**6. Detection of desmoplastic reaction in biopsy specimens of colorectal cancer**

1) Existence of carcinoma is required for detection of DR.

not signify the presence of DR.

DR, desmoplastic reaction.

culated (non-Ip) SICRCs,

retorospective study - (adapted from [23])

stains.

2) The histological findings of infiltrating carcinoma do not signify the presence of DR.

**Table 6.** Criteria for detection of desmoplastic reaction (adapted from [22])

3) DR contains areas of collagen fiber accumulation and myofibroblast proliferation, but inflammatory infiltration does

Desmoplastic Reaction in Biopsy Specimens of T1 Stage Colorectal Cancer Plays a Critical Role in Defining the Level...

4) All histological findings were determined by HE stain alone; detection of DR does notrequire the use of any special

A retorospective study for detection of desmoplastic reaction in biopsy specimens of early colorectal cancer reported that 359 patients with SICRCs, who had undergone surgical or en‐ doscopic mucosal resection, were analysed [23]. For pedunculated (Ip type) SICRCs was not significantly related to submucosal depth. However, for nonpedunculated (non-Ip type) the prevalence of DR in pretreatment biopsy specimens was significantly related to submucosal depth. When nonpedunculated(non-Ip type) SICRCs were further divided using a specific cut off value of 1000 µm for submucosal depth, the positivity ratio of DR in pretreatment biopsy specimens was significantly higher in SICRCs with an submucosal depth of 1000 µm than in cases where the submucosal depth was <1000 µm (Table 7). Detection of DR in pre‐ treatment biopsy specimens is useful for the prediction of submucosal depth in nonpedun‐

**Depth of submucosal invasion (µm) Pedunculated SICRC Nonpedunculated SICRC**

X<1000 3 (37.5) 5 (62.5) 30 (65.2) 16 (34.8) 1000 ≤ X < 2000 1 (50.0) 1 (50.0) 25 (56.8) 19 (43.2) 2000 ≤ X < 3000 5 (100) 0 (0.0) 22 (27.5) 58 (72.5) 3000 ≤ X < 4000 3 (33.3) 6 (66.7) 20 (29.9) 47 (70.1) 4000 ≤ X < 5000 0 (0.0) 1 (100) 11 (31.4) 24 (68.6) 5000 ≤ X < 6000 1 (100) 0 (0.0) 10 (43.5) 13 (56.5) 6000 ≤ X < 7000 2 (100) 0 (0.0) 5 (55.6) 4 (44.4) 7000 ≤ X < 8000 0 (0.0) 2 (100) 0 (0.0) 4 (100) 8000 ≤ X < 9000 0 (0.0) 0 (0.0) 4 (44.4) 5 (55.6) 9000 ≤ X < 10000 0 (0.0) 0 (0.0) 0 (0.0) 4 (100) 10000 ≤X 1 (50.0) 1 (50.0) 2 (33.3) 4 (66.7

**Table 7.** Relationship between depth of submucosal invasion and DR in biopsy specimens of patients with SICRCs -

**DR (-) (%) DR (+) (%) DR (-) (%) DR (+) (%)**

http://dx.doi.org/10.5772/51900

171

Early CRC, including submucosal invasive colorectal carcinoma (SICC), is defined as a tumor whose invasion is limited to the mucosa or submucosa. The endoscopic resection (EMR/ESD) have been useful for early CRC. However, endoscopic resection is applicable only to intramu‐ cosal carcinoma (Tis stage tumor), and additional surgery is required if the resected specimen reveals submucosal invasion by pathological diagnosis. However, the study of Japanese Soci‐ ety for Cancer of the Colon and Recum have recently demonstrated that the depth of submu‐ cosal invasion is closely correlated with the prevalence of lymph node metastasis in patients with SICC (See. Correlations between lymph node metastasis and early colorectal Carcinoma).

Therefore, gastrointestinal endoscopist and surgical pathologist considered that it would be very useful to predict the depth of invasion of SICC before endoscopic resection for case se‐ lection. New endoscopic systems are possible to predict submocosal invasion depth without biopsy. However, these developed systems have not been used in everywhere. On the other hand, the stromal change associated with carcinoma invasion has been called desmoplastic reaction (DR), desmoplasia, and cancer-associated fibroblast. Reported incidence of the DR is suggested to be a prognostic marker in CRC patients [16, 17]. Recently, JSCCR studies have been assessed the DR in pretreatment biopsy specimens of SICRC to predict the sub‐ mucosal depth in retorospective and prospective study.

The DR is characterized by modifications in the composition of stromal cells and extracellu‐ lar matrix (ECM) components with eosinophilic change [18, 19] (Figure. 8). The main pro‐ ducers of many ECM compounds and represent the major cellular component including DR that often show differentiated phenotype with expression of the smooth muscle actin, plate‐ let derived growth factor receptor- type I collagen [18, 20, 21].

The presence and histological findings of DR in biopsy specimens were evaluated by pathol‐ ogists at each respective institute and the criteria for assessment of DR established in con‐ sensus meeting among JSCCR members including us. (Table 6).

**Figure 8.** Histological appearance of desmoplastic reaction (DR). Note the growth of eosinophilic spindle cells (myofibroblasts.\*) with submucosal invasive carcinoma component on Hematoxylin and Eosin staining section.

1) Existence of carcinoma is required for detection of DR.

2) The histological findings of infiltrating carcinoma do not signify the presence of DR.

3) DR contains areas of collagen fiber accumulation and myofibroblast proliferation, but inflammatory infiltration does not signify the presence of DR.

4) All histological findings were determined by HE stain alone; detection of DR does notrequire the use of any special stains.

DR, desmoplastic reaction.

**6. Detection of desmoplastic reaction in biopsy specimens of colorectal**

Early CRC, including submucosal invasive colorectal carcinoma (SICC), is defined as a tumor whose invasion is limited to the mucosa or submucosa. The endoscopic resection (EMR/ESD) have been useful for early CRC. However, endoscopic resection is applicable only to intramu‐ cosal carcinoma (Tis stage tumor), and additional surgery is required if the resected specimen reveals submucosal invasion by pathological diagnosis. However, the study of Japanese Soci‐ ety for Cancer of the Colon and Recum have recently demonstrated that the depth of submu‐ cosal invasion is closely correlated with the prevalence of lymph node metastasis in patients with SICC (See. Correlations between lymph node metastasis and early colorectal Carcinoma).

Therefore, gastrointestinal endoscopist and surgical pathologist considered that it would be very useful to predict the depth of invasion of SICC before endoscopic resection for case se‐ lection. New endoscopic systems are possible to predict submocosal invasion depth without biopsy. However, these developed systems have not been used in everywhere. On the other hand, the stromal change associated with carcinoma invasion has been called desmoplastic reaction (DR), desmoplasia, and cancer-associated fibroblast. Reported incidence of the DR is suggested to be a prognostic marker in CRC patients [16, 17]. Recently, JSCCR studies have been assessed the DR in pretreatment biopsy specimens of SICRC to predict the sub‐

The DR is characterized by modifications in the composition of stromal cells and extracellu‐ lar matrix (ECM) components with eosinophilic change [18, 19] (Figure. 8). The main pro‐ ducers of many ECM compounds and represent the major cellular component including DR that often show differentiated phenotype with expression of the smooth muscle actin, plate‐

The presence and histological findings of DR in biopsy specimens were evaluated by pathol‐ ogists at each respective institute and the criteria for assessment of DR established in con‐

**Figure 8.** Histological appearance of desmoplastic reaction (DR). Note the growth of eosinophilic spindle cells (myofibroblasts.\*) with submucosal invasive carcinoma component on Hematoxylin and Eosin staining section.

mucosal depth in retorospective and prospective study.

170 Colonoscopy and Colorectal Cancer Screening - Future Directions

let derived growth factor receptor- type I collagen [18, 20, 21].

sensus meeting among JSCCR members including us. (Table 6).

**cancer**

**Table 6.** Criteria for detection of desmoplastic reaction (adapted from [22])

A retorospective study for detection of desmoplastic reaction in biopsy specimens of early colorectal cancer reported that 359 patients with SICRCs, who had undergone surgical or en‐ doscopic mucosal resection, were analysed [23]. For pedunculated (Ip type) SICRCs was not significantly related to submucosal depth. However, for nonpedunculated (non-Ip type) the prevalence of DR in pretreatment biopsy specimens was significantly related to submucosal depth. When nonpedunculated(non-Ip type) SICRCs were further divided using a specific cut off value of 1000 µm for submucosal depth, the positivity ratio of DR in pretreatment biopsy specimens was significantly higher in SICRCs with an submucosal depth of 1000 µm than in cases where the submucosal depth was <1000 µm (Table 7). Detection of DR in pre‐ treatment biopsy specimens is useful for the prediction of submucosal depth in nonpedun‐ culated (non-Ip) SICRCs,


**Table 7.** Relationship between depth of submucosal invasion and DR in biopsy specimens of patients with SICRCs retorospective study - (adapted from [23])

After retorospective study [22], same study group confirmed verification of patients SICRC with 112 nonpedunculated (non-Ip type) cases. Finally, nonpedunculated (non-Ip type) case of the prevalence of DR was significantly correlated with submucosal depth. The sensitivity and specificity of detection of DR for prediction of pSM2 (tumor invasion <1000 µm) in non‐ pedunculated SICRC were 68.6% and 92.0%, respectively.

**7. Conclusion**

to the submucosal layer.

**Acknowledgment**

In this issue, we have discussed the a critical role of pathological assessment for T1 stage color‐ ectal cancer, several problems related to the pathological diagnosis of early CRC at increased risk of lymph node metastasis and submucosal invasion. A new endoscopic systems which may prove to be reliable for detection of deep submucosal invasive CRC. Moreover, current en‐ doscopic resection (EMR/ESD) have become useful for early CRC, but, these resection is appli‐ cable only to intramucosal carcinoma, and additional surgery is required if the resected lesion reveals submucosal invasion by pathological diagnosis, because prevalence rates of lymph node metastasis about 10% of all patients with submucosal invasive colorectal carcinoma.

Desmoplastic Reaction in Biopsy Specimens of T1 Stage Colorectal Cancer Plays a Critical Role in Defining the Level...

http://dx.doi.org/10.5772/51900

173

We believe that curative endoscopic management for early CRC may be need to accurately pathological diagnosis of submucosal depth. Assessment of DR in pretreatment biopsy specimen nonpedunculated (non-Ip type) and submucosal depth with head invasion or stalk invasion in pedunculated (Ip type) for post endoscopic resection (EMR/ESD) specimen may be useful for the clinicopathological diagnosis of colorectal carcinoma with invasion in‐

The authors would like to thank member of Japanese Society for Cancer of the Colon and Recum; Professor Ajioka (Division of Molecular and Functional Pathology, Department of Cellular Function, Graduate School of Medical and Dental Sciences, Niigata University), Dr Ueno (Department of Surgery, National Defense Medical College, Saitama, Professer Oh‐ kura (Department of Pathology, Kyorin University School of Medicine), Professor Kashida (Department of Gastroenterology and Hepatology, Faculty of Medicine, Kinki University), Professor Togashi (Department of Gastroenterology, Fukushima Prefectural Aizu General Hospital), Professor Yao (Department of Human Pathology, Juntendo University School of Medicine) Dr Wada (Department of Pathology, Juntendo Shizuoka Hospital of Juntendo University School of Medicine), Professor Watanabe (Department of Surgical Oncology The University of Tokyo), Dr Ochiai (Pathology Division, Research Center for Innovative Oncol‐ ogy, National Cancer Center Hospital East), Professor Sugai (Diagnostic Pathology, Iwate Medical University, Morioka), Professor Sugihara (Surgical Oncology, Tokyo Medical and Dental University). And we also thank Dr Fujii (Center for Gastrointestinal Endoscopy, Kyo‐ to-Katsura Hospital) for his insightful comments. Moreover, the authors greatly thank Ms C. Sato -Matsuyama, A. Shimizu, M. Katayama, (Department of Surgical and Molecular Pathol‐ ogy, DOKKYO Medical University School of Medicine) for technical assistance and to Ms. S. Kidachi, A. Kikuchi (Department of Surgical and Molecular Pathology, DOKKYO Medical

University School of Medicine) for secretarial assistance in preparing the manuscript.

Japan and DOKKYO Medical University Young Investigator Award (2009, 2011, 2012).

This work was partially supported by the Grant-in-Aid for Scientific Research (C: No 18659101, 23590410) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of

Additionally, receiver operating characteristic (ROC), analysis confirmed 950 µm as the best diagnostic cut-off value of submucosal depth for DR detection, and 50 µm, which is the dif‐ ference between the value of 950 µm as determined by cut off value (COV) and 1000 µm, the defining value of pSM2, is an acceptable measurement error range.

In statistics and diagnostic testing, positive predictive value (PPV) is the proportion of pa‐ tients with positive test results who are correctly diagnosed, on the other hand, negative predictive value (NPV) is negative test results. Both of PPV and NPV are critical measure of the performance of a diagnostic method. In this studies revealed that PPV:0.93, NPV:0.59 in pedunculated and PPV:0.95, NPV:0.59 in nonpedunculated type. These results provide a ba‐ sis for assessment of DR as a good indicator of pSM2. (Table 8).


pSM2, SM invasion ≧1000 µm; pM + pSM1, SM invasion < 1000 µm.

Sensitivity: (37/37+ 17) x 100 = 68.5%.

Specificity: (24/24+ 3) x 100 = 88.9%.

DR, desmoplastic reaction.


pSM2, SM invasion ≧1000 µm; pM + pSM1, SM invasion < 1000 µm.

Sensitivity: (35/35+ 16) x100 = 68.6%.

Specificity: (23/23+ 2) x 100 = 92.0%.

DR, desmoplastic reaction.

**Table 8.** Relationship between depth of submucosal invasion and DR in biopsy specimens of patients with pedunculated (upper part) and nonpedunculated (lower part) type - prospective study - (adapted from [22])
