understanding of this likelihood function, we will describe it in terms of the MCCCS age groups. In the MCCCS, the initial age of participants varied from 28 to 90 years old, among men, and 36 to 93 years old, among women, so this is the range of 0*t* . Because the MCCCS required five annual FOBT screenings, K = 5. The <sup>0</sup> *Dk t*, is the probability that an individual will be diagnosed at the k-th scheduled exam given that she/he is in *Sp* (see equation 2 and 3); and the <sup>0</sup> *k t*, *I* is the probability of being an incident case in the k-th screening interval (see equation 4).

$$D\_{1,t\_0} = \beta\_0 \int\_0^{t\_0} w(\mathbf{x}) Q(t\_0 - \mathbf{x}) d\mathbf{x} \,\tag{2}$$

$$D\_{k,t\_0} = \beta\_{k-1} \left\{ \sum\_{i=0}^{k-2} [1 - \beta\_i] \cdots [1 - \beta\_{k-2}] \int\_{t\_{i-1}}^{t\_i} w(\mathbf{x}) Q(t\_{k-1} - \mathbf{x}) d\mathbf{x} + \int\_{t\_{k-2}}^{t\_{k-1}} w(\mathbf{x}) Q(t\_{k-1} - \mathbf{x}) d\mathbf{x} \right\}, \tag{3}$$
  $\mathbf{k} = \mathbf{2}, \ldots \mathbf{K}$ .

$$\begin{split} I\_{k,t\_0} &= \sum\_{i=0}^{k-1} [1 - \beta\_i] \cdots [1 - \beta\_{k-1}] \int\_{t\_{i-1}}^{t\_i} w(\mathbf{x}) [Q(t\_{k-1} - \mathbf{x}) - Q(t\_k - \mathbf{x})] d\mathbf{x} \\ &+ \int\_{t\_{k-1}}^{t\_k} w(\mathbf{x}) [1 - Q(t\_k - \mathbf{x})] \, d\mathbf{x} \; \; \; k = 1, \dots, \mathbf{K}. \end{split} \tag{4}$$

Where ( ) *i i t* in the above formulae. We modelled the age effect *t* and the time duration *x* in the preclinical state very carefully using a parametric model stated in equation 5.

$$\begin{split} \beta(t) &= \frac{1}{1 + \exp\{-b\_0 - b\_1 \ast (t - m)\}}, \\ w(t) &= \frac{0.1}{\sqrt{2\pi\sigma t}} \exp\{- (\log t - \mu)^2 \mid (2\sigma^2) \}. \\ q(\mathbf{x}) &= \frac{\kappa \mathbf{x}^{\kappa - 1} \boldsymbol{\rho}^{\kappa}}{\left(1 + \left(\mathbf{x} \boldsymbol{\rho}\right)^{\kappa}\right)^2}. \end{split} \tag{5}$$

Where *m* is the average age-at-entry in the study and <sup>2</sup> 0 1 *b b*, ,, ,, are unknown parameters to be estimated. For simplicity, we will include these parameters in a vector form as <sup>2</sup> 0 1 ( , ,, ,,) *b b* . If *1b 0* , then ( )*t* will be a monotone increasing function of age *t*. Usually, researchers can establish an upper bound for *w t*( ) from pilot studies or from health department data. For example, in the MCCCS, we picked 10% as a reasonable upper bound for *w t*( ) because the lifetime risk of being diagnosed with colorectal cancer reported by the National Cancer Institute (SEER Cancer Statistics Review 1975-2007, 2010) for years 2005-2007 was about 4.97% for females and 5.30% for males. Wu et al. 2005, 2009b shows the detailed justifications on how these age effect functions were chosen. Models in equation 1-5 were estimated using programs C/C++ (Silicon Graphics, I, 2003, Stroustrup, B, 2011) and we applied the likelihood separately for men and women in the MCCCS. Markov Chain Monte Carlo (MCMC) was used to generate random samples from the joint posterior distribution of the parameters in the likelihood for Bayesian inference (Wu et al., 2005, 2009b). The posterior distribution within the MCMC was partitioned into four sub-chains, e.g. sampling the posterior distribution for <sup>2</sup> 0 1 ( , ), , ,( , ) *b b* separately. Non-informative priors were used for all parameters (Wu et al., 2009b). Each MCMC was run for 20,000 steps; after a burn-in of 15,000 iterations, then posterior samples were collected every 20 steps, which finally provided 250 samples from each chain (Wu et al., 2009b). Because four overdispersed chains were simulated using MCMC, a pool of 1000 posterior samples were used for the analysis presented below. These Bayesian posterior samples are notated as \* *<sup>j</sup>* . Bayes estimates of the highest posterior density (HPD) interval were also computed, which are similar to confidence interval from a frequentist perspective and also known as credible intervals from the Bayesian perspective.

#### **2.2 Results**

476 Colorectal Cancer – From Prevention to Patient Care

at the i-th screening exam, and 0 *i t*, *r* is the number of cases diagnosed in the clinical state *Sc*

0 0 0 0

To facilitate the understanding of this likelihood function, we will describe it in terms of the MCCCS age groups. In the MCCCS, the initial age of participants varied from 28 to 90 years old, among men, and 36 to 93 years old, among women, so this is the range of 0*t* . Because the MCCCS required five annual FOBT screenings, K = 5. The <sup>0</sup> *Dk t*, is the probability that an individual will be diagnosed at the k-th scheduled exam given that she/he is in *Sp* (see equation 2 and 3); and the <sup>0</sup> *k t*, *I* is the probability of being an incident case in the k-th

*L DI(D I)*

0 <sup>0</sup> 1, 0 <sup>0</sup> <sup>0</sup> () ( ) , *<sup>t</sup> D w x Q t x dx <sup>t</sup>*

, 1 2 1 1

( )[1 ( )] , k 1, K.

*k t <sup>i</sup> <sup>k</sup> <sup>k</sup> <sup>k</sup> <sup>t</sup> <sup>i</sup>*

*i i*

*I w x Q t x Q t x dx*

duration *x* in the preclinical state very carefully using a parametric model stated in

0 1

0.1 ( ) exp{ (log ) /(2 )}, <sup>2</sup>

parameters to be estimated. For simplicity, we will include these parameters in a vector

age *t*. Usually, researchers can establish an upper bound for *w t*( ) from pilot studies or from health department data. For example, in the MCCCS, we picked 10% as a reasonable upper bound for *w t*( ) because the lifetime risk of being diagnosed with colorectal cancer reported by the National Cancer Institute (SEER Cancer Statistics Review 1975-2007, 2010) for years

*b b tm*

<sup>1</sup> ( ) , 1 exp( \* ( ))

2

[1 ] [1 ] ( )[ ( ) ( )]

*t* in the above formulae. We modelled the age effect *t* and the time

*<sup>k</sup> t t kt k <sup>i</sup> <sup>k</sup> <sup>k</sup> <sup>k</sup> t t <sup>i</sup>*

<sup>0</sup> 1 2

, 1 1

1

 

*w t t t x*

( ) . (1 ( ) )

Where *m* is the average age-at-entry in the study and <sup>2</sup>

. If *1b 0* , then

*x* 

 

*w x Q t x dx*

0 0 000

. (1)

(2)

1

(4)

(5)

are unknown

1 *k,t k,t k,t k,t k,t <sup>K</sup> s r nsr k,t k,t k,t k,t*

[1 ] [1 ] ( ) ( ) ( ) ( )

*i k i k*

, (3)

*w x Q t x dx w x Q t x dx*

2 2

 

0 1 *b b*, ,, ,, 

( )*t* will be a monotone increasing function of

within the interval <sup>1</sup> ( ,) *i i t t* , the interval cases. The likelihood function for each gender cohort is:

screening interval (see equation 4).

k 2, K.

 

*D*

Where ( ) *i i* 

form as <sup>2</sup> 0 1

 ( , ,, ,,) *b b* 

equation 5.

2

1

1

0

*k k*

*t*

<sup>0</sup> <sup>1</sup>

*<sup>k</sup> <sup>t</sup>*

*t*

*q x*

*<sup>k</sup> <sup>t</sup>*

0

<sup>0</sup> 1

*t k*

The original FOBT screening data from MCCCS for each age group, male and female, are published in table 1 and 2 in Wu et al. 2009b. The posterior estimates for parameters and the standard errors for each gender are listed in table 3 in Wu et al. 2009b. The agedependent Bayesian estimates of the sensitivity and the transition density *w t*( ) for males and females are listed in table 2 and 3 here.


aThe unit is <sup>3</sup> 10 .

Table 2. Bayesian posterior estimates of and *w t*( ) for male participants.

The sensitivity appears to increase with age for both male and female. A Bayes hypothesis test of 0 1 *H b*: 0 versus 1 1 *H b*: 0 showed that for males, the posterior probability of a positive slope was <sup>1</sup> *P b Data* ( 0| ) 0.806 ; for females, this posterior probability of a positive slope was 0.941.

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 479

is not detected by the regular screening exam but who develops clinical symptoms between exams. The distribution of the lead time is a mixture of the conditional probability *PL D* ( 0| 1) and the conditional probability density function ( | 1) *Lf zD* , for any <sup>0</sup> 0 *zTt* . Here, *T* represents the span of the human life, which is a fixed upper bound, and 0*t* is the individual's age at his/her initial screening exam. We define the same

( ), ( ), ( ), ( ) *t wt qx Qx* as in Section 2.1. The distribution for the lead time was derived and

and ( , 1) ( | 1) ( 1)

The lead time is zero if and only if the individual is an interval case, therefore the joint probability ,1 ,2 , ( 0, 1) *PL D I I I K K K K* , where *K j*, *I* is the probability of an interval

1

*I w x Q t x Q t x dx*

*i i*

*K j <sup>i</sup> <sup>j</sup> <sup>j</sup> <sup>j</sup> <sup>t</sup> <sup>i</sup>*

1

*<sup>j</sup> <sup>i</sup> t t L i ri i <sup>i</sup> t t i r*

( , 1) (1 ) (1 ) ( ) ( ) ( ) ( )

*f z D w x q t z x dx w x q t z x dx*

distribution for the male and female participants. To do this, the posterior predictive distribution of the lead time can be estimated by MCMC (Wu et al., 2009a) as stated in

<sup>1</sup> \* ( | ) ( , | ) ( | , ) ( | ) ( | ). *<sup>j</sup>*

 

*f z Data f z Data d f z Data f Data d f z <sup>n</sup>*

(1 ) (1 ) ( )[ ( ) ( )]

*<sup>t</sup> P D w x Q t x Q T x dx w x Q T x dx* , is the probability of

*L*

*L*

. (6)

(7)

(8)

*f zD f zD P D*

*t* is the sensitivity at age *it* . The joint PDF ( , 1) *Lf zD* in

1 1

*<sup>q</sup> t z x dx* , if 1 0 *z T tT t* (,) . (9)

*<sup>j</sup>* estimated from equation 5 to project the lead time

*j*

 

represents the mixture distribution in equation 6. The sample size *n* is 1000

(10)

*r i r i*

Where <sup>0</sup>

presented in equation 6 (Wu et al., 2007, 2009a).

( 0, 1) ( 0| 1) ( 1) *PL D PL D*

developing colorectal cancer in one's life time after age <sup>0</sup>*t* .

case within the interval <sup>1</sup> ( ,) *j j t t* , and it was derived as:

1

 

1

0

*j j*

*t*

equation 6 was derived and presented as:

1 1

1 0

<sup>0</sup> <sup>0</sup> <sup>0</sup>

<sup>0</sup>

0

*t*

We used the posterior samples \*

equation 10.

Where \* (| )*<sup>j</sup> f z*

in section 2 from the MCMC.

for all j=1, 2… K, with ( ) *i i*

*P D*

<sup>0</sup> <sup>0</sup> <sup>0</sup> ( 1) ( )[ ( ) ( )] ( )[1 ( )] *t T*

, 11

 

*<sup>j</sup> <sup>t</sup>*

*<sup>j</sup> <sup>t</sup>*

( )( ) .

*Lf zD wx* 

*w x q t z x dx*

( )[1 ( )] .

*w x Q t x dx*

Where <sup>1</sup> ( , ), 2, *j j z T tT t j K* ; and when j=1, it is simplified as:

0 0 <sup>0</sup> ( , 1) ( ) ( ) *<sup>t</sup>*

The age-dependent transition probability was itself a sub-pdf from our model assumption. The posterior mean transition probability varied from <sup>3</sup> 0.113 10 to <sup>3</sup> 1.707 10 for males aged 30 to 90 and varied from <sup>3</sup> 0.069 10 to <sup>3</sup> 2.009 10 for females aged 40 to 90. The transition probability was not a monotone function of age, having a single maximum at age 72 for males and a single maximum at age 75 for female.


aThe unit is <sup>3</sup> 10 .

Table 3. Bayesian posterior estimates of and *w t*( ) for female participants.

The posterior mean sojourn time was 4.08 years for males and 2.41 years for females, with a posterior median of 1.66 years for males and 1.88 years for females. The 95% HPD interval was (0.97, 20.28) for males and (1.15, 5.96) for females. This shows that CRC may have a large variation in sojourn time, as oncologists believe that CRC usually has a long sojourn time more than 5 years.

#### **3. Distribution of the lead time in colorectal cancer screening**

The goal of screening is to catch the disease before clinical symptom appears. This means that the detection and removal of any precancerous growth is important as well as the diagnosis of cancer at an early stage. To understand this, several time events are essential to prevention efforts and they will be described briefly here. If a person enters the preclinical state ( *Sp* ) at age 1*t* , and his/her clinical symptoms present later at age 2*t* , then 2 1 ( ) *t t* is the sojourn time in the preclinical state. If a person is offered a screening exam at some time point *t* within the interval 1 2 (,) *t t* , and cancer is diagnosed, then the length of the time <sup>2</sup> ( ) *t t* is the lead time. We consider lead time as the time gained by screening for that particular person.

#### **3.1 Methods**

We will briefly review the probability distribution of the lead time derived under a progressive disease model (Wu et al., 2007, 2009a). Assume there are *K* ordered screenings that, for a specific individual, occur at ages 01 1 *<sup>K</sup> tt t* . The lead time distribution is a conditional distribution given that someone will develop clinical disease before death. We let *D* represent a Bernoulli random variable, with *D* 1 indicating the development of clinical disease and 0 *D* indicating the absence of the clinical disease before death. We use *L* to denote the lead time. We consider the lead time to be zero for individuals whose disease is not detected by the regular screening exam but who develops clinical symptoms between exams. The distribution of the lead time is a mixture of the conditional probability *PL D* ( 0| 1) and the conditional probability density function ( | 1) *Lf zD* , for any <sup>0</sup> 0 *zTt* . Here, *T* represents the span of the human life, which is a fixed upper bound, and 0*t* is the individual's age at his/her initial screening exam. We define the same ( ), ( ), ( ), ( ) *t wt qx Qx* as in Section 2.1. The distribution for the lead time was derived and presented in equation 6 (Wu et al., 2007, 2009a).

$$P(L=0 \mid D=1) = \frac{P(L=0, D=1)}{P(D=1)} \text{ and } \ f\_L(z \mid D=1) = \frac{f\_L(z, D=1)}{P(D=1)}.\tag{6}$$

Where <sup>0</sup> <sup>0</sup> <sup>0</sup> <sup>0</sup> ( 1) ( )[ ( ) ( )] ( )[1 ( )] *t T <sup>t</sup> P D w x Q t x Q T x dx w x Q T x dx* , is the probability of

developing colorectal cancer in one's life time after age <sup>0</sup>*t* .

478 Colorectal Cancer – From Prevention to Patient Care

The age-dependent transition probability was itself a sub-pdf from our model assumption. The posterior mean transition probability varied from <sup>3</sup> 0.113 10 to <sup>3</sup> 1.707 10 for males aged 30 to 90 and varied from <sup>3</sup> 0.069 10 to <sup>3</sup> 2.009 10 for females aged 40 to 90. The transition probability was not a monotone function of age, having a single maximum at age

Median Mean S.E. Median Mean S.E.

45 0.333 0.418 0.315 0.113 0.162 0.161 55 0.802 0.769 0.191 0.600 0.624 0.228 65 0.976 0.920 0.122 1.413 1.427 0.273 75 0.996 0.955 0.093 1.968 2.009 0.468 85 0.999 0.964 0.098 1.931 1.946 0.392

The posterior mean sojourn time was 4.08 years for males and 2.41 years for females, with a posterior median of 1.66 years for males and 1.88 years for females. The 95% HPD interval was (0.97, 20.28) for males and (1.15, 5.96) for females. This shows that CRC may have a large variation in sojourn time, as oncologists believe that CRC usually has a long sojourn

The goal of screening is to catch the disease before clinical symptom appears. This means that the detection and removal of any precancerous growth is important as well as the diagnosis of cancer at an early stage. To understand this, several time events are essential to prevention efforts and they will be described briefly here. If a person enters the preclinical state ( *Sp* ) at age 1*t* , and his/her clinical symptoms present later at age 2*t* , then 2 1 ( ) *t t* is the sojourn time in the preclinical state. If a person is offered a screening exam at some time point *t* within the interval 1 2 (,) *t t* , and cancer is diagnosed, then the length of the time <sup>2</sup> ( ) *t t* is the lead time. We consider lead time as the time gained by screening for that

We will briefly review the probability distribution of the lead time derived under a progressive disease model (Wu et al., 2007, 2009a). Assume there are *K* ordered screenings that, for a specific individual, occur at ages 01 1 *<sup>K</sup> tt t* . The lead time distribution is a conditional distribution given that someone will develop clinical disease before death. We let *D* represent a Bernoulli random variable, with *D* 1 indicating the development of clinical disease and 0 *D* indicating the absence of the clinical disease before death. We use *L* to denote the lead time. We consider the lead time to be zero for individuals whose disease

Transition probability a for

and *w t*( ) for female participants.

72 for males and a single maximum at age 75 for female.

**3. Distribution of the lead time in colorectal cancer screening** 

Age Sensitivity

Table 3. Bayesian posterior estimates of

aThe unit is <sup>3</sup> 10 .

time more than 5 years.

particular person.

**3.1 Methods** 

The lead time is zero if and only if the individual is an interval case, therefore the joint probability ,1 ,2 , ( 0, 1) *PL D I I I K K K K* , where *K j*, *I* is the probability of an interval case within the interval <sup>1</sup> ( ,) *j j t t* , and it was derived as:

$$\begin{split} I\_{K,j} &= \sum\_{i=0}^{j-1} (\mathbf{1} - \beta\_i) \cdots (\mathbf{1} - \beta\_{j-1}) \int\_{t\_{i-1}}^{t\_i} w(\mathbf{x}) [Q(t\_{j-1} - \mathbf{x}) - Q(t\_j - \mathbf{x})] d\mathbf{x} \\ &+ \int\_{t\_{j-1}}^{t\_j} w(\mathbf{x}) [\mathbf{1} - Q(t\_j - \mathbf{x})] d\mathbf{x}. \end{split} \tag{7}$$

for all j=1, 2… K, with ( ) *i i t* is the sensitivity at age *it* . The joint PDF ( , 1) *Lf zD* in equation 6 was derived and presented as:

$$\begin{split} f\_{L}(\mathbf{z}, \mathbf{D} = \mathbf{1}) &= \sum\_{i=1}^{j-1} \beta\_{i} \left\{ \sum\_{r=0}^{i-1} (\mathbf{1} - \beta\_{r}) \cdots (\mathbf{1} - \beta\_{i-1}) \right\}\_{t\_{r-1}}^{t\_{r}} w(\mathbf{x}) q(t\_{i} + \mathbf{z} - \mathbf{x}) d\mathbf{x} + \int\_{t\_{i-1}}^{t\_{i}} w(\mathbf{x}) q(t\_{i} + \mathbf{z} - \mathbf{x}) d\mathbf{x} \right\} \\ &+ \beta\_{0} \int\_{0}^{t\_{0}} w(\mathbf{x}) q(t\_{\mathbf{y}} + \mathbf{z} - \mathbf{x}) d\mathbf{x}. \end{split} \tag{8}$$

Where <sup>1</sup> ( , ), 2, *j j z T tT t j K* ; and when j=1, it is simplified as:

$$f\_L(z, D=1) = \beta\_0 \int\_0^{t\_0} w(\mathbf{x}) q(t\_0 + z - \mathbf{x}) d\mathbf{x} \text{, if } z \in (T - t\_1, T - t\_0) \text{ .} \tag{9}$$

We used the posterior samples \* *<sup>j</sup>* estimated from equation 5 to project the lead time distribution for the male and female participants. To do this, the posterior predictive distribution of the lead time can be estimated by MCMC (Wu et al., 2009a) as stated in equation 10.

$$f(z \mid \text{Data}) = \int f(z, \theta \mid \text{Data}) d\theta = \int f(z \mid \theta, \text{Data}) f(\theta \mid \text{Data}) d\theta \approx \frac{1}{n} \sum\_{j} f(z \mid \theta\_j^{\*}). \tag{10}$$

Where \* (| )*<sup>j</sup> f z* represents the mixture distribution in equation 6. The sample size *n* is 1000 in section 2 from the MCMC.

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 481

(Day, 2005, Duffy et al., 2008, Welch & Black, 2010, Zackrisson et al., 2006), there is little reference to this problem in colorectal cancer. The flaws of using observational studies are obvious: (a) the result based on one study cannot be extended to other scenarios. The reason is that for one particular study, with one particular screening interval, the result may be correct, however, one cannot use this result to make inference for studies with different screening intervals or different cohorts without probability modelling. On the other hand, it is clear that it is of great value to policy makers to know how the proportion of over diagnosis is changing with screening frequency, sensitivity of the screening modality, and other risk factors; (b) using observational studies usually needs a long follow-up period to collect cancer incidence data from both the screening group and the control group, because most of the observational studies compares the incidence rates in the two groups to estimate over diagnosis. This is not cost effective, and the inference

To overcome these flaws, we used probability modelling, and instead of dealing with over diagnosis alone, we will address the long-term outcomes for the whole cohort, with over diagnosis as one outcome. All initially superficially healthy participants will be classified into four mutually exclusive categories: true-early-detection, no-early-detection, over




Every participant who takes part in the screening will eventually fall into one of these four outcomes. It is hoped that this will provide a systematic approach and a frame work for the

For an initially asymptomatic individual taking *K* screenings at their ages 01 1 *<sup>K</sup> tt t* , the conditional probability for each of the four cases was derived, given that their lifetime

0

*K K*

*t t K K t*

*<sup>t</sup> <sup>K</sup>*

*j t*

1

( 1: | ) 1 ( ) ( ) ( )

*P Case SympF T t x dx x Q t x dx*

0

1

 

*K*

 

1

,1 ,2 , ( 2: | ) . *P Case NoED T t I I I kK K K K* (12)

*jK k*

(1 )...(1 ) ( ) ( )

1

*w x Q t x dx*

(11)

*j*

*j*

whose disease manifest itself clinically and was not detected by screening.

diagnosis and symptom-free-life (Wu & Rosner, 2010).

evaluation of long term outcomes in cancer screening.

<sup>1</sup> ( ) *Tt t K K* and presented in equations 11-15.

appeared before death.

appeared before death.

**4.1 Methods** 

maybe biased.

#### **3.2 Results**

We applied our method to make predictive inference of the lead time using FOBT for males and females. We assumed for this simulation that the initial age is 50, and an ending age of 80. It is clear that the lead time distribution is a function of the sensitivity, the sojourn time distribution, the transition density, the screening frequency, and the initial age and ending age. Accurate estimation of the sensitivity, the sojourn time distribution and transition density were acquired from MCCCS study in section 2. Now, we plugged the estimates obtained from Section 2 into the simulation equation 10 in Section 3.1, leading to the estimation of the lead time distribution under different screening scenarios. In other words, we estimated what the results would be if people were screened at different screening intervals. The results are summarized in table 2 in Wu et al. 2009a. The time interval between screens was 6, 9, 12, 18 and 24 months, within ages 50 ( <sup>0</sup>*t* ) and 80 years (*T* ). Also, the density curves for the lead time are shown in Figure 2 and 3 in Wu et al. 2009a for different screening intervals for both males and females. From those results, if a man begins annual screening (i.e. = 12 months) when he is 50 years old and continues until he reaches 80, then there is a 18.87% chance that he will not benefit from early detection by the screening program if he develops colorectal cancer during those thirty years. His chance of no-early-detection from the screening program decreases to 6.45% for screening exams conducted 6 months apart. While for females, the chance of no-early-detection is 9.48% for annual screenings and 2.39% for screening every 6 months.

Also, Table 2 in Wu et al. 2009a showed that the mean lead time increases as the screening time interval decreases for both males and females. In other words, more screening exams will contribute to a longer lead time, which would translate to treatment of the disease at an earlier stage and, potentially improved prognosis. The increase in the mean lead time is partly due to the smaller point mass for zero lead time when screening exams are closer together. The standard error of the lead time decreases as the time between screening exams increases. Similarly, Table 2 in Wu et al. 2009a revealed that the standard deviation for the lead time was larger than the mean lead time (Wu et al., 2009a). In the same table, the mode of the lead time, which is the value that is most likely taken by the lead time when it is positive, was 0.68 years (or 8 months), corresponding to screening exams every 6 months for males, and 0.96 years (or 11.5 months) for females (Wu et al., 2009a). With annual exams, the mode value for the lead time is 0.60 years (6 months) for males and 0.78 years (9.4 months) for females.

#### **4. Evaluating long term screening outcomes in colorectal cancer**

Recently there have been heated arguments in the topic of over diagnosis, the diagnosis of ``disease" that will not cause symptoms or death during a patient's lifetime (Lichtenfeld, J L, 2010). Some profound questions should be asked with regards to over diagnosis. How do we evaluate the long-term outcomes due to continuous regular screening? Will regular screening exams contribute to a greater chance of over diagnosis? What are the percentages of over diagnosis and true-early-detection among the screen-detected cancer patients? How should the probability of no-early-detection and the probability of disease-free-life be estimated?

Some research has been done in the area of over diagnosis. However, the majority of research in this area has been based on observational studies, and mainly in breast cancer (Day, 2005, Duffy et al., 2008, Welch & Black, 2010, Zackrisson et al., 2006), there is little reference to this problem in colorectal cancer. The flaws of using observational studies are obvious: (a) the result based on one study cannot be extended to other scenarios. The reason is that for one particular study, with one particular screening interval, the result may be correct, however, one cannot use this result to make inference for studies with different screening intervals or different cohorts without probability modelling. On the other hand, it is clear that it is of great value to policy makers to know how the proportion of over diagnosis is changing with screening frequency, sensitivity of the screening modality, and other risk factors; (b) using observational studies usually needs a long follow-up period to collect cancer incidence data from both the screening group and the control group, because most of the observational studies compares the incidence rates in the two groups to estimate over diagnosis. This is not cost effective, and the inference maybe biased.

To overcome these flaws, we used probability modelling, and instead of dealing with over diagnosis alone, we will address the long-term outcomes for the whole cohort, with over diagnosis as one outcome. All initially superficially healthy participants will be classified into four mutually exclusive categories: true-early-detection, no-early-detection, over diagnosis and symptom-free-life (Wu & Rosner, 2010).


Every participant who takes part in the screening will eventually fall into one of these four outcomes. It is hoped that this will provide a systematic approach and a frame work for the evaluation of long term outcomes in cancer screening.

#### **4.1 Methods**

480 Colorectal Cancer – From Prevention to Patient Care

We applied our method to make predictive inference of the lead time using FOBT for males and females. We assumed for this simulation that the initial age is 50, and an ending age of 80. It is clear that the lead time distribution is a function of the sensitivity, the sojourn time distribution, the transition density, the screening frequency, and the initial age and ending age. Accurate estimation of the sensitivity, the sojourn time distribution and transition density were acquired from MCCCS study in section 2. Now, we plugged the estimates obtained from Section 2 into the simulation equation 10 in Section 3.1, leading to the estimation of the lead time distribution under different screening scenarios. In other words, we estimated what the results would be if people were screened at different screening intervals. The results are summarized in table 2 in Wu et al. 2009a. The time interval between screens was 6, 9, 12, 18 and 24 months, within ages 50 ( <sup>0</sup>*t* ) and 80 years (*T* ). Also, the density curves for the lead time are shown in Figure 2 and 3 in Wu et al. 2009a for different screening intervals for both males and females. From those results, if a man begins

80, then there is a 18.87% chance that he will not benefit from early detection by the screening program if he develops colorectal cancer during those thirty years. His chance of no-early-detection from the screening program decreases to 6.45% for screening exams conducted 6 months apart. While for females, the chance of no-early-detection is 9.48% for

Also, Table 2 in Wu et al. 2009a showed that the mean lead time increases as the screening time interval decreases for both males and females. In other words, more screening exams will contribute to a longer lead time, which would translate to treatment of the disease at an earlier stage and, potentially improved prognosis. The increase in the mean lead time is partly due to the smaller point mass for zero lead time when screening exams are closer together. The standard error of the lead time decreases as the time between screening exams increases. Similarly, Table 2 in Wu et al. 2009a revealed that the standard deviation for the lead time was larger than the mean lead time (Wu et al., 2009a). In the same table, the mode of the lead time, which is the value that is most likely taken by the lead time when it is positive, was 0.68 years (or 8 months), corresponding to screening exams every 6 months for males, and 0.96 years (or 11.5 months) for females (Wu et al., 2009a). With annual exams, the mode value for the lead time is 0.60 years (6 months) for males and 0.78 years (9.4 months)

**4. Evaluating long term screening outcomes in colorectal cancer** 

Recently there have been heated arguments in the topic of over diagnosis, the diagnosis of ``disease" that will not cause symptoms or death during a patient's lifetime (Lichtenfeld, J L, 2010). Some profound questions should be asked with regards to over diagnosis. How do we evaluate the long-term outcomes due to continuous regular screening? Will regular screening exams contribute to a greater chance of over diagnosis? What are the percentages of over diagnosis and true-early-detection among the screen-detected cancer patients? How should the probability of no-early-detection and the probability of disease-free-life be

Some research has been done in the area of over diagnosis. However, the majority of research in this area has been based on observational studies, and mainly in breast cancer

= 12 months) when he is 50 years old and continues until he reaches

**3.2 Results** 

annual screening (i.e.

for females.

estimated?

annual screenings and 2.39% for screening every 6 months.

For an initially asymptomatic individual taking *K* screenings at their ages 01 1 *<sup>K</sup> tt t* , the conditional probability for each of the four cases was derived, given that their lifetime <sup>1</sup> ( ) *Tt t K K* and presented in equations 11-15.

$$\begin{aligned} P(\text{Case } \ 1: \text{Sym} pF \ | \ T = t\_K) &= 1 - \int\_0^{t\_K} a(\mathbf{x}) d\mathbf{x} + \int\_{t\_{K-1}}^{t\_K} a(\mathbf{x}) Q(t\_K - \mathbf{x}) d\mathbf{x} \\ &+ \sum\_{j=0}^{K-1} (1 - \beta\_j) ... (1 - \beta\_{K-1}) \int\_{t\_{j-1}}^{t\_j} w(\mathbf{x}) Q(t\_k - \mathbf{x}) d\mathbf{x} \end{aligned} \tag{11}$$

$$P\{\text{Case}\quad \text{2}:\text{NoED}\mid T=t\_k\} = I\_{K,1} + I\_{K,2} + \cdots + I\_{K,K} \,. \tag{12}$$

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 483

. Where the event A = {an individual is asymptomatic of

0 0 0 0 0 0 ( | ) 1 () ()( ) . *t t*

probability density *w t*( ) , the sojourn time distribution *q x*( ) , a person's age at the first

the age-dependent transition probability, and the sojourn time distribution *q x*( ) , were

Given the MCCCS data, the posterior predictive probability of each case can be estimated

*P Case i T t MCCCS P Case i T t*

estimated from the MCCCS data (Wu et al., 2009a) and were given in Section 2.2.

The probability for each of the four cases is a function of the sensitivity

*P A T t w x dx w x Q t x dx* (18)

0 0 <sup>1</sup> ( | , ) ( | , ). *<sup>j</sup>*

*j*

*<sup>j</sup>* is the MCMC random sample drawn from the posterior distribution and *n* 1000

*n* 

Furthermore, we defined a diagnosed case as when either an interval clinical incident case or a screen-detected case happens in a study. Researchers may be interested in the proportion of "no-early-detection", "true-early-detection" and "over diagnosis" given that it is a diagnosed case. For example, among females, what are the estimated probabilities of "no-early-detection", given that a woman has been diagnosed with colorectal cancer, either through scheduled screening exam or not. Last but not least, researchers are most interested in the probabilities of "true-early-detection" and "over diagnosis" given that it is a screendetected case. All of these conditional probabilities were also estimated using equations 12-

distribution. A total of 1000 posterior samples were put into equation 19 to conduct the Bayesian inference. This Bayesian inference assumed that there is a program consisting of periodic screening exams from three hypothetical cohorts of asymptomatic individuals. Those cohorts have initial ages of 40, 50 and 60 at the first screening exam for males or females. For each group, we examined various screening frequencies, with screening

table from the Social Security Administration, which was published online for year 2007 (Social Security Administration, 2011). The actuarial life table is based on mortality and

Based on that life table, we derived the conditional lifetime distribution 0 (| ) *Tf tT t* (Wu and Rosner 2010) and estimated the probabilities of each of the four cases, i.e. <sup>0</sup> *P Case i A T t MCCCS* ( |, , ) , using the estimations of sensitivity, sojourn time distribution,

provides the probability of death within one year from age 0 to age 119 years old.

= 12, 18, and 24 months. For the lifetime distribution, we used the actuarial life

. The age-dependent sensitivity

\*

*<sup>j</sup>* as a MCMC random sample drawn from the posterior

( )*t* , the transition

(19)

( )*t* ,

4

1

*i*

as:

Where \* 

**4.2 Results** 

interval

is the posterior sample size.

In section 2.2 we estimated \*

0

screening 0*t* and his/her future screening interval

19 using the definition of conditional probability.

colorectal cancer before age <sup>0</sup>*t* } and

( |, ) 1

*P Case i A T t*

Where

$$\begin{split} I\_{K,j} &= \sum\_{i=0}^{j-1} (1 - \beta\_i) ... (1 - \beta\_{j-1}) \int\_{t\_{i-1}}^{t\_i} o(\mathbf{x}) [Q(t\_{j-1} - \mathbf{x}) - Q(t\_j - \mathbf{x})] d\mathbf{x} \\ &+ \int\_{t\_{j-1}}^{t\_j} o(\mathbf{x}) [1 - Q(t\_j - \mathbf{x})] d\mathbf{x}, \text{ for all } j = 1, 2 \dots, K. \end{split} \tag{13}$$

*K j*, *I* is the probability of being an interval case in the interval ( <sup>1</sup> , *j j t t* ) in a sequence of K screening exams.

$$\begin{aligned} P(\text{Case} \quad \mathfrak{Z} : \text{True} \boldsymbol{E} \boldsymbol{D} \mid \boldsymbol{T} = \boldsymbol{t}\_{K}) &= \sum\_{j=1}^{K-1} \beta\_{j} \big( \sum\_{i=0}^{j-1} (1 - \beta\_{i}) \dots (1 - \beta\_{j-1}) \big) ^{t\_{j}} \boldsymbol{\alpha}(\mathbf{x}) [Q(t\_{j} - \mathbf{x}) - Q(t\_{K} - \mathbf{x})] \mathbf{dx} \\ &+ \int\_{t\_{j-1}}^{t\_{j}} \boldsymbol{\alpha}(\mathbf{x}) [Q(t\_{j} - \mathbf{x}) - Q(t\_{K} - \mathbf{x})] d\mathbf{x}] + \beta\_{0} \int\_{0}^{t\_{0}} \boldsymbol{\alpha}(\mathbf{x}) [Q(t\_{0} - \mathbf{x}) - Q(t\_{K} - \mathbf{x})] d\mathbf{x} \end{aligned} \tag{14}$$

$$\begin{aligned} P(\text{Case} \quad 4: \text{OverD} \mid T = t\_K) &= \sum\_{j=1}^{K-1} \beta\_j \{ \sum\_{i=0}^{j-1} (1 - \beta\_i)...(1 - \beta\_{j-1}) \} \frac{t\_i}{t\_{i-1}} \alpha(\mathbf{x}) \mathbf{Q}(t\_K - \mathbf{x}) d\mathbf{x} \\ &+ \int\_{t\_{j-1}}^{t\_j} \alpha(\mathbf{x}) \mathbf{Q}(t\_K - \mathbf{x}) d\mathbf{x} + \beta\_0 \int\_0^{t\_0} \alpha(\mathbf{x}) \mathbf{Q}(t\_K - \mathbf{x}) d\mathbf{x} \} \end{aligned} \tag{15}$$

For an individual currently at age <sup>0</sup>*t* , his/her lifetime is random, and it would not be practical to fix the number of screening exams to any fixed number *K*. If, however, he/she follows a pre-planned screening schedule, or, more simply, if he/she plans to be screened every 12, 18, or 24 months, then the probability of each outcome when his/her lifetime *T* is longer than 0*t* can be obtained using equation 16.

$$P\{\text{Case}\quad \text{i} \mid T \ge t\_0\} = \bigcap\_{t\_0}^{\text{v}} P\{\text{Case}\quad \text{i} \mid \mathbb{K} = \text{K}(\text{t}) , T = t\} f\_T\{\text{t} \mid T \ge t\_0\} dt\text{ , for } \text{i} = 1, 2, 3, 4. \tag{16}$$

Where the lifetime probability density function 0 (| ) *Tf tT t* is defined in equation 17.

$$f\_T(t \mid T \ge t\_0) = \frac{f\_T(t)}{P(T > t\_0)} = \frac{f\_T(t)}{1 - F\_T(t\_0)}, \text{ if } t \ge t\_0 \text{ .} \tag{17}$$

The probability *P Case i K K t T t* ( | ( ), ) was derived in equations 11-15, and the number of screening exams *K Kt t t* ( ) ( )/ <sup>0</sup> *,* is the largest integer that is less than or equal to <sup>0</sup> ( )/ *t t* . Hence, *K* is a random variable as well, taking integer values and changing with one's lifetime*T* . It was proved that for any screening number *K* ,

colorectal cancer before age <sup>0</sup>*t* } and

482 Colorectal Cancer – From Prevention to Patient Care

1

*I x Q t x Q t x dx*

*i*

*K j i j j j*

*i*

(1 )...(1 ) ( )[ ( ) ( )]

(13)

(14)

(15)

( )[1 ( )] , for all 1, 2 , .

*K j*, *I* is the probability of being an interval case in the interval ( <sup>1</sup> , *j j t t* ) in a sequence of K

( 3 : | ) { (1 )...(1 ) ( )[ ( ) ( )]

 

*P Case TrueED T t x Q t x Q t x dx*

*j i t*

*t t*

1 1

( 4 : | ) { (1 )...(1 ) ( ) ( )

*P Case OverD T t x Q t x dx*

1 0

*t t*

For an individual currently at age <sup>0</sup>*t* , his/her lifetime is random, and it would not be practical to fix the number of screening exams to any fixed number *K*. If, however, he/she follows a pre-planned screening schedule, or, more simply, if he/she plans to be screened every 12, 18, or 24 months, then the probability of each outcome when his/her lifetime *T* is

*t K j*

1 0

*i*

*i*

0 0 0

1

 

1

, *if* <sup>0</sup> *t t* . (17)

*,* is the largest integer that is less than or equal to

*i*

*i*

1

*Kj ij j K*

 

*t K j Kj ij K j i t*

 

( )[ ( ) ( )] } ( )[ ( ) ( )]

0

, for i = 1,2,3,4. (16)

0 0

. Hence, *K* is a random variable as well, taking integer values and changing

*T*

*T T*

The probability *P Case i K K t T t* ( | ( ), ) was derived in equations 11-15, and the number

with one's lifetime*T* . It was proved that for any screening number *K* ,

0 0

*K K*

 

()( ) ()( ) }

*x Q t x dx x Q t x dx*

*j K K*

 

*x Q t x Q t x dx x Q t x Q t x dx*

*x Q t x dx j K*

, 1 1

 

*j*

1 1

1 0

1

0 0 ( | ) ( | ( ), ) ( | ) *<sup>T</sup>*

0

*P Case i T t P Case i K K t T t f t T t dt*

Where the lifetime probability density function 0 (| ) *Tf tT t* is defined in equation 17.

() () (| ) ( ) 1 ()

*<sup>f</sup> <sup>t</sup> <sup>f</sup> <sup>t</sup> f tT t PT t F t* 

*j*

*t*

*j*

*t j*

*i t*

1

1

*j*

longer than 0*t* can be obtained using equation 16.

0

*t*

*T*

of screening exams *K Kt t t* ( ) ( )/ <sup>0</sup>

<sup>0</sup> ( )/ *t t*

*t*

*j*

*j*

*t*

1

0

*j*

*t*

Where

screening exams.

$$P(A \mid T \ge t\_0) = 1 - \int\_0^{t\_0} w(\mathbf{x})d\mathbf{x} + \int\_0^{t\_0} w(\mathbf{x})Q(t\_0 - \mathbf{x})d\mathbf{x}.\tag{18}$$

The probability for each of the four cases is a function of the sensitivity ( )*t* , the transition probability density *w t*( ) , the sojourn time distribution *q x*( ) , a person's age at the first screening 0*t* and his/her future screening interval . The age-dependent sensitivity ( )*t* , the age-dependent transition probability, and the sojourn time distribution *q x*( ) , were estimated from the MCCCS data (Wu et al., 2009a) and were given in Section 2.2.

Given the MCCCS data, the posterior predictive probability of each case can be estimated as:

$$P(\text{Case} \quad \text{i} \mid T > t\_0, \text{MCCCS}) \approx \frac{1}{n} \sum\_{j} P(\text{Case} \quad \text{i} \mid T > t\_0, \theta\_j^{\*}). \tag{19}$$

Where \* *<sup>j</sup>* is the MCMC random sample drawn from the posterior distribution and *n* 1000 is the posterior sample size.

Furthermore, we defined a diagnosed case as when either an interval clinical incident case or a screen-detected case happens in a study. Researchers may be interested in the proportion of "no-early-detection", "true-early-detection" and "over diagnosis" given that it is a diagnosed case. For example, among females, what are the estimated probabilities of "no-early-detection", given that a woman has been diagnosed with colorectal cancer, either through scheduled screening exam or not. Last but not least, researchers are most interested in the probabilities of "true-early-detection" and "over diagnosis" given that it is a screendetected case. All of these conditional probabilities were also estimated using equations 12- 19 using the definition of conditional probability.

#### **4.2 Results**

In section 2.2 we estimated \* *<sup>j</sup>* as a MCMC random sample drawn from the posterior distribution. A total of 1000 posterior samples were put into equation 19 to conduct the Bayesian inference. This Bayesian inference assumed that there is a program consisting of periodic screening exams from three hypothetical cohorts of asymptomatic individuals. Those cohorts have initial ages of 40, 50 and 60 at the first screening exam for males or females. For each group, we examined various screening frequencies, with screening interval = 12, 18, and 24 months. For the lifetime distribution, we used the actuarial life table from the Social Security Administration, which was published online for year 2007 (Social Security Administration, 2011). The actuarial life table is based on mortality and provides the probability of death within one year from age 0 to age 119 years old.

Based on that life table, we derived the conditional lifetime distribution 0 (| ) *Tf tT t* (Wu and Rosner 2010) and estimated the probabilities of each of the four cases, i.e. <sup>0</sup> *P Case i A T t MCCCS* ( |, , ) , using the estimations of sensitivity, sojourn time distribution,

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 485

0.0

12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

Case 4: Over-diagnosis

12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

Case 2: No-early-detection

 0.010

 0.020

 0.030

Fig. 1. The box plot of the estimated percentage for each outcomes, with 0*t* =60.

**5. Discussion** 

0.94

0.01

 0.03

 0.05 12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

Case 3: True-early-detection

12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

Case 1: Symptom-free-life

 0.96

 0.98

estimates of these parameters.

Figure 2 shows the probabilities of "true-early-detection" and the probability of "over diagnosis" among those whose cancer would be diagnosed by regular screening exam for the initial-age-60 group of both genders. In Figure 2, the screening time interval for males and females are presented for 12, 18 and 24 months. The estimated mean percentage for "true-early-detection" and "over diagnosis" given that it is a screen-detected case was similar for males and females. However, the 95% C.I. for males were much larger than that for females; this indicates that there is more uncertainty of these probabilities for males.

0.0 0.005

 0.015

We presented some results in probability modelling and statistical inference in colorectal cancer screening, using FOBT as an example. As we have shown in section 2, the three key parameters are the sensitivity of the screening modality, the transition probability of the disease, and the sojourn time distribution. All other parameters of interests can be expressed as a function of these three key parameters, hence accurate estimation of them is very important. These three key parameters are the building blocks in the cancer screening model, many researchers are striving to improve the modelling and get more accurate

and transition probability obtained from the MCCCS data for males and females. Overall, the proportion of over diagnosis was very small, less than 0.3% for any age and gender.

The probability of "true-early-detection" for males was between 1.91% (at 60 years old, with 24 months as screening interval) and 3.28% for 40 years old, with 12 months as screening interval. Correspondingly, the probability of "true-early-detection" for females was between 2.75% for 60 years old, with screening interval of 24 months and 3.76% for 40 years old, with screening interval of 12 months. Regardless of the age, the probability of "true-early-detection" slightly decreased as the screening time interval increased and overall, the probability of "true-early-detection" was consistently lower for males than for females.

The probability of "no-early-detection" for males was between 0.53% for 60 years old, with screening interval of 12 months to 1.95% for 40 years old, with screening interval of 24 months. The probability of "no-early-detection" for females was between 0.29% for 60 years old, with screening interval of 12 months to 1.34% for 40 years old, with screening interval of 12 months. In general, the probability of "no-early-detection" slightly increased as the screening interval increased, and slightly decreased as the age at initial screening was older. The probability of "symptom-free-life" was very large (e.g. over 95%). Regardless of age or gender, the probability of "symptom-free-life" was almost constant for any number of months between two screenings. For example, among 50 years old males, the probability of "symptom-free-life" was 95.84% if 12 months was the screening time interval; 95.87% if 18 months was the screening time interval; and 95.90% if 24 months was the screening time interval. The sum of the probability of all four cases should add up to 1, and it was observed in the simulation, the total probability is above 0.998, due to simulation accuracy, this is clinically insignificant.

The box plot of the probability for each case when 0*t* =60 is given in Figure 1. Within each box plot, the three left-hand-side boxes represent females and the three right-hand-side boxes represent males, and these probabilities are presented at different screening intervals. We decided to present the box plots when the initial screening age was 60 but similar box plots were observed for 0*t* = 40 and 50. Again, we see in Figure 1 that the probability of "symptom-free-life" and the probability of "over diagnosis" are pretty stable over the screening time intervals, regardless of gender. The probability of "noearly-detection" increased monotonically with the screening time interval, while the probability of "true-early-detection" decreases monotonically with the length of the screening time interval.

The estimated conditional probabilities of "no-early-detection", "true-early-detection" and "over diagnosis", given that it is a diagnosed case, for females and males were estimated. Among the initial age of 40 years-old women group, the percentage of over diagnosis given that it was a diagnosed cancer case was 5.04%, if she was screened every 24 months apart; and 6.50%, if she was screened every 12 months apart. Similarly, the estimated conditional probabilities of "true-early-detection" and "over diagnosis" given that it is a screen-detected case, for females and males were also estimated. Among the 40 years-old women initial age group, the percentage of over diagnosis among the screen-detected cases was 6.75% (95% HPD: 2.56%-19.27%), if screened every 24 months apart, and 7.12% (95% HPD: 2.76%- 19.91%), if screened every 12 months apart.

12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

12 mo. 18 mo. 24 mo. m-12mo.m-18mo.m-24mo.

0.0

Figure 2 shows the probabilities of "true-early-detection" and the probability of "over diagnosis" among those whose cancer would be diagnosed by regular screening exam for the initial-age-60 group of both genders. In Figure 2, the screening time interval for males and females are presented for 12, 18 and 24 months. The estimated mean percentage for "true-early-detection" and "over diagnosis" given that it is a screen-detected case was similar for males and females. However, the 95% C.I. for males were much larger than that for females; this indicates that there is more uncertainty of these probabilities for males.

Fig. 1. The box plot of the estimated percentage for each outcomes, with 0*t* =60.

#### **5. Discussion**

484 Colorectal Cancer – From Prevention to Patient Care

and transition probability obtained from the MCCCS data for males and females. Overall, the proportion of over diagnosis was very small, less than 0.3% for any age and gender. The probability of "true-early-detection" for males was between 1.91% (at 60 years old, with 24 months as screening interval) and 3.28% for 40 years old, with 12 months as screening interval. Correspondingly, the probability of "true-early-detection" for females was between 2.75% for 60 years old, with screening interval of 24 months and 3.76% for 40 years old, with screening interval of 12 months. Regardless of the age, the probability of "true-early-detection" slightly decreased as the screening time interval increased and overall, the probability of "true-early-detection" was consistently lower for males than for

The probability of "no-early-detection" for males was between 0.53% for 60 years old, with screening interval of 12 months to 1.95% for 40 years old, with screening interval of 24 months. The probability of "no-early-detection" for females was between 0.29% for 60 years old, with screening interval of 12 months to 1.34% for 40 years old, with screening interval of 12 months. In general, the probability of "no-early-detection" slightly increased as the screening interval increased, and slightly decreased as the age at initial screening was older. The probability of "symptom-free-life" was very large (e.g. over 95%). Regardless of age or gender, the probability of "symptom-free-life" was almost constant for any number of months between two screenings. For example, among 50 years old males, the probability of "symptom-free-life" was 95.84% if 12 months was the screening time interval; 95.87% if 18 months was the screening time interval; and 95.90% if 24 months was the screening time interval. The sum of the probability of all four cases should add up to 1, and it was observed in the simulation, the total probability is above 0.998, due to simulation accuracy, this is

The box plot of the probability for each case when 0*t* =60 is given in Figure 1. Within each box plot, the three left-hand-side boxes represent females and the three right-hand-side boxes represent males, and these probabilities are presented at different screening intervals. We decided to present the box plots when the initial screening age was 60 but similar box plots were observed for 0*t* = 40 and 50. Again, we see in Figure 1 that the probability of "symptom-free-life" and the probability of "over diagnosis" are pretty stable over the screening time intervals, regardless of gender. The probability of "noearly-detection" increased monotonically with the screening time interval, while the probability of "true-early-detection" decreases monotonically with the length of the

The estimated conditional probabilities of "no-early-detection", "true-early-detection" and "over diagnosis", given that it is a diagnosed case, for females and males were estimated. Among the initial age of 40 years-old women group, the percentage of over diagnosis given that it was a diagnosed cancer case was 5.04%, if she was screened every 24 months apart; and 6.50%, if she was screened every 12 months apart. Similarly, the estimated conditional probabilities of "true-early-detection" and "over diagnosis" given that it is a screen-detected case, for females and males were also estimated. Among the 40 years-old women initial age group, the percentage of over diagnosis among the screen-detected cases was 6.75% (95% HPD: 2.56%-19.27%), if screened every 24 months apart, and 7.12% (95% HPD: 2.76%-

females.

clinically insignificant.

screening time interval.

19.91%), if screened every 12 months apart.

We presented some results in probability modelling and statistical inference in colorectal cancer screening, using FOBT as an example. As we have shown in section 2, the three key parameters are the sensitivity of the screening modality, the transition probability of the disease, and the sojourn time distribution. All other parameters of interests can be expressed as a function of these three key parameters, hence accurate estimation of them is very important. These three key parameters are the building blocks in the cancer screening model, many researchers are striving to improve the modelling and get more accurate estimates of these parameters.

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 487

al., 1997). Our estimates are more accurate for different age groups as reported in section 2. There are other data sets that were used to estimate the FOBT screening sensitivity and mean sojourn time. For example, French data reported by Launoy et al. (1997) estimated the FOBT mass-screening sensitivity to be about 50% (Launoy et al., 1997). Their estimated mean sojourn time was longer than our results, between 4.5 and 5 years for all combined cancer cases. Also, these researches showed that the estimation of the sensitivity and the sojourn time maybe negatively correlated (Launoy et al., 1997). Better modelling strategies are needed to handle this situation. We plan to explore solutions accounting for the negative

There is little research in the topic of lead time bias or the lead time distribution except in Wu et al. (2009) (Wu et al., 2009a). Since there is convincing evidence that FOBT and/or other colorectal screening modalities can significantly reduce mortality (Mandel et al., 1999), the U.S. Preventive Services Task Force has recommended screening people between 50-75 years-old since 2008 (U.S.Preventive Services Task Force, 2008). Unfortunately, the compliance to colorectal cancer screening is low in the U.S. and the world (Sarfaty & Wender, 2007). We hope the lead time results from our simulations and models (Section 3) can provide some helpful information to general audiences about the benefit of taking

There is almost no research in the topic of over diagnosis or long-term outcomes in colorectal cancer, to our knowledge. As the first of the baby boomer generation turns age 65 this year, evaluating the long-term outcomes will provide useful information and great insights to policy makers. We hope our method will provide a frame work and a systematic approach for evaluation purposes. To explore this topic more, we will need to obtain more recent screening data. We are exploring if data from the Prostate, Lung Colorectal and Ovarian (PLCO) cancer screening trial can be released to us (National Cancer Institute

Our future research topic includes three areas: (1) exploring the relationship between sensitivity and the sojourn time distribution, and building up a better modelling strategy for these three parameters; (2) exploring the optimal screening interval based on an individual's screening history; and (3) exploring the survival benefit from screening after removing the lead time bias, hence we can have a better understanding of what we gained from screening.

The authors thank Laura Littler for her research assistant help and in proof-reading this manuscript. We thank the anonymous reviewers for their valuable suggestions and comments for improving this paper. The second author was supported by a research supplement (3R37CA057030-20S1) from the National Cancer Institute and by a grant from

American Cancer Society. (2011). Colorectal Cancer Facts & Figures 2011-2013,In: American Cancer Society, Date Accessed: 6/15/2011, Available from:

We hope the research will benefit the health of the general population.

the Michael and Susan Dell Foundation during the writing of this chapter.

correlation between the sensitivity and the sojourn time to solve this problem.

screening exams.

Division of Cancer Prevention, 2011).

**6. Acknowledgment** 

**7. References** 

Fig. 2. Estimated percentage of true-early-detection and over diagnosis with its 95%HPD for <sup>0</sup>*t* =60.

Although, other researchers have also estimated the sensitivity and the mean sojourn time in fecal Hemoccult testing, using data from Calvados, France between 1991 and 1994 (Prevost et al., 1998), their models are different from the progressive model that we used here. Prevost et al. (1998) modelled the incidence of cancer as a Poisson random variable, with different parameter value for the mean of the Poisson distribution (Prevost et al., 1998). Another difference is that their sojourn time was assumed to follow an exponential distribution. They reported that the mean sojourn time increases with age, which is approximately two years among 45-54 years-old, 3 years among 55-64 years-old, and 6 years among 65-74 years-old (Prevost et al., 1998). Their estimation of sensitivity decreases with age, which is approximately 75% among 45-54 years-old, 50% among 55-64 years-old, and 40% among 65-74 years-old (Prevost et al., 1998).

Church et al (1997) used the same Minnesota study (MCCCS) to estimate the sensitivity. However, their estimate of program sensitivity is about 90%, regardless of age (Church et al., 1997). Our estimates are more accurate for different age groups as reported in section 2. There are other data sets that were used to estimate the FOBT screening sensitivity and mean sojourn time. For example, French data reported by Launoy et al. (1997) estimated the FOBT mass-screening sensitivity to be about 50% (Launoy et al., 1997). Their estimated mean sojourn time was longer than our results, between 4.5 and 5 years for all combined cancer cases. Also, these researches showed that the estimation of the sensitivity and the sojourn time maybe negatively correlated (Launoy et al., 1997). Better modelling strategies are needed to handle this situation. We plan to explore solutions accounting for the negative correlation between the sensitivity and the sojourn time to solve this problem.

There is little research in the topic of lead time bias or the lead time distribution except in Wu et al. (2009) (Wu et al., 2009a). Since there is convincing evidence that FOBT and/or other colorectal screening modalities can significantly reduce mortality (Mandel et al., 1999), the U.S. Preventive Services Task Force has recommended screening people between 50-75 years-old since 2008 (U.S.Preventive Services Task Force, 2008). Unfortunately, the compliance to colorectal cancer screening is low in the U.S. and the world (Sarfaty & Wender, 2007). We hope the lead time results from our simulations and models (Section 3) can provide some helpful information to general audiences about the benefit of taking screening exams.

There is almost no research in the topic of over diagnosis or long-term outcomes in colorectal cancer, to our knowledge. As the first of the baby boomer generation turns age 65 this year, evaluating the long-term outcomes will provide useful information and great insights to policy makers. We hope our method will provide a frame work and a systematic approach for evaluation purposes. To explore this topic more, we will need to obtain more recent screening data. We are exploring if data from the Prostate, Lung Colorectal and Ovarian (PLCO) cancer screening trial can be released to us (National Cancer Institute Division of Cancer Prevention, 2011).

Our future research topic includes three areas: (1) exploring the relationship between sensitivity and the sojourn time distribution, and building up a better modelling strategy for these three parameters; (2) exploring the optimal screening interval based on an individual's screening history; and (3) exploring the survival benefit from screening after removing the lead time bias, hence we can have a better understanding of what we gained from screening. We hope the research will benefit the health of the general population.

#### **6. Acknowledgment**

486 Colorectal Cancer – From Prevention to Patient Care

Fig. 2. Estimated percentage of true-early-detection and over diagnosis with its 95%HPD for

Although, other researchers have also estimated the sensitivity and the mean sojourn time in fecal Hemoccult testing, using data from Calvados, France between 1991 and 1994 (Prevost et al., 1998), their models are different from the progressive model that we used here. Prevost et al. (1998) modelled the incidence of cancer as a Poisson random variable, with different parameter value for the mean of the Poisson distribution (Prevost et al., 1998). Another difference is that their sojourn time was assumed to follow an exponential distribution. They reported that the mean sojourn time increases with age, which is approximately two years among 45-54 years-old, 3 years among 55-64 years-old, and 6 years among 65-74 years-old (Prevost et al., 1998). Their estimation of sensitivity decreases with age, which is approximately 75% among 45-54 years-old, 50% among 55-64 years-old, and

Church et al (1997) used the same Minnesota study (MCCCS) to estimate the sensitivity. However, their estimate of program sensitivity is about 90%, regardless of age (Church et

<sup>0</sup>*t* =60.

40% among 65-74 years-old (Prevost et al., 1998).

The authors thank Laura Littler for her research assistant help and in proof-reading this manuscript. We thank the anonymous reviewers for their valuable suggestions and comments for improving this paper. The second author was supported by a research supplement (3R37CA057030-20S1) from the National Cancer Institute and by a grant from the Michael and Susan Dell Foundation during the writing of this chapter.

#### **7. References**

American Cancer Society. (2011). Colorectal Cancer Facts & Figures 2011-2013,In: American Cancer Society, Date Accessed: 6/15/2011, Available from:

Modelling and Inference in Screening: Exemplification with the Faecal Occult Blood Test 489

Sarfaty, M. & Wender, R. (2007). How to Increase Colorectal Cancer Screening Rates in

Screen for Life: National Colorectal Cancer Action Campaign. (2009). Colorectal Cancer

SEER Cancer Statistics Review 1975-2007. (2010). Lifetime Risk Tables. *National Cancer* 

Silicon Graphics, I. (2003). C Language Reference Manual, In: Silicon Graphics, Inc, Date

Social Security Administration. (4/5/2011). Period Life Table., In: Actuarial Publications,

Stroustrup, B. (2011). C++ Programming Language, In: The C++ Resources Network, Date

U.S.Preventive Services Task Force. (10/1/2008). Screening for colorectal cancer: U.S.

World Health Organization. (3/4/2003). Global cancer rates could increase by 50% to 15

World Health Organization. (2011). Cancer; World Cancer Day 2011, In: World Health

Wu, D. & Rosner, G.L. (2010). A projection of true-early-detection, no-early-detection,

Wu, D., Erwin, D., & Rosner, G.L. (2009a). A projection of benefits due to fecal occult blood

Wu, D., Erwin, D., & Rosner, G.L. (2009b). Estimating key parameters in FOBT screening for

Wu, D., Rosner, G.L., & Broemeling, L. (2005). MLE and Bayesian inference of age-

Vol. 61, No. 4, (December 2005), pp. (1056-1063), ISBN: 0006-341X.

Publication 08-051124-EF-3, Date Accessed: 6/23/2011, Available from: http://www.uspreventiveservicestaskforce.org/uspstf08/colocancer/colors.htm. Welch, H.G. & Black, W.C. (2010). Overdiagnosis in cancer. [Review] [43 refs]. *Journal of the* 

Preventive Services Task Force Recommendation Statement., In: AHRQ

million by 2020, In: World Health Organization, Date Accessed: 6/15/2011,

overdiagnosis and not-so-necessary probabilities in tumor screening. *In JSM Proceedings, Biopharmaceutical Section, VA: American Statistical Associtation*, pp. (1144-

test for colorectal cancer. *Cancer Epidemiology*, Vol. 33, No. 3-4, (October 2009a), pp.

colorectal cancer. *Cancer Causes & Control*, Vol. 20, No. 1, (February 2009b), pp. (41-

dependent sensitivity and transition probability in periodic screening. *Biometrics*,

Accessed: 7/1/2011, Available from: http://www.cplusplus.com/.

*National Cancer Institute*, Vol. 102, No. 9, (May 2010), pp. (605-613).

http://www.who.int/mediacentre/news/releases/2003/pr27/en/.

Organization, Date Accessed: 6/15/2011, Available from:

Accessed: 7/1/2011, ISBN: ISBN 007-0701-150, Available from: http://techpubs.sgi.com/library/manuals/0000/007-0701-150/pdf/007-0701-

(354-366), ISBN: 1542-4863.

*Institute*, pp. (1-78).

150.pdf.

Available from:

1157).

46).

(212-215).

http://www.who.int/cancer/en/.

ISBN: CDC Publication #99-6949.

Date Accessed: 6/23/2011, Available from: http://www.ssa.gov/oact/STATS/table4c6.html.

Practice. *CA: A Cancer Journal for Clinicians*, Vol. 57, No. 6, (September 2008), pp.

Screening, basic facts sheet. *Centers for Disease Contral and Prevention*, (June 2009),

 http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/docume nts/document/acspc-028323.pdf.


http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/docume

ARUP Laboratories. (2010). Septin 9 (SEPT9) methylated DNA detection by real-time PCR:

Colon Cancer Prevention Project. (2011). 6th Annual Walk Away from Colon Cacer & 5K

Day, N.E. (2005). Overdiagnosis and breast cancer screening. *Breast Cancer Research*, Vol. 7,

Duffy, S.W., Lynge, E., Jonsson, H., Ayyaz, S., & Olsen, A.H. (2008). Complexities in the

Kronborg, O., Fenger, C., Olsen, J., Jorgensen, O.D., & Sondergaard, O. (1996). Randomised

Launoy, G., Smith, T.C., Duffy, S.W., & Bouvier, V. (1997). Colorectal cancer mass-screening:

Lichtenfeld, J L. (4/22/2010). Crux of the dilemma: Whose cancer is overdiagnosed?, In: Dr Len's Cancer Blog, Date Accessed: 6/23/2011, Available from: http://www.cancer.org/AboutUs/DrLensBlog/post/2010/04/22/Crux-Of-The-

Mandel, J.S., Bond, J.H., Church, T.R., Snover, D.C., Bradley, G.M., Schuman, L.M., &

Mandel, J.S., Church, T.R., Ederer, F., & Bond, J.H. (1999). Colorectal cancer mortality:

National Cancer Institute. (4/15/2011). SEER Stat Fact Sheets: Colon and Rectum, In:

National Cancer Institute Division of Cancer Prevention. (1/6/2011). Prostate, Lung

Health, Date Accessed: 6/23/2011, Available from: www.cancer.gov/plco. Prevost, T.C., Launoy, G., Duffy, S.W., & Chen, H.H. (1998). Estimating sensitivity and

cancer screening. *National Reference Laboratory*, (August 2010), pp. (1-2). Church, T.R., Ederer, F., & Mandel, J.S. (1997). Fecal occult blood screening in the minnesota

Detection of circulating methlyated septin 9 DNA from blood plasma for colorectal

study: Sensitivity of the screening test. *Journal of National Cancer Institute*, Vol. 89,

Run, In: Colon Cancer Prevention Project, Date Accessed: 6/15/2011, Available

estimation of overdiagnosis in breast cancer screening. *British Journal of Cancer*, Vol.

study of screening for colorectal cancer with faecal-occult-blood test. *Lancet*, Vol.

Estimation of faecal occult blood test sensitivity, taking into account cancer mean sojourn time. *International Journal of Cancer*, Vol. 73, No. 2, (October 1997), pp. (220-

Ederer, F. (1993). Reducing mortality from colorectal cancer by screening for fecal occult blood. *New England Journal of Medicine*, Vol. 329, No. 19, (May 1993), pp.

Effectiveness of biennial screening for fecal occult blood. *Journal of National Cancer* 

Surveillance Epidemiology and End Resuls, Date Accessed: 6/15/2011, Available

Colorectal and Ovarian (PLCO) cancer screening trial, In: U.S.National Institutes of

sojourn time in screening for colorectal cancer: a comparison of statistical approaches. *American Journal of Epidemiology*, Vol. 148, No. 6, (September 1998), pp.

nts/document/acspc-028323.pdf.

No. 19, (October 1997), pp. (1440-1448).

No. 5, (August 2005), pp. (228-229).

224), ISBN: 1097-0215.

(609-619), ISBN: 0002-9262.

(1365-1371).

from: http://www.coloncancerpreventionproject.org/.

DilemmaWhose-Cancer-Is-Overdiagnosed.aspx.

*Institute*, Vol. 91, No. 5, (March 1999), pp. (434-437).

from: http://seer.cancer.gov/statfacts/html/colorect.html.

99, No. 7, (September 2008), pp. (1176-1178), ISBN: 0007-0920.

348, No. 9040, (November 1996), pp. (1467-1471), ISBN: 0140-6736.


http://www.who.int/mediacentre/news/releases/2003/pr27/en/.


**25** 

*1,3China 2Japan*

**Dietary Risks: Folate, Alcohol** 

Zi-Yuan Zhou1, Keitaro Matsuo2,Wen-Chang Wang3,

*Aichi Cancer Center Research Institute, Chikusa-ku, Nagoya, 3Department of Statistics, College of Preventive Medicine, The Third Military Medical University, ChongQing,* 

*(Key Laboratory of The Ministry of Education of China), College of Preventive Medicine,* 

Folate, as a member of the water-soluble B-group vitamins, is found widely in foodstuffs. Folate cannot be synthesized by human therefore dietary intake is the only source for human to obtain folate. The pteropolyglumates, usually with 1~6 glutamic acid molecules, are the major forms of natural food folates (Lucock 2000) and the 5-methyl tetrahydrofolate (5-metTHF), derived from hydrolization of absorbed folate and folic acid (the synthetic form of folate) as well, is the primary form in circulation (Ulrich 2005; Pietrzik, Bailey et al. 2010). Through the transmembrane transportation, the 5-metTHF in cell can be reduced by dihydrafolate reductase to tetrahydrofolate (THF) that is directly involved in metabolic process (Lucock 2000; Pietrzik, Bailey et al. 2010), and then performs biological functions in several ways. THF can be metabolized to 5,10-methylene-THF and further be irreversibly reduced into 5-metTHF which is the key step in one-carbon unit metabolism that is catalyzed by the enzyme methylenetetrahydrofolate reductase (MTHFR). By using the methyl donated by 5-methyltetrahydrofolate, the enzyme methionine synthase (MS) converts homocystine to methionine and then the *de novo* synthesized methionine can be catalyzed by the methionine adenosyl transferase to yield S-adenosylmethionine which directly provides methyl for a variety of important *in vivo* methylation reactions (Lucock 2000; Sanderson, Stone et al. 2007)*.* By using 5,10-methylene-THF as methyl donor, the enzyme thymidylate synthase converts deoxyuridylate (dUMP) to deoxythymidylate (dTMP), meanwhile the 5,10-formyltetrahydrofolate from 5,10-methylene-THF is involved in the production of both adenosine and guanosine, all are physiological building blocks of DNA replication (Bollheimer, Buettner et al. 2005; Duthie 2011). Thus, the most prominent function of folate is to transfer and process the one-carbon unit which is needed for methylation reactions and synthesis of thymine and purines. Consequently, folate deficiency

**1. Introduction** 

**and Gene Polymorphisms**

Huan Yang1, Kazuo Tajima2 and Jia Cao1

*The Third Military Medical University, ChongQing,* 

*1Department of Hygiene Toxicology ,* 

*2Division of Epidemiology and Prevention,* 


### **Dietary Risks: Folate, Alcohol and Gene Polymorphisms**

Zi-Yuan Zhou1, Keitaro Matsuo2,Wen-Chang Wang3, Huan Yang1, Kazuo Tajima2 and Jia Cao1 *1Department of Hygiene Toxicology , (Key Laboratory of The Ministry of Education of China), College of Preventive Medicine, The Third Military Medical University, ChongQing, 2Division of Epidemiology and Prevention, Aichi Cancer Center Research Institute, Chikusa-ku, Nagoya, 3Department of Statistics, College of Preventive Medicine, The Third Military Medical University, ChongQing, 1,3China 2Japan*

#### **1. Introduction**

490 Colorectal Cancer – From Prevention to Patient Care

Wu, D., Rosner, G.L., & Broemeling, L.D. (2007). Bayesian inference for the lead time in

Zackrisson, S., Andersson, I., Janzon, L., Manjer, J., & Garne, J.P. (2006). Rate Of Over-

880), ISBN: 0006-341X.

ISBN: 09598138.

periodic cancer screening. *Biometrics*, Vol. 63, No. 3, (September 2007), pp. (873-

Diagnosis Of Breast Cancer 15 Years After End Of Malmö Mammographic Screening Trial: Follow-Up Study. Vol. 332, No. 7543, (March 2006), pp. (689-691),

> Folate, as a member of the water-soluble B-group vitamins, is found widely in foodstuffs. Folate cannot be synthesized by human therefore dietary intake is the only source for human to obtain folate. The pteropolyglumates, usually with 1~6 glutamic acid molecules, are the major forms of natural food folates (Lucock 2000) and the 5-methyl tetrahydrofolate (5-metTHF), derived from hydrolization of absorbed folate and folic acid (the synthetic form of folate) as well, is the primary form in circulation (Ulrich 2005; Pietrzik, Bailey et al. 2010). Through the transmembrane transportation, the 5-metTHF in cell can be reduced by dihydrafolate reductase to tetrahydrofolate (THF) that is directly involved in metabolic process (Lucock 2000; Pietrzik, Bailey et al. 2010), and then performs biological functions in several ways. THF can be metabolized to 5,10-methylene-THF and further be irreversibly reduced into 5-metTHF which is the key step in one-carbon unit metabolism that is catalyzed by the enzyme methylenetetrahydrofolate reductase (MTHFR). By using the methyl donated by 5-methyltetrahydrofolate, the enzyme methionine synthase (MS) converts homocystine to methionine and then the *de novo* synthesized methionine can be catalyzed by the methionine adenosyl transferase to yield S-adenosylmethionine which directly provides methyl for a variety of important *in vivo* methylation reactions (Lucock 2000; Sanderson, Stone et al. 2007)*.* By using 5,10-methylene-THF as methyl donor, the enzyme thymidylate synthase converts deoxyuridylate (dUMP) to deoxythymidylate (dTMP), meanwhile the 5,10-formyltetrahydrofolate from 5,10-methylene-THF is involved in the production of both adenosine and guanosine, all are physiological building blocks of DNA replication (Bollheimer, Buettner et al. 2005; Duthie 2011). Thus, the most prominent function of folate is to transfer and process the one-carbon unit which is needed for methylation reactions and synthesis of thymine and purines. Consequently, folate deficiency

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 493

folate metabolism are likely presented in the etiology of CRC risk, but the published results were not consistent (Giovannucci 2004; Mason and Choi 2005; Matsuo, Ito et al. 2005; Kim 2007; Kim 2007). Several published studies have investigated the role of folate metabolizing gene polymorphisms and their interactions with folate intake or alcohol drinking in the etiology of CRC, but few in Chinese populations. Therefore, we performed a case-control study to assess the effect of folate intake and some reported functional polymorphisms in genes involved in folate metabolism in the etiology of CRC in Chinese populations, either for their individual effects or the joint effects with alcohol consumption and tobacco

All cases and controls were recruited from those who were registered into three hospitals of Chongqing City, Southwest China, between January 2001 and September 2004. All cases were newly diagnosed and histopathologically confirmed as having CRC, without any prior cancer history or any chemo-radiotherapies. Controls were cancer-free inpatients from those who had no other severe diseases (i.e., severe cardiovascular diseases, diabetes, severe hypertension, fatty liver and hepatocirrhosis), and without cancer history. Cases and controls were frequency matched by sex, age (±5 years), residence (the same city or county). All subjects were aged between 30-80 years and asked during personal interview to provide a one-time 2~5 ml peripheral blood sample and to complete a questionnaire that elicited information about lifestyles including alcohol drinking and tobacco smoking (1 year prior to the diagnosis for the cases and the time at recruitment for the controls). This study was approved by the Research Ethics Committee of The Third Military Medical University. All subjects provided a signed written informed consent or oral consent if illiterate. Finally, of a total of 1082 cases and 949 controls, we recruited 478 eligible cases (185 colon and 293 rectal cancer) and 838 eligible controls, who had consented to the present study, completed the

Information about folate intake and alcohol consumption one year prior to CRC diagnosis (for cases) or the reference date at recruitment (for controls) was obtained by using the 119-item semi-quantitative food frequency questionnaire developed specifically for Chongqing middle-aged population in our previous work as described elsewhere (Zhou, Takezaki et al. 2004). Briefly, according to the folate content listed in China Food Composition 2002 (Institute of Nutrition and Food Safety. China Center of Disease Control 2002), the frequencies of consumed portion of each food were converted into nutrients; for example, a crude mean of daily folate intake was calculated by multiplying the daily various food intake by its folate content (per 100 grams). The main sources of folate included in the questionnaire were cereals, beans, legumes, nuts, eggs, meats, fishes, bread, edible roots, melons, mushrooms, vegetables and fruits. Similarly, all subjects were also asked to provide detailed information about dietary supplements consumed in the period of one year before diagnosis or recruitment. For alcohol drinking, those who consumed alcohol more than 50 grams each week for more than 6 months were defined as "drinkers". Consumption of all kinds of beverage (beer, alcohol, and wine) was

smoking.

**2.1 Subjects** 

**2. Materials and methods** 

questionnaires, and provided blood samples.

**2.2 Assessment of folate intake and alcohol consumption** 

may biologically implicated in physiological processes including base misincorporation and DNA strand breaks, insufficient *de nove* nucleotide synthesis, as well as impaired DNA repair and methylation (Lucock 2000; Ames 2001; Kim 2003; Sanderson, Stone et al. 2007; Duthie 2011).

Therefore, folate has been implicated in colorectal cancer (CRC) because that the steps of folate metabolism may be involved in distinct biological process. A number of epidemiologic and experimental studies concluded that folate may have an inverse association with risk of CRC, however, the results are not consistent and it is argued that too much folate may be unfavourable for preventing the development of CRC especially in those with precursor lesions such as invisible minor adenoma (Giovannucci 2002; Sharp and Little 2004; Strohle, Wolters et al. 2005; Kim 2006; Sanderson, Stone et al. 2007; Sauer, Mason et al. 2009; Kennedy, Stern et al. 2011). The distinct effects of folate on the development of CRC in populations with diversely genetic background suggest that genetic factors, as well as the interaction with folate intake and other coenzymatical factors, may also play a role in the prevention or promotion of colorectal carcinogenesis (Giovannucci 2002; Sharp and Little 2004; Kim 2006; Arasaradnam, Commane et al. 2008; Hubner and Houlston 2009). Growing evidence revealed that the polymorphisms in key folate-metabolism genes may also modify CRC risk in relation to folate intake; but the results are not consistent. Studies suggested that several functional polymorphisms in key genes involved in folate metabolism, such as *MS* A2756G, *MTHFR* C677T and A1298C, may associate with risk of CRC (Giovannucci 2002; Sharp and Little 2004; Kim 2007; Sanderson, Stone et al. 2007; Yu, Zhang et al. 2010). For the *MS* gene, it is still debatable to what extent can the *MS* 2756G variant modulate enzyme activity and plasma homocysteine levels, though evidence from epidemiological studies suggests an association between *MS* A2756G polymorphism and risk of CRC (DeVos, Chanson et al. 2008; Yu, Zhang et al. 2010). To date, several published studies have investigated the role of the *MS* 2756G variant and its interaction with folate intake in the etiology of CRC, mainly in western but few in Chinese populations (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999; Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009; Yu, Zhang et al. 2010). For the *MTHFR* gene, the 677T variant encodes an enzyme that is thermolabile, and the heterozygous CT or homozygous TT genotype have nearly 35% or 70% reduction in normal function of the enzyme *in vitro*, respectively (Molloy, Daly et al. 1997). Similarly, the 1298A>C change leads to a decrease in the enzyme activity reportedly *in vitro*, but to a lesser extent, compared with the 677T variant (Weisberg, Jacques et al. 2001). Although the association between *MTHFR* C677T or A1298C polymorphism and CRC risk has also been extensively investigated, the results are not consistent (Chen, Giovannucci et al. 1996; Ma, Stampfer et al. 1997; Chen, Giovannucci et al. 1998; Chen, Ma et al. 2002; Sharp and Little 2004; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Huang, Han et al. 2007; Kim 2007). In addition, alcohol drinking, as one of the known risk factors for CRC, can interfere with the metabolism of folate and one-carbon unit and thus may alter CRC risk in subjects carrying different genotypes. Some earlier studies on folate intake reported that the favourable effects of folate or some genotypes (such as the *MTHFR* 677TT genotype) can be conversely modified by alcohol drinking (Giovannucci 2004; Mason and Choi 2005; Matsuo,

Ito et al. 2005). Many, but not all, epidemiological studies further suggested that the interactions among folate intake, alcohol drinking and polymorphisms in genes involved in folate metabolism are likely presented in the etiology of CRC risk, but the published results were not consistent (Giovannucci 2004; Mason and Choi 2005; Matsuo, Ito et al. 2005; Kim 2007; Kim 2007). Several published studies have investigated the role of folate metabolizing gene polymorphisms and their interactions with folate intake or alcohol drinking in the etiology of CRC, but few in Chinese populations. Therefore, we performed a case-control study to assess the effect of folate intake and some reported functional polymorphisms in genes involved in folate metabolism in the etiology of CRC in Chinese populations, either for their individual effects or the joint effects with alcohol consumption and tobacco smoking.

#### **2. Materials and methods**

#### **2.1 Subjects**

492 Colorectal Cancer – From Prevention to Patient Care

may biologically implicated in physiological processes including base misincorporation and DNA strand breaks, insufficient *de nove* nucleotide synthesis, as well as impaired DNA repair and methylation (Lucock 2000; Ames 2001; Kim 2003; Sanderson, Stone et al. 2007;

Therefore, folate has been implicated in colorectal cancer (CRC) because that the steps of folate metabolism may be involved in distinct biological process. A number of epidemiologic and experimental studies concluded that folate may have an inverse association with risk of CRC, however, the results are not consistent and it is argued that too much folate may be unfavourable for preventing the development of CRC especially in those with precursor lesions such as invisible minor adenoma (Giovannucci 2002; Sharp and Little 2004; Strohle, Wolters et al. 2005; Kim 2006; Sanderson, Stone et al. 2007; Sauer, Mason et al. 2009; Kennedy, Stern et al. 2011). The distinct effects of folate on the development of CRC in populations with diversely genetic background suggest that genetic factors, as well as the interaction with folate intake and other coenzymatical factors, may also play a role in the prevention or promotion of colorectal carcinogenesis (Giovannucci 2002; Sharp and Little 2004; Kim 2006; Arasaradnam, Commane et al. 2008; Hubner and Houlston 2009). Growing evidence revealed that the polymorphisms in key folate-metabolism genes may also modify CRC risk in relation to folate intake; but the results are not consistent. Studies suggested that several functional polymorphisms in key genes involved in folate metabolism, such as *MS* A2756G, *MTHFR* C677T and A1298C, may associate with risk of CRC (Giovannucci 2002; Sharp and Little 2004; Kim 2007; Sanderson, Stone et al. 2007; Yu, Zhang et al. 2010). For the *MS* gene, it is still debatable to what extent can the *MS* 2756G variant modulate enzyme activity and plasma homocysteine levels, though evidence from epidemiological studies suggests an association between *MS* A2756G polymorphism and risk of CRC (DeVos, Chanson et al. 2008; Yu, Zhang et al. 2010). To date, several published studies have investigated the role of the *MS* 2756G variant and its interaction with folate intake in the etiology of CRC, mainly in western but few in Chinese populations (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999; Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009; Yu, Zhang et al. 2010). For the *MTHFR* gene, the 677T variant encodes an enzyme that is thermolabile, and the heterozygous CT or homozygous TT genotype have nearly 35% or 70% reduction in normal function of the enzyme *in vitro*, respectively (Molloy, Daly et al. 1997). Similarly, the 1298A>C change leads to a decrease in the enzyme activity reportedly *in vitro*, but to a lesser extent, compared with the 677T variant (Weisberg, Jacques et al. 2001). Although the association between *MTHFR* C677T or A1298C polymorphism and CRC risk has also been extensively investigated, the results are not consistent (Chen, Giovannucci et al. 1996; Ma, Stampfer et al. 1997; Chen, Giovannucci et al. 1998; Chen, Ma et al. 2002; Sharp and Little 2004; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Huang, Han et al. 2007; Kim 2007). In addition, alcohol drinking, as one of the known risk factors for CRC, can interfere with the metabolism of folate and one-carbon unit and thus may alter CRC risk in subjects carrying different genotypes. Some earlier studies on folate intake reported that the favourable effects of folate or some genotypes (such as the *MTHFR* 677TT genotype) can be conversely modified by alcohol drinking (Giovannucci 2004; Mason and Choi 2005; Matsuo, Ito et al. 2005). Many, but not all, epidemiological studies further suggested that the interactions among folate intake, alcohol drinking and polymorphisms in genes involved in

Duthie 2011).

All cases and controls were recruited from those who were registered into three hospitals of Chongqing City, Southwest China, between January 2001 and September 2004. All cases were newly diagnosed and histopathologically confirmed as having CRC, without any prior cancer history or any chemo-radiotherapies. Controls were cancer-free inpatients from those who had no other severe diseases (i.e., severe cardiovascular diseases, diabetes, severe hypertension, fatty liver and hepatocirrhosis), and without cancer history. Cases and controls were frequency matched by sex, age (±5 years), residence (the same city or county). All subjects were aged between 30-80 years and asked during personal interview to provide a one-time 2~5 ml peripheral blood sample and to complete a questionnaire that elicited information about lifestyles including alcohol drinking and tobacco smoking (1 year prior to the diagnosis for the cases and the time at recruitment for the controls). This study was approved by the Research Ethics Committee of The Third Military Medical University. All subjects provided a signed written informed consent or oral consent if illiterate. Finally, of a total of 1082 cases and 949 controls, we recruited 478 eligible cases (185 colon and 293 rectal cancer) and 838 eligible controls, who had consented to the present study, completed the questionnaires, and provided blood samples.

#### **2.2 Assessment of folate intake and alcohol consumption**

Information about folate intake and alcohol consumption one year prior to CRC diagnosis (for cases) or the reference date at recruitment (for controls) was obtained by using the 119-item semi-quantitative food frequency questionnaire developed specifically for Chongqing middle-aged population in our previous work as described elsewhere (Zhou, Takezaki et al. 2004). Briefly, according to the folate content listed in China Food Composition 2002 (Institute of Nutrition and Food Safety. China Center of Disease Control 2002), the frequencies of consumed portion of each food were converted into nutrients; for example, a crude mean of daily folate intake was calculated by multiplying the daily various food intake by its folate content (per 100 grams). The main sources of folate included in the questionnaire were cereals, beans, legumes, nuts, eggs, meats, fishes, bread, edible roots, melons, mushrooms, vegetables and fruits. Similarly, all subjects were also asked to provide detailed information about dietary supplements consumed in the period of one year before diagnosis or recruitment. For alcohol drinking, those who consumed alcohol more than 50 grams each week for more than 6 months were defined as "drinkers". Consumption of all kinds of beverage (beer, alcohol, and wine) was

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 495

scale in the unconditional logistic regression model with and without adjustment for sex, age, family cancer history of first and second degree relatives. When analyzing the joint effects of two polymorphisms, we used the combined common genotypes as the reference. Hardy-Weinberg equilibrium in the controls was checked for all genotyping data with the

All statistical analyses were performed by using SAS (version 8.0, SAS institute, Cary, NC),

Table 1 shows the characteristics of 478 cases and 838 controls included in the final analysis. Overall, the cases were slightly older than the controls with a mean age of 54.3 years for cases and 52.0 years for controls. There was no difference in the distributions of sex, education, and BMI. However, these differences were further adjusted for their residual

N=478

Mean 54.3±12.4 52.0±11.3 ≤50 Years 171 (35.8) 358 (42.7) 51-60 Years 137 (28.7) 287 (34.3) .≥61 Years 170 (35.4) 193 (23.0)

> Male 273 (57.1) 462 (55.1) Female 205 (42.9) 376 (44.9)

Illiterate 51 (10.7) 75 (8.9)

High school 98 (20.5) 151 (18.0) College or upper 42 (8.8) 86 (10.3)

> Mean 21.6±2.4 21.6±2.5 < 20.0 121 (25.3) 216 (25.8) 20-22.5 200 (41.8) 345 (41.2) 22.5-25.0 113 (23.6) 207 (24.7) > 25.0 44 (9.2) 70 (8.4)

Primary, middle school 287 (60.0) 526 (62.8)

The results of multiple logistic regression analyses of risk of CRC and folate intake, alcohol

Controls (%) N=838

test, and the exact *P* value was used to assess any departure of genotypes.

Variables Cases (%)

with two-sided tests and a significance level of 0.05.

**3. Results** 

**3.1 Baseline characteristics** 

effects in the later analyses.

Age

Gender

Educations

BMI (10 years ago)

Table 1. Characteristics of cases and controls in the present study

consumption, and tobacco smoking are summarized in Table 2.

**3.2 Folate intake, alcohol consumption, genotypes, and CRC risk** 

calculated as pure alcohol volume by their alcohol concentrations (%). For smoking, those who smoked more than four cigarettes each week in average for more than six months were defined as "smokers". Pack-years were calculated by multiplying the total years of smoking by the average packs smoked each day.

#### **2.3 Genotyping**

DNA was extracted for each subject from the buffy coat fraction with the Promega DNA Purification Wizard kit (Promega Co. Madison, WI). Genotyping was performed by the polymerase chain reaction restriction-fragments-length polymorphism analysis according to methods described previously for *MS* A2756G (De Marco, Calevo et al. 2002), *MTHFR*  C677T (Chen, Giovannucci et al. 1998) and A1298C (Yi, Pogribny et al. 2002), respectively. For the *MS* A2756G polymorphism, a 285-bp PCR product was digested with HaeIII at 37°C and visualized after electrophoresis; the genotypes identified were: 265bp for AA, 265bp, 185bp, 80bp for AG, and 185bp, 80bp for GG. For *MTHFR* C677T polymorphism, a 198-bp PCR product was digested with Hinf I at 37°C and visualized after electrophoresis; the genotypes identified as follows: 198bp for CC, 198bp, 175bp, 23bp for CT, and 175bp, 23bp for TT. For *MTHFR* A1298C polymorphism, a 128-bp PCR product was digested with MboII at 37°C and visualized after electrophoresis; the genotypes identified were: 72bp, 28bp, 28bp for AA, 100bp, 72bp, 28bp for AC, and 100bp, 28bp for CC. The 23bp and 28bp fragments had been electrophoresed out of the gel and cannot be seen. Two cases and four controls failed to be amplified for *MTHFR* and *MS* polymorphisms, possibly due to poor quality of DNA. For genotyping quality control, electrophoresis results of genotypes were identified by a double-blind check and tested for Hardy-Weinberg equilibrium. Furthermore, randomly selected 92 cases and 136 controls were re-genotyped, and there were no discrepancies between the original and repeated genotyping results.

#### **2.4 Statistical analysis**

Energy-adjusted daily folate intake was categorized into quartiles based on the distribution in the controls, odds ratios (ORs) and their 95% confidence intervals (CI) were calculated by unconditional logistic regression models to estimate the strength of association between CRC risk and folate intake, the lowest quartile was used as the reference in OR calculation and was further adjusted for sex, age (in years), family cancer history of first degree relatives (yes vs. no) and second degree relatives (yes vs. no). Alcohol consumption status was divided into four categories based on alcohol (g/d) consumed: non-drinker (0), <30 g/d, 30~100 g/d, and >100 g/d for estimation of ORs (95% CI), and the drinking status was only classified into non-drinker and drinker when exploring the interaction with genotypes. Similarly, smoking status was also divided into four categories based on pack years smoked: non-smoker (0), ~10, 10~20, and >20 pack years for estimation of ORs (95% CI), and the tobacco exposure status was only classified into non-smoker and smoker when exploring the interaction with genotypes.

For evaluating the association between CRC risk and polymorphisms of *MS* (A2756G) and *MTHFR* (C677T and A1298C) genes, both ORs (95% CI) with and without adjustment for sex, age, family cancer history of first and second degree relatives, BMI of 10 years ago (divided into 4 subgroups: <20, 20~22.5, 22.5~25.0, >25.0), alcohol drinking and smoking status (yes vs. no) were calculated by using unconditional logistic regression models. ORs (95% CIs) for gene-gene or gene-environment interactions were assessed on a multiplicative scale in the unconditional logistic regression model with and without adjustment for sex, age, family cancer history of first and second degree relatives. When analyzing the joint effects of two polymorphisms, we used the combined common genotypes as the reference. Hardy-Weinberg equilibrium in the controls was checked for all genotyping data with the test, and the exact *P* value was used to assess any departure of genotypes.

All statistical analyses were performed by using SAS (version 8.0, SAS institute, Cary, NC), with two-sided tests and a significance level of 0.05.

#### **3. Results**

494 Colorectal Cancer – From Prevention to Patient Care

calculated as pure alcohol volume by their alcohol concentrations (%). For smoking, those who smoked more than four cigarettes each week in average for more than six months were defined as "smokers". Pack-years were calculated by multiplying the total years of

DNA was extracted for each subject from the buffy coat fraction with the Promega DNA Purification Wizard kit (Promega Co. Madison, WI). Genotyping was performed by the polymerase chain reaction restriction-fragments-length polymorphism analysis according to methods described previously for *MS* A2756G (De Marco, Calevo et al. 2002), *MTHFR*  C677T (Chen, Giovannucci et al. 1998) and A1298C (Yi, Pogribny et al. 2002), respectively. For the *MS* A2756G polymorphism, a 285-bp PCR product was digested with HaeIII at 37°C and visualized after electrophoresis; the genotypes identified were: 265bp for AA, 265bp, 185bp, 80bp for AG, and 185bp, 80bp for GG. For *MTHFR* C677T polymorphism, a 198-bp PCR product was digested with Hinf I at 37°C and visualized after electrophoresis; the genotypes identified as follows: 198bp for CC, 198bp, 175bp, 23bp for CT, and 175bp, 23bp for TT. For *MTHFR* A1298C polymorphism, a 128-bp PCR product was digested with MboII at 37°C and visualized after electrophoresis; the genotypes identified were: 72bp, 28bp, 28bp for AA, 100bp, 72bp, 28bp for AC, and 100bp, 28bp for CC. The 23bp and 28bp fragments had been electrophoresed out of the gel and cannot be seen. Two cases and four controls failed to be amplified for *MTHFR* and *MS* polymorphisms, possibly due to poor quality of DNA. For genotyping quality control, electrophoresis results of genotypes were identified by a double-blind check and tested for Hardy-Weinberg equilibrium. Furthermore, randomly selected 92 cases and 136 controls were re-genotyped, and there were no

Energy-adjusted daily folate intake was categorized into quartiles based on the distribution in the controls, odds ratios (ORs) and their 95% confidence intervals (CI) were calculated by unconditional logistic regression models to estimate the strength of association between CRC risk and folate intake, the lowest quartile was used as the reference in OR calculation and was further adjusted for sex, age (in years), family cancer history of first degree relatives (yes vs. no) and second degree relatives (yes vs. no). Alcohol consumption status was divided into four categories based on alcohol (g/d) consumed: non-drinker (0), <30 g/d, 30~100 g/d, and >100 g/d for estimation of ORs (95% CI), and the drinking status was only classified into non-drinker and drinker when exploring the interaction with genotypes. Similarly, smoking status was also divided into four categories based on pack years smoked: non-smoker (0), ~10, 10~20, and >20 pack years for estimation of ORs (95% CI), and the tobacco exposure status was only classified into non-smoker and smoker when exploring

For evaluating the association between CRC risk and polymorphisms of *MS* (A2756G) and *MTHFR* (C677T and A1298C) genes, both ORs (95% CI) with and without adjustment for sex, age, family cancer history of first and second degree relatives, BMI of 10 years ago (divided into 4 subgroups: <20, 20~22.5, 22.5~25.0, >25.0), alcohol drinking and smoking status (yes vs. no) were calculated by using unconditional logistic regression models. ORs (95% CIs) for gene-gene or gene-environment interactions were assessed on a multiplicative

smoking by the average packs smoked each day.

discrepancies between the original and repeated genotyping results.

**2.3 Genotyping** 

**2.4 Statistical analysis** 

the interaction with genotypes.

#### **3.1 Baseline characteristics**

Table 1 shows the characteristics of 478 cases and 838 controls included in the final analysis. Overall, the cases were slightly older than the controls with a mean age of 54.3 years for cases and 52.0 years for controls. There was no difference in the distributions of sex, education, and BMI. However, these differences were further adjusted for their residual effects in the later analyses.


Table 1. Characteristics of cases and controls in the present study

#### **3.2 Folate intake, alcohol consumption, genotypes, and CRC risk**

The results of multiple logistic regression analyses of risk of CRC and folate intake, alcohol consumption, and tobacco smoking are summarized in Table 2.

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 497

highest (673.7 μg/d) folate intake levels were 0.55 (0.40-0.76), 0.53 (0.38-0.75) and 0.46 (0.31- 0.67), respectively (*P*trend = 0.004), and the same trend kept significance in either males or females as described in table 2. Alcohol users had significantly increased CRC risks by nearly 1.6-2.0 folds higher than those who never drank. The ORs (95% CI) for those who consumed alcohol <30g/d, 30-100g/d, and >100g/d were 1.99 (1.38-2.88), 1.64 (1.10-2.44) and 1.98 (1.21- 3.25), respectively (*P*trend=0.001). A longer drinking time was associated with significant increased CRC risk (*P*trend=0.001). Compared with those who drank less than five years, those who drank 6-20 years or more than 20 years had higher CRC risks with ORs (95% CI) of 2.35 (1.50-3.69) and 1.67 (1.20-2.31), respectively. However, we did not observe statistical evidence

Table 3 describes the distributions of *MS* A2756G, *MTHFR* C677T and A1298C polymorphisms and their associations with CRC risk. Those rare *MS* 2756G genotypes carriers had a 1.5 fold increased risk (OR=1.49; 95% CI: 1.11-1.96; *P*trend=0.017), and the same trend was observed in the rare homozygous GG carriers though without significant difference. However, Both the C677T and A1298C polymorphisms of the *MTHFR* gene were not associated with CRC risk in this study population. We also tested the joint effect or locus-locus interaction. By using the combination of any two common genotypes as the reference, there were no meaningful interactions between *MS* A2756G and *MTHFR* C677T /

Genotypes Cases (%) Controls (%) ORa (95%CI) ORb (95%CI)

AG 113 (23.8) 141 (16.9) 1.52 (1.15-2.01) 1.50 (1.13-1.99) GG 4 (0.8) 6 (0.7) 1.06 (0.29-3.88) 1.12 (0.30-4.11) AG+GG 117( 24.6) 147 (17.6) 1.50 (1.13-1.99) 1.49 (1.11-1.96)

CT 210 (43.9) 362 (43.4) 0.97 (0.76-1.25) 0.99 (0.77-1.27) TT 66 (13.8) 123 (14.7) 0.90 (0.63-1.28) 0.87 (0.61-1.24) CT+TT 276 (57.7) 485 (58.2) 0.96 (0.76-1.21) 0.96 (0.76-1.22)

AC 158 (33.2) 282 (33.8) 0.96 (0.75-1.22) 0.96 (0.75-1.22) CC 23 (4.8) 41 (4.9) 0.95 (0.55-1.62) 0.96 (0.56-1.65) AC+CC 181 (38.0) 323 (38.7) 0.95 (0.75-1.21) 0.96 (0.76-1.21)

AA 359 (75.4) 688 (82.4) 1.00 1.00

*P* trend= 0.009 0.017

*P* trend= 0.684 0.609

*P* trend= 0.634 0.766

Table 3. CRC risk associated with genotypes of *MS* and *MTHFR* gene

a Adjusted for age, gender, cancer history of first and second degree relatives and total energy intake; b Adjusted for age, gender, cancer history of first and second degree relatives, total energy intake, BMI of

We further evaluated whether the impaction of folate intake on CRC risk was modulated by *MS* and *MTFHR* genes, alcohol drinking or tobacco smoking as summarized in Table 4.

AA 295 (62.0) 512 (61.3) 1.00 1.00

CC 202 (42.3) 349 (41.9) 1.00 1.00

of an association between smoking and CRC risk in this study population.

A1298C polymorphisms.

*MS* A2756G

*MTHFR* C677T

*MTHFR* A1298C

10 years ago, alcohol drinking and smoking


a Adjusted for age, gender, cancer history of first and second degree relatives and total energy intake.

Table 2. Odds Ratios (95% CI) of CRC associated with folate intake, alcohol consumption and smoking

In the four groups categorized by the quartile of folate intake, the CRC risk decreased significantly with the increasing folate intake in a dose-response manner (the same trend kept significance in either males or females). Compared with the lowest quartile (a mean intake of 324.1 μg/d), the ORs (95% CI) for the lower (442.3 μg/d), middle (522.8 μg/d) and

Cases/ Controls

56/112 0.47

64/116 0.52

68/142 0.45

60/78 1.93

62/88 1.62

35/47 1.95

41/49 2.30

109/155 1.58

41/71 1.45

51/73 1.54

104/162 1.40

ORa (95%CI)

(0.30-0.76)

(0.32-0.84)

(0.28-0.74)

(1.26-2.95)

(1.07-2.45)

(1.17-3.24)

(1.38-3.81)

(1.11-2.23)

(0.88-2.41)

(0.96-2.47)

(0.95-2.07)

Cases/ Controls

47/98 0.60

36/94 0.52

19/67 0.41

15/12 2.69

0/1 NA

1/0 NA

7/6 3.04

7/4 3.85

1/7 0.28

1/2 0.77

0/2 NA

ORa (95%CI)

(0.38-0.95)

(0.31-0.86)

(0.21-0.82)

(1.20-6.04)

(0.96-9.57)

(1.07-13.81)

(0.03-2.34)

(0.07-8.91)

Total Male Female

Lowest. (324.0) 188/209 1.00 85/92 1.00 103/117 1.00

*P*trend= 0.004 0.005 0.002

Non-Drinkers 305/612 1.00 116/249 1.00 189/363 1.00

*P*trend= 0.001 0.002 0.011

~5 years 314/624 1.00 123/258 1.00 191/366 1.00

Ptrend= 0.001 0.005 0.009

Non-Smoker 280/521 1.00 77/156 1.00 203/365 1.00

*P*trend= 0.187 0.087 0.169 a Adjusted for age, gender, cancer history of first and second degree relatives and total energy intake. Table 2. Odds Ratios (95% CI) of CRC associated with folate intake, alcohol consumption

In the four groups categorized by the quartile of folate intake, the CRC risk decreased significantly with the increasing folate intake in a dose-response manner (the same trend kept significance in either males or females). Compared with the lowest quartile (a mean intake of 324.1 μg/d), the ORs (95% CI) for the lower (442.3 μg/d), middle (522.8 μg/d) and

ORa (95%CI)

(0.40-0.76)

(0.38-0.75)

(0.31-0.67)

(1.38-2.88)

(1.10-2.44)

(1.21-3.25)

(1.50-3.69)

(1.20-2.31)

(0.75-1.91)

(0.87-2.15)

(0.87-1.83)

52/75 1.37

62/89 1.64

Cases/ Controls

Lower (442.3) 103/210 0.55

Middle (522.8) 100/210 0.53

Highest (673.7) 87/209 0.46

Light ~30 g/d 75/90 1.99

Heavy >100 g/d 36/47 1.98

6~20 years 48/55 2.35

20~ years 116/159 1.67

Light ~10PY 42/78 1.20

Heavy >20PY 104/164 1.26

Environmental

Folate intake (mean, μg/d)

Alcohol drinking

Moderate ~100

Years of alcohol drinking

g/d

Smoking

Moderate 10~20PY

and smoking

Factor

highest (673.7 μg/d) folate intake levels were 0.55 (0.40-0.76), 0.53 (0.38-0.75) and 0.46 (0.31- 0.67), respectively (*P*trend = 0.004), and the same trend kept significance in either males or females as described in table 2. Alcohol users had significantly increased CRC risks by nearly 1.6-2.0 folds higher than those who never drank. The ORs (95% CI) for those who consumed alcohol <30g/d, 30-100g/d, and >100g/d were 1.99 (1.38-2.88), 1.64 (1.10-2.44) and 1.98 (1.21- 3.25), respectively (*P*trend=0.001). A longer drinking time was associated with significant increased CRC risk (*P*trend=0.001). Compared with those who drank less than five years, those who drank 6-20 years or more than 20 years had higher CRC risks with ORs (95% CI) of 2.35 (1.50-3.69) and 1.67 (1.20-2.31), respectively. However, we did not observe statistical evidence of an association between smoking and CRC risk in this study population.

Table 3 describes the distributions of *MS* A2756G, *MTHFR* C677T and A1298C polymorphisms and their associations with CRC risk. Those rare *MS* 2756G genotypes carriers had a 1.5 fold increased risk (OR=1.49; 95% CI: 1.11-1.96; *P*trend=0.017), and the same trend was observed in the rare homozygous GG carriers though without significant difference. However, Both the C677T and A1298C polymorphisms of the *MTHFR* gene were not associated with CRC risk in this study population. We also tested the joint effect or locus-locus interaction. By using the combination of any two common genotypes as the reference, there were no meaningful interactions between *MS* A2756G and *MTHFR* C677T / A1298C polymorphisms.


a Adjusted for age, gender, cancer history of first and second degree relatives and total energy intake; b Adjusted for age, gender, cancer history of first and second degree relatives, total energy intake, BMI of 10 years ago, alcohol drinking and smoking

Table 3. CRC risk associated with genotypes of *MS* and *MTHFR* gene

We further evaluated whether the impaction of folate intake on CRC risk was modulated by *MS* and *MTFHR* genes, alcohol drinking or tobacco smoking as summarized in Table 4.

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 499

Lowest 124/144 Ref. 64/65 1.06

Lower 67/134 0.57 (0.38-0.85) 36/76 0.55

Higher 52/131 0.45 (0.30-0.70) 48/79 0.72

Highest 37/112 0.39 (0.24-0.65) 50/97 0.59

a Adjusted for age, gender, cancer history of first and second degree relatives, total energy intake and

Table 4. CRC risk associated with interactions between folate intake and genotypes (*MS* A2756G and *MTHFR* C677T and A1298C) or environmental factors (alcohol and smoke)

Foalte intake was inversely associated in a dose-dependent manner with CRC risk independent of the three genotypes of *MS* A2756G, *MTHFR* C677T and A1298C. Results in groups stratified by each genotype (common or rare) were similar, though the protective effect of folate intake almost disappeared in those carrying rare *MS* 2756 AG or GG genotype; there was no evidence of an interaction between folate intake and each polymorphism in a multiplicative model. When non-drinkers having the lowest level of folate intake was used as the reference, however, folate intake (from lower to highest) was associated with significantly decreasing CRC risk among non-drinkers (OR=0.39, 95% CI: 0.24-0.62 for the highest level), whereas the significance of protective effect of folate intake shown in Table 2 almost disappeared among drinkers who even had the highest level of folate intake (OR=0.86, 95% CI: 0.52-1.44); similarly, the protective effect of folate intake varied in smokers. Though alcohol drinking or tobacco smoking appeared to have an attenuated protective effect of folate intake, we failed to observe a statistically significant interaction with either drinking (*P*interaction=0.188) or smoking (*P*interaction=0.208) (Table 4).

Results of further analyses stratified by alcohol and smoking status are shown in Table 5. Here, we did observe a statistically significant interaction between the *MS* A2756G polymorphism and alcohol intake. An increased risk of CRC was observed in those alcohol drinkers carrying AG or GG genotype, whereas no significant association with alcohol drinking was observed among those carrying the AA genotype (*Pinteraction*=0.04); Compared with non-drinkers carrying *MS* 2756 AA, the ORs (95% CI) for AG or GG genotype carriers who were drinkers of light (~30 g/d), moderate (30~100 g/d) and highest (~100 g/d) level were 2.84 (1.44-5.60), 3.14 (1.44-6.83) and 4.40 (1.88-10.32), respectively. We also observed a borderline significant interaction between *MTHFR* A1298C polymorphism and alcohol intake (*P*interaction=0.07). For the 1298 AA genotype carriers, the CRC risk of drinkers was 2.0-

ORa (95% CI) Cases/

Controls

Smokers

ORa (95% CI) *P*interaction

0.208

(0.65-1.73)

(0.32-0.96)

(0.42-1.21)

(0.35-1.01)

Interactions Cases/

Folate × Smoking Non-

**3.3 Gene-environment interaction** 

Controls

Smokers

also adjusted for smoking status or drinking status, whenever appropriate.


Lowest 145/179 Ref. 42/28 1.84

Lower 74/168 0.55 (0.38-0.79) 29/42 0.84

Higher 78/177 0.54 (0.37-0.79) 21/32 0.83

Highest 62/164 0.46 (0.30-0.71) 25/45 0.65

Lowest 77/80 Ref. 111/127 0.93

Lower 44/88 0.55 (0.33-0.90) 59/122 0.50

Higher 36/91 0.43 (0.25-0.72) 64/117 0.58

Highest 45/90 0.54 (0.32-1.02) 42/119 0.35

1298AA AC+CC

Lowest 111/138 Ref. 76/69 1.34

Lower 66/120 0.69 (0.46-1.04) 37/90 0.50

Higher 61/124 0.63 (0.41-0.97) 38/85 0.53

Highest 57/130 0.54 (0.34-0.85) 30/79 0.47

Lowest 134/163 Ref. 54/46 1.63

Lower 72/154 0.58 (0.40-0.84) 31/56 0.80

Higher 58/159 0.46 (0.31-0.69) 42/51 1.17

Highest 41/136 0.39 (0.24-0.62) 46/73 0.86

ORa (95% CI) Cases/

Controls

ORa (95% CI) *P*interaction

0.556

0.629

0.205

0.188

(1.07-3.16)

(0.49-1.45)

(0.45-1.53)

(0.37-1.16)

(0.62-1.41)

(0.32-0.79)

(0.36-0.92)

(0.21-0.60)

(0.88-2.05)

(0.31-0.81)

(0.32-0.87)

(0.27-0.80)

(1.00-2.65)

(0.46-1.38)

(0.69-1.99)

(0.52-1.44)

Drinkers

Interactions Cases/

*1298* 

Folate × *MTHFR* 

Folate × Drinking Non-

Drinkers

Controls

Folate × *MS* 2756 2756AA AG+GG

Folate × *MTHFR 677* 677CC CT+TT


a Adjusted for age, gender, cancer history of first and second degree relatives, total energy intake and also adjusted for smoking status or drinking status, whenever appropriate.

Table 4. CRC risk associated with interactions between folate intake and genotypes (*MS* A2756G and *MTHFR* C677T and A1298C) or environmental factors (alcohol and smoke)

Foalte intake was inversely associated in a dose-dependent manner with CRC risk independent of the three genotypes of *MS* A2756G, *MTHFR* C677T and A1298C. Results in groups stratified by each genotype (common or rare) were similar, though the protective effect of folate intake almost disappeared in those carrying rare *MS* 2756 AG or GG genotype; there was no evidence of an interaction between folate intake and each polymorphism in a multiplicative model. When non-drinkers having the lowest level of folate intake was used as the reference, however, folate intake (from lower to highest) was associated with significantly decreasing CRC risk among non-drinkers (OR=0.39, 95% CI: 0.24-0.62 for the highest level), whereas the significance of protective effect of folate intake shown in Table 2 almost disappeared among drinkers who even had the highest level of folate intake (OR=0.86, 95% CI: 0.52-1.44); similarly, the protective effect of folate intake varied in smokers. Though alcohol drinking or tobacco smoking appeared to have an attenuated protective effect of folate intake, we failed to observe a statistically significant interaction with either drinking (*P*interaction=0.188) or smoking (*P*interaction=0.208) (Table 4).

#### **3.3 Gene-environment interaction**

Results of further analyses stratified by alcohol and smoking status are shown in Table 5. Here, we did observe a statistically significant interaction between the *MS* A2756G polymorphism and alcohol intake. An increased risk of CRC was observed in those alcohol drinkers carrying AG or GG genotype, whereas no significant association with alcohol drinking was observed among those carrying the AA genotype (*Pinteraction*=0.04); Compared with non-drinkers carrying *MS* 2756 AA, the ORs (95% CI) for AG or GG genotype carriers who were drinkers of light (~30 g/d), moderate (30~100 g/d) and highest (~100 g/d) level were 2.84 (1.44-5.60), 3.14 (1.44-6.83) and 4.40 (1.88-10.32), respectively. We also observed a borderline significant interaction between *MTHFR* A1298C polymorphism and alcohol intake (*P*interaction=0.07). For the 1298 AA genotype carriers, the CRC risk of drinkers was 2.0-

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 501

2.4 folds of non-drinkers; however, for those carrying 1298 AC or CC genotype, alcohol drinking was non-significantly associated with CRC risk, compared with non-drinkers

We also tested the gene-smoking interaction. The patterns of risk associated with *MS* A2756G genotypes seemed to vary by smoking status. For example, smoking was found to be associated with significantly increased CRC risk in *MS* 2756G carriers but not in 2756AA carriers, and there was evidence of an interaction (*P*interaction=0.006). Compared with nonsmokers carrying the AA genotype, an OR of 2.90 (1.61-5.22) was observed for those with AG or GG genotype and the highest smoking level (>20 pack-years). However, there were no evidence of an interaction between the genotypes of *MTHFR* C677T or A1298C and

**4.1 Folate intake, alcohol drinking,** *MS* **A2756G polymorphism and CRC risk** 

Folate is traditionally regarded as a protective factor for CRC, and many studies have reported a beneficial role in reducing CRC risk, especially in some large-scale case-control or cohort studies (Giovannucci 2002; Terry, Jain et al. 2002; Sanjoaquin, Allen et al. 2005; Strohle, Wolters et al. 2005; Kennedy, Stern et al. 2011), but in recent years some clinical intervention trials have raised the controversy that an increased CRC risk may be produced when folate, especially fortified or supplemental folic acid (synthetic), was administered in an excessive dose and was inopportunely administered when there has some existing lesions (such as undetectable small cancer or precursors) (Strohle, Wolters et al. 2005; Hubner and Houlston 2009; Sauer, Mason et al. 2009). Nonetheless, there is no confirmative evidence against the hypothesis that the loss of homeostasis of folate-mediated one-carbon metabolism can cause abnormal DNA methylation or DNA misincorporation, thus resulting in colorectal neoplasia, but some studies argued that the folate (natural or synthetic) per se can indeed contribute to the reduction of CRC risk (Bollheimer, Buettner et al. 2005; Strohle, Wolters et al. 2005; Kim 2007; Sauer, Mason et al. 2009). The present study investigated the association between folate intake and the risk of CRC in a Chinese population, in which no one had the habit of daily use of any vitamin supplement; therefore, the "folate intake" evaluated in this study means only from natural food, and our results showed a significant association between higher folate intake and lower CRC risk (Table 2). Such a protective effect did not change substantially before and after multivariate adjustment, even in subgroups of colon or rectal cancer (data not shown). Therefore, the present study, generally consistent with many previously published reports (Giovannucci 2002; Terry, Jain et al. 2002; Sanjoaquin, Allen et al. 2005; Kennedy, Stern et al. 2011), provides a further insight and a support for an inverse association between folate (from food) intake and CRC risk in

Although folate intake alone showed a significant protection against CRC risk, the variation in *MS* and *MTHFR* genes may also play important roles in the folate-mediated methyl cycles, and both alcohol drinking and cigarette smoking are known to impair the absorption and biological actions of folate. Although there was no evidence for an interaction in the present study, there was a trend that the protective effect of the folate appeared to be more obvious in those who were not exposed to the known risk factors drinking or smoking) compared with those who were exposed; in fact, the significant inverse association between folate intake and CRC risk was observed in 2756AA carriers or non-drinkers (non-smokers)

carrying the 1298 AA genotype (Table 5).

smoking (Table 5).

**4. Discussion** 

Chinese populations.


a Adjusted for age, cancer history of first and second degree relatives and total energy intake, smoking status or drinking status, whenever appropriate.

Table 5. CRC risk associated with interactions between genotypes (*MS* A2756G and *MTHFR* C677T and A1298C) and environmental factors (drinking and smoking)

2.4 folds of non-drinkers; however, for those carrying 1298 AC or CC genotype, alcohol drinking was non-significantly associated with CRC risk, compared with non-drinkers carrying the 1298 AA genotype (Table 5).

We also tested the gene-smoking interaction. The patterns of risk associated with *MS* A2756G genotypes seemed to vary by smoking status. For example, smoking was found to be associated with significantly increased CRC risk in *MS* 2756G carriers but not in 2756AA carriers, and there was evidence of an interaction (*P*interaction=0.006). Compared with nonsmokers carrying the AA genotype, an OR of 2.90 (1.61-5.22) was observed for those with AG or GG genotype and the highest smoking level (>20 pack-years). However, there were no evidence of an interaction between the genotypes of *MTHFR* C677T or A1298C and smoking (Table 5).

#### **4. Discussion**

500 Colorectal Cancer – From Prevention to Patient Care

Control ORa (95% CI) *P*interaction

Control ORa (95% CI) Case/

Heavy >100 g/d 21/37 1.46 (0.81-2.63) 15/10 4.40 (1.88-10.32) 0.041

Heavy >20PY 66/138 0.96 (0.64-1.45) 37/26 2.90 (1.61-5.22) 0.006

Heavy >100 g/d 15/20 1.83 (0.87-3.82) 21/27 1.82 (0.95-3.47) 0.780

Heavy >20PY 38/60 1.17 (0.69-1.98) 66/103 1.10 (0.70-1.72) 0.268

Heavy >100 g/d 24/28 2.39 (1.30-4.42) 12/19 1.64 (0.75-3.57) 0.069

Heavy >20PY 39/102 1.38 (0.89-2.13) 34/62 1.12 (0.66-1.89) 0.377 a Adjusted for age, cancer history of first and second degree relatives and total energy intake, smoking

Table 5. CRC risk associated with interactions between genotypes (*MS* A2756G and *MTHFR*

Non-Drinker 242/502 1.00 62/107 1.19 (0.83-1.71) Light ~30 g/d 52/73 1.83 (1.20-2.78) 22/17 2.84 (1.44-5.60) Moderate ~100 g/d 44/76 1.43 (0.92-2.24) 18/13 3.14 (1.44-6.83)

Non-Smoker 224/427 1.00 55/91 1.13 (0.77-1.65) Light ~10PY 32/62 1.16 (0.70-1.95) 10/16 1.40 (0.60-3.28) Moderate 10~20PY 37/61 1.25 (0.76-2.06) 15/14 1.95 (0.87-4.35)

Non-Drinker 133/250 1.00 172/358 0.89 (0.671.18) Light ~30 g/d 29/47 1.39 (0.81-2.39) 46/43 2.32 (1.42-3.82) Moderate ~100 g/d 25/32 1.80 (0.97-3.31) 37/57 1.36 (0.82-2.25)

Non-Smoker 130/214 1.00 150/304 0.81 (0.60-1.09) Light ~10PY 18/36 0.96 (0.50-1.83) 24/42 1.14 (0.62-2.09) Moderate 10~20PY 16/39 0.78 (0.39-1.54) 36/36 1.63 (0.92-2.89)

Non-Drinker 179/375 1.00 125/234 1.14 (0.85-1.52) Light ~30 g/d 46/50 2.40 (1.51-3.81) 28/40 1.66 (0.85-2.88) Moderate ~100 g/d 46/59 1.97 (1.23-3.14) 16/30 1.22 (0.61-2.44)

Non-Smoker 165/323 1.00 114/195 1.09 (0.81-1.49) Light ~10PY 27/42 1.50 (0.85-2.66) 15/36 0.92 (0.46-1.82) Moderate 10~20PY 34/45 1.46 (0.84-2.51) 18/30 1.32 (0.67-2.60)

C677T and A1298C) and environmental factors (drinking and smoking)

*MTHFR* 1298 1298AA AC+CC

*MS* 2756 2756AA AG+GG

status or drinking status, whenever appropriate.

*MTHFR* 677 677CC CT+TT

Interactions Case/

*MS* 2756×Drinking

*MS* 2756×Smoking

*MTHFR* 677×Drinking

*MTHFR* 677×Smoking

*MTHFR* 1298×Drinking

*MTHFR* 1298×Smoking

#### **4.1 Folate intake, alcohol drinking,** *MS* **A2756G polymorphism and CRC risk**

Folate is traditionally regarded as a protective factor for CRC, and many studies have reported a beneficial role in reducing CRC risk, especially in some large-scale case-control or cohort studies (Giovannucci 2002; Terry, Jain et al. 2002; Sanjoaquin, Allen et al. 2005; Strohle, Wolters et al. 2005; Kennedy, Stern et al. 2011), but in recent years some clinical intervention trials have raised the controversy that an increased CRC risk may be produced when folate, especially fortified or supplemental folic acid (synthetic), was administered in an excessive dose and was inopportunely administered when there has some existing lesions (such as undetectable small cancer or precursors) (Strohle, Wolters et al. 2005; Hubner and Houlston 2009; Sauer, Mason et al. 2009). Nonetheless, there is no confirmative evidence against the hypothesis that the loss of homeostasis of folate-mediated one-carbon metabolism can cause abnormal DNA methylation or DNA misincorporation, thus resulting in colorectal neoplasia, but some studies argued that the folate (natural or synthetic) per se can indeed contribute to the reduction of CRC risk (Bollheimer, Buettner et al. 2005; Strohle, Wolters et al. 2005; Kim 2007; Sauer, Mason et al. 2009). The present study investigated the association between folate intake and the risk of CRC in a Chinese population, in which no one had the habit of daily use of any vitamin supplement; therefore, the "folate intake" evaluated in this study means only from natural food, and our results showed a significant association between higher folate intake and lower CRC risk (Table 2). Such a protective effect did not change substantially before and after multivariate adjustment, even in subgroups of colon or rectal cancer (data not shown). Therefore, the present study, generally consistent with many previously published reports (Giovannucci 2002; Terry, Jain et al. 2002; Sanjoaquin, Allen et al. 2005; Kennedy, Stern et al. 2011), provides a further insight and a support for an inverse association between folate (from food) intake and CRC risk in Chinese populations.

Although folate intake alone showed a significant protection against CRC risk, the variation in *MS* and *MTHFR* genes may also play important roles in the folate-mediated methyl cycles, and both alcohol drinking and cigarette smoking are known to impair the absorption and biological actions of folate. Although there was no evidence for an interaction in the present study, there was a trend that the protective effect of the folate appeared to be more obvious in those who were not exposed to the known risk factors drinking or smoking) compared with those who were exposed; in fact, the significant inverse association between folate intake and CRC risk was observed in 2756AA carriers or non-drinkers (non-smokers)

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 503

findings suggested this variant may play a role in the etiology of CRC in Chinese populations, possibly a risk factor of CRC for Asia populations, in contrast to a protective effect in other ethnic populations. However, this finding needs to be further validated in

In the present study we found that *MTHFR* 677 or 1298 variants were non-significantly associated with decreased CRC risk which is consistent in trend with other earlier epidemiological studies (Chen, Giovannucci et al. 1996; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Huang, Han et al. 2007; Kim 2007). The frequencies of 677T or 1298C alleles in our controls were very close to those of other Chinese populations and different Asia populations, such as Japanese or Korean, although the 1298C allele frequency was a slightly lower compared with western white populations (Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Kim 2007). Laboratory evidence suggested that the rare 677T or 1298C allele can result in decreased enzyme activity *in vitro* (Molloy, Daly et al. 1997; Weisberg, Jacques et al. 2001) which seemed to favor an increased CRC risk, but on the contrary, most reported studies have not found an significant association between these *MTHFR* polymorphisms and CRC risk, and some earlier studies even reported an inverse association especially in white populations that were likely to have a relatively higher average total folate intake, partly due to use of vitamin supplements (Chen, Giovannucci et al. 1996; Ma, Stampfer et al. 1997). One-carbon unit metabolism may depend on a series of enzymatic steps forming a complex biochemical network, in which multiple dietary or environmental factors (e.g., vitamin B2, B12, and alcohol) may interact with folate, therefore the genetic variations of the *MTHFR* gene alone might not be sufficient to influence colorectal tumorigenesis during the onecarbon unit metabolism. Larger studies are required to further evaluate gene-gene and geneenvironment interactions in the association between *MTHFR* C677T or A1298C

Though our study was relatively smaller, we did find some evidence of interactions between the *MS* A2756G polymorphism and three environmental factors (folate intake, alcohol use, and tobacco smoking) in the CRC etiology. Our study provided the first report of an effect of the *MS* 2756 A>G polymorphism and its interactions with dietary folate intake, alcohol

Epidemiological studies have linked heavy alcohol use to increased risk of CRC, (Giovannucci 2002; Cho, Smith-Warner et al. 2004). Because alcohol can break the folate or disturb the one-carbon unit metabolism and thus may cause abnormal DNA methylation, DNA repair, or increase the activation of precarcinogen in liver by inducing cytochrome p-450 (Giovannucci 2004; Sharp and Little 2004), drinkers carrying rare *MS* 2756G*,* MTHFR *677T* or 1298C alleles may have additional CRC risk caused by abnormal folate metabolism (Giovannucci 2002; Sharp and Little 2004; Matsuo, Ito et al. 2005; Yamaji, Iwasaki et al. 2009). Our results supported such an association as well as a possible interaction between these

Cigarette smoking may play a role in CRC but is not a major recognized risk factor, even after a long period of exposure (Giovannucci 2001; Anderson, Attam et al. 2003), and this was also true in the present study. However, we found an interaction between *MS A2756G* genotypes and smoking; compared with non-smokers carrying the *MS* 2756AA genotype, the AG or GG carriers of heavy smokers (>20 pack-years) had a 3-fold increased CRC risk. It

consumption or cigarette smoking on CRC risk in a Chinese population.

larger studies of Asia populations.

polymorphisms and CRC risk in Chinese populations.

**4.3 Gene-environment interactions and CRC risk** 

polymorphisms and alcohol use in CRC risk.

but not in either 2756G carriers or drinkers (or smokers). It seems that, to some extent, the favourable effect of folate may be impaired by the *MS* 2756G allele or drinking (or smoking), a finding consistent with other published studies (Kim 2007; Kim 2007). However, because our study was relatively small, larger studies, especially in Chinese populations, are needed to validate such an interaction between folate intake and *MS* 2756 AG+GG genotypes or drinking (smoking).

#### **4.2 Gene polymorphisms and CRC risk**

Several studies have investigated the association between the *MS* 2756 A>G polymorphism and CRC risk but generated conflicting results (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999; Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009; Yu, Zhang et al. 2010). Using the common AA genotype as the reference, four studies reported no overall effect but a non-significantly association between CRC risk and the 2756G genotypes (Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009). Recently, a Japanese study and an American study with 257/771 and 513/609 cases/controls, respectively, also supported the trend that the *MS* 2756G genotypes can elevate the risk of colorectal cancer or adenomas (Goode, Potter et al. 2004; Matsuo, Ito et al. 2005). One population-based case-control study of colon, but not rectal, cancer found no association (Ulrich, Curtin et al. 2005); only two earlier studies (one cancer and one adenomas) found significantly reduced risk among AG or GG carriers (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999), and one large-scale nested casecontrol study reported an inverse association between the G allele and CRC risk in Norwegians (Ulvik, Vollset et al. 2004). The present study was the first to explore the effect of the *MS* polymorphism on CRC risk in a Chinese population, and we found that the *MS* 2756 AG or GG genotypes were significantly associated with increased risk of CRC, further supporting a positive association between rare 2756 AG or GG genotypes and CRC risk.

We also observed another interesting finding that the frequency of the *MS* 2756G allele (9.2%) among controls was materially different from those (15%~20%) among different ethnic populations reported by other studies, and the frequency of 2756GG genotype in our study was less than 1% in both cases and controls, much lower than 3%-5% in other populations, such as Americans, Europeans and other Asia populations of Japanese or Hindoo (Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Diwakar, Rudresh Kumar et al. 2008; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009). However, our results, especially for the frequencies of AG and GG genotypes among controls, were very similar to other studies in Chinese populations that investigated the association between the *MS* A2756G polymorphism and Alzheimer disease or lung cancer, in which the G allele frequency for controls was 8.5% and 9.8%, respectively (Liu, Jin et al. 2008; Zhao, Li et al. 2008). Therefore, it is likely that the allele frequency of the 2756G in Chinese is quite different from that of western or other Asia populations such as Japanese or Hindoo. It is still unclear whether the *MS* 2756A>G polymorphism has any functional consequences in its enzyme activity, but it was suggested that this polymorphism may probably decrease the enzyme activity, since the polymorphic site lies in a region connecting the vitamin B12 binding domain and the activation domain (Matthews, Sheppard et al. 1998). Considering epidemiologic evidence and the rare 2756G allele frequency in Chinese populations, our

but not in either 2756G carriers or drinkers (or smokers). It seems that, to some extent, the favourable effect of folate may be impaired by the *MS* 2756G allele or drinking (or smoking), a finding consistent with other published studies (Kim 2007; Kim 2007). However, because our study was relatively small, larger studies, especially in Chinese populations, are needed to validate such an interaction between folate intake and *MS* 2756 AG+GG genotypes or

Several studies have investigated the association between the *MS* 2756 A>G polymorphism and CRC risk but generated conflicting results (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999; Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009; Yu, Zhang et al. 2010). Using the common AA genotype as the reference, four studies reported no overall effect but a non-significantly association between CRC risk and the 2756G genotypes (Koushik, Kraft et al. 2006; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009). Recently, a Japanese study and an American study with 257/771 and 513/609 cases/controls, respectively, also supported the trend that the *MS* 2756G genotypes can elevate the risk of colorectal cancer or adenomas (Goode, Potter et al. 2004; Matsuo, Ito et al. 2005). One population-based case-control study of colon, but not rectal, cancer found no association (Ulrich, Curtin et al. 2005); only two earlier studies (one cancer and one adenomas) found significantly reduced risk among AG or GG carriers (Chen, Giovannucci et al. 1998; Ma, Stampfer et al. 1999), and one large-scale nested casecontrol study reported an inverse association between the G allele and CRC risk in Norwegians (Ulvik, Vollset et al. 2004). The present study was the first to explore the effect of the *MS* polymorphism on CRC risk in a Chinese population, and we found that the *MS* 2756 AG or GG genotypes were significantly associated with increased risk of CRC, further supporting a positive association between rare 2756 AG or GG genotypes and CRC risk. We also observed another interesting finding that the frequency of the *MS* 2756G allele (9.2%) among controls was materially different from those (15%~20%) among different ethnic populations reported by other studies, and the frequency of 2756GG genotype in our study was less than 1% in both cases and controls, much lower than 3%-5% in other populations, such as Americans, Europeans and other Asia populations of Japanese or Hindoo (Goode, Potter et al. 2004; Ulvik, Vollset et al. 2004; Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Ulrich, Curtin et al. 2005; Koushik, Kraft et al. 2006; Diwakar, Rudresh Kumar et al. 2008; Theodoratou, Farrington et al. 2008; Yamaji, Iwasaki et al. 2009). However, our results, especially for the frequencies of AG and GG genotypes among controls, were very similar to other studies in Chinese populations that investigated the association between the *MS* A2756G polymorphism and Alzheimer disease or lung cancer, in which the G allele frequency for controls was 8.5% and 9.8%, respectively (Liu, Jin et al. 2008; Zhao, Li et al. 2008). Therefore, it is likely that the allele frequency of the 2756G in Chinese is quite different from that of western or other Asia populations such as Japanese or Hindoo. It is still unclear whether the *MS* 2756A>G polymorphism has any functional consequences in its enzyme activity, but it was suggested that this polymorphism may probably decrease the enzyme activity, since the polymorphic site lies in a region connecting the vitamin B12 binding domain and the activation domain (Matthews, Sheppard et al. 1998). Considering epidemiologic evidence and the rare 2756G allele frequency in Chinese populations, our

drinking (smoking).

**4.2 Gene polymorphisms and CRC risk** 

findings suggested this variant may play a role in the etiology of CRC in Chinese populations, possibly a risk factor of CRC for Asia populations, in contrast to a protective effect in other ethnic populations. However, this finding needs to be further validated in larger studies of Asia populations.

In the present study we found that *MTHFR* 677 or 1298 variants were non-significantly associated with decreased CRC risk which is consistent in trend with other earlier epidemiological studies (Chen, Giovannucci et al. 1996; Ulvik, Vollset et al. 2004; Matsuo, Ito et al. 2005; Huang, Han et al. 2007; Kim 2007). The frequencies of 677T or 1298C alleles in our controls were very close to those of other Chinese populations and different Asia populations, such as Japanese or Korean, although the 1298C allele frequency was a slightly lower compared with western white populations (Chen, Jiang et al. 2005; Matsuo, Ito et al. 2005; Kim 2007). Laboratory evidence suggested that the rare 677T or 1298C allele can result in decreased enzyme activity *in vitro* (Molloy, Daly et al. 1997; Weisberg, Jacques et al. 2001) which seemed to favor an increased CRC risk, but on the contrary, most reported studies have not found an significant association between these *MTHFR* polymorphisms and CRC risk, and some earlier studies even reported an inverse association especially in white populations that were likely to have a relatively higher average total folate intake, partly due to use of vitamin supplements (Chen, Giovannucci et al. 1996; Ma, Stampfer et al. 1997). One-carbon unit metabolism may depend on a series of enzymatic steps forming a complex biochemical network, in which multiple dietary or environmental factors (e.g., vitamin B2, B12, and alcohol) may interact with folate, therefore the genetic variations of the *MTHFR* gene alone might not be sufficient to influence colorectal tumorigenesis during the onecarbon unit metabolism. Larger studies are required to further evaluate gene-gene and geneenvironment interactions in the association between *MTHFR* C677T or A1298C polymorphisms and CRC risk in Chinese populations.

#### **4.3 Gene-environment interactions and CRC risk**

Though our study was relatively smaller, we did find some evidence of interactions between the *MS* A2756G polymorphism and three environmental factors (folate intake, alcohol use, and tobacco smoking) in the CRC etiology. Our study provided the first report of an effect of the *MS* 2756 A>G polymorphism and its interactions with dietary folate intake, alcohol consumption or cigarette smoking on CRC risk in a Chinese population.

Epidemiological studies have linked heavy alcohol use to increased risk of CRC, (Giovannucci 2002; Cho, Smith-Warner et al. 2004). Because alcohol can break the folate or disturb the one-carbon unit metabolism and thus may cause abnormal DNA methylation, DNA repair, or increase the activation of precarcinogen in liver by inducing cytochrome p-450 (Giovannucci 2004; Sharp and Little 2004), drinkers carrying rare *MS* 2756G*,* MTHFR *677T* or 1298C alleles may have additional CRC risk caused by abnormal folate metabolism (Giovannucci 2002; Sharp and Little 2004; Matsuo, Ito et al. 2005; Yamaji, Iwasaki et al. 2009). Our results supported such an association as well as a possible interaction between these polymorphisms and alcohol use in CRC risk.

Cigarette smoking may play a role in CRC but is not a major recognized risk factor, even after a long period of exposure (Giovannucci 2001; Anderson, Attam et al. 2003), and this was also true in the present study. However, we found an interaction between *MS A2756G* genotypes and smoking; compared with non-smokers carrying the *MS* 2756AA genotype, the AG or GG carriers of heavy smokers (>20 pack-years) had a 3-fold increased CRC risk. It

Dietary Risks: Folate, Alcohol and Gene Polymorphisms 505

Cancer from the Ministry of Education, Culture, Sports, Science and Technology of Japan (12670383), and also supported in part by the Doctoral Innovation Foundation of TMMU

Ames, B. N. (2001). "DNA damage from micronutrient deficiencies is likely to be a major

Anderson, J. C., R. Attam, et al. (2003). "Prevalence of colorectal neoplasia in smokers." *Am J* 

Arasaradnam, R. P., D. M. Commane, et al. (2008). "A review of dietary factors and its

Bollheimer, L. C., R. Buettner, et al. (2005). "Folate and its preventive potential in colorectal

Brown, K. S., L. A. Kluijtmans, et al. (2004). "The 5,10-methylenetetrahydrofolate reductase

Chen, J., E. Giovannucci, et al. (1998). "A prospective study of methylenetetrahydrofolate

Chen, J., E. Giovannucci, et al. (1996). "A methylenetetrahydrofolate reductase polymorphism and the risk of colorectal cancer." *Cancer Res* 56(21): 4862-4. Chen, J., J. Ma, et al. (2002). "Linkage disequilibrium between the 677C>T and 1298A>C

contributions to risk of colorectal cancer." *Pharmacogenetics* 12(4): 339-42. Chen, K., Q. T. Jiang, et al. (2005). "Relationship between metabolic enzyme polymorphism

Cho, E., S. A. Smith-Warner, et al. (2004). "Alcohol intake and colorectal cancer: a pooled

De Marco, P., M. G. Calevo, et al. (2002). "Study of MTHFR and MS polymorphisms as risk factors for NTD in the Italian population." *J Hum Genet* 47(6): 319-24. DeVos, L., A. Chanson, et al. (2008). "Associations between single nucleotide

Duthie, S. J. (2011). "Folate and cancer: how DNA damage, repair and methylation impact on

Giovannucci, E. (2001). "An updated review of the epidemiological evidence that cigarette

Giovannucci, E. (2002). "Epidemiologic studies of folate and colorectal neoplasia: a review." *J* 

and colorectal cancer." *World J Gastroenterol* 11(3): 331-5.

colon carcinogenesis." *J Inherit Metab Dis* 34(1): 101-9.

analysis of 8 cohort studies." *Ann Intern Med* 140(8): 603-13.

influence on DNA methylation in colorectal carcinogenesis." *Epigenetics* 3(4): 193-

carcinogenesis. How strong is the biological and epidemiological evidence?" *Crit* 

C677T polymorphism interacts with smoking to increase homocysteine."

reductase and methionine synthase gene polymorphisms, and risk of colorectal

polymorphisms in human methylenetetrahydrofolate reductase gene and their

polymorphisms in folate uptake and metabolizing genes with blood folate, homocysteine, and DNA uracil concentrations." *Am J Clin Nutr* 88(4): 1149-58. Diwakar, L., K. J. Rudresh Kumar, et al. (2008). "The influence of MTR A2756G

polymorphism on plasma homocysteine in young south Indians." *Clin Chim Acta*

smoking increases risk of colorectal cancer." *Cancer Epidemiol Biomarkers Prev* 10(7):

cause of cancer." *Mutat Res* 475(1-2): 7-20.

*Gastroenterol* 98(12): 2777-83.

*Rev Oncol Hematol* 55(1): 13-36.

*Atherosclerosis* 174(2): 315-22.

395(1-2): 172-4.

*Nutr* 132(8 Suppl): 2350S-2355S.

725-31.

adenoma." *Carcinogenesis* 19(12): 2129-32.

(200403).

**7. References** 

8.

was reported that *MTHFR* 677T allele caused an increased plasm homocysteine concentration in heavy smokers than in moderate or non-smokers (Brown, Kluijtmans et al. 2004), and other studies found that an interaction between smoking and *MTHFR C677T* genotypes can be determinants of adenomatous and hyperplastic polyps of colorectum (Ulvik, Evensen et al. 2001). In the present study, however, we did not find any evidence of an interaction between smoking and *MTHFR* C677T or A1298C genotypes.

Overall, as Hubner mentioned very recently (Hubner and Houlston 2009), the existing evidence is still insufficient to confirm a protective effect of folate intake, and variants of the key metabolic-enzyme genes add the complexity to the unresolved problem of how and when the folate can have an effect on CRC risk. Our results suggested that geneenvironment interactions may affect CRC risk more profoundly than the individual effect of folate intake or any of other known risk factors in this study population.

There are some potential limitations in the present study. First of all, because we did not have serum levels of folate or homocysteine, there may be biases in categorizing actual folate intake levels that were solely based on questionnaire data. Second, this hospital-based case-control study may have introduced some unknown selection biases. However, we reasonably believe that our cases and controls came from the same population base served by the hospitals, because all cases were newly diagnosed and most of controls were also registered at hospitals for the first time. Third, there was inherent recall bias in case-control studies; however, our interviewers did not know case-control status of the subjects. Lastly, our relatively smaller sample size may not have sufficient study power to detect interactions among folate, alcohol, smoking and the studied genotypes on CRC risk.

#### **5. Conclusion**

The present study suggested that sufficient folate intake may reduce the risk of CRC, and alcohol use can significantly increase CRC risk in the study population. The *MS* 2756 AG+GG genotypes may be associated with an increased CRC risk; our data further suggested that the interaction between *MS* 2756 A>G polymorphism and alcohol use may result in further increased CRC risk in this Chinese populations. How and to what extent can these joint effects modify the CRC risk need additional larger epidemiological studies especially in other Chinese populations.

#### **6. Acknowledgment**

The authors would like to thank the research staff from the Preventive Medicine College, The Third Military Medical University (TMMU) and the staff from the Department of General Surgery, Orthopaedics, and Clinical Trauma of Southwest hospital, Xin-qiao hospital and Da-ping hospital. We are also grateful to Dr. Man-tian Mi from Food and Nutrition Department, TMMU for supporting the data of folate content, to Mrs. Xiao-li Shen and Qing Liu for interviewing, and to Mrs. Ya-jing Li and Xue-zheng Li for their help for preparing for laboratory experiments in this study. We thank Dr. Ana Neumann and Dr. Qingyi Wei of The University of Texas M. D. Anderson Cancer Center for reviewing and scientifically editing the manuscript. This work was supported in part by a Major International (Regional) Joint Research Projects (30320140461) and General Programs (30771841 and 30700676) from the National Natural Science Foundation of China (NSFC), and by a Grant-in Aid for Scientific Research on Special Priority Areas of Cancer from the Ministry of Education, Culture, Sports, Science and Technology of Japan (12670383), and also supported in part by the Doctoral Innovation Foundation of TMMU (200403).

#### **7. References**

504 Colorectal Cancer – From Prevention to Patient Care

was reported that *MTHFR* 677T allele caused an increased plasm homocysteine concentration in heavy smokers than in moderate or non-smokers (Brown, Kluijtmans et al. 2004), and other studies found that an interaction between smoking and *MTHFR C677T* genotypes can be determinants of adenomatous and hyperplastic polyps of colorectum (Ulvik, Evensen et al. 2001). In the present study, however, we did not find any evidence of

Overall, as Hubner mentioned very recently (Hubner and Houlston 2009), the existing evidence is still insufficient to confirm a protective effect of folate intake, and variants of the key metabolic-enzyme genes add the complexity to the unresolved problem of how and when the folate can have an effect on CRC risk. Our results suggested that geneenvironment interactions may affect CRC risk more profoundly than the individual effect of

There are some potential limitations in the present study. First of all, because we did not have serum levels of folate or homocysteine, there may be biases in categorizing actual folate intake levels that were solely based on questionnaire data. Second, this hospital-based case-control study may have introduced some unknown selection biases. However, we reasonably believe that our cases and controls came from the same population base served by the hospitals, because all cases were newly diagnosed and most of controls were also registered at hospitals for the first time. Third, there was inherent recall bias in case-control studies; however, our interviewers did not know case-control status of the subjects. Lastly, our relatively smaller sample size may not have sufficient study power to detect interactions

The present study suggested that sufficient folate intake may reduce the risk of CRC, and alcohol use can significantly increase CRC risk in the study population. The *MS* 2756 AG+GG genotypes may be associated with an increased CRC risk; our data further suggested that the interaction between *MS* 2756 A>G polymorphism and alcohol use may result in further increased CRC risk in this Chinese populations. How and to what extent can these joint effects modify the CRC risk need additional larger epidemiological studies

The authors would like to thank the research staff from the Preventive Medicine College, The Third Military Medical University (TMMU) and the staff from the Department of General Surgery, Orthopaedics, and Clinical Trauma of Southwest hospital, Xin-qiao hospital and Da-ping hospital. We are also grateful to Dr. Man-tian Mi from Food and Nutrition Department, TMMU for supporting the data of folate content, to Mrs. Xiao-li Shen and Qing Liu for interviewing, and to Mrs. Ya-jing Li and Xue-zheng Li for their help for preparing for laboratory experiments in this study. We thank Dr. Ana Neumann and Dr. Qingyi Wei of The University of Texas M. D. Anderson Cancer Center for reviewing and scientifically editing the manuscript. This work was supported in part by a Major International (Regional) Joint Research Projects (30320140461) and General Programs (30771841 and 30700676) from the National Natural Science Foundation of China (NSFC), and by a Grant-in Aid for Scientific Research on Special Priority Areas of

an interaction between smoking and *MTHFR* C677T or A1298C genotypes.

folate intake or any of other known risk factors in this study population.

among folate, alcohol, smoking and the studied genotypes on CRC risk.

**5. Conclusion** 

**6. Acknowledgment** 

especially in other Chinese populations.


Dietary Risks: Folate, Alcohol and Gene Polymorphisms 507

Molloy, A. M., S. Daly, et al. (1997). "Thermolabile variant of 5,10-methylenetetrahydrofolate

Pietrzik, K., L. Bailey, et al. (2010). "Folic acid and L-5-methyltetrahydrofolate: comparison

Sanderson, P., E. Stone, et al. (2007). "Folate and colo-rectal cancer risk." *Br J Nutr* 98(6):

Sanjoaquin, M. A., N. Allen, et al. (2005). "Folate intake and colorectal cancer risk: a meta-

Sauer, J., J. B. Mason, et al. (2009). "Too much folate: a risk factor for cancer and

Sharp, L. and J. Little (2004). "Polymorphisms in genes involved in folate metabolism and colorectal neoplasia: a HuGE review." *Am J Epidemiol* 159(5): 423-43. Strohle, A., M. Wolters, et al. (2005). "Folic acid and colorectal cancer prevention: molecular mechanisms and epidemiological evidence (Review)." *Int J Oncol* 26(6): 1449-64. Terry, P., M. Jain, et al. (2002). "Dietary intake of folic acid and colorectal cancer risk in a

Theodoratou, E., S. M. Farrington, et al. (2008). "Dietary vitamin B6 intake and the risk of

Ulrich, C. M. (2005). "Nutrigenetics in cancer research--folate metabolism and colorectal

Ulrich, C. M., K. Curtin, et al. (2005). "Polymorphisms in the reduced folate carrier,

Ulvik, A., E. T. Evensen, et al. (2001). "Smoking, folate and methylenetetrahydrofolate

Ulvik, A., S. E. Vollset, et al. (2004). "Colorectal cancer and the methylenetetrahydrofolate

Weisberg, I. S., P. F. Jacques, et al. (2001). "The 1298A-->C polymorphism in

Yamaji, T., M. Iwasaki, et al. (2009). "Methionine synthase A2756G polymorphism interacts

Yi, P., I. Pogribny, et al. (2002). "Multiplex PCR for simultaneous detection of 677 C-->T and

Yu, K., J. Zhang, et al. (2010). "Methionine synthase A2756G polymorphism and cancer risk:

Zhao, H. L., X. Q. Li, et al. (2008). "Association analysis of methionine synthase gene 2756

thymidylate synthase, or methionine synthase and risk of colon cancer." *Cancer* 

reductase status as interactive determinants of adenomatous and hyperplastic

reductase 677C -> T and methionine synthase 2756A -> G polymorphisms: a study of 2,168 case-control pairs from the JANUS cohort." *Cancer Epidemiol Biomarkers Prev*

methylenetetrahydrofolate reductase (MTHFR): in vitro expression and association

with alcohol and folate intake to influence the risk of colorectal adenoma." *Cancer* 

1298 A-->C polymorphisms in methylenetetrahydrofolate reductase gene for

A>G polymorphism and Alzheimer disease in a Chinese population." *Brain Res*

colorectal cancer." *Cancer Epidemiol Biomarkers Prev* 17(1): 171-82.

cardiovascular disease?" *Curr Opin Clin Nutr Metab Care* 12(1): 30-6.

recommendations." *Lancet* 349(9065): 1591-3.

analytical approach." *Int J Cancer* 113(5): 825-8.

cohort of women." *Int J Cancer* 97(6): 864-7.

*Epidemiol Biomarkers Prev* 14(11 Pt 1): 2509-16.

polyps of colorectum." *Am J Med Genet* 101(3): 246-54.

with homocysteine." *Atherosclerosis* 156(2): 409-15.

a meta-analysis." *Eur J Hum Genet* 18(3): 370-8.

population studies of cancer risk." *Cancer Lett* 181(2): 209.

*Epidemiol Biomarkers Prev* 18(1): 267-74.

cancer." *J Nutr* 135(11): 2698-702.

13(12): 2175-80.

1204: 118-22.

48.

1299-304.

reductase associated with low red-cell folates: implications for folate intake

of clinical pharmacokinetics and pharmacodynamics." *Clin Pharmacokinet* 49(8): 535-


Giovannucci, E. (2004). "Alcohol, one-carbon metabolism, and colorectal cancer: recent

Goode, E. L., J. D. Potter, et al. (2004). "Methionine synthase D919G polymorphism, folate

Huang, Y., S. Han, et al. (2007). "Different roles of MTHFR C677T and A1298C

Hubner, R. A. and R. S. Houlston (2009). "Folate and colorectal cancer prevention." *Br J* 

Institute of Nutrition and Food Safety. China Center of Disease Control (2002). *China Food* 

Kennedy, D. A., S. J. Stern, et al. (2011). "Folate intake and the risk of colorectal cancer: a

Kim, D. H. (2007). "The interactive effect of methyl-group diet and polymorphism of

Kim, Y. I. (2003). "Role of folate in colon cancer development and progression." *J Nutr* 133(11

Kim, Y. I. (2006). "Folate: a magic bullet or a double edged sword for colorectal cancer

Kim, Y. I. (2007). "Folate and colorectal cancer: an evidence-based critical review." *Mol Nutr* 

Koushik, A., P. Kraft, et al. (2006). "Nonsynonymous polymorphisms in genes in the one-

Liu, H., G. Jin, et al. (2008). "Association of polymorphisms in one-carbon metabolizing

Lucock, M. (2000). "Folic acid: nutritional biochemistry, molecular biology, and role in

Ma, J., M. J. Stampfer, et al. (1999). "A polymorphism of the methionine synthase gene:

Ma, J., M. J. Stampfer, et al. (1997). "Methylenetetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer." *Cancer Res* 57(6): 1098-102. Mason, J. B. and S. W. Choi (2005). "Effects of alcohol on folate metabolism: implications for

Matsuo, K., H. Ito, et al. (2005). "One-carbon metabolism related gene polymorphisms

Matthews, R. G., C. Sheppard, et al. (1998). "Methylenetetrahydrofolate reductase and

carbon metabolism pathway and associations with colorectal cancer." *Cancer* 

genes and lung cancer risk: a case-control study in Chinese population." *Lung* 

association with plasma folate, vitamin B12, homocyst(e)ine, and colorectal cancer

interact with alcohol drinking to influence the risk of colorectal cancer in Japan."

methionine synthase: biochemistry and molecular biology." *Eur J Pediatr* 157 Suppl

methylenetetrahydrofolate reductase on the risk of colorectal cancer." *Mutat Res*

*Composition Table 2002*. Beijing, Peking University Medical Press.

systematic review and meta-analysis." *Cancer Epidemiol* 35(1): 2-10.

metabolism, and colorectal adenoma risk." *Cancer Epidemiol Biomarkers Prev* 13(1):

polymorphisms in colorectal adenoma and colorectal cancer: a meta-analysis." *J* 

insights from molecular studies." *J Nutr* 134(9): 2475S-2481S.

157-62.

*Hum Genet* 52(1): 73-85.

*Cancer* 100(2): 233-9.

622(1-2): 14-8.

Suppl 1): 3731S-3739S.

*Food Res* 51(3): 267-92.

*Cancer* 61(1): 21-9.

prevention?" *Gut* 55(10): 1387-9.

*Epidemiol Biomarkers Prev* 15(12): 2408-17.

disease processes." *Mol Genet Metab* 71(1-2): 121-38.

risk." *Cancer Epidemiol Biomarkers Prev* 8(9): 825-9.

carcinogenesis." *Alcohol* 35(3): 235-41.

*Carcinogenesis* 26(12): 2164-71.

2: S54-9.


**26** 

*Japan* 

**The Prognostic Significance of Number of** 

**Lymph Node Metastasis in Colon Cancer –** 

**and Handling of Resected Specimens** 

*Kurume University School of Medicine, Department of Surgery, Fukuoka,* 

Yoshito Akagi, Romeo Kansakar and Kazuo Shirouzu

**Based on Japanese Techniques of Resection** 

Staging systems for cancer reflects the prognosis of the disease and it is used to choose the modality of treatment. The TNM classification has mainly been used in the west. In Japan, Japanese classification according to General Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus(JGR)(Japanese Society for cancer of colon and rectum, 2009) is used. The degree of the lymph node metastasis in each staging system has

In 2009, 7th edition of (JGR, 2009) was revised to make it uniform with the 6th edition of TNM classification (Sobin & Wittekond, 2002). However, the 7th edition of TNM classification (Sobin et al., 2009) was further revised where the category of nodal status was subdivided (Table 1) on the basis of number of positive lymph nodes. The validity of which is based on the pooled SEER database of 109,953 cases of colorectal cancer lymph node metastases (Gunderson et al., 2010). Japanese classification of nodal status takes into account not only the number of positive lymph nodes but also the site from where they are retrieved according to the location of the tumor. Our study showed recategorization of lymph nodes such as 1, 2 to 6 and 7 or more lymph nodes with metastasis reflected the prognosis of the disease (Akagi et al., 2010). Thus, the number of lymph nodes retrieved plays a vital role in the staging system and is one of the main prognostic indicators of the disease. The various techniques of resection and handling of resected specimens may also vary according to different institutions and countries. The number of lymph nodes retrieved can depend on different factors like the surgical technique, length of resection, mesocolic excision, lymph node dissection, handling of resected specimen and criteria for pathological diagnosis which has some differences in Japan as compared to the west. Moreover, chemotherapy protocols and treatment of recurrence also may vary in different places. This can alter the stage, recurrence rate, as well as the outcome of the disease. Therefore, here we have elaborated our technique of resection, specimen handling and nodal dissection which is uniformly practiced in all centers of Japan and present data from our center where these techniques

**1. Introduction** 

some variations.

have been carried out consecutively.

Zhou, Z. Y., T. Takezaki, et al. (2004). "Development of a semi-quantitative food frequency questionnaire to determine variation in nutrient intakes between urban and rural areas of Chongqing, China." *Asia Pac J Clin Nutr* 13(3): 273-83.

### **The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer – Based on Japanese Techniques of Resection and Handling of Resected Specimens**

Yoshito Akagi, Romeo Kansakar and Kazuo Shirouzu *Kurume University School of Medicine, Department of Surgery, Fukuoka, Japan* 

#### **1. Introduction**

508 Colorectal Cancer – From Prevention to Patient Care

Zhou, Z. Y., T. Takezaki, et al. (2004). "Development of a semi-quantitative food frequency

areas of Chongqing, China." *Asia Pac J Clin Nutr* 13(3): 273-83.

questionnaire to determine variation in nutrient intakes between urban and rural

Staging systems for cancer reflects the prognosis of the disease and it is used to choose the modality of treatment. The TNM classification has mainly been used in the west. In Japan, Japanese classification according to General Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus(JGR)(Japanese Society for cancer of colon and rectum, 2009) is used. The degree of the lymph node metastasis in each staging system has some variations.

In 2009, 7th edition of (JGR, 2009) was revised to make it uniform with the 6th edition of TNM classification (Sobin & Wittekond, 2002). However, the 7th edition of TNM classification (Sobin et al., 2009) was further revised where the category of nodal status was subdivided (Table 1) on the basis of number of positive lymph nodes. The validity of which is based on the pooled SEER database of 109,953 cases of colorectal cancer lymph node metastases (Gunderson et al., 2010). Japanese classification of nodal status takes into account not only the number of positive lymph nodes but also the site from where they are retrieved according to the location of the tumor. Our study showed recategorization of lymph nodes such as 1, 2 to 6 and 7 or more lymph nodes with metastasis reflected the prognosis of the disease (Akagi et al., 2010). Thus, the number of lymph nodes retrieved plays a vital role in the staging system and is one of the main prognostic indicators of the disease. The various techniques of resection and handling of resected specimens may also vary according to different institutions and countries. The number of lymph nodes retrieved can depend on different factors like the surgical technique, length of resection, mesocolic excision, lymph node dissection, handling of resected specimen and criteria for pathological diagnosis which has some differences in Japan as compared to the west. Moreover, chemotherapy protocols and treatment of recurrence also may vary in different places. This can alter the stage, recurrence rate, as well as the outcome of the disease. Therefore, here we have elaborated our technique of resection, specimen handling and nodal dissection which is uniformly practiced in all centers of Japan and present data from our center where these techniques have been carried out consecutively.

The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer –

from location of the feeding arteries where the tumor was located.

P

I

P

P

M : Main lymph nodes, I : Intermediate lymph nodes P : Pericolic lymph nodes

M I

M

M

I I

M

I

P

P

P

I

I

P

P

P

Fig. 1.

Based on Japanese Techniques of Resection and Handling of Resected Specimens 511

excision by sharp dissection of the entire mesocolon with intact facial layers and ligation of the supplying vessels at its origin was performed (Hohenberger et al. 2009). The pedicle of artery of the main lymph nodes was ligated and cut. In principle, the extent of mesocolon supplied by the feeding artery and all regional lymph nodes were removed en block for advanced cancer (Table 2.). For early cancer, the intermediate and pericolic lymph nodes of feeding artery were removed. Length of bowel resection was 10cm proximally and distally


LN; Lymph node, LNM; Lymph node metastasis, JGR - Japanese classification according to General Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus

Table 1. The changes of lymph nodes category between former and current system

#### **2. Methods**

#### **2.1 Patients**

A total of 1107 patients with primary colon cancer treated by curative resection from January 1, 1985 to December 31, 2006, were identified from the colorectal cancer database of Kurume University, Fukuoka, Japan. Of these patients, 361 patients with Dukes C colon cancer located from caecum to recto sigmoid junction were included in this study. Patients who underwent neoadjuvavnt chemo-radiotherapy, patients with familial adenomatous polyposis (FAP) or inflammatory bowel disease (IBD) and patients with rectal cancer were excluded from the study. The median age of patients was 66 years (64.9±12.6) and 213 (59%) patients were male. Almost all patients were administered oral prodrug of 5-fluorouracil as postoperative adjuvant chemotherapy. The median number of nodes examined was 28 (range, 5-108, average, 30.7 ±16.5) and the median duration of follow-up was 68 months (24- 186 months, average, 60.5±22.3) from the date of their initial surgery.

#### **2.2 Surgical technique for resection of colon and staging of lymph node status**

Surgery for colorectal cancer was performed by only certified colorectal surgeons. A similar protocol for length of resection and lymph node dissection for colon cancer was followed by all surgeons. The extent of the resection was determined by the location of cancer, its feeding arteries, cancer staging and the pattern of potential lymphatic spread. The feeding arteries were superior and inferior mesenteric artery and its branches such as ileocolic, right colic, middle colic, left colic and sigmoid arteries. The regional lymph nodes consisted of three groups; main, intermediate and pericolic lymph nodes (Figure 1.) Complete mesocolic

6th 7th LN category stage LN category stage

pericolic LN IIIa N1 1-3 pericolic/perirectal,

LN N3 Main LNM

Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus

186 months, average, 60.5±22.3) from the date of their initial surgery.

**2.2 Surgical technique for resection of colon and staging of lymph node status** 

Surgery for colorectal cancer was performed by only certified colorectal surgeons. A similar protocol for length of resection and lymph node dissection for colon cancer was followed by all surgeons. The extent of the resection was determined by the location of cancer, its feeding arteries, cancer staging and the pattern of potential lymphatic spread. The feeding arteries were superior and inferior mesenteric artery and its branches such as ileocolic, right colic, middle colic, left colic and sigmoid arteries. The regional lymph nodes consisted of three groups; main, intermediate and pericolic lymph nodes (Figure 1.) Complete mesocolic

Table 1. The changes of lymph nodes category between former and current system

N2

paraaortic LN IV M1 Metastasis beyond regional

LN; Lymph node, LNM; Lymph node metastasis, JGR - Japanese classification according to General

A total of 1107 patients with primary colon cancer treated by curative resection from January 1, 1985 to December 31, 2006, were identified from the colorectal cancer database of Kurume University, Fukuoka, Japan. Of these patients, 361 patients with Dukes C colon cancer located from caecum to recto sigmoid junction were included in this study. Patients who underwent neoadjuvavnt chemo-radiotherapy, patients with familial adenomatous polyposis (FAP) or inflammatory bowel disease (IBD) and patients with rectal cancer were excluded from the study. The median age of patients was 66 years (64.9±12.6) and 213 (59%) patients were male. Almost all patients were administered oral prodrug of 5-fluorouracil as postoperative adjuvant chemotherapy. The median number of nodes examined was 28 (range, 5-108, average, 30.7 ±16.5) and the median duration of follow-up was 68 months (24-

N1a 1 regional LNM

nodes

N1b 2-3 regional LNM

N2a 4-6 regional LNM

4 or more

N1c Satellite without regional

N2b 7 or more regional LNM IIIB<sup>~</sup>

pericolic/perirectal,

(include Lateral LN)

intermediate LNM IIIa

intermediate LNM IIIb

LN IV

IIIA~ IIIC

IIIC

TNM

JGR

**2. Methods 2.1 Patients** 

N1 1- 3 regional LNM IIIA<sup>~</sup>

LNM IIIC

intermediate LN IIIb

N2 4 or more regional

N1 Metastasis in

N2 Metastasis in

N4 Metastasis in

N3 Metastasis in main

IIIB

excision by sharp dissection of the entire mesocolon with intact facial layers and ligation of the supplying vessels at its origin was performed (Hohenberger et al. 2009). The pedicle of artery of the main lymph nodes was ligated and cut. In principle, the extent of mesocolon supplied by the feeding artery and all regional lymph nodes were removed en block for advanced cancer (Table 2.). For early cancer, the intermediate and pericolic lymph nodes of feeding artery were removed. Length of bowel resection was 10cm proximally and distally from location of the feeding arteries where the tumor was located.

M : Main lymph nodes, I : Intermediate lymph nodes P : Pericolic lymph nodes Fig. 1.

The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer –

a) b)

d) The tumor sectioned at 5mm intervals e) Pathological specimen for examination.

Fig. 2. Handling of fresh specimen after ileocecal resection

c)

Based on Japanese Techniques of Resection and Handling of Resected Specimens 513


The intestine with tumor and adjacent mesocolon ( : Lymph nodes are included in this) removed after ligation of pedicle ( ).

Table 2. Operation for advanced colon cancer

#### **2.3 Handling of resected specimen**

Lymph node dissection was carried out by the surgeon prior to formalin fixation in fresh resected specimens. The lymph nodes along the feeding vessels were picked up from the mesocolon and kept separately according to the lymph node stations and fixed in formalin (Fig. 2a,b,c). The pericolic nodes in the fat tissue beside the tumor were left intact for the correct judgment of depth of invasion. The opened intestine was placed on a board with the mucosal side up and the edge stretched and pinned to reproduce its original appearance. After formalin fixation for several days the tumor was sectioned at 5 mm intervals (Fig. 2d). One of the deepest invasive specimens was examined by expert pathologist (Fig. 2e). The final decision of histological examination of specimen and lymph node metastasis was made by the surgical colorectal pathologist (K. Shirouzu; co-author).

The intestine with tumor and adjacent mesocolon ( : Lymph nodes are included in this) removed after

Lymph node dissection was carried out by the surgeon prior to formalin fixation in fresh resected specimens. The lymph nodes along the feeding vessels were picked up from the mesocolon and kept separately according to the lymph node stations and fixed in formalin (Fig. 2a,b,c). The pericolic nodes in the fat tissue beside the tumor were left intact for the correct judgment of depth of invasion. The opened intestine was placed on a board with the mucosal side up and the edge stretched and pinned to reproduce its original appearance. After formalin fixation for several days the tumor was sectioned at 5 mm intervals (Fig. 2d). One of the deepest invasive specimens was examined by expert pathologist (Fig. 2e). The final decision of histological examination of specimen and lymph node metastasis was made

ligation of pedicle ( ).

Table 2. Operation for advanced colon cancer

by the surgical colorectal pathologist (K. Shirouzu; co-author).

**2.3 Handling of resected specimen** 


d) The tumor sectioned at 5mm intervals e) Pathological specimen for examination.

Fig. 2. Handling of fresh specimen after ileocecal resection

The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer –

**3.2 Recategorization and its association with survival** 

and more LNM were considered to be poor prognostic factors.

(a) According to number of lymph node metastasis (LNM). A: 1 LNM, B: 2 LNM, C: 3 LNM, D: 2-6 LNM, E: 7 and more LNM

(b) According to recategorized group A: 1 LNM, B: 2-6 LNM, C: 7 or more LNM Fig. 3. Cumulative survival curves

Based on Japanese Techniques of Resection and Handling of Resected Specimens 515

The survival curves of five groups based on the former method (Akagi et al., 2010) were compared. Group A consisted of cases of 1 LNM, group B 2 LNM, group C 3 LNM, group D 4-6 LNM, and group E 7 or more LNM. Survival curves for group B, C and D were similar (Fig. 3a). Based on the above-mentioned results, the survival curve of each group, survival rate, patient number and current classification was considered and integrated and reorganized into three groups (Table 3, Figure 3b). The new classification used 1 LNM for group A (143 cases), 2-6 LNM for group B (184 cases), and 7LNM for group C (38 cases). In brief, N category seemed the most appropriate when the number of LNM was classified as 1, 2-6, and 7. The 5-year survival rate for each group was A; 81.6%, B; 70.9%, and C; 44.1%, respectively (Table 3). This classification more accurately reflected the prognosis as compared to the conventional categories for LNM. Factors influencing prognosis were extracted from clinicopathological factors of every group by univariate analysis. Then, correlative prognostic factors were included in the new classification of LNM degree was estimated with multivariate analysis (Table 4). As for independent prognostic factors, degree of venous invasion and seven

#### **2.4 Statistical analysis**

The cases were classified according to the number of metastatic lymph nodes. Survival rate for each group was assessed. A new classification was then considered to recategorize the lymph nodes from cases with similar survival rates. In the new classification, survival rate was assessed with prognostic-relate factors inferred statistically. Analysis of variance or a ttest was used to analyze continuous variables. 2 test was used for categorical variables. Five-year survival rates and prognostic factors were estimated using Kaplan-Meier survival method and Cox proportional hazard regression model, respectively. Log-rank test was used to assess whether survival differences were significant. Values of *p*<0.05 was considered statistically significant. Statistical analysis was performed using JMP software ver. 8.0 (SAS Institute, Cary, NC, USA).

#### **3. Results**

#### **3.1 Number of lymph nodes retrieved and Lymph Node Metastasis (LNM)**

Table 3. shows the number of cases, average number of lymph nodes retrieved, five year overall survival and recurrence rate in each group divided by number of LNM. The most common was the group which had one LNM. The median number of LNM was three (range, 1-14). The average number of lymph nodes retrieved in each group was from 26 to 47, and there is significant relation between number of LNM and lymph nodes retrieved. The more the number of lymph nodes retrieved increased, the number of LNM also increased. Similarly the recurrence rate was higher when the number of lymph nodes retrieved was more. However, it did not show statistical significance. The 5 year survival rate for each group classified by the number of LNM is shown in Table 3. 5-year survival rate of the group with 1 LNM was significantly better than that of other groups and the group with 7 LNM showed significantly worst survival than groups with 6 LNM.


LNM: Lymph nodes metastasis, Ave.: average (range), 5 yr OS: 5 year overall survival \*: *p*<0.001 (number of LNM vs Ave. of LNs retrieved, \*\*: *p*= 0.763 (Ave. of LNs retrieved vs recurrence rate)

Table 3. Recurrence and Survival on the number of lymph nodes metastasis

#### **3.2 Recategorization and its association with survival**

514 Colorectal Cancer – From Prevention to Patient Care

The cases were classified according to the number of metastatic lymph nodes. Survival rate for each group was assessed. A new classification was then considered to recategorize the lymph nodes from cases with similar survival rates. In the new classification, survival rate was assessed with prognostic-relate factors inferred statistically. Analysis of variance or a ttest was used to analyze continuous variables. 2 test was used for categorical variables. Five-year survival rates and prognostic factors were estimated using Kaplan-Meier survival method and Cox proportional hazard regression model, respectively. Log-rank test was used to assess whether survival differences were significant. Values of *p*<0.05 was considered statistically significant. Statistical analysis was performed using JMP software

**3.1 Number of lymph nodes retrieved and Lymph Node Metastasis (LNM)** 

group with 7 LNM showed significantly worst survival than groups with 6 LNM.

LNM: Lymph nodes metastasis, Ave.: average (range), 5 yr OS: 5 year overall survival

Table 3. Recurrence and Survival on the number of lymph nodes metastasis

\*: *p*<0.001 (number of LNM vs Ave. of LNs retrieved, \*\*: *p*= 0.763 (Ave. of LNs retrieved vs recurrence rate)

Table 3. shows the number of cases, average number of lymph nodes retrieved, five year overall survival and recurrence rate in each group divided by number of LNM. The most common was the group which had one LNM. The median number of LNM was three (range, 1-14). The average number of lymph nodes retrieved in each group was from 26 to 47, and there is significant relation between number of LNM and lymph nodes retrieved. The more the number of lymph nodes retrieved increased, the number of LNM also increased. Similarly the recurrence rate was higher when the number of lymph nodes retrieved was more. However, it did not show statistical significance. The 5 year survival rate for each group classified by the number of LNM is shown in Table 3. 5-year survival rate of the group with 1 LNM was significantly better than that of other groups and the

**2.4 Statistical analysis** 

**3. Results** 

ver. 8.0 (SAS Institute, Cary, NC, USA).

The survival curves of five groups based on the former method (Akagi et al., 2010) were compared. Group A consisted of cases of 1 LNM, group B 2 LNM, group C 3 LNM, group D 4-6 LNM, and group E 7 or more LNM. Survival curves for group B, C and D were similar (Fig. 3a). Based on the above-mentioned results, the survival curve of each group, survival rate, patient number and current classification was considered and integrated and reorganized into three groups (Table 3, Figure 3b). The new classification used 1 LNM for group A (143 cases), 2-6 LNM for group B (184 cases), and 7LNM for group C (38 cases). In brief, N category seemed the most appropriate when the number of LNM was classified as 1, 2-6, and 7. The 5-year survival rate for each group was A; 81.6%, B; 70.9%, and C; 44.1%, respectively (Table 3). This classification more accurately reflected the prognosis as compared to the conventional categories for LNM. Factors influencing prognosis were extracted from clinicopathological factors of every group by univariate analysis. Then, correlative prognostic factors were included in the new classification of LNM degree was estimated with multivariate analysis (Table 4). As for independent prognostic factors, degree of venous invasion and seven and more LNM were considered to be poor prognostic factors.

(a) According to number of lymph node metastasis (LNM). A: 1 LNM, B: 2 LNM, C: 3 LNM, D: 2-6 LNM, E: 7 and more LNM (b) According to recategorized group A: 1 LNM, B: 2-6 LNM, C: 7 or more LNM

Fig. 3. Cumulative survival curves

The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer –

also depend on the facility, institutional protocol or the individual surgeon.

level is similar to the LN category in the 7th edition of TNM classification.

correlate with higher nodal metastasis rate. (Kim et al., 2006)

technique of dissection and the treatment of specimen.

lymph node metastasis.

Based on Japanese Techniques of Resection and Handling of Resected Specimens 517

al., 2009). Only in stage III colon cancer does the prognosis depend upon the number of

Lymph node (LN) involvement is an important prognostic indicator of carcinomas arising in the colon and the rectum. It also influences treatment decisions, as patients with nodepositive colorectal carcinoma (CRC) are generally advised for systemic adjuvant chemotherapy. Thus, the accuracy of number of lymph node metastasis becomes a very important factor. However, when the precision of diagnosis of metastasis to lymph nodes is concerned factors such as length of dissection of bowel, lymph node dissection techniques and treatment of the specimen needs to be considered. In addition, the retrieval method may

If the degree of LNM is inadequately assessed this changes the stage of the disease which in turn reflects on the further inappropriate treatment protocol and prognosis. Then the question of stage migration arises. However, inappropriate retrieval of lymph node and thus incorrect staging cannot be blamed on stage migration. Rectal cancer was excluded from this study as most of our patients did not undergo neoadjuvant chemo radiotherapy as patients in the west but underwent pelvic lymphadenectomy which increased the number of lymph nodes retrieved that may alter the surgical procedure preoperative stage and thus the prognosis. There have been several papers regarding relationship between degree of LNM and its prognosis like the number of LNM, the site of metastatic lymph nodes, number of lymph nodes retrieved, and lymph node ratio (LNR). Our data showed colon cancer with only 1 LNM had significantly better prognosis than 2 or more LNM and patients with 7 or more LNM had the worst prognosis. Our recategorization considered here with both number and

There are some reports that the prognosis of colon cancer is associated with number of LNM (Vaccaro et al., 2004) or classification by the number of LNM predicts prognosis better than classification by level of LNM (Carlos et al., 2004). On the other hand, Newland et al showed the level of LNM rather than the number of LNM is the most important variable associated with prognosis. (Newland et al., 1994) Tapper and Nelson et al. mentioned that staging for colorectal cancer required retrieval of 12-17 lymph nodes. (Tapper et al., 2001; Nelson et al., 2001) Kim et al. mentioned that retrieval of >10 lymph nodes offered almost certain identification of metastasis to lymph nodes, and tumor differentiation and T stage seemed to

The 7th ed. TNM classification mentions that histological examination of a regional lymphadenectomy specimen will ordinarily include 12 or more lymph nodes. The most accepted limit for accurate staging seems to be at least 12 nodes, as also suggested by other current node metastasis related publications (Wittekend et al., 2003; Greene et al., 2002). Cserni suggested not only the minimum number of LNs should be considered in terms of staging but some qualitative features may also influence the accurate staging. The question arises whether accurate staging can be reached with fewer than 12 LNs or not (Cserni et al.,1999). However, the number of lymph nodes obtained in specimens of colorectal cancer is significantly associated with the length of resected bowel, patient age and tumor location (Shen et al., 2009). Recent study by Cserni et al have mentioned that nodal status of CRCs may be adequately assessed by examining the lymph nodes from the close fraction around the tumor and the 3 cm side long bowel segment in both directions (Cserni et al., 2011). Thus, the retrievable number of lymph nodes depends on different factors like stage,


LN: lymph node, se: serosa, ss: sub serosa, poor: poorly differentiated adenocarcinoma, muci: mucinous carcinoma, well: well differentiated adenocarcinoma, mod: moderately differentiated adenocarcinoma, ly0-1, v0-1: negative to minimal invasion, ly2-3, v2-3: moderate to severe invasion, CEA:carcinoembryonic antigen, CI; confidence interval

Table 4. Independent Prognostic Factor for Desease Specific Survival using Cox Regression Analysis

#### **4. Discussion**

Depth of invasion and number of regional lymph node metastasis(LNM) are known to be important prognostic factors for colorectal cancer, and these factors are used to determine the stage of the disease (Chapuis et al.,1985; Vaccaro et al., 2004; Choen et al., 1991). In Japan, on the basis of clinical studies on colorectal cancer the general rules for clinical and pathological studies of cancer of the colon, rectum and anus which has been modified continuously (JGR, 1977). Based on these data the surgical procedure has been standardized with en bloc resection of tumor, distal and proximal normal colon, mesocolon along with apical vessels of feeding artery.


Japanese Society for cancer of colon and rectum, 1991-1996

Table 5. Recurrence rate of each stage of colorectal cancer after curative resection

A large difference has been found in the recurrence rate and prognosis in stage II and III colorectal cancer according to the report of 2004 in Japan. (Table 5.) (Japanese Society for Cancer of the Colon and Rectum, [JSCCR] Guidelines 2010 for the Treatment of Colorectal Cancer. (In Japanese). Similar data has also been published from other countries (Andre et

LN: lymph node, se: serosa, ss: sub serosa, poor: poorly differentiated adenocarcinoma, muci: mucinous carcinoma, well: well differentiated adenocarcinoma, mod: moderately differentiated adenocarcinoma,

Table 4. Independent Prognostic Factor for Desease Specific Survival using Cox Regression

Depth of invasion and number of regional lymph node metastasis(LNM) are known to be important prognostic factors for colorectal cancer, and these factors are used to determine the stage of the disease (Chapuis et al.,1985; Vaccaro et al., 2004; Choen et al., 1991). In Japan, on the basis of clinical studies on colorectal cancer the general rules for clinical and pathological studies of cancer of the colon, rectum and anus which has been modified continuously (JGR, 1977). Based on these data the surgical procedure has been standardized with en bloc resection of tumor, distal and proximal normal colon, mesocolon along with

I 3.70% 90.60% 89.30% 90.60% II 12.50% 83.60% 76.40% 81.20% IIIa 24.10% 76.10% 64.70% 71.40% IIIb 40.80% 62.10% 47.10% 56%

A large difference has been found in the recurrence rate and prognosis in stage II and III colorectal cancer according to the report of 2004 in Japan. (Table 5.) (Japanese Society for Cancer of the Colon and Rectum, [JSCCR] Guidelines 2010 for the Treatment of Colorectal Cancer. (In Japanese). Similar data has also been published from other countries (Andre et

colon rectum total

ly0-1, v0-1: negative to minimal invasion, ly2-3, v2-3: moderate to severe invasion,

stage\* recurrence rate 5 year OS

Table 5. Recurrence rate of each stage of colorectal cancer after curative resection

CEA:carcinoembryonic antigen, CI; confidence interval

Japanese Society for cancer of colon and rectum, 1991-1996

Analysis

**4. Discussion** 

apical vessels of feeding artery.

al., 2009). Only in stage III colon cancer does the prognosis depend upon the number of lymph node metastasis.

Lymph node (LN) involvement is an important prognostic indicator of carcinomas arising in the colon and the rectum. It also influences treatment decisions, as patients with nodepositive colorectal carcinoma (CRC) are generally advised for systemic adjuvant chemotherapy. Thus, the accuracy of number of lymph node metastasis becomes a very important factor. However, when the precision of diagnosis of metastasis to lymph nodes is concerned factors such as length of dissection of bowel, lymph node dissection techniques and treatment of the specimen needs to be considered. In addition, the retrieval method may also depend on the facility, institutional protocol or the individual surgeon.

If the degree of LNM is inadequately assessed this changes the stage of the disease which in turn reflects on the further inappropriate treatment protocol and prognosis. Then the question of stage migration arises. However, inappropriate retrieval of lymph node and thus incorrect staging cannot be blamed on stage migration. Rectal cancer was excluded from this study as most of our patients did not undergo neoadjuvant chemo radiotherapy as patients in the west but underwent pelvic lymphadenectomy which increased the number of lymph nodes retrieved that may alter the surgical procedure preoperative stage and thus the prognosis.

There have been several papers regarding relationship between degree of LNM and its prognosis like the number of LNM, the site of metastatic lymph nodes, number of lymph nodes retrieved, and lymph node ratio (LNR). Our data showed colon cancer with only 1 LNM had significantly better prognosis than 2 or more LNM and patients with 7 or more LNM had the worst prognosis. Our recategorization considered here with both number and level is similar to the LN category in the 7th edition of TNM classification.

There are some reports that the prognosis of colon cancer is associated with number of LNM (Vaccaro et al., 2004) or classification by the number of LNM predicts prognosis better than classification by level of LNM (Carlos et al., 2004). On the other hand, Newland et al showed the level of LNM rather than the number of LNM is the most important variable associated with prognosis. (Newland et al., 1994) Tapper and Nelson et al. mentioned that staging for colorectal cancer required retrieval of 12-17 lymph nodes. (Tapper et al., 2001; Nelson et al., 2001) Kim et al. mentioned that retrieval of >10 lymph nodes offered almost certain identification of metastasis to lymph nodes, and tumor differentiation and T stage seemed to correlate with higher nodal metastasis rate. (Kim et al., 2006)

The 7th ed. TNM classification mentions that histological examination of a regional lymphadenectomy specimen will ordinarily include 12 or more lymph nodes. The most accepted limit for accurate staging seems to be at least 12 nodes, as also suggested by other current node metastasis related publications (Wittekend et al., 2003; Greene et al., 2002). Cserni suggested not only the minimum number of LNs should be considered in terms of staging but some qualitative features may also influence the accurate staging. The question arises whether accurate staging can be reached with fewer than 12 LNs or not (Cserni et al.,1999). However, the number of lymph nodes obtained in specimens of colorectal cancer is significantly associated with the length of resected bowel, patient age and tumor location (Shen et al., 2009). Recent study by Cserni et al have mentioned that nodal status of CRCs may be adequately assessed by examining the lymph nodes from the close fraction around the tumor and the 3 cm side long bowel segment in both directions (Cserni et al., 2011). Thus, the retrievable number of lymph nodes depends on different factors like stage, technique of dissection and the treatment of specimen.

The Prognostic Significance of Number of Lymph Node Metastasis in Colon Cancer –

*Oncol* , Vol. 28, No. 2,(Jan 2010), pp. 264-271, ISSN19949014.

ISSN0021-9746.

ISSN 19016817

In Japanese.

2006), pp. 902-905, ISSN 17058731

2082, ISSN 8156513

968-978, ISSN 19092341

2007), pp. 827-831, ISSN 18005779

(7th ed.) Wiley-Liss, New York

York: Springer, 2002)

Based on Japanese Techniques of Resection and Handling of Resected Specimens 519

Ga´bor Cserni, Rita Bori, Istva´n Sejben, Limited lymph-node recovery based on lymph-

Greene FL, Page DL, Morrow M, et al, eds. AJCC Cancer Staging Manual. 6th edn. New

Gunderson, LL.; Jessup, JM.; Sargent, DJ.; Greene, FL. & Stewart, AK. (2010). Revised TN

Hohenberger, W.; Weber, K.; Matzel, K.; Papadopoulos, T. & Merkel, S. (2009). Standardized

Japanese Society for cancer of colon and rectum. (2009). *Japanese classification of colorectal* 

Japanese Society for cancer of the colon and rectum. (1977). *General rules for clinical and* 

Japanese Society for cancer of colon and rectum*.* (2009)*. General Rules for Clinical and* 

Japanese Society for Cancer of the Colon and Rectum. JSCCR Guidelines 2010 for the

Kim, J.; Huynh, R.; Abraham, I.; Kim, E. & Kumar, RR. (2006). Number of lymph nodes

Nelson, H.; Petrelli, N.; Carlin, A.; Couture, J.; Fleshman, J.; Guillem, J.; Miedema, B.; Ota, D.

Newland, RC.; Dent, OF.; Chapuis, PH. & Bokey, EL. (1994). Pathologic determinants of

Rosenberg, R.; Friederichs, J.; Schuster, T.; Gertler, R.; Maak, M.; Becker. K.; Grebner, A.;

Schumacher, P.; Dineen, S.; Barnett, C. Jr; Fleming, J. & Anthony, T. (2007). The metastatic

Shen, SS.; Haupt, BX.; Ro, JY.; Bailey, HR. & Schwartz, MR. (2009). Number of lymph nodes

*Pathol Lab Med*, Vol. 133, No.5, (May 2009), pp. 781-786, ISSN 19415953 Sobin, L.; Gospodarowicz, M. & Wittekond, C. (2009). *TNM classification of malignant tumours*

Kanehara & Co., Ltd., Tokyo. In Japanese. ISBN9784307202541

Vol. 93, No.8, (Apr 2001), pp. 583-596, ISSN 11309435

Treatment of Colorectal Cancer. (In Japanese). ISBN 9784307202794

node localisation is sufficient for accurate staging. *J Clin Pathol* 2011;64:13-15.

Categorization for colon cancer based on National Survival Outcomes data. *J Clin* 

surgery for colonic cancer: complete mesocolic excision and central ligation – technical notes and outcome. *Colorectal Disease*, Vol. 11, (May 2009), pp. 354–365,

*carcinoma* (2nd English ed.), Kanehara & Co. Ltd.,Tokyo. ISBN9784307202442 C3047

*pathological studies on cancer of the colon, rectum and anus*. (1st ed), Kanehara, Tokyo.

*Pathological Studies on Cancer of the Colon, Rectum and Anus* (7th ed., revised version),

examined and its impact on colorectal cancer staging. *Am Surg*, Vol. 72, No. 10, (Oct

& Sargent, D. (2001). Guidelines 2000 for colon and rectal surgery. *J Natl Cancer Inst*,

survival associated with colorectal cancer with lymph node metastases: a multivariate analysis of 579 patients. *Cancer*, Vol. 73, No. 8, (Apr 1994), pp. 2076–

Ulm, K.; Höfler, H.; Nekarda, H. & Siewert, JR. (2008). Prognosis of patients with colorectal cancer is associated with lymph node ratio: a single-center analysis of 3,026 patients over a 25-year time period. *Ann Surg*, Vol.248, No. 6, (Dec 2006), pp.

lymph node ratio predicts survival in colon cancer. *Am J Surg*, Vol. 194, No.6, (Dec

examined and associated clinicopathologic factors in colorectal carcinoma. *Arch* 

In recent years, it has been reported that lymph node ratio (LNR) i.e., the number of tumor infiltrated nodes divided by the total number of resected nodes, is associated with prognosis (Schumacher et al., 2007; Vaccaro et al., 2009). LNR is a more accurate prognostic parameter than just the presence of lymph nodes metastasis (Rosenberg et al., 2008). However, this idea needs more verification as LNR changes according to the extracted lymph nodes number. The size of the LNs is one such possible qualifier in the study about diagnosis for lymph node metastasis. Cserni et al found that the evaluation of the seven largest LNs gives a correct qualitative (negative vs positive) nodal status in 97% of the cases. (Cserni G, 2002). Thus, the diagnosis and staging the degree of lymph node metastasis seems to be still controversial.

#### **5. Conclusion**

LNM staging reflects the prognosis of the disease. The method of evaluation of LNM varies according to surgical treatment, the handling of specimens, pre and post operative chemoradiotherapy protocols etc. which varies between different institutions and countries. Therefore, it is difficult to compare every data in detail. Since, staging systems are based on depth of invasion and lymph node metastasis which influences the management and prognosis of the disease thus a standard protocol reflecting the best method for dissection of nodes and handling of specimens is necessary which reflects the accurate stage and thus the prognosis of the disease. Further studies for lymph node staging are thus necessary to find a universally accepted technique and staging system with maximum validity and reliability.

#### **6. References**


In recent years, it has been reported that lymph node ratio (LNR) i.e., the number of tumor infiltrated nodes divided by the total number of resected nodes, is associated with prognosis (Schumacher et al., 2007; Vaccaro et al., 2009). LNR is a more accurate prognostic parameter than just the presence of lymph nodes metastasis (Rosenberg et al., 2008). However, this idea needs more verification as LNR changes according to the extracted lymph nodes number. The size of the LNs is one such possible qualifier in the study about diagnosis for lymph node metastasis. Cserni et al found that the evaluation of the seven largest LNs gives a correct qualitative (negative vs positive) nodal status in 97% of the cases. (Cserni G, 2002). Thus, the diagnosis and staging the degree of lymph node metastasis seems to be still controversial.

LNM staging reflects the prognosis of the disease. The method of evaluation of LNM varies according to surgical treatment, the handling of specimens, pre and post operative chemoradiotherapy protocols etc. which varies between different institutions and countries. Therefore, it is difficult to compare every data in detail. Since, staging systems are based on depth of invasion and lymph node metastasis which influences the management and prognosis of the disease thus a standard protocol reflecting the best method for dissection of nodes and handling of specimens is necessary which reflects the accurate stage and thus the prognosis of the disease. Further studies for lymph node staging are thus necessary to find a universally accepted technique and staging system with maximum validity and reliability.

Akagi, Y.; Fukushima, T.; Mizobe, T.; Shiratsuchi, I.; Ryu, Y.; Yoshida, T.; Ishibashi, N.;

Andre´ Thierry, Corrado Boni, Matilde Navarro, Josep Tabernero, Tamas Hickish, Clare

Chapuis, PH.; Dent, OF.; Fisher, R.; Newland, RC.; Pheils, MT,; Smyth, E. & Colquhoun, K.

Choen, AM.; Tremiterra, S.; Candela, F.; Thaler, HT.; & Sigurdson, ER. (1991). Prognosis of

Cserni G, Vajda K, Tarja´n M, et al. Nodal staging of colorectal carcinomas from quantitative

Cserni G. The influence of nodal size on the staging of colorectal carcinomas. *J Clin Pathol* 

Kinugasa, T. & Shirouzu, K. (2010). Challenges in staging systems of colorectal cancer: clinical significance of metastatic lymph node number in colorectal cancer and mesorectal extension in rectal cancer. *Digestion*, Vol. 82, No. 3, (Jun 2010), pp.

Topham, Andrea Bonetti, Philip Clingan, John Bridgewater, Fernando Rivera, and Aimery de Gramont (2009). Improved Overall Survival With Oxaliplatin, Fluorouracil, and Leucovorin As Adjuvant Treatment in Stage II or III Colon Cancer in the MOSAIC Trial *J Clin Oncol*, vol.27,No. 19(July 2009), pp. 3109-

(1985). A multivariate analysis of clinical and pathological variables in prognosis after resection of large bowel cancer. *Br J Surg*, Vol. 72, No. 9, (Sep 1985), pp. 698-

node-positive colon cancer. *Cancer*, Vol. 67, No. 7, (Apr 1991), pp. 1859-1861, ISSN

and qualitative aspects. Can lymphatic mapping help staging? Pathol Oncol Res

**5. Conclusion** 

**6. References** 

192-197, ISSN 20588033

3116.ISSN0732183X

702, ISSN 4041728

1999;5:291-6.ISSN12194956

2002;55:386-90. ISSN0021-9746.

2004298


**27** 

*Italy* 

**Minimally Invasive Robot –** 

Annibale D'Annibale, Graziano Pernazza,

Vito Pende and Igor Monsellato *San Giovanni – Addolorata Hospital, Rome,* 

**Assisted Colorectal Resections** 

Minimally invasive techniques have revolutionized general surgery, especially in the field of

Many authors argue that the era of laparoscopic technique had begun in 1987, when Mouret performed the first laparoscopic cholecystectomy (Koopmann et al., 2008; Law et al., 2007). Since that point, laparoscopic technique has become the first choice for a multitude of surgical procedures: cholecystectomy, gastric bypass, fundoplication and its variants are some examples of procedures which are currently performed laparoscopically (Stage et al., 1997; Lacy et al., 2002). This spread has been fostered by the advantages of laparoscopic technique: reduced postoperative pain, decreased hospital stay and faster postoperative recovery, reduced incidence of postoperative complications, improved cosmetic outcome, decreased incidence of incisional hernias (Stage et al., 1997; Lacy et al., 2002; Guillou et al., 2005; Jayne et al., 2007; Fleshman et al., 2007; Nelson et al., 2004; Veldkamp et al., 2005;

The first hemicolectomy was performed laparoscopically in 1990 (Weber et al., 2002; Delaney et al., 2003). Since then, the introduction of this technique for colorectal disease was gradual, especially in malignancy, because of early skepticism towards this technique. The major questions have arisen about the treatment of malignant disease. The oncological adequacy was analysed in terms of lymph node dissection, resection margins and

Since 2002 a series of randomized clinical trials compared the laparoscopic and the open technique, the results of which definitely eliminated any doubts concerning the oncological adequacy of laparoscopic technique (Lacy et al., 2002; Jayne et al., 2007; Poulin et al., 1999; Hasegawa et al., 2003; Kaiser et al., 2004; Milsom et al., Tang et al., 2001; Champault et al., 2002). The advantages of minimally invasive approach in colorectal cancer surgery have been demonstrated in both pathophysiological (decreased inflammatory response → decreased immune response → decreased postoperative morbidity / comorbidity) and oncological aspects (similar results in terms of survival and outcome) (Leung et al., 2000; Delgado et al., 2001; Hu et al., 2003; Poulin et al., 1999; Hasegawa et al., 2003; Kaiser et al.,

The laparoscopic approach for colorectal disease, however, has both technical and "anatomical" disadvantages: the need of a long learning curve, the presence of a large

2004; Milsom et al., 1998; Tang et al., 2001; Champault et al., 2002).

**1. Introduction** 

gastrointestinal surgery.

Ballantyne et al., 2001).

intraoperative tumor dissemination.


### **Minimally Invasive Robot – Assisted Colorectal Resections**

Annibale D'Annibale, Graziano Pernazza, Vito Pende and Igor Monsellato *San Giovanni – Addolorata Hospital, Rome, Italy* 

#### **1. Introduction**

520 Colorectal Cancer – From Prevention to Patient Care

Sobin, L. & Wittekond, C. (2002). *TNM classification of malignant tumours* (6th ed.), Wiley-

Tepper, JE.; O'Connel, MJ.; Niedzwiecki, D.; Hollis, D.; Compton, C.; Benson, AB. 3rd;

Vaccaro, CA.; Bonadeo, FA.; Benati, ML.; Quintana, GM.; Rubinstein, F.; Mullen, E.; Telenta,

Wittekind Ch, Greene FL, Henson DE, et al, eds. TNM Supplement dA Commentary on

Uniform Use. 3rd edn. New York: John Wiley and Sons, 2003.

Vol. 19, No. 1, (Jan 2001), pp. 157-163, ISSN 11134208

Cummings, B.; Gunderson, L.; Macdonald, JS. & Mayer, RJ. (2001). Impact of number of nodes retrieved on outcome in patients with rectal cancer. *J Clin Oncol*,

M. & Lastiri, JM. (2004). Colorectal cancer staging: reappraisal of N/PN classification. *Dis Colon Rectum,* Vol. 47, No. 1, ( Jan 2004), pp.66-69, ISSN 14719153 Vaccaro, CA.; Im, V.; Rossi, GL.; Quintana, GO.; Benati, ML.; Perez de Arenaza, D. &

Bonadeo, FA. (2009). Lymph node ratio as prognosis factor for colon cancer treated by colorectal surgeons. *Dis Colon Rectum,* Vol.52, No.7, (Jul 2009), pp. 1244-1250,

Liss, New York

ISSN 19571700

Minimally invasive techniques have revolutionized general surgery, especially in the field of gastrointestinal surgery.

Many authors argue that the era of laparoscopic technique had begun in 1987, when Mouret performed the first laparoscopic cholecystectomy (Koopmann et al., 2008; Law et al., 2007). Since that point, laparoscopic technique has become the first choice for a multitude of surgical procedures: cholecystectomy, gastric bypass, fundoplication and its variants are some examples of procedures which are currently performed laparoscopically (Stage et al., 1997; Lacy et al., 2002). This spread has been fostered by the advantages of laparoscopic technique: reduced postoperative pain, decreased hospital stay and faster postoperative recovery, reduced incidence of postoperative complications, improved cosmetic outcome, decreased incidence of incisional hernias (Stage et al., 1997; Lacy et al., 2002; Guillou et al., 2005; Jayne et al., 2007; Fleshman et al., 2007; Nelson et al., 2004; Veldkamp et al., 2005; Ballantyne et al., 2001).

The first hemicolectomy was performed laparoscopically in 1990 (Weber et al., 2002; Delaney et al., 2003). Since then, the introduction of this technique for colorectal disease was gradual, especially in malignancy, because of early skepticism towards this technique. The major questions have arisen about the treatment of malignant disease. The oncological adequacy was analysed in terms of lymph node dissection, resection margins and intraoperative tumor dissemination.

Since 2002 a series of randomized clinical trials compared the laparoscopic and the open technique, the results of which definitely eliminated any doubts concerning the oncological adequacy of laparoscopic technique (Lacy et al., 2002; Jayne et al., 2007; Poulin et al., 1999; Hasegawa et al., 2003; Kaiser et al., 2004; Milsom et al., Tang et al., 2001; Champault et al., 2002). The advantages of minimally invasive approach in colorectal cancer surgery have been demonstrated in both pathophysiological (decreased inflammatory response → decreased immune response → decreased postoperative morbidity / comorbidity) and oncological aspects (similar results in terms of survival and outcome) (Leung et al., 2000; Delgado et al., 2001; Hu et al., 2003; Poulin et al., 1999; Hasegawa et al., 2003; Kaiser et al., 2004; Milsom et al., 1998; Tang et al., 2001; Champault et al., 2002).

The laparoscopic approach for colorectal disease, however, has both technical and "anatomical" disadvantages: the need of a long learning curve, the presence of a large

Minimally Invasive Robot – Assisted Colorectal Resections 523

matched. Adhesions, localization and peritoneal carcinomatosis are not considered a

The patient undergoes bowel preparation with Polyethylene glycol one day before the procedure and a slag-free diet three days before the procedure. The banding technique of the inferior arms associated to administration of a low-molecular-weight heparin 12 hours

Before procedure starts, a venous access is needed to maintain correct blood volume and hydration of the patient and to infuse anesthetic agents during the whole procedure. The preferred choices of venous access are: right superior arm (18G catheter) for right colectomy and left superior arm for left colectomy and rectal resection (16/18 gauge catheter). If peripheral venous access is unavailable, a central catheter should be inserted before the day of the procedure. A heating system for venous access is needed in order to maintain the thermal homeostasis. Keeping patients warm has been associated with a threefold decrease in the rate of wound infection, reduction in operative blood loss, decrease in untoward cardiac events and patient discomfort. Maintenance of a normal temperature during surgery

General anesthesia with orotracheal intubation and mechanical dynamic ventilation is the preferred technique among anesthesiologists. Currently, both intravenous and gasintravenous (Desforane, Sevorane) techniques are used by anesthesiologists as there is no evidence of better results by the one or the other procedure. The usual scheme provides an association with epidural access in order to better control postoperative pain. N2O (nitrous oxide) is contraindicated as it may increase the risk of pulmonary embolism. An optimal anesthesia may provide an acceptable muscular relaxation, better ventilation, thus minimizing the risk of pulmonary embolism and hemodynamics alteration, granting a

ECG+HR, O2 saturation, Airways Pressure (PAW – Pressure plateau), capnography (EtCO2)

Neuromuscular function (TOF), Diuresis monitoring (bladder catheterization), Swan-Ganz

A peripheral arterial access is mandatory for monitoring blood pressure in high-risk

catheter, PeakPressure Monitoring, arterial catheterization, haemo-gas analysis.

is important to reduce the stress of the surgical procedure and organ dysfunction.

wider operating field thus reducing the operative time.

Intraoperative monitoring differs between ASA I-III or III+ patients:

before the procedure is performed for thromboembolism prophylaxis.

contraindication by the majority of experts.

**3. Patient preparation** 

**3.1 Venous access** 

**3.2 Types of anesthesia** 

**3.3 Intraoperative monitoring** 

**3.4 Surgical instrumentation** 

In ASA III+ in addition to previous tests:

Stereoscopic endoscope (da Vinci Surgical System)

In ASA I-III:

patients.

CO2 insufflator

surgical field that requires a dynamic view and consequently, a skilled camera assistant surgeon, a constant dialogue between the operator and the assistant, the loss of threedimensional vision, reduced ergonomics during specific phases of the procedure (need to take preternatural positions), poor dexterity of the laparoscopic instruments and decreased range of motion due to the rigidity of the insertion of the trocarts site, amplification of physiological tremor and the fulcrum effect. The so-called "surgical robots" or otherwise called "computer-assisted telemanipulators" have been introduced to overcome these "limits" in the practice of surgery (Mirnezami et al., 2009; Weber et al., 2002; Cadiere et al., 2001; Garca-Ruiz et al., 1998; Lanfranco et al., 2004; Morino et al., 2006; Horgan et al., 2001; Ballantyne et al., 2002; Hanly et al., 2004; Moorthy et al., 2005; Baik et al., 2008; Ballantyne et al., 2001).

#### **2. Criteria in selecting patient for minimally invasive surgery**

Patient's selection is based on pathophysiological and pathological conditions. Absolute contraindications in colorectal cancer robotic surgery, reflect those for whole minimally invasive surgery. Nevertheless, indications to minimally invasive approach are expanding rapidly suggesting the establishment of international guidelines in patient's selection.

There are several physiological parameters, which have been analyzed for patient's selection:


#### **Pathological criteria:**

Characteristics of tumor may influence the surgical approach. A tumor invading adjacent structures (T4) is considered an "absolute" contraindication to minimally invasive approach, as the principles of atraumatic manipulation and wide resection margins could be not matched. Adhesions, localization and peritoneal carcinomatosis are not considered a contraindication by the majority of experts.

#### **3. Patient preparation**

522 Colorectal Cancer – From Prevention to Patient Care

surgical field that requires a dynamic view and consequently, a skilled camera assistant surgeon, a constant dialogue between the operator and the assistant, the loss of threedimensional vision, reduced ergonomics during specific phases of the procedure (need to take preternatural positions), poor dexterity of the laparoscopic instruments and decreased range of motion due to the rigidity of the insertion of the trocarts site, amplification of physiological tremor and the fulcrum effect. The so-called "surgical robots" or otherwise called "computer-assisted telemanipulators" have been introduced to overcome these "limits" in the practice of surgery (Mirnezami et al., 2009; Weber et al., 2002; Cadiere et al., 2001; Garca-Ruiz et al., 1998; Lanfranco et al., 2004; Morino et al., 2006; Horgan et al., 2001; Ballantyne et al., 2002; Hanly et al., 2004; Moorthy et al., 2005; Baik et al., 2008; Ballantyne et

Patient's selection is based on pathophysiological and pathological conditions. Absolute contraindications in colorectal cancer robotic surgery, reflect those for whole minimally invasive surgery. Nevertheless, indications to minimally invasive approach are expanding rapidly suggesting the establishment of international guidelines in patient's selection. There are several physiological parameters, which have been analyzed for patient's




Characteristics of tumor may influence the surgical approach. A tumor invading adjacent structures (T4) is considered an "absolute" contraindication to minimally invasive approach, as the principles of atraumatic manipulation and wide resection margins could be not

considered (younger, middle-aged, elderly). Age is not a contraindication.

**2. Criteria in selecting patient for minimally invasive surgery** 

al., 2001).

selection:

perfusion mismatch.

after open surgery.

**Pathological criteria:** 

The patient undergoes bowel preparation with Polyethylene glycol one day before the procedure and a slag-free diet three days before the procedure. The banding technique of the inferior arms associated to administration of a low-molecular-weight heparin 12 hours before the procedure is performed for thromboembolism prophylaxis.

#### **3.1 Venous access**

Before procedure starts, a venous access is needed to maintain correct blood volume and hydration of the patient and to infuse anesthetic agents during the whole procedure. The preferred choices of venous access are: right superior arm (18G catheter) for right colectomy and left superior arm for left colectomy and rectal resection (16/18 gauge catheter). If peripheral venous access is unavailable, a central catheter should be inserted before the day of the procedure. A heating system for venous access is needed in order to maintain the thermal homeostasis. Keeping patients warm has been associated with a threefold decrease in the rate of wound infection, reduction in operative blood loss, decrease in untoward cardiac events and patient discomfort. Maintenance of a normal temperature during surgery is important to reduce the stress of the surgical procedure and organ dysfunction.

#### **3.2 Types of anesthesia**

General anesthesia with orotracheal intubation and mechanical dynamic ventilation is the preferred technique among anesthesiologists. Currently, both intravenous and gasintravenous (Desforane, Sevorane) techniques are used by anesthesiologists as there is no evidence of better results by the one or the other procedure. The usual scheme provides an association with epidural access in order to better control postoperative pain. N2O (nitrous oxide) is contraindicated as it may increase the risk of pulmonary embolism. An optimal anesthesia may provide an acceptable muscular relaxation, better ventilation, thus minimizing the risk of pulmonary embolism and hemodynamics alteration, granting a wider operating field thus reducing the operative time.

#### **3.3 Intraoperative monitoring**

Intraoperative monitoring differs between ASA I-III or III+ patients: In ASA I-III:

ECG+HR, O2 saturation, Airways Pressure (PAW – Pressure plateau), capnography (EtCO2) In ASA III+ in addition to previous tests:

Neuromuscular function (TOF), Diuresis monitoring (bladder catheterization), Swan-Ganz catheter, PeakPressure Monitoring, arterial catheterization, haemo-gas analysis.

A peripheral arterial access is mandatory for monitoring blood pressure in high-risk patients.

#### **3.4 Surgical instrumentation**

Stereoscopic endoscope (da Vinci Surgical System) CO2 insufflator

Minimally Invasive Robot – Assisted Colorectal Resections 525

hypochondrium for electrocautery/ultrasonic instruments and in the left iliac fossa for Cadiére grasper under direct vision. An additional 12-mm port is inserted in the left flank. This port is used by the assistant to help the surgeon during some steps of the procedure and to introduce the linear stapler for vascular, transverse colon and ileum resection. An additional robotic port is inserted in the right iliac fossa for the fourth robotic arm. It may be useful to provide effective and stable retraction during several steps of the procedure ( i.e. to grasp the ileocecal valve and place the ileocolic vascular pedicle under tension, to lift up the

The procedure is carried out with a full robotic technique. The robotic cart approaches from

The procedure begins by grasping upward and laterally the mesentery of the last ileal loop. This maneuver, performed with the forceps mounted on the fourth arm, enhances the prominence of the ileocolic vessels and provides stable and durable retraction. The peritoneal layer of the mesentery is incised just below this salience, and an accurate lymphadenectomy is performed along the superior mesenteric axis. Then, the ileocolic

Dissection of the right mesocolon follows a caudal-cranial pathway, along the right side of the superior mesenteric axis. Following this path, it is possible to remove the lymphatic tissue completely, safely identifying the inconstant right colic vessels, which may be sectioned at their origin, until reaching the root of the transverse mesocolon. Dissection along the lateral margin of the middle colic vessels allows the right branch of the middle colic vessels to be reached more easily, which is then treated as in the standard right colectomy (R1). Resection of the whole pedicle of the middle colic vessels is performed only for localization at the right colic flexure, for which extended right colectomy (R2) is needed.

the right side of the patient, and the three operative arms are connected to the ports.

hepatic flexure during the dissection of transverse mesocolon, etc.).

vessels are isolated and separately ligated and sectioned (fig. 2).

**4.3 Description of the procedure** 

Fig. 2. Intraoperative view: Ileocolic vessels


#### **4. Robotic right colectomy**

#### **4.1 Patient position and operating room setup**

Patient is placed in supine reverse Trendelenburg position (15° to 20°), with 10° to 15° left lateral rotation and shoulder supports. The legs are secured at the thigh and calf with straps. The table is tilted to the left to allow the small intestine to fall off from the midline. The assistant surgeon stands on the patient's left side. The robotic cart is approached from the patient's right side. The operating room scheme is shown in fig. 1.

Fig. 1. Operating room setup and trocarts position

#### **4.2 Trocarts position**

A conventional 12mm port is placed by open technique on the lateral margin of the left rectal muscle, 1-2 cm above the transverse umbilical line, and pneumoperitoneum is induced until reaching a 12mmHg endoabdominal pressure. Then the 30° robotic stereo endoscope is inserted. Two daVinci 8-mm ports are inserted respectively in the left hypochondrium for electrocautery/ultrasonic instruments and in the left iliac fossa for Cadiére grasper under direct vision. An additional 12-mm port is inserted in the left flank. This port is used by the assistant to help the surgeon during some steps of the procedure and to introduce the linear stapler for vascular, transverse colon and ileum resection. An additional robotic port is inserted in the right iliac fossa for the fourth robotic arm. It may be useful to provide effective and stable retraction during several steps of the procedure ( i.e. to grasp the ileocecal valve and place the ileocolic vascular pedicle under tension, to lift up the hepatic flexure during the dissection of transverse mesocolon, etc.).

#### **4.3 Description of the procedure**

524 Colorectal Cancer – From Prevention to Patient Care

1 45-mm laparoscopic stapler device with 2-3 cartridges (intestinal and vascular type)

Patient is placed in supine reverse Trendelenburg position (15° to 20°), with 10° to 15° left lateral rotation and shoulder supports. The legs are secured at the thigh and calf with straps. The table is tilted to the left to allow the small intestine to fall off from the midline. The assistant surgeon stands on the patient's left side. The robotic cart is approached from the

A conventional 12mm port is placed by open technique on the lateral margin of the left rectal muscle, 1-2 cm above the transverse umbilical line, and pneumoperitoneum is induced until reaching a 12mmHg endoabdominal pressure. Then the 30° robotic stereo endoscope is inserted. Two daVinci 8-mm ports are inserted respectively in the left

Irrigation/suction system device

3 Laparoscopic forceps type Johann

1 Wound protector for specimen extraction

**4.1 Patient position and operating room setup** 

Fig. 1. Operating room setup and trocarts position

**4.2 Trocarts position** 

patient's right side. The operating room scheme is shown in fig. 1.

1 Laparoscopic clip applier

1 Robotic ultrasound device 1 Robotic electrocautery hook 1 Robotic bipolar forceps

**4. Robotic right colectomy** 

Video processor 1 Hasson-type trocart 1 10/12mm trocart 3 8mm robotic trocart 2 Robotic Cadiere's graspers 2 Robotic needle-holders 1 Laparoscopic needle-holder 1 Laparoscopic dissector

> The procedure is carried out with a full robotic technique. The robotic cart approaches from the right side of the patient, and the three operative arms are connected to the ports.

> The procedure begins by grasping upward and laterally the mesentery of the last ileal loop. This maneuver, performed with the forceps mounted on the fourth arm, enhances the prominence of the ileocolic vessels and provides stable and durable retraction. The peritoneal layer of the mesentery is incised just below this salience, and an accurate lymphadenectomy is performed along the superior mesenteric axis. Then, the ileocolic vessels are isolated and separately ligated and sectioned (fig. 2).

Fig. 2. Intraoperative view: Ileocolic vessels

Dissection of the right mesocolon follows a caudal-cranial pathway, along the right side of the superior mesenteric axis. Following this path, it is possible to remove the lymphatic tissue completely, safely identifying the inconstant right colic vessels, which may be sectioned at their origin, until reaching the root of the transverse mesocolon. Dissection along the lateral margin of the middle colic vessels allows the right branch of the middle colic vessels to be reached more easily, which is then treated as in the standard right colectomy (R1). Resection of the whole pedicle of the middle colic vessels is performed only for localization at the right colic flexure, for which extended right colectomy (R2) is needed.

Minimally Invasive Robot – Assisted Colorectal Resections 527

assistant surgeon stands on the patient's right side. The robotic cart is approached from the

A conventional 12mm port is placed by open technique 2-cm right from the umbilicus along the umbilical transverse line, and pneumoperitoneum is induced until reaching a 12mmHg intrabdominal pressure. Then the 30° robotic stereo endoscope is inserted. Two da Vinci 8 mm ports are inserted under direct vision respectively in the epigastrium just 2 cm right from the midline and in the right flank. An additional 12-mm port is inserted between the two robotic trocarts. This port is used by the assistant to help the surgeon during some steps of the procedure and to introduce the linear stapler for vascular and left colon resection.

The procedure is carried out with a full robotic technique. The robotic cart approaches from

The dissection begins with the identification of the inferior mesenteric vein at the level of the inferior margin of the pancreas and the incision of the peritoneum at the origin of the mesocolon under the salience of the vein, on the right side. Then, the inferior mesenteric

A sharp dissection is performed in a cranial-caudal and medial-to-lateral direction between the anterior and the posterior layer of Toldt's fascia. An incision of the peritoneum is performed from the promontory up to the origin of the inferior mesenteric artery, identifying and preserving the preaortic nerves. The dissection is carried out up to join the previous plane of dissection identifying and preserving the left ureter and the gonadal vessels. Then, the inferior mesenteric artery is exposed and sectioned at its origin by linear

Afterwards, the colon is freed laterally by the incision along Monk's line, from the sigmoid colon upward: the splenic flexure is taken down if necessary. The colon is divided distally to the tumor at the level of the promontory. Then, the specimen is usually extracted through a mini-Pfannenstiel incision, after the insertion of a wound protector. Before closing the minilaparotomy, the anvil of the circular stapler is inserted at the distal margin of the

Re-induction of the pneumoperitoneum is performed. The proximal colon is joined to the rectum by circular stapler, and a mechanical termino-terminal colorectal anastomosis is

Two 10F Jackson-Pratt drainages are placed anteriorly and posteriorly to the anastomosis

Patient is placed in a lithotomy position with his legs apart and no modification of position will occur during the whole procedure. The legs are secured at the thigh and calf with straps. The table is tilted to the right (15° – 20°) to allow the small intestine to fall off from the midline. The assistant stands on the patient's right side. The robotic cart then approaches to the operative bed by patient's left side, with a 60 degrees angle, following the imaginary

line passing through the umbilicus and the left anterosuperior iliac spine.

the left side of the patient, and the two operative arms are connected to the ports.

vein is exposed and sectioned by linear stapler or between clips.

patient's right side.

**5.2 Trocarts position** 

**5.3 Description of the procedure** 

stapler or between clips.

fashioned by Knight & Griffen technique.

through one of the lower trocart access.

**6.1 Patient position and operating room setup** 

proximal colon.

**6. Rectal surgery** 

Mobilization of the colon is performed in a medial-to-lateral direction in the avascular plane between Gerota's and Toldt's fasciae. During this step, the knee of the duodenum constitutes an important landmark to drive the dissection upward, over the duodenal third portion and the pancreatic head, along Fredet's fascia. The right ureter and the gonadic vessels are left below the plane of dissection. The hepatic flexure is then mobilized, sectioning the lateral portion of the gastrocolic and the hepatocolic ligaments. This step enables the resection to join the previously dissected plane and to complete the lymphadenectomy around the gastroepiploic vessels.

The transverse colon and the last ileal loop are finally sectioned by linear stapler. The ileum and the transverse colon are joined with a running suture, and an intracorporeal isoperistaltic double layer side-to-side ileocolic anastomosis is fashioned using a 3-0 absorbable monofilament suture (fig.3).

We performed an extracorporeal anastomosis in the first five cases: the daVinci system is disengaged from the patient, then a median supraumbilical minilaparotomy is performed, through which an isoperistaltic side-to-side ileocolic anastomosis is fashioned.

The specimen is retrieved at the end of the procedure through a small muscle-splitting Pfannenstiel minilaparotomy. This incision is protected from potential contamination by a wound protector.

A 10F Jackson-Pratt drain is placed laterally to the anastomosis through one of the lower trocart access.

Fig. 3. Intraoperative view: intracorporeal ileocolic robot-assisted anastomosis

#### **5. Left colectomy**

#### **5.1 Patient position and operating room setup**

Patient is placed in supine Trendelenburg position (15° to 20°), with 10° to 15° right lateral rotation and shoulder supports. The legs are secured at the thigh and calf with straps. The table is tilted to the right to allow the small intestine to fall off from the midline. The assistant surgeon stands on the patient's right side. The robotic cart is approached from the patient's right side.

#### **5.2 Trocarts position**

526 Colorectal Cancer – From Prevention to Patient Care

Mobilization of the colon is performed in a medial-to-lateral direction in the avascular plane between Gerota's and Toldt's fasciae. During this step, the knee of the duodenum constitutes an important landmark to drive the dissection upward, over the duodenal third portion and the pancreatic head, along Fredet's fascia. The right ureter and the gonadic vessels are left below the plane of dissection. The hepatic flexure is then mobilized, sectioning the lateral portion of the gastrocolic and the hepatocolic ligaments. This step enables the resection to join the previously dissected plane and to complete the

The transverse colon and the last ileal loop are finally sectioned by linear stapler. The ileum and the transverse colon are joined with a running suture, and an intracorporeal isoperistaltic double layer side-to-side ileocolic anastomosis is fashioned using a 3-0

We performed an extracorporeal anastomosis in the first five cases: the daVinci system is disengaged from the patient, then a median supraumbilical minilaparotomy is performed,

The specimen is retrieved at the end of the procedure through a small muscle-splitting Pfannenstiel minilaparotomy. This incision is protected from potential contamination by a

A 10F Jackson-Pratt drain is placed laterally to the anastomosis through one of the lower

through which an isoperistaltic side-to-side ileocolic anastomosis is fashioned.

Fig. 3. Intraoperative view: intracorporeal ileocolic robot-assisted anastomosis

Patient is placed in supine Trendelenburg position (15° to 20°), with 10° to 15° right lateral rotation and shoulder supports. The legs are secured at the thigh and calf with straps. The table is tilted to the right to allow the small intestine to fall off from the midline. The

lymphadenectomy around the gastroepiploic vessels.

absorbable monofilament suture (fig.3).

wound protector.

**5. Left colectomy** 

**5.1 Patient position and operating room setup** 

trocart access.

A conventional 12mm port is placed by open technique 2-cm right from the umbilicus along the umbilical transverse line, and pneumoperitoneum is induced until reaching a 12mmHg intrabdominal pressure. Then the 30° robotic stereo endoscope is inserted. Two da Vinci 8 mm ports are inserted under direct vision respectively in the epigastrium just 2 cm right from the midline and in the right flank. An additional 12-mm port is inserted between the two robotic trocarts. This port is used by the assistant to help the surgeon during some steps of the procedure and to introduce the linear stapler for vascular and left colon resection.

#### **5.3 Description of the procedure**

The procedure is carried out with a full robotic technique. The robotic cart approaches from the left side of the patient, and the two operative arms are connected to the ports.

The dissection begins with the identification of the inferior mesenteric vein at the level of the inferior margin of the pancreas and the incision of the peritoneum at the origin of the mesocolon under the salience of the vein, on the right side. Then, the inferior mesenteric vein is exposed and sectioned by linear stapler or between clips.

A sharp dissection is performed in a cranial-caudal and medial-to-lateral direction between the anterior and the posterior layer of Toldt's fascia. An incision of the peritoneum is performed from the promontory up to the origin of the inferior mesenteric artery, identifying and preserving the preaortic nerves. The dissection is carried out up to join the previous plane of dissection identifying and preserving the left ureter and the gonadal vessels. Then, the inferior mesenteric artery is exposed and sectioned at its origin by linear stapler or between clips.

Afterwards, the colon is freed laterally by the incision along Monk's line, from the sigmoid colon upward: the splenic flexure is taken down if necessary. The colon is divided distally to the tumor at the level of the promontory. Then, the specimen is usually extracted through a mini-Pfannenstiel incision, after the insertion of a wound protector. Before closing the minilaparotomy, the anvil of the circular stapler is inserted at the distal margin of the proximal colon.

Re-induction of the pneumoperitoneum is performed. The proximal colon is joined to the rectum by circular stapler, and a mechanical termino-terminal colorectal anastomosis is fashioned by Knight & Griffen technique.

Two 10F Jackson-Pratt drainages are placed anteriorly and posteriorly to the anastomosis through one of the lower trocart access.

#### **6. Rectal surgery**

#### **6.1 Patient position and operating room setup**

Patient is placed in a lithotomy position with his legs apart and no modification of position will occur during the whole procedure. The legs are secured at the thigh and calf with straps. The table is tilted to the right (15° – 20°) to allow the small intestine to fall off from the midline. The assistant stands on the patient's right side. The robotic cart then approaches to the operative bed by patient's left side, with a 60 degrees angle, following the imaginary line passing through the umbilicus and the left anterosuperior iliac spine.

Minimally Invasive Robot – Assisted Colorectal Resections 529

The dissection thus continues distal to the rectoprostatic (or rectovaginal) septum: the anterior mesorectal excision is completed, and the rectum is exposed anteriorly to the anorectal junction. A gentle traction of the rectum takes place laterally by a Cadiere on the arm #1, on either side, and the dissection is performed by the electrocautery hook until reaching the lateral ligaments of the rectum (LLR). At this level, the dissection is carried out close to the rectum, in order to avoid the injury of the inferior hypogastric plexus (IHP). The dissection must include only the rectal branches from the IHP and the small rectal branches

The total mesorectal excision (TME) is finally completed and the rectum is sectioned at its

Fig. 5. Intraoperative view: Incision of Denonvillier's Fascia.

Fig. 6. Intraoperative view: lateral dissection of the rectum

from the middle rectal artery (MRA) (fig. 6).

distal end by linear stapler.

#### **6.2 Trocarts position**

We usually perform a 5-ports technique. The first 12-mm periumbilical port (C) is placed by "open-laparoscopy" technique, for the stereoscopic endoscope. Then a 12-mmHg pneumoperitoneum is gained and other three 8-mm robotic ports are added: the first port (R1) is inserted in the right iliac fossa on an imaginary line between the anterosuperior iliac spine and the umbilicus. The second port (R2) is inserted in the right hypochondrium, and the third port (R3) is placed in patient's left flank. Last, a second 12-mm port is inserted in patient's right hip between R1 and R2 for the assistant surgeon (A). Trocarts' position is performed laparoscopically, and an exploration of the abdominal cavity precedes the robotic technique, with an adhesiolysis in case of visceral adhesions. Hence, the small bowel is displaced right in order to expose the Treitz ligament and the plane of the inferior mesenteric vein (IMV).

#### **6.3 Description of the procedure**

The procedure starts with the "vascular phase": the inferior mesenteric vein is identified at the level of the inferior margin of the pancreas. The peritoneum under the inferior mesenteric vein is then incised, and a smooth dissection is performed in a medial-to-lateral direction along the avascular plane between the Toldt's, above, and the Gerota's fascia, below, up to the left abdominal wall. The inferior mesenteric vein is subsequently divided between clips or by linear stapler. The dissection is prolonged up downward. The left ureter and the gonadic vessels are previously identified and preserved. The incision of the peritoneum at the level of the mesosigmoid is performed in order to reach the deeper plane of the inferior mesenteric artery. The iliac vessels and the left ureter are covered by the prerenal fascia. The dissection of the plane covering the inferior mesenteric artery is then performed with an accurate regional lymphadenectomy, preserving the preaortic nerves and the superior hypogastric plexus. The inferior mesenteric artery is then sectioned by stapler or between clips. A complete lateral dissection of the colon is carried out from the sigmoid to the splenic flexure. Flexure takedown is performed if the descending colon is needed to be used for the anastomosis.

Then, arm #2 is switched from R2 to R3. The assistant surgeon could use R2 and A trocarts in helping the surgeon during the total mesorectal excision (TME). Dissection starts posteriorly, at the level of the promontory, along the plane between the fascia recti propria, anteriorly, (peritoneum) and the presacral fascia (Waldeyer's fascia), posteriorly. Rectum is lifted up and laterally by a Cadiere on the arm #2 and dissection is carried out by the electrocautery hook or ultrasound device on the arm #1: the retrorectal plane has been opened. Care should be taken to preserve the inferior hypogastric nerves lying laterally along this plane. Pneumoperitoneum also helps the dissection between these two layers. At the level of the fourth sacral vertebra, the rectosacral fascia is incised in order to better mobilize the rectum and to access to the inferior part of the retrorectal space, in case of lower tumor localization. The mesorectal dissection has been completed behind the tip of the coccyx as the pelvic floor curves upward anteriorly toward the anorectal junction. Anteriorly, the rectum is retracted cranially and posteriorly by a Cadiere on the arm #2. The anterior peritoneal brim is incised by the electrocautery hook on the arm #1, and the dissection continues along the plane between the Denonvillier's fascia (or the rectovaginal fascia) and the fascia propria recti. At the level of the base of the prostate, Denonvillier's fascia is sectioned in order to preserve rectoprostatic (or rectovaginal) blood vessels and branches of the cavernous nerves (fig. 5).

We usually perform a 5-ports technique. The first 12-mm periumbilical port (C) is placed by "open-laparoscopy" technique, for the stereoscopic endoscope. Then a 12-mmHg pneumoperitoneum is gained and other three 8-mm robotic ports are added: the first port (R1) is inserted in the right iliac fossa on an imaginary line between the anterosuperior iliac spine and the umbilicus. The second port (R2) is inserted in the right hypochondrium, and the third port (R3) is placed in patient's left flank. Last, a second 12-mm port is inserted in patient's right hip between R1 and R2 for the assistant surgeon (A). Trocarts' position is performed laparoscopically, and an exploration of the abdominal cavity precedes the robotic technique, with an adhesiolysis in case of visceral adhesions. Hence, the small bowel is displaced right in order to expose the Treitz ligament and the plane of the inferior

The procedure starts with the "vascular phase": the inferior mesenteric vein is identified at the level of the inferior margin of the pancreas. The peritoneum under the inferior mesenteric vein is then incised, and a smooth dissection is performed in a medial-to-lateral direction along the avascular plane between the Toldt's, above, and the Gerota's fascia, below, up to the left abdominal wall. The inferior mesenteric vein is subsequently divided between clips or by linear stapler. The dissection is prolonged up downward. The left ureter and the gonadic vessels are previously identified and preserved. The incision of the peritoneum at the level of the mesosigmoid is performed in order to reach the deeper plane of the inferior mesenteric artery. The iliac vessels and the left ureter are covered by the prerenal fascia. The dissection of the plane covering the inferior mesenteric artery is then performed with an accurate regional lymphadenectomy, preserving the preaortic nerves and the superior hypogastric plexus. The inferior mesenteric artery is then sectioned by stapler or between clips. A complete lateral dissection of the colon is carried out from the sigmoid to the splenic flexure. Flexure takedown is performed if the descending colon is

Then, arm #2 is switched from R2 to R3. The assistant surgeon could use R2 and A trocarts in helping the surgeon during the total mesorectal excision (TME). Dissection starts posteriorly, at the level of the promontory, along the plane between the fascia recti propria, anteriorly, (peritoneum) and the presacral fascia (Waldeyer's fascia), posteriorly. Rectum is lifted up and laterally by a Cadiere on the arm #2 and dissection is carried out by the electrocautery hook or ultrasound device on the arm #1: the retrorectal plane has been opened. Care should be taken to preserve the inferior hypogastric nerves lying laterally along this plane. Pneumoperitoneum also helps the dissection between these two layers. At the level of the fourth sacral vertebra, the rectosacral fascia is incised in order to better mobilize the rectum and to access to the inferior part of the retrorectal space, in case of lower tumor localization. The mesorectal dissection has been completed behind the tip of the coccyx as the pelvic floor curves upward anteriorly toward the anorectal junction. Anteriorly, the rectum is retracted cranially and posteriorly by a Cadiere on the arm #2. The anterior peritoneal brim is incised by the electrocautery hook on the arm #1, and the dissection continues along the plane between the Denonvillier's fascia (or the rectovaginal fascia) and the fascia propria recti. At the level of the base of the prostate, Denonvillier's fascia is sectioned in order to preserve rectoprostatic (or rectovaginal) blood vessels and

**6.2 Trocarts position** 

mesenteric vein (IMV).

**6.3 Description of the procedure** 

needed to be used for the anastomosis.

branches of the cavernous nerves (fig. 5).

Fig. 5. Intraoperative view: Incision of Denonvillier's Fascia.

The dissection thus continues distal to the rectoprostatic (or rectovaginal) septum: the anterior mesorectal excision is completed, and the rectum is exposed anteriorly to the anorectal junction. A gentle traction of the rectum takes place laterally by a Cadiere on the arm #1, on either side, and the dissection is performed by the electrocautery hook until reaching the lateral ligaments of the rectum (LLR). At this level, the dissection is carried out close to the rectum, in order to avoid the injury of the inferior hypogastric plexus (IHP). The dissection must include only the rectal branches from the IHP and the small rectal branches from the middle rectal artery (MRA) (fig. 6).

The total mesorectal excision (TME) is finally completed and the rectum is sectioned at its distal end by linear stapler.

Fig. 6. Intraoperative view: lateral dissection of the rectum

Minimally Invasive Robot – Assisted Colorectal Resections 531

Twenty-three male and twenty female patients underwent surgery for left-sided colon cancer. Mean age was 60 ± 12 years. Median operative time was 220 (215 – 230) min. One conversion occurred for a splenic injury at the beginning of the experience. Specimen length was 24 ± 7.7 (range 10 – 49) cm. Number of harvested lymph nodes was 12 ± 7 (6 – 24). Surgery-related morbidity was 1/43 (2,3%): an anastomotic dehiscence in one of the first cases. All patients were included in a follow-up regimen. No 30-day mortality occurred. Median hospital stay was 7 d (range, 6 – 11), oral diet resumption was 3 d (range, 2 – 10). All patients were treated with curative-intent surgery and adjuvant chemotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months, disease-free survival was 88.3% (38/43), overall survival was 90% (38/43), and disease-related mortality was 7% (3/43) (Table 1). One

Thirty male and eighteen female patients underwent rectal surgery: 45 rectal anterior resections (RAR) with TME and 3 abdominalperineal resections (APA). Mean age was 67 ± 12 years. Median operative time was 270 (240 – 315) min. Specimen length was 23 (19 – 27,5) cm. Number of harvested lymph nodes was 15 (12 – 20). Circumferential margins are shown

> Cm Upper rectum Middle rectum Lower rectum < 0.2 0 0 0 0.2 – 0.6 1 7 0.6 – 1.0 0 12 > 1.0 11 17

Median longitudinal distal margin was 3 (2 – 4) cm. Surgery-related morbidity was 8% (4/48): there were four anastomotic leakages: two were treated laparoscopically only by peritoneal washing and drainage. Two were treated conservatively. No 30-day mortality occurred. Median hospital stay was 8 d (range, 8 – 11), oral diet resumption was 3 d (range, 2 – 13). All patients were treated with curative-intent surgery and neoadjuvant chemoradiotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months, disease-free survival was 67.8% (38/48), overall survival was 87.5% (42/48), and disease-related mortality was 8.3% (4/48) (Table 3). One

> Disease free survival 38 (67.8%) Alive w/recurrence 3 (13.0%) Drop-out 1 (4.3%)

Deceased 6 (24.7% - 5 Cancer-related; 1 others)

**7.1.2 Left colectomy, hartmann procedure, sigmoidectomy** 

patient was lost during follow-up.

Table 2. Circumferential margins

patient was lost during follow-up.

Table 3. Robotic rectal surgery Follow-up

**7.1.3 Rectal surgery** 

in table 2.

Once the distal rectal transection ended, the robotic cart is disengaged, and a suprapubic minilaparotomy (Pfannenstiel) is performed. The specimen is then extracted, after protecting the minilaparotomy by a wound protector. The descending colon is transected, and the anvil of the stapler is inserted at the end of the colon.

The bowel reconstruction is conducted laparoscopically, and an intracorporeal mechanical colorectal termino-terminal anastomosis is performed by Knight & Griffen technique with a circular stapler. In case of coloanal anastomosis, the specimen is retrieved through the anal canal by a pull-trough technique and a manual colo-anal anastomosis is performed.

In case of ultra-low anterior resections a diverting loop-ileostomy is fashioned enlarging the R3 trocart site.

#### **7. Postoperative patients' care, outcome and technical results**

#### **7.1 Technical results and outcome**

Our study is based on about 300 consecutive robotic colorectal procedures structured as follows: 140 colic resections for cancer (84 right colectomies, 43 left colectomies, 13 others), 48 rectal resections and 110 colorectal resections for benign disease.

#### **7.1.1 Right colectomy**

Forty-three male and forty-one female patients underwent robotic right colectomy. Mean age was 73.34 ± 11 years. Median operative time was 213.50 (180–250) min. No conversion occurred. Specimen length was 28 ± 5 cm (range 21–50 cm). Number of harvested lymph nodes was 19.70 ± 7.2 (range 12–44), and mean number of positive lymph nodes was 1.65 ± 3 (range 0–17). Surgery-related morbidity was 2/84 (2,3%): one twisting of the mesentery in one of the first cases with extracorporeal anastomosis and a dehiscence of the colic stump in a patient with Crohn disease. All patients were included in a follow-up regimen. Neither conversions nor 30-day mortality occurred. Oral re-intake was on day 3.47 ± 0.6 (range 2–4) and length of stay was 7 ± 1.2 days (range 5–9 days). All patients were treated with curative-intent surgery and adjuvant chemotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months (range 6–96 months), disease-free survival was 90% (76/84), overall survival was 92% (78/84), and disease-related mortality was 4% (3/84) (Table 1). Stages included in the survival analyses were II, III, and IV. Disease-free survival was 90% (72/80), overall survival was 92% (73/80) and cancer-related mortality was 13% (11/80) at a median 3 year follow-up. Overall survival for stage II, III, and IV was 94.1%, 92.3%, and 66.7%, respectively. Disease-free survival for stage II and III was 100% and 84.6%, respectively (table 1).


Table 1. Robotic right colectomy Follow-up

#### **7.1.2 Left colectomy, hartmann procedure, sigmoidectomy**

Twenty-three male and twenty female patients underwent surgery for left-sided colon cancer. Mean age was 60 ± 12 years. Median operative time was 220 (215 – 230) min. One conversion occurred for a splenic injury at the beginning of the experience. Specimen length was 24 ± 7.7 (range 10 – 49) cm. Number of harvested lymph nodes was 12 ± 7 (6 – 24). Surgery-related morbidity was 1/43 (2,3%): an anastomotic dehiscence in one of the first cases. All patients were included in a follow-up regimen. No 30-day mortality occurred. Median hospital stay was 7 d (range, 6 – 11), oral diet resumption was 3 d (range, 2 – 10). All patients were treated with curative-intent surgery and adjuvant chemotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months, disease-free survival was 88.3% (38/43), overall survival was 90% (38/43), and disease-related mortality was 7% (3/43) (Table 1). One patient was lost during follow-up.

#### **7.1.3 Rectal surgery**

530 Colorectal Cancer – From Prevention to Patient Care

Once the distal rectal transection ended, the robotic cart is disengaged, and a suprapubic minilaparotomy (Pfannenstiel) is performed. The specimen is then extracted, after protecting the minilaparotomy by a wound protector. The descending colon is transected,

The bowel reconstruction is conducted laparoscopically, and an intracorporeal mechanical colorectal termino-terminal anastomosis is performed by Knight & Griffen technique with a circular stapler. In case of coloanal anastomosis, the specimen is retrieved through the anal canal by a pull-trough technique and a manual colo-anal anastomosis is

In case of ultra-low anterior resections a diverting loop-ileostomy is fashioned enlarging the

Our study is based on about 300 consecutive robotic colorectal procedures structured as follows: 140 colic resections for cancer (84 right colectomies, 43 left colectomies, 13 others),

Forty-three male and forty-one female patients underwent robotic right colectomy. Mean age was 73.34 ± 11 years. Median operative time was 213.50 (180–250) min. No conversion occurred. Specimen length was 28 ± 5 cm (range 21–50 cm). Number of harvested lymph nodes was 19.70 ± 7.2 (range 12–44), and mean number of positive lymph nodes was 1.65 ± 3 (range 0–17). Surgery-related morbidity was 2/84 (2,3%): one twisting of the mesentery in one of the first cases with extracorporeal anastomosis and a dehiscence of the colic stump in a patient with Crohn disease. All patients were included in a follow-up regimen. Neither conversions nor 30-day mortality occurred. Oral re-intake was on day 3.47 ± 0.6 (range 2–4) and length of stay was 7 ± 1.2 days (range 5–9 days). All patients were treated with curative-intent surgery and adjuvant chemotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months (range 6–96 months), disease-free survival was 90% (76/84), overall survival was 92% (78/84), and disease-related mortality was 4% (3/84) (Table 1). Stages included in the survival analyses were II, III, and IV. Disease-free survival was 90% (72/80), overall survival was 92% (73/80) and cancer-related mortality was 13% (11/80) at a median 3 year follow-up. Overall survival for stage II, III, and IV was 94.1%, 92.3%, and 66.7%, respectively. Disease-free survival for stage II and III was 100% and 84.6%, respectively

> Disease free survival 76 (90%) Alive w/recurrence 5 (6%) Drop-out 0

Deceased 3 (4% - 3 cancer-related)

**7. Postoperative patients' care, outcome and technical results** 

48 rectal resections and 110 colorectal resections for benign disease.

and the anvil of the stapler is inserted at the end of the colon.

performed.

R3 trocart site.

**7.1.1 Right colectomy** 

(table 1).

**7.1 Technical results and outcome** 

Table 1. Robotic right colectomy Follow-up

Thirty male and eighteen female patients underwent rectal surgery: 45 rectal anterior resections (RAR) with TME and 3 abdominalperineal resections (APA). Mean age was 67 ± 12 years. Median operative time was 270 (240 – 315) min. Specimen length was 23 (19 – 27,5) cm. Number of harvested lymph nodes was 15 (12 – 20). Circumferential margins are shown in table 2.


Table 2. Circumferential margins

Median longitudinal distal margin was 3 (2 – 4) cm. Surgery-related morbidity was 8% (4/48): there were four anastomotic leakages: two were treated laparoscopically only by peritoneal washing and drainage. Two were treated conservatively. No 30-day mortality occurred. Median hospital stay was 8 d (range, 8 – 11), oral diet resumption was 3 d (range, 2 – 13). All patients were treated with curative-intent surgery and neoadjuvant chemoradiotherapy (CHT) according to current international guidelines for colorectal cancer. At median follow-up of 36 months, disease-free survival was 67.8% (38/48), overall survival was 87.5% (42/48), and disease-related mortality was 8.3% (4/48) (Table 3). One patient was lost during follow-up.


Table 3. Robotic rectal surgery Follow-up

Minimally Invasive Robot – Assisted Colorectal Resections 533

We consider the rectal anterior resection the procedure in which robotic system better

The current technique of TME was developed to reduce local recurrences and improve overall survival while maintaining an adequate quality of life. The concept of TME is founded on the anatomical dissection along the embryologic avascular areolar plane between the fascia propria recti and the parietal endopelvic fascia. The integrity of the mesorectum as well as clear circumferential and distal margins are important oncological and surgical end-points. Moreover, the complexity of the regional anatomy requires a precise and a sharp dissection under direct vision following anatomical pathways in order to preserve the autonomic innervation. All the advantages may contribute to improve oncological adequacy and nerve preservation during this procedure. The first step of TME starts with the incision of the posterior peritoneum at the level of the promontory on the bifurcation of the aorta into the common iliac arteries. At this level, the 3-D view allows the surgeon to better identify and preserve the preaortic nerves and the superior hypogastric plexus (SHP). The use of the articulated monopolar cautery hook helps to obtain a better energy delivery control, avoiding inopportune cauterization of the nervous bundle. Moreover, the steady image and the view magnification allow a correct identification of both fasciae and a sharp dissection of the "holy plane". Any dissection strayed to the presacral fascia may lead to injuries to the ureters, autonomic nerves and

The second step of the TME consists in the anterior dissection, following the plane between the Denonvilliers' fascia above and the fascia recti below. Denonvillier's fascia can be easily identified by robotic view, helping the surgeon to carry out a precise incision of this fascia at the level of the seminal vesicles, avoiding gross manipulation of the tissue and unintentional injuries to the posterior capsule of the prostate (male) or posterior vaginal wall (female). The third step of TME includes the lateral mobilization of the rectum by incision of the lateral ligaments. A dissection close to the rectal wall avoids injuries to this nervous bundle. Moreover, a gentle counter traction of the rectum may help opening the dissection plane: this maneuver seems to be improved by robotic assistance thanks to the stability and motion scaling of the robotic arms. Excessive traction may lead to risk of injury to the pelvic splanchnic nerve. Robotic stereoscopic view, in addition, makes these structures more clearly visible. Middle rectal artery or its branches may be easily identified and cauterized without any peculiar difficulty. The tip articulation of the instruments facilitates the TME also, allowing a fine and precise dissection even in a narrow space, where dissection may result difficult by conventional laparoscopy or open surgery. This aspect is important as the reduction of the local recurrence rate is directly related to the optimization of the free surgical margins with recognition of the importance of clear radial (CRM) and distal mesorectal margins and of the distance from the tumor rim. The extent of circumferential tumor clearance after rectal cancer excision impacts long-term oncologic outcomes. In our experience, robotic assistance allowed us to achieve a 0% CRM < 2 mm rate and a correct mesorectal excision in all cases. Moreover, median lymph node number was 15.60 (12 – 21) and median specimen length was 23 (19 – 27,50). Length of stay was similar to laparoscopic series and shorter than open experiences. Operative time was 270 (240 – 315) min. A comparison of our results to main robotic experiences in Literature reveals similar trends in terms of

length of stay, pathological findings and short-term outcome (table 4).

expresses its potential advantages.

presacral veins.

#### **8. Discussion**

Colorectal cancer is still the third leading cause of death in the US, even though death rates have also been declining by 2,2% per year since 1998.

The Medical Research Council Conventional versus Laparoscopic-Assisted Surgery In Colorectal Cancer (MRC CLASICC) trial was set up in 1996 to evaluate the technical and oncological safety and efficacy of laparoscopically assisted surgery in comparison with conventional open surgery for the treatment of colorectal cancer.

The last update of the CLASICC Trial showed the oncological adequacy of the laparoscopic technique compared to the open one. Moreover, minimally invasive surgery has general benefits such as less blood loss, postoperative pain, and use of anesthetics, as well as fewer early and late wound complications, a shorter hospital stay, and better aesthetic outcomes. However, there are some limitations of the laparoscopic surgery: tremor, unstable twodimensional view, and limited degree of freedom of the instruments.

Robotic surgery is spreading all over the world for many surgical procedures ranging from cardiac to general and urologic surgery thanks to its potential advantages overcoming the negative aspects of laparoscopic approach. It provides the surgeon with a 3-dimension display which enhances depth perception, allows the surgeon to operate in a comfortable, seated position with eyes, hands and operative field in line. Furthermore, the robotic instruments contain articulation which recall human wrist movements with 7 degrees of freedom to improve dexterity.

These characteristics of the robotic system may improve dissection and consequently oncological outcome. It is acquired that presence of nodal metastasis and mainly its distribution are key factors in predicting disease-free and long-term survival and for deciding on postoperative adjuvant therapy. The American Joint Committee on Cancer (AJCC) and College of American Pathologists (CAP) recommend evaluation of a minimum of 12 lymph nodes. In right colon cancer, we were able to perform a correct right colectomy easily identifying the major colic vessels and carrying out accurate lymphadenectomy, taking advantage of the steady, 3-D image view and of the articulation of the robotic instruments, which allowed us to manage organs such as the pancreas or the duodenum gently. The average length of the resected specimen in this series was 28 ± 5 cm, and the mean number of harvested lymph nodes was 19.70 ± 7.2, above the minimum recommended by the AJCC. To our knowledge our experience on robotic right colon resection is the largest published in the literature. We report a median follow-up of 36 months. Disease-free and overall survival were 90% and 92%, respectively; survival rates for stage II and III was 94.1% and 92.3%, and disease-free survival was 100% and 93%, respectively. Recent studies have shown 3-year overall survival varying from 68% to 100% for stage II and from 68% to 97% for stage III, and 5-year survival rates for stage II and III of about 72–90% and 44–72%, respectively (Gattaj et al., 2003; Roxburgh, 2209; Japan National Cancer Center, 2010). A comparison of our results with the literature shows that robotic right colic resection is able to offer the same short-term outcome as right colic resection performed by conventional laparoscopy or laparotomy. Moreover, we assert that the da Vinci System allows better standardization of the surgical technique of right colectomy, positively increasing the percentage of correct lymphatic resections.

We agree with other authors (deSouza et al., 2010) that among all robotic colorectal resections, right colectomy may be also considered the ideal procedure for the surgeon at the beginning of the learning curve as the robotic left colectomy.

Colorectal cancer is still the third leading cause of death in the US, even though death rates

The Medical Research Council Conventional versus Laparoscopic-Assisted Surgery In Colorectal Cancer (MRC CLASICC) trial was set up in 1996 to evaluate the technical and oncological safety and efficacy of laparoscopically assisted surgery in comparison with

The last update of the CLASICC Trial showed the oncological adequacy of the laparoscopic technique compared to the open one. Moreover, minimally invasive surgery has general benefits such as less blood loss, postoperative pain, and use of anesthetics, as well as fewer early and late wound complications, a shorter hospital stay, and better aesthetic outcomes. However, there are some limitations of the laparoscopic surgery: tremor, unstable two-

Robotic surgery is spreading all over the world for many surgical procedures ranging from cardiac to general and urologic surgery thanks to its potential advantages overcoming the negative aspects of laparoscopic approach. It provides the surgeon with a 3-dimension display which enhances depth perception, allows the surgeon to operate in a comfortable, seated position with eyes, hands and operative field in line. Furthermore, the robotic instruments contain articulation which recall human wrist movements with 7 degrees of

These characteristics of the robotic system may improve dissection and consequently oncological outcome. It is acquired that presence of nodal metastasis and mainly its distribution are key factors in predicting disease-free and long-term survival and for deciding on postoperative adjuvant therapy. The American Joint Committee on Cancer (AJCC) and College of American Pathologists (CAP) recommend evaluation of a minimum of 12 lymph nodes. In right colon cancer, we were able to perform a correct right colectomy easily identifying the major colic vessels and carrying out accurate lymphadenectomy, taking advantage of the steady, 3-D image view and of the articulation of the robotic instruments, which allowed us to manage organs such as the pancreas or the duodenum gently. The average length of the resected specimen in this series was 28 ± 5 cm, and the mean number of harvested lymph nodes was 19.70 ± 7.2, above the minimum recommended by the AJCC. To our knowledge our experience on robotic right colon resection is the largest published in the literature. We report a median follow-up of 36 months. Disease-free and overall survival were 90% and 92%, respectively; survival rates for stage II and III was 94.1% and 92.3%, and disease-free survival was 100% and 93%, respectively. Recent studies have shown 3-year overall survival varying from 68% to 100% for stage II and from 68% to 97% for stage III, and 5-year survival rates for stage II and III of about 72–90% and 44–72%, respectively (Gattaj et al., 2003; Roxburgh, 2209; Japan National Cancer Center, 2010). A comparison of our results with the literature shows that robotic right colic resection is able to offer the same short-term outcome as right colic resection performed by conventional laparoscopy or laparotomy. Moreover, we assert that the da Vinci System allows better standardization of the surgical technique of right colectomy, positively increasing the

We agree with other authors (deSouza et al., 2010) that among all robotic colorectal resections, right colectomy may be also considered the ideal procedure for the surgeon at

**8. Discussion** 

freedom to improve dexterity.

percentage of correct lymphatic resections.

the beginning of the learning curve as the robotic left colectomy.

have also been declining by 2,2% per year since 1998.

conventional open surgery for the treatment of colorectal cancer.

dimensional view, and limited degree of freedom of the instruments.

We consider the rectal anterior resection the procedure in which robotic system better expresses its potential advantages.

The current technique of TME was developed to reduce local recurrences and improve overall survival while maintaining an adequate quality of life. The concept of TME is founded on the anatomical dissection along the embryologic avascular areolar plane between the fascia propria recti and the parietal endopelvic fascia. The integrity of the mesorectum as well as clear circumferential and distal margins are important oncological and surgical end-points. Moreover, the complexity of the regional anatomy requires a precise and a sharp dissection under direct vision following anatomical pathways in order to preserve the autonomic innervation. All the advantages may contribute to improve oncological adequacy and nerve preservation during this procedure. The first step of TME starts with the incision of the posterior peritoneum at the level of the promontory on the bifurcation of the aorta into the common iliac arteries. At this level, the 3-D view allows the surgeon to better identify and preserve the preaortic nerves and the superior hypogastric plexus (SHP). The use of the articulated monopolar cautery hook helps to obtain a better energy delivery control, avoiding inopportune cauterization of the nervous bundle. Moreover, the steady image and the view magnification allow a correct identification of both fasciae and a sharp dissection of the "holy plane". Any dissection strayed to the presacral fascia may lead to injuries to the ureters, autonomic nerves and presacral veins.

The second step of the TME consists in the anterior dissection, following the plane between the Denonvilliers' fascia above and the fascia recti below. Denonvillier's fascia can be easily identified by robotic view, helping the surgeon to carry out a precise incision of this fascia at the level of the seminal vesicles, avoiding gross manipulation of the tissue and unintentional injuries to the posterior capsule of the prostate (male) or posterior vaginal wall (female). The third step of TME includes the lateral mobilization of the rectum by incision of the lateral ligaments. A dissection close to the rectal wall avoids injuries to this nervous bundle. Moreover, a gentle counter traction of the rectum may help opening the dissection plane: this maneuver seems to be improved by robotic assistance thanks to the stability and motion scaling of the robotic arms. Excessive traction may lead to risk of injury to the pelvic splanchnic nerve. Robotic stereoscopic view, in addition, makes these structures more clearly visible. Middle rectal artery or its branches may be easily identified and cauterized without any peculiar difficulty. The tip articulation of the instruments facilitates the TME also, allowing a fine and precise dissection even in a narrow space, where dissection may result difficult by conventional laparoscopy or open surgery. This aspect is important as the reduction of the local recurrence rate is directly related to the optimization of the free surgical margins with recognition of the importance of clear radial (CRM) and distal mesorectal margins and of the distance from the tumor rim. The extent of circumferential tumor clearance after rectal cancer excision impacts long-term oncologic outcomes. In our experience, robotic assistance allowed us to achieve a 0% CRM < 2 mm rate and a correct mesorectal excision in all cases. Moreover, median lymph node number was 15.60 (12 – 21) and median specimen length was 23 (19 – 27,50). Length of stay was similar to laparoscopic series and shorter than open experiences. Operative time was 270 (240 – 315) min. A comparison of our results to main robotic experiences in Literature reveals similar trends in terms of length of stay, pathological findings and short-term outcome (table 4).

Minimally Invasive Robot – Assisted Colorectal Resections 535

resection in a high fraction of cases. Moreover, if we compare our oncological results in terms of overall and disease-free survival to those in literature, it is clear how robotic right colic resection is able to offer the same short-term outcome as right colic resection performed by conventional laparoscopy or laparotomy not only by an oncological point of view but also by recovery time duration. In our experience, indeed, hospital stay was shorter than

In left colectomy procedures, the sole advantage of the robotic system consists in IMA dissection: the 3-D view and the Endowrist articulation allow the surgeon to better identify the preaortic parasympathetic fibers which may be incorrectly manipulated and injured, increasing the risk of sexual or urinary dysfunctions. Moreover, thanks to a stable and tridimensional view, it is possible to decrease the risk of vascular injuries. Besides this aspect, we believe that robotic left colectomy is to be considered as an initial step in the

In our opinion, the predominant procedure which best enhances the advantages of the robot is TME, and several other authors have reported their experience with the robot in TME. The main concerns about laparoscopic techniques relate to the poor dexterity and the rigidity of the instruments, the 2-dimensional view and the camera stability depending by the assistant skillness. The robot overcomes these limitations and allows for more precise oncologic dissection. In our experience, circumferential margins were acceptable and none of the analyzed specimens presented an infiltrated circumferential margin or less than 2 mm from the tumor bed. Moreover, the magnified, stable, 3-D view and the articulation of the tip of the robotic instruments allowed us to better identify the planes of dissection, so performing a correct nerve sparing resection and a correct TME as showed by the pathological reports. The advantages of the robotic system are emphasized especially in men, in which the narrow structure of the pelvis makes the dissection difficult by laparoscopy approach and "blinded" by open approach. Operative time is longer than the laparoscopic one, but we believe it may be reduced by experience. Moreover, the robot setup we adopted allows to reduce operative time by switching only one robotic arm from one trocart to another one, avoiding disengagement and re-engagement of the robotic system, as described by initial experiences. Recovery time was shorter than in open surgery, and morbidity was acceptable,

In conclusion, robotic assistance may help the surgeon in performing colorectal procedures

Ballantyne GH, Merola P, Weber A, Wasielewski A. Robotic solutions to the pitfalls of

Ballantyne GH. Robotic surgery, telerobotic surgery, telepresence, and telementoring.

Ballantyne GH. The pitfalls of laparoscopic surgery: challenges for robotics and telerobotic

Braumann C, Jacobi CA, Menenakos C, Borchert U, Rueckert JC, Mueller JM. Computer-

assisted laparoscopic colon resection with the Da Vinci system: our first

Review of early clinical results. Surg Endosc 2002; 16: 1389–1402.

surgery. Surg Laparosc Endosc Percutan Tech 2002;12:1–5.

experiences. Dis Colon Rectum. 2005 Sep;48(9):1820-7.

and improve the patient outcomes and provides acceptable oncological results.

Baik SH. Robotic colorectal surgery. Yonsei Med J. 2008 Dec 31;49(6):891-6.

laparoscopic colectomy. Osp Ital Chir. 2001;7:405–12.

open one and comparable to laparoscopy.

learning curve of robotic surgery for a surgeon.

confirming the safety and feasibility of robotic assistance in TME.

**10. References** 


Table 4. Robotic rectal surgery experiences. LN: lymph nodes. LOS: length of stay

#### **9. Conclusions**

Laparoscopic colorectal surgery has become a mainstay in the treatment of benign and malignant colorectal diseases. Recently, a new update of the CLASICC trial has confirmed the oncological adequacy and the safety of laparoscopic colorectal surgery (Jayne et al., 2010).

There are some drawbacks, however, of the laparoscopic technique such as unstable video camera platform, limited motion of straight instruments, two-dimensional imaging, and poor ergonomics for the surgeon. Robotic surgery is spreading all over the world for many surgical procedures ranging from cardiac to general and urologic surgery thanks to its potential advantages overcoming the negative aspects of laparoscopic approach (Piazza et al., 1999; Reichenspurner et al., 2000; Kappert et al., 2000; Gill et al., 2000; Chen et al., 2009).

The da Vinci surgical system (Intuitive Surgical Inc., Sunnyvale, CA, USA) was the first telerobotic system approved for intra-abdominal surgery in the USA by the Food and Drug Administration (FDA, 2000). The first robot-assisted colectomy was reported by Ballantyne et al. in 2001 (Ballantyne et al., 2001). Since then, several surgeons have performed robotic colorectal surgery. The advantageous features of the robotic system are the physical separation of the surgeon from the patient, six degrees of freedom plus grasping of the robotic arms, hand-like motions of the instruments offering the surgeon the impression of an open access, elimination of tremor, optional motion downscaling (2:1 to 5:1), and three dimensional stereoscopic image (Ballantyne et al., 2002) The surgeons console and the projected three-dimensional virtual operative field offer an ergonomically comfortable position with minimum fatigue (Braumann et al., 2005).

In right colectomy procedures, we were able to perform correct R1 and R2 right colectomy, easily identifying the major colic vessels and carrying out accurate lymphadenectomy over the plane of the superior mesenteric axis, taking advantage of the steady, 3-D image view and of the articulation (Endowrist) of the robotic instruments, which allowed us to manage organs such as the pancreas or the duodenum gently. Our observation is that robotic technique could allow better standardization, leading to improved performance of minimally invasive right colic resection, especially in terms of achieving correct lymphatic

Time LN Positive

<sup>2011</sup>36 337.9 (81.8) 15 (7.8) 0 11 (30.6) 7.0 (5.8) Baek 2010 41 296 13.1 1 9 6.5

<sup>2010</sup>25 240\* 18\* 0 4 6.5\* Park 2010 41 231.9 (61.4) 17.3 (7.7) 2 12 9.9 (4.2)

<sup>2010</sup>44 347\*0 14 (5–45) 0 11 5 Koh 2010 21 292.3 ± 32.6 17.8 ± 7.1 1 5 6.4 ± 4.1 Luca 2009 28 290 ± 69 18.5 ±8.3 0 12 7.5 ± 2.8

e 2011 48 270\* 15 0 4 8\*

Laparoscopic colorectal surgery has become a mainstay in the treatment of benign and malignant colorectal diseases. Recently, a new update of the CLASICC trial has confirmed the oncological adequacy and the safety of laparoscopic colorectal surgery (Jayne et al.,

There are some drawbacks, however, of the laparoscopic technique such as unstable video camera platform, limited motion of straight instruments, two-dimensional imaging, and poor ergonomics for the surgeon. Robotic surgery is spreading all over the world for many surgical procedures ranging from cardiac to general and urologic surgery thanks to its potential advantages overcoming the negative aspects of laparoscopic approach (Piazza et al., 1999; Reichenspurner et al., 2000; Kappert et al., 2000; Gill et al., 2000; Chen et al., 2009). The da Vinci surgical system (Intuitive Surgical Inc., Sunnyvale, CA, USA) was the first telerobotic system approved for intra-abdominal surgery in the USA by the Food and Drug Administration (FDA, 2000). The first robot-assisted colectomy was reported by Ballantyne et al. in 2001 (Ballantyne et al., 2001). Since then, several surgeons have performed robotic colorectal surgery. The advantageous features of the robotic system are the physical separation of the surgeon from the patient, six degrees of freedom plus grasping of the robotic arms, hand-like motions of the instruments offering the surgeon the impression of an open access, elimination of tremor, optional motion downscaling (2:1 to 5:1), and three dimensional stereoscopic image (Ballantyne et al., 2002) The surgeons console and the projected three-dimensional virtual operative field offer an ergonomically comfortable

In right colectomy procedures, we were able to perform correct R1 and R2 right colectomy, easily identifying the major colic vessels and carrying out accurate lymphadenectomy over the plane of the superior mesenteric axis, taking advantage of the steady, 3-D image view and of the articulation (Endowrist) of the robotic instruments, which allowed us to manage organs such as the pancreas or the duodenum gently. Our observation is that robotic technique could allow better standardization, leading to improved performance of minimally invasive right colic resection, especially in terms of achieving correct lymphatic

Table 4. Robotic rectal surgery experiences. LN: lymph nodes. LOS: length of stay

position with minimum fatigue (Braumann et al., 2005).

RM Complications LOS

Author

Ashwin

Bianchi

deSouza

D'Annibal

2010).

**9. Conclusions** 

Year Procedures Operative

resection in a high fraction of cases. Moreover, if we compare our oncological results in terms of overall and disease-free survival to those in literature, it is clear how robotic right colic resection is able to offer the same short-term outcome as right colic resection performed by conventional laparoscopy or laparotomy not only by an oncological point of view but also by recovery time duration. In our experience, indeed, hospital stay was shorter than open one and comparable to laparoscopy.

In left colectomy procedures, the sole advantage of the robotic system consists in IMA dissection: the 3-D view and the Endowrist articulation allow the surgeon to better identify the preaortic parasympathetic fibers which may be incorrectly manipulated and injured, increasing the risk of sexual or urinary dysfunctions. Moreover, thanks to a stable and tridimensional view, it is possible to decrease the risk of vascular injuries. Besides this aspect, we believe that robotic left colectomy is to be considered as an initial step in the learning curve of robotic surgery for a surgeon.

In our opinion, the predominant procedure which best enhances the advantages of the robot is TME, and several other authors have reported their experience with the robot in TME. The main concerns about laparoscopic techniques relate to the poor dexterity and the rigidity of the instruments, the 2-dimensional view and the camera stability depending by the assistant skillness. The robot overcomes these limitations and allows for more precise oncologic dissection. In our experience, circumferential margins were acceptable and none of the analyzed specimens presented an infiltrated circumferential margin or less than 2 mm from the tumor bed. Moreover, the magnified, stable, 3-D view and the articulation of the tip of the robotic instruments allowed us to better identify the planes of dissection, so performing a correct nerve sparing resection and a correct TME as showed by the pathological reports. The advantages of the robotic system are emphasized especially in men, in which the narrow structure of the pelvis makes the dissection difficult by laparoscopy approach and "blinded" by open approach. Operative time is longer than the laparoscopic one, but we believe it may be reduced by experience. Moreover, the robot setup we adopted allows to reduce operative time by switching only one robotic arm from one trocart to another one, avoiding disengagement and re-engagement of the robotic system, as described by initial experiences. Recovery time was shorter than in open surgery, and morbidity was acceptable, confirming the safety and feasibility of robotic assistance in TME.

In conclusion, robotic assistance may help the surgeon in performing colorectal procedures and improve the patient outcomes and provides acceptable oncological results.

#### **10. References**

Baik SH. Robotic colorectal surgery. Yonsei Med J. 2008 Dec 31;49(6):891-6.


Minimally Invasive Robot – Assisted Colorectal Resections 537

Kappert U, Cichon R, Schneider J, et al. Closed-chest coronary artery surgery on the beating

Koopmann MC, Heise CP. Laparoscopic and minimally invasive resection of malignant

Lacy AM, Garcia-Valdecasas JC, Delgado S, et al. Laparoscopy assisted colectomy versus

Lanfranco AR, Castellanos AE, Desai JP, Meyers WC. Robotic surgery: a current perspective.

Law WL, Lee YM, Choi HK, et al. Impact of laparoscopic resection for colorectal cancer on

Leung KL, Lai PB, Ho RL, et al. Systemic cytokine response after laparoscopic-assisted

Milsom JW, Bohm B, Hammerhofer KA, et al. A prospective, randomized trial comparing

Mirnezami AH, Mirnezami R, Venkatasubramaniam AK, Chandrakumaran K, Cecil TD,

Morino M, Pellegrino L, Giaccone C, Garrone C, Rebecchi F. Randomized clinical trial of robot-assisted vs. laparoscopic Nissen fundoplication. Br J Surg 2006; 93: 553–558. Nelson H, Sargent DJ, Wieand HS, et al. A comparison of laparoscopically assisted and open

Piazza L, Caragliano P, Scardilli M, Sgroi AV, Marino G, Giannone G. Laparoscopic robot-

Poulin EC, Mamazza J, Schlachta CM, et al. Laparoscopic resection does not adversely affect

Reichenspurner H, Boehm DH, Gulbins H, et al. Three-dimensional video and robotassisted port-access mitral valve operation. Ann Thorac Surg. 2000;69(4):1176–81. Roxburgh CS, Salmond JM, Horgan PG, Oien KA, McMillan DC. The relationship between

colorectal disease. Surg Clin North Am. 2008;88:1047–72.

operative outcomes and survival. Ann Surg. 2007;245(1):1–7.

preliminary report. J Am Coll Surg 1998;187(1):46–54.

simulated operating theatre. Ann Surg 2005;242: 631-39.

Lancet. 2002; 359(9325):2224–9.

Engl J Med. 2004;350(20):2050–9.

adenocarcinoma. Ann Surg 1999;229(4):487–92.

colorectal cancer. Br J Surg 2001;88(6):801–7.

1999;51(6):465–6.

Ann Surg 2004; 239: 14–21.

2000; 231(4):506–11.

11.

heart with the use of a robotic system. J Thorac Cardiovasc Surg. 2000;120(4):809–

open colectomy for treatment of non-metastatic colon cancer: a randomised trial.

resection of rectosigmoid carcinoma: a prospective randomized trial. Ann Surg

laparoscopic versus conventional techniques in colorectal cancer surgery: a

Moran BJ. Robotic colorectal surgery: hype or new hope? A systematic review of robotics in colorectal surgery. Colorectal Dis 2009 Jul 6 [Epub ahead of print] Moorthy K, munz Y, Adams S, Pandey V, Darzi A. A human factors analysis of technical

and team skills among surgical trainees during procedural simulations in a

colectomy for colon cancer. Clinical Outcomes of Surgical Therapy Study Group. N

assisted right adrenalectomy and left ovariectomy (case reports). Chir Ital.

early survival curves in patients undergoing surgery for colorectal

the local and systemic inflammatory responses and survival in patients undergoing curative surgery for colon and rectal cancers. J Gastrointest Surg. 2009;13:2011–8. Stage JG, Schulze S, Moller P, et al. Prospective randomized study of laparoscopic versus open colonic resection for adenocarcinoma. Br J Surg. 1997;84(3):391–6. Tang CL, Eu KW, Tai BC, et al. Randomized clinical trial of the effect of open versus

laparoscopically assisted colectomy on systemic immunity in patients with


Cadiere GB, Himpens J, Germay O, et al. Feasibility of robotic laparoscopic surgery: 146

Champault GG, Barrat C, Raselli R, Elizalde A, Catheline J-M. Laparoscopic versus open

Chen CC, Falcone T. Robotic gynecologic surgery: past, present, and future. Clin Obstet

Delaney CP, Lynch AG, Senagore AJ, Fazio VW. Comparison of robotically performed and traditional laparoscopic colorectal surgery. Dis Colon Rectum. 2003;46:1633–9. Delgado S, Lacy AM, Filella X, et al. Acute phase response in laparoscopic and open

deSouza AL, Prasad LM, Park JJ, Marecik SJ, Blumetti J, Abcarian H. Robotic assistance in right hemicolectomy: Is there a role? Dis Colon Rectum. 2010;53(7):1000-6. Fleshman J, Sargent DJ, Green E, et al. Laparoscopic colectomy for cancer is not inferior to

Garca-Ruiz A, Gagner M, Miller J, Steiner C, Hahn JF. Manual vs robotic assisted

Gattaj G, Ciccolallo L. Differences in colorectal cancer survival between European and US

Gill IS, Sung GT, Hsu TH, Meraney AM. Robotic remote laparoscopic nephrectomy and

Guillou PJ, Quirke P, Thorpe H, et al. Short-term endpoints of conventional versus

Hasegawa H, Kabeshima Y, WatanabeM, et al. Randomized controlled trial of laparoscopic

Horgan S, Vanuno D. Robots in laparoscopic surgery. J Laparoendosc Adv Surg Tech A

Hu JK, Zhou ZG, Chen ZX, et al. Comparative evaluation of immune response after

Jayne DG, Thorpe HC, Copeland J, Quirke P, Brown JM, Guillou PJ. Five-year follow-up of

open surgery for colorectal cancer. Br J Surg. 2010 Nov;97(11):1638-45. Kaiser AM, Kang JC, Chan LS, et al. Laparoscopic-assisted vs. open colectomy for colon

in patients with rectal cancer. World J Gastroenterol 2003;9(12):2690–4. Jayne DG, Gullou PJ, Thorpe H, et al. Randomized trial of laparoscopic-assisted resection of

adrenalectomy: the initial experience. J Urol. 2000;164(6):2082–5.

Hanly EJ, Talamini MA. Robotic abdominal surgery. Am J Surg 2004; 188: 19S–26S.

surgery for colorectal carcinoma. A prospective clinical trial involving 157 cases with mean follow-up of 5 years. Surg Laparosc Endosc Percutan Tech 2002;12:88–

colectomy in colon cancer: randomized study. Dis Colon Rectum 2001;44(5):638–46.

open surgery based on 5-year data from the COST study group trial. Ann Surg.

laparoscopic surgery in the performance of basic manipulation and suturing tasks.

populations: the importance of subsite and morphology. Eur J Cancer.

laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): multicentre, randomized controlled trial. Lancet. 2005;365(9472):1718–26.

versus open colectomy for advanced colorectal cancer. Surg Endosc 2003;17(4):636–

laparoscopical and open total mesorectal excisions with anal sphincter preservation

colorectal carcinoma: 3-year results of the UK MRC CLASICC trial group. J Clin

the Medical Research Council CLASICC trial of laparoscopically assisted versus

cancer: a prospective randomized trial. J Laparoendosc Adv Surg

cases. World J Surg 2001;25:1467–77.

Gynecol. 2009;52(3):335–43.

2007;246(4):655–62.

2003;39:2214–22.

2001; 11: 415–419.

Oncol. 2007;25(21):3061–8.

TechA2004;14(6):329–34.

40.

Arch Surg 1998;133:957– 61.

95.


Veldkamp R, Kuhry E, Hop WC, et al. Laparoscopic surgery versus open surgery for colon cancer: short-term outcomes of a randomised trial. Lancet Oncol. 2005;6(7):477–84. Weber PA, Merola S, Wasielewski A, Ballantyne GH. Telerobotic –assisted laparoscopic

Dec;45(12):1689-94; discussion 1695-6.

right and sigmoid colectomies for benign disease. Dis Colon Rectum. 2002

### *Edited by Rajunor Ettarh*

The projections for future growth in the number of new patients with colorectal cancer in most parts of the world remain unfavorable. When we consider the substantial morbidity and mortality that accompanies the disease, the acute need for improvements and better solutions in patient care becomes evident. This volume, organized in five sections, represents a synopsis of the significant efforts from scientists, clinicians and investigators towards finding improvements in different patient care aspects including nutrition, diagnostic approaches, treatment strategies with the addition of some novel therapeutic approaches, and prevention. For scientists involved in investigations that explore fundamental cellular events in colorectal cancer, this volume provides a framework for translational integration of cell biological and clinical information. Clinicians as well as other healthcare professionals involved in patient management for colorectal cancer will find this volume useful.

Colorectal Cancer - From Prevention to Patient Care

Colorectal Cancer

From Prevention to Patient Care

*Edited by Rajunor Ettarh*

Photo by wildpixel / iStock