**Meet the editor**

Mariastefania Antica graduated from the Faculty of Science, Zagreb University, Croatia in 1981. She obtained and her PhD in Immunology from Ludwig Maximilian's University, Munich, Germany in 1987 prior to becoming a visiting scientist (WEHI, Melbourne, Australia, 1990-1994, Institute of Immunology, Munich, Germany, 1996, Mayo Clinic, Rochester, USA, 1998), and a principal investigator

of national and international scientific projects. She is employed as Senior Scientist at the Rudjer Boskovic Institute, and a professor at the Faculty of Science in Zagreb, Croatia, teaching both undergraduate and graduate studies at the Faculty and Medical School, Zagreb University. In regard to her scientific achievements, she was awarded the National Science Award for young scientists (1984), the Award of the Academy of Sciences and Arts (1999), and the Award of Croatian Government (2000). She was the vice president of the Croatian Immunological Society from 1996 – 2001, editor of the Internet Journal of Hematology, and member of the Croatian Medical Journal Editorial Board. She is a life member of the UICC, and referee for international journals as well as European grant applications.

Contents

**Preface IX** 

Chapter 1 **Classification of Acute Leukemia 3**  Gamal Abdul-Hamid

**Under-Resourced Laboratories 19** 

**Part 2 Molecular Mechanisms and Markers 41** 

Roel Polak and Miranda Buitenhuis

**Ikaros Transcription Factors 69** 

**Therapy in Acute Leukemia 75**  Xudong Ma, Shaohong Jiang, Yiqun Huang, Yong Zou, Ruiji Zheng and Liyun Xiao

**Stratification in Adult Acute Myeloid Leukemia with Normal Cytogenetics 95** 

**Chemokine Receptor Pathways that** 

**Modulate Leukemia Cell Dissemination 137**  Seiji Fukuda, Chie Onishi and Louis M. Pelus

**Normal and Malignant Hematopoiesis 43** 

Chapter 2 **Diagnosis of Acute Leukemia in** 

Abbas H. Abdulsalam

Chapter 3 **The PI3K/PKB Signaling Module in** 

Chapter 4 **Lymphocyte Commitment and** 

Mariastefania Antica

Chapter 5 **Epigenetics and Targeted** 

Chapter 6 **Molecular Markers for Risk** 

Chapter 7 **Trafficking of Acute Leukemia Cells –** 

Ota Fuchs

**Part 1 Introduction 1** 

### Contents

#### **Preface XI**

**Part 1 Introduction 1** 

	- **Part 2 Molecular Mechanisms and Markers 41**

X Contents


	- **Part 4 Treatment and Future Prospects 313**

### Preface

Remarkable advances have been made in the understanding and treatment of leukemia since its first description by Alfred Velpeau in 1827 and Alfred Donné in 1844. John Hughes Bennett gave the first official diagnosis back in 1845, and in 1856, the pathologist Rudolf Virchow coined the term leukemia from the Greek words "leukos" and "heima," also meaning "white blood". In 1970, it was first confirmed that some patients could be cured of leukemia, and by 1990s, the cure rate for leukemia was around 70 percent. Recent advances in the diagnosis and treatment of childhood acute lymphoblastic leukemia have since achieved a success rate of some 80 percent.

This book gives a comprehensive overview of basic mechanisms underlying acute leukemia, current advances, and future directions in the management of this disease. It presents a collection of articles on acute leukemia and the most important advances made in recent years. Employing the principles of molecular biology and understanding the basic mechanisms of cell proliferation and development are essential for dealing with leukemia. The book brings together the expert knowledge from more than 40 internationally renowned scientists, and conveys the basic information, from classification, analysis and treatment, to novel molecular mechanism and principles observed in acute leukemia. It combines and assembles scientific groups worldwide dealing with acute leukemia, from the molecular to the clinical point of view. After a thorough revision of more than 30 reviews submitted, only about 50 percent were selected for the first phase of this editorial process.

The book is divided in four chapters: 1) Introduction to acute leukemia, 2) Molecular mechanisms and markers, 3) Pediatric acute leukemia, and 4) Treatment and future prospects. The articles synthesize an enormous amount of scientific and clinical data and give a comprehensive overview to create state-of-the-art descriptions of acute leukemia.

Objective of the book Acute Leukemia – The Scientist's Perspective and Challenge represents an extremely aggressive, malignant transformation of an early hematopoieetic precursor into an immature blast form. It may be derived from myeloid cell lines, resulting in acute myeloid leukemia (AML), or from lymphoid cell lines resulting in acute lymphoblastic leukemia (ALL). This general division has implications for different management approaches. The undifferentiated malignant

#### X Preface

clone proliferates abnormally, accumulates in bone marrow, and results in progressive hematopoietic failure with anemia, thrombocytopenia, and granulocytopenia. It may also infiltrate different organs, including liver, spleen, lymph nodes, CNS, kidneys, and gonads. Treatment options for patients with acute leukemia include chemotherapy, radiation therapy, targeted therapy, immunotherapy and stem cell or bone marrow transplantation. Finding a cure is a realistic goal for both ALL and AML, especially in younger patients. However, long-term survival is reachable in about one half of patients with acute lymphoblastic leukemia and in the minority of patients with acute myeloid leukemia. Current research, such as new genes whose protein products are suitable for targeted therapy and new strategies of immunotherapy, is focalized to improve therapy for this challenging disease.

#### **Acknowledgments**

Most important acknowledgements go to all the contributing authors for their assistance, eagerness, support, and their expert scientific reviews. I thank my colleagues, Lipa Cicin-Sain, Maja Matulic and Mladen Paradzik for the motivating and inspiring discussions. Thanks to my husband Darko for taking care of our children during the long days and weekends while I worked on this project, and for his understanding and encouragement. Last but not least, I am most grateful to the publishing process managers Petra Nenadic and Ana Pantar for their outstanding work, great support and indispensable help at every phase in the preparation of this book.

> **Prof. Mariastefania Antica, PhD**  Rudjer Boskovic Institute,

Zagreb, Croatia

X Preface

clone proliferates abnormally, accumulates in bone marrow, and results in progressive hematopoietic failure with anemia, thrombocytopenia, and granulocytopenia. It may also infiltrate different organs, including liver, spleen, lymph nodes, CNS, kidneys, and gonads. Treatment options for patients with acute leukemia include chemotherapy, radiation therapy, targeted therapy, immunotherapy and stem cell or bone marrow transplantation. Finding a cure is a realistic goal for both ALL and AML, especially in younger patients. However, long-term survival is reachable in about one half of patients with acute lymphoblastic leukemia and in the minority of patients with acute myeloid leukemia. Current research, such as new genes whose protein products are suitable for targeted therapy and new strategies of immunotherapy, is focalized to

Most important acknowledgements go to all the contributing authors for their assistance, eagerness, support, and their expert scientific reviews. I thank my colleagues, Lipa Cicin-Sain, Maja Matulic and Mladen Paradzik for the motivating and inspiring discussions. Thanks to my husband Darko for taking care of our children during the long days and weekends while I worked on this project, and for his understanding and encouragement. Last but not least, I am most grateful to the publishing process managers Petra Nenadic and Ana Pantar for their outstanding work, great support and indispensable help at every phase in the preparation of this

> **Prof. Mariastefania Antica, PhD**  Rudjer Boskovic Institute,

> > Zagreb, Croatia

improve therapy for this challenging disease.

**Acknowledgments** 

book.

**Part 1** 

**Introduction** 

## **Part 1**

### **Introduction**

**1** 

*Yemen* 

Gamal Abdul-Hamid

*University of Aden/Hematology Unit* 

**Classification of Acute Leukemia** 

Acute leukemia is a proliferation of immature bone marrow-derived cells (blasts) that may also involve peripheral blood or solid organs. The percentage of bone marrow blast cells required for a diagnosis of acute leukemia has traditionally been set arbitrarily at 30% or more. However, more recently proposed classification systems have lowered the blast cell count to 20% for many leukemia types, and do not require any minimum blast cell percentage when

(1) by morphology and cytochemistry supplemented by immunophenotyping, as proposed by French-American-British (FAB) group (Bennet et al 1976); (2) Proposed World Health Organization Classification of Acute Leukemia (Harris et al 1999); (3) by immunophenotyping alone, as proposed by the European Group for the immunological classification of leukemias

The traditional classification of acute leukemia used criteria proposed by the French– American–British Cooperative Group (FAB) , using the 30% bone marrow blast cell cutoff (Bennett *et al*, 1985). This classification system originally distinguished different leukemia types by morphologic features and cytochemical studies, particularly myeloperoxidase (or Sudan black B) and non-specific esterase staining. It was revised to include leukemia types that could only be accurately identified with the addition of immunophenotyping or electron microscopic studies (Bennett *et al.*, 1991). Although the FAB classification failed to distinguish immunophenotypic groups of acute lymphoblastic leukemias, did not recognize the significance of myelodysplastic changes in acute myeloid leukemias or cytogenetic abnormalities in either leukemia type, and resulted in some subcategories of little clinical significance, this system provided very clear guidelines for classification. In addition, some distinct leukemia subtypes, particularly acute promyelocytic leukemia and acute myeloid leukemia with abnormal eosinophils, were found to correlate with specific cytogenetic aberrations and had unique clinical features, and those remain in recently proposed

Acute myelogenous leukemia (AML) was based on how leukemic blasts, the predominant cell in the disease process, recapitulate normal hematopoiesis. Are blasts in a given case myeloblasts, monoblasts, megakaryoblasts, etc., and are they un-, minimally, or moderately

certain morphologic and cytogenetic features are present.

(EGIL) (Bene et al 1995 & Hayhoe FG 1988).

classification systems.

differentiated.

**2. Acute leukemia can be classified in many ways** 

**1. Introduction** 

### **Classification of Acute Leukemia**

### Gamal Abdul-Hamid

*University of Aden/Hematology Unit Yemen* 

#### **1. Introduction**

Acute leukemia is a proliferation of immature bone marrow-derived cells (blasts) that may also involve peripheral blood or solid organs. The percentage of bone marrow blast cells required for a diagnosis of acute leukemia has traditionally been set arbitrarily at 30% or more. However, more recently proposed classification systems have lowered the blast cell count to 20% for many leukemia types, and do not require any minimum blast cell percentage when certain morphologic and cytogenetic features are present.

#### **2. Acute leukemia can be classified in many ways**

(1) by morphology and cytochemistry supplemented by immunophenotyping, as proposed by French-American-British (FAB) group (Bennet et al 1976); (2) Proposed World Health Organization Classification of Acute Leukemia (Harris et al 1999); (3) by immunophenotyping alone, as proposed by the European Group for the immunological classification of leukemias (EGIL) (Bene et al 1995 & Hayhoe FG 1988).

The traditional classification of acute leukemia used criteria proposed by the French– American–British Cooperative Group (FAB) , using the 30% bone marrow blast cell cutoff (Bennett *et al*, 1985). This classification system originally distinguished different leukemia types by morphologic features and cytochemical studies, particularly myeloperoxidase (or Sudan black B) and non-specific esterase staining. It was revised to include leukemia types that could only be accurately identified with the addition of immunophenotyping or electron microscopic studies (Bennett *et al.*, 1991). Although the FAB classification failed to distinguish immunophenotypic groups of acute lymphoblastic leukemias, did not recognize the significance of myelodysplastic changes in acute myeloid leukemias or cytogenetic abnormalities in either leukemia type, and resulted in some subcategories of little clinical significance, this system provided very clear guidelines for classification. In addition, some distinct leukemia subtypes, particularly acute promyelocytic leukemia and acute myeloid leukemia with abnormal eosinophils, were found to correlate with specific cytogenetic aberrations and had unique clinical features, and those remain in recently proposed classification systems.

Acute myelogenous leukemia (AML) was based on how leukemic blasts, the predominant cell in the disease process, recapitulate normal hematopoiesis. Are blasts in a given case myeloblasts, monoblasts, megakaryoblasts, etc., and are they un-, minimally, or moderately differentiated.

Classification of Acute Leukemia 5

It is found in all aged groups with highest incidence seen in adult and in infants less than a year old. Leukocytes in about 50% of patients at the time of diagnosis was increased. The predominant cell in the peripheral blood is usually a poorly differentiated myeloblast with finely reticulated chromatin and prominent nucleoli. Auer rods are found in the blast of 50% of the M1. If no evidence of granules or Auer rods is present, the blasts may resemble L2 lymphoblast. The myeloperoxidase or Sudan black B stains are positive in more than 3% of the blasts indicating granulocytes differentiation, the diagnosis is more likely AML-M1 than ALL (Bennett et al, 1976). PAS and alpha-naphthyl acetate esterase and naphthol AS-Desterase are negative. About 50% of the patients will have acquired clonal chromosome aberrations in the leukemic cells. CD13, 14, 15, 33 and CD34 myeloid antigens are frequently positive in M1 leukemia. The most common cytogenetic abnormalities are: t (9; 22) (q34; q11)

The presenting symptoms for M2 AML are similar to those of the M1 type. Leukocytes increased in 50% of patients. Myeloblast can usually be found in the blood smears and may be the predominant cell type. Pseudopelger–Huet and hypogranular neutrophils being most

Fig. 1. Acute myeloblastic leukemia AML –M0

Fig. 2. Acute myeloblastic leukemia **AML-M1** 

common cells are seen in M2.

**2.3 M2 Acute myeloblastic leukemia with maturation** 

**2.2 M1 Acute myeloblastic leukemia without maturation** 

The original classification scheme proposed by the French-American-British (FAB) Cooperative Group divides AML into 8 subtypes (M0 to M7) and ALL into 3 subtypes (L1 to L3). Although AML blasts evolve from common myeloid precursors, the 8 subtypes differ in degree of maturation (Table 1). As specified in the table, M0 designates AML with minimal morphologic or cytochemical differentiation, M1–2 AML with minimal or moderate granulocytic differentiation, M3 acute promyelocytic leukemia (APL), M4 AML with mixed myelomonocytic differentiation, M5a and M5b monoblastic leukemia with minimal or moderate differentiation, M6a myeloid leukemia with dysplastic background erythropoiesis, M6b acute erythroblastic leukemia, and M7 acute megakaryoblastic leukemia. The FAB classification of ALL includes 3 subtypes (L1 to L3), which are differentiated based on morphology, including cell size, prominence of nucleoli, and the amount and appearance of cytoplasm (Table 2). Approximately 75% of adult ALL cases have blasts with the B -cell phenotype, and 25% have blasts with the T-cell phenotype. The FAB classification of ALL and AML is based on morphology and cytochemical staining of blasts (Cheson et al 1990).

M0 AML with no Romanowsky or cytochemical evidence of differentiation M1 Myeloblastic leukemia with little maturation M2 Myeloblastic leukemia with maturation M3 Acute promyelocytic leukemia (APL) M3h APL, hypergranular variant M3v APL, microgranular variant M4 Acute myelomonocytic leukemia (AMML) M4eo AMML with dysplastic marrow eosinophils M5 Acute monoblastic leukemia (AMoL) M5a AMoL, poorly differentiated M5b AMoL, differentiated M6 "Erythroleukemia" M6a AML with erythroid dysplasia M6b Erythroleukemia M7 Acute megakaryoblastic leukemia (AMkL)

Table 1. French-American-British (FAB) Classification of Acute Myelogenous Leukemia

#### **2.1 M0 Acute myeloblastic leukemia with minimally differentiated**

AML-M0 is most common in adult patients. Accounts for approximately 5-10% of all AML patients. WBCs show Leukocytosis in 40% and > 50% with leukocytopenia. The diagnosis is made if less than 3% of the blasts are positive for peroxidase or the Sudan black B reaction and if the Blasts are positive for the myeloid-associated markers CD13, 14, CD15 or CD33, CD34 and negative for B or T lineage marker (CD3, CD10, CD19 and CD5). Bone marrow aspirated was hypercellular in all patients and contained a large number of leukemic blasts (Bennette JM 1991). Almost no mature myeloid cells were seen. The blasts were small to medium-sized round cells with an eccentric nucleus. The nucleus often had a flattened shape and was sometimes lobulated or cleaved and contained fine chromatin with several distinct nucleoli. The cytoplasm was lightly basophilic without granules. Auer rods are not found**.**

The original classification scheme proposed by the French-American-British (FAB) Cooperative Group divides AML into 8 subtypes (M0 to M7) and ALL into 3 subtypes (L1 to L3). Although AML blasts evolve from common myeloid precursors, the 8 subtypes differ in degree of maturation (Table 1). As specified in the table, M0 designates AML with minimal morphologic or cytochemical differentiation, M1–2 AML with minimal or moderate granulocytic differentiation, M3 acute promyelocytic leukemia (APL), M4 AML with mixed myelomonocytic differentiation, M5a and M5b monoblastic leukemia with minimal or moderate differentiation, M6a myeloid leukemia with dysplastic background erythropoiesis, M6b acute erythroblastic leukemia, and M7 acute megakaryoblastic leukemia. The FAB classification of ALL includes 3 subtypes (L1 to L3), which are differentiated based on morphology, including cell size, prominence of nucleoli, and the amount and appearance of cytoplasm (Table 2). Approximately 75% of adult ALL cases have blasts with the B -cell phenotype, and 25% have blasts with the T-cell phenotype. The FAB classification of ALL and

AML is based on morphology and cytochemical staining of blasts (Cheson et al 1990).

Table 1. French-American-British (FAB) Classification of Acute Myelogenous Leukemia

AML-M0 is most common in adult patients. Accounts for approximately 5-10% of all AML patients. WBCs show Leukocytosis in 40% and > 50% with leukocytopenia. The diagnosis is made if less than 3% of the blasts are positive for peroxidase or the Sudan black B reaction and if the Blasts are positive for the myeloid-associated markers CD13, 14, CD15 or CD33, CD34 and negative for B or T lineage marker (CD3, CD10, CD19 and CD5). Bone marrow aspirated was hypercellular in all patients and contained a large number of leukemic blasts (Bennette JM 1991). Almost no mature myeloid cells were seen. The blasts were small to medium-sized round cells with an eccentric nucleus. The nucleus often had a flattened shape and was sometimes lobulated or cleaved and contained fine chromatin with several distinct nucleoli. The cytoplasm was lightly basophilic without granules. Auer rods are not

**2.1 M0 Acute myeloblastic leukemia with minimally differentiated**

M0 AML with no Romanowsky or cytochemical evidence of differentiation

M1 Myeloblastic leukemia with little maturation M2 Myeloblastic leukemia with maturation M3 Acute promyelocytic leukemia (APL)

M4 Acute myelomonocytic leukemia (AMML) M4eo AMML with dysplastic marrow eosinophils

M7 Acute megakaryoblastic leukemia (AMkL)

M5 Acute monoblastic leukemia (AMoL)

M3h APL, hypergranular variant M3v APL, microgranular variant

M5a AMoL, poorly differentiated M5b AMoL, differentiated M6 "Erythroleukemia"

M6a AML with erythroid dysplasia

M6b Erythroleukemia

found**.**

Fig. 1. Acute myeloblastic leukemia AML –M0

#### **2.2 M1 Acute myeloblastic leukemia without maturation**

It is found in all aged groups with highest incidence seen in adult and in infants less than a year old. Leukocytes in about 50% of patients at the time of diagnosis was increased. The predominant cell in the peripheral blood is usually a poorly differentiated myeloblast with finely reticulated chromatin and prominent nucleoli. Auer rods are found in the blast of 50% of the M1. If no evidence of granules or Auer rods is present, the blasts may resemble L2 lymphoblast. The myeloperoxidase or Sudan black B stains are positive in more than 3% of the blasts indicating granulocytes differentiation, the diagnosis is more likely AML-M1 than ALL (Bennett et al, 1976). PAS and alpha-naphthyl acetate esterase and naphthol AS-Desterase are negative. About 50% of the patients will have acquired clonal chromosome aberrations in the leukemic cells. CD13, 14, 15, 33 and CD34 myeloid antigens are frequently positive in M1 leukemia. The most common cytogenetic abnormalities are: t (9; 22) (q34; q11)

Fig. 2. Acute myeloblastic leukemia **AML-M1** 

#### **2.3 M2 Acute myeloblastic leukemia with maturation**

The presenting symptoms for M2 AML are similar to those of the M1 type. Leukocytes increased in 50% of patients. Myeloblast can usually be found in the blood smears and may be the predominant cell type. Pseudopelger–Huet and hypogranular neutrophils being most common cells are seen in M2.

Classification of Acute Leukemia 7

(ATRA) has improved significantly the treatment of AML-M3 in this regard; early mortality

Cytochemistry: Peroxidase (MPO) and Sudan black B are strong positive. The periodic acid Schiff (PAS) is negative and Nonspecific esterase is also weak positive . The MPO reaction is

Immunological studies demonstrate positivity with CD13, CD15, CD1 and CD33 myeloid antigens. Cytogenetic studies have revealed a high prevalence (almost 50%) of the chromosomal translocation t(15; 17) associated with both AML M3 and M3 variant . **M3** AML with t(15;17) is usually characterized by the association of the lymphoid marker, CD2

and CD19, with myeloid markers and the negativity of HLA-DR and CD34.

as a result of DIC is substantially reduced.

also strong positive in the AML-M3 variant.

Fig. 4. Promyelocytic leukemia AML-M3

Fig. 5. Acute myelomonocytic leukemia M4

**2.5 M4 Acute myelomonocytic leukemia (AMML)** 

It is distinguished from M1, M2, and M3 by an increased proportion of leukemia monocytic cells in the bone marrow or blood or both. Gingival hyperplasia with gingival bleeding is present. Serum and urine levels of muramidase (lysozyme) are usually elevated because of the monocytic proliferation. The leukocyte count is usually increased monocytic cells

The bone marrow is hypercellular and types I and II myeloblasts make up from 30-83% of the promyelocytes to mature segmented cells. The monocytic component is less than 20%, differentiating M2 from M4. Basophils in some patient (M2 baso) was increased. Eosinophils and their precursors may be abundant, and in some cases accounts for up to 15% of myelogram (Berger and Flandrin, 1984). The characteristic that distinguishes AML-M2 from AML-M1 is the presence of maturation at or beyond the promyelocyte stage. Abnormal neutrophil maturation appears to be an integral part of AML-M2 with t(8;21) translocation. The neutrophils may show many abnormal nuclear segmentations and Auer rods.

Cytochemistry; Myeloperoxidase (MPO) reaction in blast cells gives the same result as in AML-M1, but the reaction is often of little practical value because the granulocytic nature of AML-M2 is usually demonstrated clearly by the presence of maturing cells in the granulocytic series. Sodium fluoride does not inhibit esterase. PAS and nonspecific esterase are negative. Positive reaction with CD13 and CD15 antigens are frequently seen in cases of M2.

Fig. 3. Acute myeloblastic leukemia **AML-M2**

#### **2.4 M3 Acute promyelocytic leukemia (APL)**

The median age and survival average of APL is about 18 months and occurred in younger adult. M3 is of particular interest because it results in the fusion of a truncated retinoic acid receptor alpha (RAR-alpha) gene on chromosome 17 to a transcription unit called PML (for promyelocytic leukemia) on chromosome 15. It is interesting to note that high doses of the vitamin A derivative all-trans-retinoic acid are able to overcome thus block in differentiation both in vitro and in vivo and this agent has been successfully used to induce remission in patients.

A "variant" form of M3 (Bennett et al, 1980) is characterized by paucity of granules within the promyelocytic blasts and should not be confused with monocytic leukemia. The blasts are large with abundant cytoplasm, and the nucleus is usually irregular. The nucleus is often bilobed or markedly indented and a nucleolus can be seen in each lobe. The cytoplasm is completely occupied by closely packed large granules, staining bright pink, red or purple. Cells containing bundles of Auer rods "faggots" randomly distributed in the cytoplasm are characteristic, but are not present in all cases.

It is believed that the release of large numbers of promyelocytic granules containing a procoagulant initiate disseminated intravascular clotting (DIC). This is the most serious complication of M3 AML occurs frequently in both AML-M3 as well as AML-M3 variant (McKenna et al, 1982) . Initial therapy with the differentiating agent all-trans-retinoic acid

The bone marrow is hypercellular and types I and II myeloblasts make up from 30-83% of the promyelocytes to mature segmented cells. The monocytic component is less than 20%, differentiating M2 from M4. Basophils in some patient (M2 baso) was increased. Eosinophils and their precursors may be abundant, and in some cases accounts for up to 15% of myelogram (Berger and Flandrin, 1984). The characteristic that distinguishes AML-M2 from AML-M1 is the presence of maturation at or beyond the promyelocyte stage. Abnormal neutrophil maturation appears to be an integral part of AML-M2 with t(8;21) translocation.

Cytochemistry; Myeloperoxidase (MPO) reaction in blast cells gives the same result as in AML-M1, but the reaction is often of little practical value because the granulocytic nature of AML-M2 is usually demonstrated clearly by the presence of maturing cells in the granulocytic series. Sodium fluoride does not inhibit esterase. PAS and nonspecific esterase are negative.

The median age and survival average of APL is about 18 months and occurred in younger adult. M3 is of particular interest because it results in the fusion of a truncated retinoic acid receptor alpha (RAR-alpha) gene on chromosome 17 to a transcription unit called PML (for promyelocytic leukemia) on chromosome 15. It is interesting to note that high doses of the vitamin A derivative all-trans-retinoic acid are able to overcome thus block in differentiation both in vitro and in vivo and this agent has been successfully used to induce remission in

A "variant" form of M3 (Bennett et al, 1980) is characterized by paucity of granules within the promyelocytic blasts and should not be confused with monocytic leukemia. The blasts are large with abundant cytoplasm, and the nucleus is usually irregular. The nucleus is often bilobed or markedly indented and a nucleolus can be seen in each lobe. The cytoplasm is completely occupied by closely packed large granules, staining bright pink, red or purple. Cells containing bundles of Auer rods "faggots" randomly distributed in the cytoplasm are

It is believed that the release of large numbers of promyelocytic granules containing a procoagulant initiate disseminated intravascular clotting (DIC). This is the most serious complication of M3 AML occurs frequently in both AML-M3 as well as AML-M3 variant (McKenna et al, 1982) . Initial therapy with the differentiating agent all-trans-retinoic acid

The neutrophils may show many abnormal nuclear segmentations and Auer rods.

Positive reaction with CD13 and CD15 antigens are frequently seen in cases of M2.

Fig. 3. Acute myeloblastic leukemia **AML-M2**

**2.4 M3 Acute promyelocytic leukemia (APL)** 

characteristic, but are not present in all cases.

patients.

(ATRA) has improved significantly the treatment of AML-M3 in this regard; early mortality as a result of DIC is substantially reduced.

Cytochemistry: Peroxidase (MPO) and Sudan black B are strong positive. The periodic acid Schiff (PAS) is negative and Nonspecific esterase is also weak positive . The MPO reaction is also strong positive in the AML-M3 variant.

Immunological studies demonstrate positivity with CD13, CD15, CD1 and CD33 myeloid antigens. Cytogenetic studies have revealed a high prevalence (almost 50%) of the chromosomal translocation t(15; 17) associated with both AML M3 and M3 variant . **M3** AML with t(15;17) is usually characterized by the association of the lymphoid marker, CD2 and CD19, with myeloid markers and the negativity of HLA-DR and CD34.

Fig. 4. Promyelocytic leukemia AML-M3

Fig. 5. Acute myelomonocytic leukemia M4

#### **2.5 M4 Acute myelomonocytic leukemia (AMML)**

It is distinguished from M1, M2, and M3 by an increased proportion of leukemia monocytic cells in the bone marrow or blood or both. Gingival hyperplasia with gingival bleeding is present. Serum and urine levels of muramidase (lysozyme) are usually elevated because of the monocytic proliferation. The leukocyte count is usually increased monocytic cells

Classification of Acute Leukemia 9

M6 is a rare form of leukemia that primarily affects the peripheral cells. It is nonsexist in children. The clinical manifestations are similar to other types of AML. The most frequent presentation is bleeding. The most dominant changes in the peripheral blood are anemia with sticking poikilocytosis and anisocytosis. Nucleated red cells demonstrate abnormal nuclear configuration. The leukocytes and platelets are usually decreased. The diagnosis of erythroleukemia can be made when more than 50% of all nucleated bone marrow cells are erythroid and 30% or more of all remaining nonerythroid cells are type I or type II blast cells (Bennett et al, 1985). The erythroblast is abnormal with bizarre morphologic features. Giant multilobular or multinucleated forms are common. Other features are; fragmentation, Howell-Jolly bodies, ring sideroblast, megaloblastic and dyserythropoiesis changes are common. The cytochemistry of erythroblasts are normally PAS negative but in AML-M6, erythroblasts especially pronormoblast demonstrates coarse positivity of PAS. Blast cells express a variety of myeloid associated antigens such as CD13, CD33, anti-MPO with or without expression of precursor-cell markers as CD34, HLA-Dr determinants as for blast cells from other AML subtypes. In M6-variant forms, the more differentiated cells can be

detected by the expression of glycophorin A and the absence of myeloid markers.

M7 is rare. It occurs as a leukemia transformation of chronic granulocytic leukemia (CGL) and myelodysplastic syndrome (MDS). Pancytopenia is characteristic at initial diagnosis. Peripheral blood shows micromegakaryocytes and undifferentiated blasts. Bone marrow dry tap is common. Bone marrow biopsy show increased fibroblasts and/or increased reticulin and presence of greater than 30% blast cells. The diagnosis of M7 should be suspected when the blast cells show cytoplasmic protrusion or budding. As bone marrow smears obtained by aspiration may not be adequate to make a diagnosis, the peripheral blood films must be examined carefully for the presence of micromegakaryoblasts. Bone marrow biopsy sections are usually necessary and show a prominent reticulin fibrosis and

Cytochemistry: Peroxidase is negative, PAS +/-, Esterase +/- and positive acid phospatase. Cytochemical positivity for -naphthyl acetate esterase reaction and negative reaction with -naphthyl butyrate esterase is unique to megakaryoblast. (Monocytes react positively with

**2.7 M6 "Erythroleukemia"** 

Fig. 7. M6 "Erythroleukemia"

excessive numbers of small blasts.

both esterase substrates).

**2.8 M7 Acute megakaryoblastic leukemia (AMkL)** 

(monoblast, promoncytes, monocytes), are increased to 5000/L or more. Anemia and thrombocytopenia are present in almost all cases. The marrow differs from M1, M2 and M3 in those monocytic cells exceed 20% of the nonerythroid nucleated cells. The sum of the myelocytic cells including myeloblasts, promyelocytes and later granulocytes is >20% and <80% of nonerythroid cells. This bone marrow picture together with a peripheral blood monocyte count of 5000/L or more is compatible with a diagnosis of M4.

Confirmation of the monocytic component of this subgroup requires cytochemistry. The profile includes positive reactions for sudan black B or peroxidase and both specific and non-specific esterase. A few cases of M4 AML are characterized by increased marrow eosinophils and classified as M4e (Berger et al 1985) . Immunological studies demonstrate positivity with CD13, CD33, CD11b and CD14. Cytogenetic: inv(16) (p13; q22) and del (16)(q22) .

#### **2.6 M5 Acute monoblastic leukemia (AMoL)**

Common findings are weakness, bleeding and a diffuse erythematous skin rash. There is a high frequency of extramedulary infiltration of the lungs, colon, meninges, lymphnodes, bladder and larynx and gingival hyperplasia. Serum and urinary muramidase levels are often extremely high. The one criterion for a diagnosis of M5 is that 80% or more of all nonerythroid cells in the bone marrow are monocytic cells. There are two distinct forms 5a (maturation index <4%) and 5b (maturation index > 4%).M5a: Granulocyte <20% and Monocyte >80% >80% monoblast. M5b: Granulocyte <20% and Monocyte >80% <80% monoblast (Characterized by the presence of all developmental stages of monocytes; monoblast, promonocyte, monocyte)

Cytochemistry: Non-specific esterase stains and alpha-naphthyl esterase are positive and PAS is negative. Myeloperoxidase and Sudan black are weak diffuse activity in the monoblast. The use of alph-naphthyl butyrate esterase (ANBE) is advantageous because of its greater degree of specificity and stronger reaction, and also because sodium fluoride inhibition is not required (Shibata et al, 1985). Immunological studies demonstrate positivity with CD11b and CD14. There is a strong association between AML M5/M4 and deletion and translocations involving band 11q23.

Fig. 6. M5 Acute monoblastic leukemia (AMoL)

(monoblast, promoncytes, monocytes), are increased to 5000/L or more. Anemia and thrombocytopenia are present in almost all cases. The marrow differs from M1, M2 and M3 in those monocytic cells exceed 20% of the nonerythroid nucleated cells. The sum of the myelocytic cells including myeloblasts, promyelocytes and later granulocytes is >20% and <80% of nonerythroid cells. This bone marrow picture together with a peripheral blood

Confirmation of the monocytic component of this subgroup requires cytochemistry. The profile includes positive reactions for sudan black B or peroxidase and both specific and non-specific esterase. A few cases of M4 AML are characterized by increased marrow eosinophils and classified as M4e (Berger et al 1985) . Immunological studies demonstrate positivity with CD13, CD33, CD11b and CD14. Cytogenetic: inv(16) (p13; q22) and del

Common findings are weakness, bleeding and a diffuse erythematous skin rash. There is a high frequency of extramedulary infiltration of the lungs, colon, meninges, lymphnodes, bladder and larynx and gingival hyperplasia. Serum and urinary muramidase levels are often extremely high. The one criterion for a diagnosis of M5 is that 80% or more of all nonerythroid cells in the bone marrow are monocytic cells. There are two distinct forms 5a (maturation index <4%) and 5b (maturation index > 4%).M5a: Granulocyte <20% and Monocyte >80% >80% monoblast. M5b: Granulocyte <20% and Monocyte >80% <80% monoblast (Characterized by the presence of all developmental stages of monocytes;

Cytochemistry: Non-specific esterase stains and alpha-naphthyl esterase are positive and PAS is negative. Myeloperoxidase and Sudan black are weak diffuse activity in the monoblast. The use of alph-naphthyl butyrate esterase (ANBE) is advantageous because of its greater degree of specificity and stronger reaction, and also because sodium fluoride inhibition is not required (Shibata et al, 1985). Immunological studies demonstrate positivity with CD11b and CD14. There is a strong association between AML M5/M4 and deletion

monocyte count of 5000/L or more is compatible with a diagnosis of M4.

**2.6 M5 Acute monoblastic leukemia (AMoL)** 

monoblast, promonocyte, monocyte)

and translocations involving band 11q23.

Fig. 6. M5 Acute monoblastic leukemia (AMoL)

(16)(q22) .

#### **2.7 M6 "Erythroleukemia"**

M6 is a rare form of leukemia that primarily affects the peripheral cells. It is nonsexist in children. The clinical manifestations are similar to other types of AML. The most frequent presentation is bleeding. The most dominant changes in the peripheral blood are anemia with sticking poikilocytosis and anisocytosis. Nucleated red cells demonstrate abnormal nuclear configuration. The leukocytes and platelets are usually decreased. The diagnosis of erythroleukemia can be made when more than 50% of all nucleated bone marrow cells are erythroid and 30% or more of all remaining nonerythroid cells are type I or type II blast cells (Bennett et al, 1985). The erythroblast is abnormal with bizarre morphologic features. Giant multilobular or multinucleated forms are common. Other features are; fragmentation, Howell-Jolly bodies, ring sideroblast, megaloblastic and dyserythropoiesis changes are common. The cytochemistry of erythroblasts are normally PAS negative but in AML-M6, erythroblasts especially pronormoblast demonstrates coarse positivity of PAS. Blast cells express a variety of myeloid associated antigens such as CD13, CD33, anti-MPO with or without expression of precursor-cell markers as CD34, HLA-Dr determinants as for blast cells from other AML subtypes. In M6-variant forms, the more differentiated cells can be detected by the expression of glycophorin A and the absence of myeloid markers.

Fig. 7. M6 "Erythroleukemia"

#### **2.8 M7 Acute megakaryoblastic leukemia (AMkL)**

M7 is rare. It occurs as a leukemia transformation of chronic granulocytic leukemia (CGL) and myelodysplastic syndrome (MDS). Pancytopenia is characteristic at initial diagnosis. Peripheral blood shows micromegakaryocytes and undifferentiated blasts. Bone marrow dry tap is common. Bone marrow biopsy show increased fibroblasts and/or increased reticulin and presence of greater than 30% blast cells. The diagnosis of M7 should be suspected when the blast cells show cytoplasmic protrusion or budding. As bone marrow smears obtained by aspiration may not be adequate to make a diagnosis, the peripheral blood films must be examined carefully for the presence of micromegakaryoblasts. Bone marrow biopsy sections are usually necessary and show a prominent reticulin fibrosis and excessive numbers of small blasts.

Cytochemistry: Peroxidase is negative, PAS +/-, Esterase +/- and positive acid phospatase. Cytochemical positivity for -naphthyl acetate esterase reaction and negative reaction with -naphthyl butyrate esterase is unique to megakaryoblast. (Monocytes react positively with both esterase substrates).

Classification of Acute Leukemia 11

confused with the blasts of acute myeloid leukemia. Approximately 66% of these cases of

**ALL-L3: Burkitt's lymphoma type**: Cells are large and homogenous in size, nuclear shape is round or oval. One to three prominent nucleoli and sometimes to 5 nuleoli are visible. Cytoplasm is deeply basophilic with vacuoles often prominent. Intense cytoplasmic basophilia is present in every cell, with prominent vacuolation in most. A high mitotic index is characteristic with presence of varying degrees of macrophage activity. Mature B-

ALL in patients older than 15 years are of type 2.

Fig. 9. Acute lymphoblastic leukemia L1

Fig. 10. Acute lymphoblastic leukemia L2

lymphoid markers are expressed by most cases.

Fig. 11. Acute lymphoblastic leukemia L3

The monoclonal antibodies that reacts with platelet glycoprotein Ib, IIb/IIIa and IIIb, using immunologic technique as well as CD41, CD42 and CD61 positivity.

There is no unique chromosomal abnormality associated with acute megakaryoblastic leukemia, with the exception of t(1;22)(p13;q13), which has been found almost exclusively in young children, less than 18 months old who do not have Downís syndrome.

Fig. 8. M7 Acute megakaryoblastic leukemia (AMkL)


Table 2. Morphologic Classification of Acute Lymphocytic Leukemia

**Acute lymphoblastic leukemia** (ALL) is divided in FAB L1 (children), L2 (older children and adult), and L3 (patients with leukemia secondary to Burkitt's lymphoma. These types are defined according to two criteria (1) the occurrence of individual cytologic features and (2) the degree of heterogeneity among the leukemic cells. These features considered are cell size, chromatin, nuclear shape, nucleoli, and degree of basophilia in the cytoplasm and the presence of cytoplasmic vacuolation (Bennett et al 1976).

**ALL-L1: Homogenous** cells **(Small cell):** One population of cells within the case. Small cells predominant, nuclear shape is regular with occasional cleft. Nuclear contents are rarely visible. Cytoplasm is moderately basophilic. L1 accounts 70% of patients. The L1 type is the acute leukemia that is common in childhood, with 74% of these cases occurring in children 15 years of age or younger.

**ALL-L2: Heterogeneous cells:** Large cells with an irregular nuclear shape, cleft in the nucleus are common. One or more large nucleoli are visible. Cytoplasm varies in colour and nuclear membrane irregularities. L2 accounts 27% of ALL patients. The FAB-L2 blast may be

The monoclonal antibodies that reacts with platelet glycoprotein Ib, IIb/IIIa and IIIb, using

There is no unique chromosomal abnormality associated with acute megakaryoblastic leukemia, with the exception of t(1;22)(p13;q13), which has been found almost exclusively in

**Morphologic Classification** 

**L3** Large cells with moderately abundant cytoplasm; regular, oval-to-round

prominent cytoplasmic basophilia and cytoplasmic vacuoles

**Acute lymphoblastic leukemia** (ALL) is divided in FAB L1 (children), L2 (older children and adult), and L3 (patients with leukemia secondary to Burkitt's lymphoma. These types are defined according to two criteria (1) the occurrence of individual cytologic features and (2) the degree of heterogeneity among the leukemic cells. These features considered are cell size, chromatin, nuclear shape, nucleoli, and degree of basophilia in the cytoplasm and the

**ALL-L1: Homogenous** cells **(Small cell):** One population of cells within the case. Small cells predominant, nuclear shape is regular with occasional cleft. Nuclear contents are rarely visible. Cytoplasm is moderately basophilic. L1 accounts 70% of patients. The L1 type is the acute leukemia that is common in childhood, with 74% of these cases occurring in children

**ALL-L2: Heterogeneous cells:** Large cells with an irregular nuclear shape, cleft in the nucleus are common. One or more large nucleoli are visible. Cytoplasm varies in colour and nuclear membrane irregularities. L2 accounts 27% of ALL patients. The FAB-L2 blast may be

clefting and indentation of nucleus; large and prominent nucleoli

**L1** Small cells with scant cytoplasm; nucleoli indistinct and not visible **L2** Large, heterogeneous cells with moderately abundant cytoplasm;

immunologic technique as well as CD41, CD42 and CD61 positivity.

Fig. 8. M7 Acute megakaryoblastic leukemia (AMkL)

**FAB Type Features of Blasts** 

nucleus; prominent nucleoli;

presence of cytoplasmic vacuolation (Bennett et al 1976).

15 years of age or younger.

Table 2. Morphologic Classification of Acute Lymphocytic Leukemia

young children, less than 18 months old who do not have Downís syndrome.

confused with the blasts of acute myeloid leukemia. Approximately 66% of these cases of ALL in patients older than 15 years are of type 2.

Fig. 9. Acute lymphoblastic leukemia L1

Fig. 10. Acute lymphoblastic leukemia L2

**ALL-L3: Burkitt's lymphoma type**: Cells are large and homogenous in size, nuclear shape is round or oval. One to three prominent nucleoli and sometimes to 5 nuleoli are visible. Cytoplasm is deeply basophilic with vacuoles often prominent. Intense cytoplasmic basophilia is present in every cell, with prominent vacuolation in most. A high mitotic index is characteristic with presence of varying degrees of macrophage activity. Mature Blymphoid markers are expressed by most cases.

Fig. 11. Acute lymphoblastic leukemia L3

Classification of Acute Leukemia 13

addition to discarding the L1-L3 terms, the new classification characterizes these heterogenous diseases based upon immunophenotype into 3 basic categories: precursor B-cell ALL, precursor T-cell ALL, and mature B-cell ALL (Burkitt lymphoma/leukemia) (Jaffe et al 2001)

WHO classification

 Hypodiploid Hyperdiploid, >50

Precursor B-cell ALL/LBL Cytogenetic subgroups t(9;22)(q34,q11),BCR/ABL t(v;11q23);MLL rearranged t(1;19)(q23;p13);PBX1/E2A t(12;21)(p13;q22);TEL/AML1

Precursor T-cell ALL/LBL

Mature B-cell leukemia/lymphoma ALL= acute lymphoblastic leukemia; LBL= lymphoblastic lymphoma; MLL= mixed lineage leukemia

Table 4. World Health Organization classification of acute lymphoblastic leukemia

infancy and are associated with both acute lymphoid and myeloid leukemias.

Abnormalities in chromosome number as well as structural rearrangements (translocations) occur commonly in ALL. Important cytogenetic abnormalities in precursor B-cell ALL that are associated with a poor prognosis include t(9;22) or Philadelphia chromosome-positive (Ph+) ALL, which increases in frequency with age; t(4;11); hypodiploidy, especially if <45% chromosomes ; and trisomy 8 in adult ALL. The t(4;11) results from a balanced translocation involving a gene on the long arm of chromosome 11 at band q23 (11q23), known as the mixed lineage leukemia (MLL) gene. MLL gene translocations occur most commonly in

**3. European Group for the Immunological classification of Leukemias (EGIL)** The European Group for the Immunological Classification of Leukemias (EGIL)(Bene MC et al 1995 & Hoelzer et al 2002)) has proposed that acute leukaemia be classified on the basis of immunophenotype alone. This classification has the strength that it suggests standardized criteria for defining a leukaemia as myeloid, T lineage, B lineage, or biphenotypic. It also suggests criteria for distinguishing biphenotypic leukaemia from AML with aberrant expression of lymphoid antigens, and from ALL with aberrant expression of myeloid antigens. However, a purely immunological classification has the disadvantage that discrete entities may fall into one of two categories; for example some cases of AML of FAB M2 subtype associated with t(8;21)(q22;q22) would be classified as "AML of myelomonocytic lineage", while others would be classified as "AML with lymphoid antigen expression," depending on whether or not a case showed aberrant expression of CD19. In addition, rare cases of acute leukaemia have been described which were clearly myeloid when assessed by cytology and cytochemistry but which did not express any of the commonly investigated myeloid antigens.

**Acute Myeloid Leukemia (AML) and Related Precursor Neoplasm AML with recurrent genetic abnormalities** AML with t(8;21)(q22;q22); RUNX1-RUNX1T1 AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11 Acute promyelocytic leukemia with t(15;17)(q22;q12); PML-RARA AML with t(9 ;11)(p22;q23); MLLT3-MLL AML with t(6;9)(p23;q34); DEK-NUP214 AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1-EVI1 AML with mutated NPM1 AML with mutated CEBPA AML with myelodysplasia-related changes Therapy-related myeloid neoplasms Myeloid sarcoma Myeloid proliferations related to Down syndrome Transient abnormal myelopoiesis Myeloid leukemia associated with Down syndrome Blastic plasmacytoid denderitic cell neoplasm

Table 3. World Health Organization Classification of Acute Myelogenous Leukemia (2008)

**AML** defined as ≥20% blasts in blood or bone marrow; however, clonal, recurring cytogenetic abnormalities should be considered AML regardless of blast percentage. Ongoing clinical trials may continue to use French-American-British (FAB) criteria of ≥30% blasts until completion of trial. FAB classification identified as M0 through M7.

The classification schemes by the World Health Organization (WHO) require the additional evaluation of the leukemic blasts by molecular analysis and flow cytometry (Harris NL 1997 & Brunangelo Falini 2010, Sachdeva et al 2006). The results of these 4 methods of evaluation (i.e, morphology, staining, molecular analysis, flow cytometry) not only differentiate ALL from AML, but also categorize the subtypes of acute leukemia. Table 3 summarizes the new classification of AML as proposed by WHO, Knowing the subtype of a patient's leukemia helps in predicting the clinical behavior of the disease and the prognosis, and in making treatment recommendations. This classification also improves the reproducibility of diagnoses and stresses the heterogeneity of the subtypes of AML and ALL (Vardiman 2009). Recent advances in molecular biology have shown that various subtypes of AML and ALL behave differently and should not be treated similarly. For example, the identification of M3 AML (acute promyelocytic leukemia) is crucial because it is associated with disseminated intravascular coagulation (DIC), and retinoic acid, in addition to chemotherapy, is the treatment of choice.

The two most significant differences between the FAB and the WHO classifications are:

(a) A lower blast threshold for the diagnosis of AML: The WHO defines AML when the blast percentage reaches 20% in the bone marrow.

(b) Patients with recurring clonal cytogenetic abnormalities should be considered to have AML regardless of the blast percentage (8;21)(q22;q22), t(16;16)(p13;q22), inv(16)(p13;q22), or t(15;17)(q22;q12) (Arber DA et al 2008abc, Weinberg OK et al 2009).

The world Health Organization (WHO) classification has changed the grouping of ALL to reflect increased understanding of the biology and molecular pathogenesis of the diseases. In

Table 3. World Health Organization Classification of Acute Myelogenous Leukemia (2008) **AML** defined as ≥20% blasts in blood or bone marrow; however, clonal, recurring cytogenetic abnormalities should be considered AML regardless of blast percentage. Ongoing clinical trials may continue to use French-American-British (FAB) criteria of ≥30%

The classification schemes by the World Health Organization (WHO) require the additional evaluation of the leukemic blasts by molecular analysis and flow cytometry (Harris NL 1997 & Brunangelo Falini 2010, Sachdeva et al 2006). The results of these 4 methods of evaluation (i.e, morphology, staining, molecular analysis, flow cytometry) not only differentiate ALL from AML, but also categorize the subtypes of acute leukemia. Table 3 summarizes the new classification of AML as proposed by WHO, Knowing the subtype of a patient's leukemia helps in predicting the clinical behavior of the disease and the prognosis, and in making treatment recommendations. This classification also improves the reproducibility of diagnoses and stresses the heterogeneity of the subtypes of AML and ALL (Vardiman 2009). Recent advances in molecular biology have shown that various subtypes of AML and ALL behave differently and should not be treated similarly. For example, the identification of M3 AML (acute promyelocytic leukemia) is crucial because it is associated with disseminated intravascular coagulation (DIC), and retinoic acid, in addition to chemotherapy, is the

The two most significant differences between the FAB and the WHO classifications are: (a) A lower blast threshold for the diagnosis of AML: The WHO defines AML when the

or t(15;17)(q22;q12) (Arber DA et al 2008abc, Weinberg OK et al 2009).

(b) Patients with recurring clonal cytogenetic abnormalities should be considered to have AML regardless of the blast percentage (8;21)(q22;q22), t(16;16)(p13;q22), inv(16)(p13;q22),

The world Health Organization (WHO) classification has changed the grouping of ALL to reflect increased understanding of the biology and molecular pathogenesis of the diseases. In

blasts until completion of trial. FAB classification identified as M0 through M7.

**Acute Myeloid Leukemia (AML) and Related Precursor Neoplasm** 

AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB-MYH11 Acute promyelocytic leukemia with t(15;17)(q22;q12); PML-RARA

AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1-EVI1

**AML with recurrent genetic abnormalities** AML with t(8;21)(q22;q22); RUNX1-RUNX1T1

AML with t(9 ;11)(p22;q23); MLLT3-MLL AML with t(6;9)(p23;q34); DEK-NUP214

AML with myelodysplasia-related changes Therapy-related myeloid neoplasms

Myeloid proliferations related to Down syndrome

Myeloid leukemia associated with Down syndrome Blastic plasmacytoid denderitic cell neoplasm

blast percentage reaches 20% in the bone marrow.

AML with mutated NPM1 AML with mutated CEBPA

Transient abnormal myelopoiesis

Myeloid sarcoma

treatment of choice.

addition to discarding the L1-L3 terms, the new classification characterizes these heterogenous diseases based upon immunophenotype into 3 basic categories: precursor B-cell ALL, precursor T-cell ALL, and mature B-cell ALL (Burkitt lymphoma/leukemia) (Jaffe et al 2001)


Table 4. World Health Organization classification of acute lymphoblastic leukemia

Abnormalities in chromosome number as well as structural rearrangements (translocations) occur commonly in ALL. Important cytogenetic abnormalities in precursor B-cell ALL that are associated with a poor prognosis include t(9;22) or Philadelphia chromosome-positive (Ph+) ALL, which increases in frequency with age; t(4;11); hypodiploidy, especially if <45% chromosomes ; and trisomy 8 in adult ALL. The t(4;11) results from a balanced translocation involving a gene on the long arm of chromosome 11 at band q23 (11q23), known as the mixed lineage leukemia (MLL) gene. MLL gene translocations occur most commonly in infancy and are associated with both acute lymphoid and myeloid leukemias.

### **3. European Group for the Immunological classification of Leukemias (EGIL)**

The European Group for the Immunological Classification of Leukemias (EGIL)(Bene MC et al 1995 & Hoelzer et al 2002)) has proposed that acute leukaemia be classified on the basis of immunophenotype alone. This classification has the strength that it suggests standardized criteria for defining a leukaemia as myeloid, T lineage, B lineage, or biphenotypic. It also suggests criteria for distinguishing biphenotypic leukaemia from AML with aberrant expression of lymphoid antigens, and from ALL with aberrant expression of myeloid antigens. However, a purely immunological classification has the disadvantage that discrete entities may fall into one of two categories; for example some cases of AML of FAB M2 subtype associated with t(8;21)(q22;q22) would be classified as "AML of myelomonocytic lineage", while others would be classified as "AML with lymphoid antigen expression," depending on whether or not a case showed aberrant expression of CD19. In addition, rare cases of acute leukaemia have been described which were clearly myeloid when assessed by cytology and cytochemistry but which did not express any of the commonly investigated myeloid antigens.

Classification of Acute Leukemia 15

lymphoid and myeloid lineages among those of earlier stages of cell differentiation, plus some non-specific but stem-cell markers. The system introduced a modified terminology specific to each 'maturation' step within the B- or T-cell lineage (EGIL 1995) and was confirmed as

Biphenotypic acute leukemia (BAL), or acute leukemia with a single population of blasts

The scoring systems proposed by Catovsky *et al*. and by the EGIL (Bene 1995) allowed for a better definition of biphenotypic acute leukemia (BAL), clearly distinguishing them from classical AL expressing aberrantly one or two markers of another lineage. However, increasing evidence suggests that this system has limitations, as acknowledged by the 2008 World Health Organization (WHO) Classification of Tumors of Hematopoietic and Lymphoid Tissues. Although substantially improved in relation to the EGIL, the new WHO Classification is still not optimal for guiding the clinical management of patients with BAL. Typical examples of such aberrations are the expression of CD15 on B-ALL(Maynadie et al 1997), or of CD2 on acute promyelocytic –AML( Albano 2006). In 1998, the EGIL further refined this scoring system by attributing one point for the expression of CD117, after showing the strong relationship of this marker with engagement in the myeloid lineage (Bene MC et al 1998) . To identify BAL, it is therefore necessary to consider aberrant coexpression of markers usually associated to different lineages, with a score higher than 2 in

B-lineage T-lineage Myeloid lineage

IgM Anti TCR Lisozyme

CD10 CD5 CD33 CD20 CD8 CD65 CD10

CD24 CD17 CD15

Biphenotypic acute leukemia is defined when scores are >2 for the myeloid lineage and >1 for the lymphoid lineage. In some T-ALL cases, clonality of TCR alphabeta rearrangements

The prognosis of biphenotypic acute leukemia patients is poor when compared with de novo acute myeloid leukemia or acute lymphoblastic leukemia. Biphenotypic acute leukemia patients showed a much higher incidence of CD34 antigen expression, complex abnormal karyotype, extramedullary infiltration, relapse, and resistance to therapy after

CD10 CD64, CD117

adequate for both diagnosis and subclassification of ALL (Thalhammer-Scherrer 2002).

**4. European Group for the Immunological characterization of Leukemias** 

**(EGIL) classification for biphenotypic acute leukemia**

more than one lineage (Zhao XF et al 2009)

CD22

relapse (Xu XQ et al 2009).

coexpressing markers of two different lineages, is a rare clinical entity.

2 point CD79 CD3 MPO

1 point CD19 CD2 CD13

0.5 point TdT TdT CD14

can now be assessed cytoflourmetrically (Langerak 2001, Xu XQ et al 2009).

Table 6. EGIL Scoring system for biphenotypic acute leukemia

**Precursor B-lymphoblastic leukemia** ( HLA-DR+, TdT+, CD19+, and/or CD79a+, and/or CD22+, and/or CD34+). This type of ALL accounts for around 75% of adult cases and is subdivided into the following groups:


#### **Precursor T-lymphoblastic leukemia**

Cells are TdT+ in addition to cytoplasmic CD3+ and CD34+. This type of ALL accounts for around 25% of adult cases and is subdivided into:


Table 5. European Group for the Immunological Characterization of Leukemias (EGIL) classification of acute lymphocytic leukemia (Hoelzer 2002)

The consensus considers a 20% minimum threshold to define a positive reaction of blast cells to a given monoclonal antibody. Roughly 75% of cases of adult ALL are of B-cell lineage. B-lineage markers are CD19, CD20, CD22, CD24, and CD79a (Huh 2000).

The earliest B-lineage markers are CD19, CD22 (membrane and cytoplasm) and CD79a (Campana 1988). A positive reaction for any two of these three markers, without further differentiation markers, identifies pro-B ALL. The presence of CD10 antigen (CALLA) defines the "common" ALL subgroup. Cases with additional identification of cytoplasmatic IgM constitute the pre-B group, whereas the presence of surface immunglobulin light chains defines mature B-ALL.

T-cell ALL constitutes approximately 25% of all adult cases of ALL. T-cell markers are CD1a, CD2, CD3 (membrane and cytoplasm), CD4, CD5, CD7 and CD8. CD2, CD5 and CD7 antigens are the most immature T-cell markers, but none of them is absolutely lineagespecific, so that the unequivocal diagnosis of T-ALL rests on the demonstration of surface/cytoplasmic CD3.

ALL of B or T lineage can additionally express myeloid antigens or stem-cell antigen CD34. The latter has little diagnostic relevance but can be prognostically important (De Waele 2001) The scoring system recently proposed by the EGIL group addressed the characterization of the acute leukemia as B or T lineage ALL, or AML by including the most specific markers for the

**Precursor B-lymphoblastic leukemia** ( HLA-DR+, TdT+, CD19+, and/or CD79a+, and/or CD22+, and/or CD34+). This type of ALL accounts for around 75% of

immunoglobulin negative; represents approximately 10% of adult ALL. b. Common ALL is characterized by the presence of CD10, cytoplasmic

Cells are TdT+ in addition to cytoplasmic CD3+ and CD34+. This type of ALL

a. Pro T-ALL CD2-, CD7+, CD4-, CD8- seen in around 7% of adult ALL.

TdT/CD34/CD1a- and make up approximately 1% of adult ALL.

Table 5. European Group for the Immunological Characterization of Leukemias (EGIL)

The consensus considers a 20% minimum threshold to define a positive reaction of blast cells to a given monoclonal antibody. Roughly 75% of cases of adult ALL are of B-cell

The earliest B-lineage markers are CD19, CD22 (membrane and cytoplasm) and CD79a (Campana 1988). A positive reaction for any two of these three markers, without further differentiation markers, identifies pro-B ALL. The presence of CD10 antigen (CALLA) defines the "common" ALL subgroup. Cases with additional identification of cytoplasmatic IgM constitute the pre-B group, whereas the presence of surface immunglobulin light chains

T-cell ALL constitutes approximately 25% of all adult cases of ALL. T-cell markers are CD1a, CD2, CD3 (membrane and cytoplasm), CD4, CD5, CD7 and CD8. CD2, CD5 and CD7 antigens are the most immature T-cell markers, but none of them is absolutely lineagespecific, so that the unequivocal diagnosis of T-ALL rests on the demonstration of

ALL of B or T lineage can additionally express myeloid antigens or stem-cell antigen CD34. The latter has little diagnostic relevance but can be prognostically important (De Waele 2001) The scoring system recently proposed by the EGIL group addressed the characterization of the acute leukemia as B or T lineage ALL, or AML by including the most specific markers for the

c. Cortical T-ALL or Thymic ALL (Thy ALL) is CD1a+ and accounts for 17% of

immunoglobulin negative; comprises greater than 50% of adult cases of ALL. c. Pre B-ALL is characterised by the expression of cytoplasmic immunoglobulin and CD10; this subtype of ALL is identified in nearly 10% of adult cases. d. Mature B-ALL is found in approximately 4% of adult ALL patients. The blast cells express surface antigens of mature B cells, including surface membrane immunoglobulin (SmIg+). They are typically TdT and CD34 negative and have L3 morphology. This category overlaps with Burkitt lymphoma, which is

a. Pro B-ALL expresses HLA-DR, TdT, and CD19. CD10-, cytoplasmic

adult cases and is subdivided into the following groups:

included under the mature B-cell neoplasms.

adult ALL CD7+, CD2+, CD5+, CD4+, CD8+

accounts for around 25% of adult cases and is subdivided into:

d. Mature T-ALL are surface CD3+, CD2+, CD7+, CD4 or 8, and

lineage. B-lineage markers are CD19, CD20, CD22, CD24, and CD79a (Huh 2000).

**Precursor T-lymphoblastic leukemia** 

b. Pre T-ALL CD2+, CD7+, CD4-, CD8-.

classification of acute lymphocytic leukemia (Hoelzer 2002)

defines mature B-ALL.

surface/cytoplasmic CD3.

lymphoid and myeloid lineages among those of earlier stages of cell differentiation, plus some non-specific but stem-cell markers. The system introduced a modified terminology specific to each 'maturation' step within the B- or T-cell lineage (EGIL 1995) and was confirmed as adequate for both diagnosis and subclassification of ALL (Thalhammer-Scherrer 2002).

#### **4. European Group for the Immunological characterization of Leukemias (EGIL) classification for biphenotypic acute leukemia**

Biphenotypic acute leukemia (BAL), or acute leukemia with a single population of blasts coexpressing markers of two different lineages, is a rare clinical entity.

The scoring systems proposed by Catovsky *et al*. and by the EGIL (Bene 1995) allowed for a better definition of biphenotypic acute leukemia (BAL), clearly distinguishing them from classical AL expressing aberrantly one or two markers of another lineage. However, increasing evidence suggests that this system has limitations, as acknowledged by the 2008 World Health Organization (WHO) Classification of Tumors of Hematopoietic and Lymphoid Tissues. Although substantially improved in relation to the EGIL, the new WHO Classification is still not optimal for guiding the clinical management of patients with BAL. Typical examples of such aberrations are the expression of CD15 on B-ALL(Maynadie et al 1997), or of CD2 on acute promyelocytic –AML( Albano 2006). In 1998, the EGIL further refined this scoring system by attributing one point for the expression of CD117, after showing the strong relationship of this marker with engagement in the myeloid lineage (Bene MC et al 1998) . To identify BAL, it is therefore necessary to consider aberrant coexpression of markers usually associated to different lineages, with a score higher than 2 in more than one lineage (Zhao XF et al 2009)


Table 6. EGIL Scoring system for biphenotypic acute leukemia

Biphenotypic acute leukemia is defined when scores are >2 for the myeloid lineage and >1 for the lymphoid lineage. In some T-ALL cases, clonality of TCR alphabeta rearrangements can now be assessed cytoflourmetrically (Langerak 2001, Xu XQ et al 2009).

The prognosis of biphenotypic acute leukemia patients is poor when compared with de novo acute myeloid leukemia or acute lymphoblastic leukemia. Biphenotypic acute leukemia patients showed a much higher incidence of CD34 antigen expression, complex abnormal karyotype, extramedullary infiltration, relapse, and resistance to therapy after relapse (Xu XQ et al 2009).

Classification of Acute Leukemia 17

Benett JM, Catovsky D, Daniel MT et al(1980) The French-American-British (FAB) Cooperative Group. Corresponence; a variant form of hypergranular promyelocytic

Benette JM, CatoSky D, Daniel MT et al(1982) Proposals for the classification of

Bennett JM, Catovsky D, Daniel M-T, et al. Criteria for the diagnosis of acute leukemia of

Bennett JM, Catovsky D, Daniel MT, et al. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML MO). Brj Haematol 1991;78:325-9. Berger R, Flandrin G (1984) Determining the nature of cells studied cytogenetically. Cancer

Berger R, Bernheim A, Daniel MT, et al(1985) Cytogenetic studies on acute myelomonocytic

Campana D, Janossy G. Proliferation of normal and malignant human immature lymphoid

Catovsky D, Matutes E, Buccheri V, Shetty V, Hanslip J, Yoshida N, Morilla R. A classification of acute leukaemia for the 1990s. *Ann Hematol.* 1991;62:16–21. Cheson BD, Cassileth PA, Head DR, et al. Report of the National Cancer Institute-sponsored

De Waele M, Renmans W, Vander GK, Jochmans K, Schots R, Otten J, et al. Growth factor

European Group for the Immunological Characterization of leukemia (EGIL), Bene MC,

Falini B, Tiacci E, Martelli MP, Ascni S, Pileri SA. New classification of acute myeloid

First MIC Cooperative Study Group. Morphologic, immunologic, and cytogenetic (MIC)

Harris NL, Jaffe ES, Diebold J, et al. World Health Organization classification of neoplastic

Hoelzer D., GokbugerN: Diagnostik und Therapie der akuten lymphatischen Leukaemia des

Huh YO, Ibrahim S. Immunophenotypes in adult acute lymphocytic leukemia. Role of flow

Jaffe ES, Harris NL, Stein H et al: *World Health Organization Classification of Tumours of* 

workshop on definitions of diagnosis and response in acute myeloid leukemia. J

receptor profile of CD34+ cells in AML and B-lineage ALL and in their normal bone

Castoldi G, Knapp W, et al. Proposals for the immunological classification of acute

leukemia and precursor-related neoplasms: changes and unsolved issues. Discov

working classification of acute lymphoblastic leukaemias. Cancer Genet

diseases of the hematopoietic and lymphoid tissues: report of the ClinicalAdvisory Committee meeting—Airlie House, Virginia, November 1997. J Clin Oncol

cytometry in diagnosis and monitoring of disease. Hematol Oncol Clin North Am

megakaryocytic lineage (M7): a report of the French-American-British cooperative

myelodsplastic syndromes. Br J Haematol 51; 189-99

leukemia (M4) with eosinophilia. Leuk Res 9: 279-88

marrow counterparts. Eur J Haematol 2001; 66: 178-187

Hayhoe FG. The classification of acute leukaemia. Blood Rev 1988;2:186-93.

*haematopoietic and lymphoid tissue f.* Lyon, IARC Press, 2001.

Erwachsenen. Onkologie 8 (7) (2002) 672-85

group. Ann Intern Med 1985;103:460-2.

cells. Blood 1988; 71: 1201-1210

leukemias. Leukemia 1995;9:1783-6.

Med. 2010 Oct;10(53):281-92.

Cytogenetic 1986;23:189-97.

1999;17:3835–49.

2000; 14: 1251-1265

Clin Oncol 1990;8:813–9.

leukemia (M3). Br J Haematol 44: 169-70

Surveys 3:423-38

#### **5. Conclusion**

Acute leukaemia can be classified in many ways. An ideal classification is one which recognizes real entities with fundamental biological differences. The FAB classification of ALL and AML is based on morphology and cytochemical staining of blasts. However, the recent classification schemes proposed by the World Health Organization (WHO) require the additional evaluation of the leukemic blasts by molecular analysis and flow cytometry. The results of these 4 methods of evaluation (ie, morphology, staining, molecular analysis, flow cytometry) not only differentiate ALL from AML, but also categorize the subtypes of acute leukemia . Recent advances in molecular biology have shown that various subtypes of AML and ALL behave differently and should not be treated similarly. For example, the identification of M3 AML (acute promyelocytic leukemia) is crucial because it is associated with disseminated intravascular coagulation (DIC), and retinoic acid, in addition to chemotherapy, is the treatment of choice.

The European Group for the Immunological Classification of Leukemias (EGIL) has proposed that acute leukaemia be classified on the basis of immunophenotype alone. This classification has the strength that it suggests standardized criteria for defining a leukaemia as myeloid, T lineage, B lineage, or biphenotypic. It also suggests criteria for distinguishing biphenotypic leukaemia from AML with aberrant expression of lymphoid antigens, and from ALL with aberrant expression of myeloid antigens.

#### **6. References**


Acute leukaemia can be classified in many ways. An ideal classification is one which recognizes real entities with fundamental biological differences. The FAB classification of ALL and AML is based on morphology and cytochemical staining of blasts. However, the recent classification schemes proposed by the World Health Organization (WHO) require the additional evaluation of the leukemic blasts by molecular analysis and flow cytometry. The results of these 4 methods of evaluation (ie, morphology, staining, molecular analysis, flow cytometry) not only differentiate ALL from AML, but also categorize the subtypes of acute leukemia . Recent advances in molecular biology have shown that various subtypes of AML and ALL behave differently and should not be treated similarly. For example, the identification of M3 AML (acute promyelocytic leukemia) is crucial because it is associated with disseminated intravascular coagulation (DIC), and retinoic acid, in addition to

The European Group for the Immunological Classification of Leukemias (EGIL) has proposed that acute leukaemia be classified on the basis of immunophenotype alone. This classification has the strength that it suggests standardized criteria for defining a leukaemia as myeloid, T lineage, B lineage, or biphenotypic. It also suggests criteria for distinguishing biphenotypic leukaemia from AML with aberrant expression of lymphoid antigens, and

Albano F, Mestice A, Pannunzio A, Lanza F, Martino B, Pastore D, et al. The biological

Arber DA, Brunning RD, Le Beau MM, Falini B, Vardiman JW, Porwit A, Thiele J,

Arber DA, Brunning RD, Orazi A, Bain BJ, Porwit A, Vardiman JW, Le Beau MM, Greenberg

Arber DA, Brunning RD, Orazi A, Porwit A, Peterson L, Thiele J, Le Beau MM. Acute

Bene MC, Castoldi G, Knapp W, et al. Proposals for the immunological classification of

Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the acute

leukaemias (FAB cooperative group). Brj Haematol 1976;33:451-8.

characteristics of CD34+ CD2+ adult acute promyelocytic leukemia and the CD34 CD2 hypergranular (M3) and microgranular (M3v) phenotypes. *Haematologica.* 

Bloomfield CD. Acute myeloid leukaemia with recurrent genetic abnormalities. In: *WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues.* 4th edition. pp. 110-123. Swerdlow SH, Campo E, Harris NL et al. (eds.). International Agency

PL. Acute myeloid leukaemia with myelodysplasia-related changes. In: *WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues.* 4th edition. pp. 124- 126. Swerdlow SH, Campo E, Harris NL et al. (eds.). International Agency for

myeloid leukemia, not otherwise specified. In: *WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues.* 4th edition. pp. 130-139. Swerdlow SH, Campo E, Harris NL et al. (eds.). International Agency for Research on Cancer (IARC),

**5. Conclusion** 

**6. References** 

2006;91:311–6.

Lyon, France, 2008c.

chemotherapy, is the treatment of choice.

from ALL with aberrant expression of myeloid antigens.

for Research on Cancer (IARC), Lyon, France, 2008a.

Research on Cancer (IARC), Lyon, France, 2008b.

acute leukemias. Leukemia 1995;9:1783-6

Benett JM, Catovsky D, Daniel MT et al(1980) The French-American-British (FAB) Cooperative Group. Corresponence; a variant form of hypergranular promyelocytic leukemia (M3). Br J Haematol 44: 169-70


**2** 

 *Iraq* 

Abbas H. Abdulsalam

*Al-Yarmouk Teaching Hospital, Baghdad* 

**Diagnosis of Acute Leukemia in** 

**Under-Resourced Laboratories** 

Laboratory diagnosis of acute leukemia in modern hematology practice is increasingly relying on guidelines that require the availability of relatively expensive machines with

In under-resourced hematology laboratories there is usually a missing step in the battery of required investigations. Moreover, when some of the advanced diagnostic instruments can be found then the problem of chronic inadequate and irregular supply of kits and services would supervene. Therefore, the laboratory diagnosis would mostly depend on the more basic, but consistently available and well controlled, laboratory techniques that should at least include complete blood count (CBC) and peripheral blood morphology, after which a

Moreover, in specialized hematology centers there may be a routine availability of few special stains, a very limited immunophenotyping CD markers panel, cytogenetics and PCR

The aim of diagnosis, lineage assignment and sub-classification of acute leukemia in these

Sketching rational systematic schemes for optimum use of the locally available investigation options would usually permit the diagnosis of most varieties of acute leukemias with a very acceptable level of reliability. Also these schemes may provide invaluable information regarding the prospect of update and future plans for laboratory development as it can

The WHO classification of this pre-AML disorder (Table 1) can be applied in most underresourced laboratories as it only entails the use of peripheral blood morphology and bone marrow aspirate morphology with Perl's reaction (diagnosis of rare hypoplastic and myelofibrotic MDS would require also bone marrow trephine biopsy) with the sole addition

Even when cytogenetic testing is not available, still the WHO classification can be reliably applied for the diagnosis of most of the MDS subcategories, with the exceptions of the otherwise provisional diagnosis of MDS-5q- syndrome (which is characteristically found in a middle age or an elderly female with peripheral blood macrocytic anemia and upper

consistent need for continuous quality control, kits supply and maintenance.

bone marrow study with aspirate and sometimes a trephine biopsy will follow.

or FISH testing mostly for BCR-ABL1 oncogene.

laboratories should immediately serve a clear therapeutic goal.

show clearly where the weak joints are (Abdulsalam, 2010).

**2. Basis of diagnosis of myelodysplastic syndrome (MDS)** 

of cytogenetics, preferably performed on marrow aspirate sample.

**1. Introduction** 


### **Diagnosis of Acute Leukemia in Under-Resourced Laboratories**

Abbas H. Abdulsalam *Al-Yarmouk Teaching Hospital, Baghdad Iraq* 

#### **1. Introduction**

18 Acute Leukemia – The Scientist's Perspective and Challenge

Langerak AW, van Den BR, I.L., Boor PP, van Lochem EG, Hooijkaas H, et al. Molecular and

Maynadié M, Campos L, Moskovtchenko P, Sabido O, Aho S, Lenormand B, et al.

Sachdeva MU, Ahluwalia J, Das R, et al (2006) Role of FAB classification of acute leukemias in era of immunophenotyping.Indian J Pathol Microbiol: 2006 Oct;49(4):524-7. Shibata A, Bennett JM, Castoldi GL et al (1985) Recommended methods for cytological

Thalhammer-Scherrer R, Mitterbauer G, Simonitsch I, Jaeger U, Lechner K, Schneider B, et

Vardiman JW. The World Health Organization (WHO) classification of tumors of the

Weinberg OK, Seetharam M, Ren K, Seo K, Merker JD, Gotlib J, Zehnder JL, Arber DA.

Xu XQ, Wang JM, Lu SQ, Chen L, Yang JM, Zhang WP, Song XM, Hou J, Ni X, Qiu HY.

TCRalphabeta T-cell proliferations. Blood 2001; 98: 165-173

procedures in haematology. Clin Lab Haematol 7:55-74

neoplasms. Chem Biol Interact 2010 Mar 19;184(1-2):16-20.

Haematol 50:201-14

380-389

27.

26;113(9):1906-8.

flow cytometric analysis of the Vbeta repertoire for clonality assessment in mature

Heterogenous expression of CD15 in acute lymphoblastic leukemia: a study of ten anti-CD15 monoclonal antibodies in 158 patients. *Leuk Lymphoma.* 1997;25:135–43. McKenna RW, Parkin J, Bllomfield CD et al (1982)Acute promyelocytic leukemia: a study of

39 caes with identification of a hyperbasophilic microgranular variant. Br J

al. The immunophenotype of 325 adult acute leukemias: relationship to morphologic and molecular classification and proposal for a minimal screening program highly predictive for lineage discrimination. Am J Clin Pathol 2002; 117:

hematopoietic and lymphoid tissues: an overview with emphasis on the myeloid

Clinical characterization of acute myeloid leukemia with myelodysplasia-related changes as defined by the 2008 WHO classification system. Blood. 2009 Feb

Clinical and biological characteristics of adult biphenotypic acute leukemia in comparison with that of acute myeloid leukemia and acute lymphoblastic leukemia: a case series of a Chinese population. Haematologica. 2009 Jul;94(7):919Laboratory diagnosis of acute leukemia in modern hematology practice is increasingly relying on guidelines that require the availability of relatively expensive machines with consistent need for continuous quality control, kits supply and maintenance.

In under-resourced hematology laboratories there is usually a missing step in the battery of required investigations. Moreover, when some of the advanced diagnostic instruments can be found then the problem of chronic inadequate and irregular supply of kits and services would supervene. Therefore, the laboratory diagnosis would mostly depend on the more basic, but consistently available and well controlled, laboratory techniques that should at least include complete blood count (CBC) and peripheral blood morphology, after which a bone marrow study with aspirate and sometimes a trephine biopsy will follow.

Moreover, in specialized hematology centers there may be a routine availability of few special stains, a very limited immunophenotyping CD markers panel, cytogenetics and PCR or FISH testing mostly for BCR-ABL1 oncogene.

The aim of diagnosis, lineage assignment and sub-classification of acute leukemia in these laboratories should immediately serve a clear therapeutic goal.

Sketching rational systematic schemes for optimum use of the locally available investigation options would usually permit the diagnosis of most varieties of acute leukemias with a very acceptable level of reliability. Also these schemes may provide invaluable information regarding the prospect of update and future plans for laboratory development as it can show clearly where the weak joints are (Abdulsalam, 2010).

#### **2. Basis of diagnosis of myelodysplastic syndrome (MDS)**

The WHO classification of this pre-AML disorder (Table 1) can be applied in most underresourced laboratories as it only entails the use of peripheral blood morphology and bone marrow aspirate morphology with Perl's reaction (diagnosis of rare hypoplastic and myelofibrotic MDS would require also bone marrow trephine biopsy) with the sole addition of cytogenetics, preferably performed on marrow aspirate sample.

Even when cytogenetic testing is not available, still the WHO classification can be reliably applied for the diagnosis of most of the MDS subcategories, with the exceptions of the otherwise provisional diagnosis of MDS-5q- syndrome (which is characteristically found in a middle age or an elderly female with peripheral blood macrocytic anemia and upper

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 21

stains and immunophenotyping (preferably using flowcytometry, or if it is not available then one can rely on either immunocytochemistry with/without immunohistochemistry).

Later on, a follow up for minimal residual disease can be performed using the same genetic abnormality found at diagnosis, i.e., cytogenetic remission, or more accurately, real time quantitative (RQ) PCR for quantization of the characteristic translocation that was found positive (but without quantization using conventional "qualitative" PCR) at diagnosis.

Laboratory risk stratification relies on cytogenetics and multiplex conventional PCR.

**leukemia Subtype Morphology Additional tests** 

high N:C ratio blasts

Larger, heterogeneous, nucleolated with low N:C ratio blasts

Large, homogeneous and nucleolated blasts with basophilic and vacuolated cytoplasm

Undifferentiated blasts ± dysplastic myeloid differentiation

maturation

Faggot cells

bilobed nuclei

Peripheral blood monocytes ≥5.0 \*109/l ± bone marrow monocytic lineage ≥20%

Bone marrow monocytic lineage ≥80%

>50% erythroblasts

and bone marrow fibrosis CD41

M3 Characteristic morphology,

M3 variant Characteristic morphology,

M6 Trilineage dysplasia and

M7 Blast with cytoplasmic blebs

Table 2. Diagnosis of acute leukemia based on FAB groups

SBB stain M2 Myeloblasts with myeloid

M0 Undifferentiated blasts Anti-MPO, CD117, CD33, CD68

TdT, CD3, CD79a, CD20, CD10

Cytogenetics, ISH or PCR

or Lysozyme

NSE confirmation of monocytic lineage

SBB, Glycophorin + anti-MPO, avoid CD34 as it would stain both myeloblasts and proerythroblasts

L1 Small, homogeneous with

**Acute** 

ALL

AML

L2

Leukemic phase of Burkitt's lymphoma (L3)

M1

M4

M5a (Monoblastic), M5b (Monocytic) and M5c (Histiocytic)


\* Dysplasia is considered significant only if it is present in >10% of the cells.

\*\* If Auer rods or pseudo-Chediak Higashi inclusions are present then MDS RAEB-2 is diagnosed even when the peripheral blood and bone marrow blasts are <5% and <10% respectively.

\*\*\* If blasts > 5% then it is classified as RAEB, although still lenalidomide treatment should be tried.

Table 1. WHO classification of MDS, 2008 (Vardiman, et al, 2009)

normal platelet count or even thrombocytosis, bone marrow megakaryocytes of normal overall size but with a relatively small mono- or bi-lobed nucleus and a very good response to a therapeutic trial of lenalidomide) (Kelaidi et al, 2008) and MDS-Unclassified (which can be diagnosed on follow up when the disease persists or even progress). This classification, unlike the late FAB classification, would keep with the newest 20% blasts cut-off point to diagnose acute leukemia, coping with the worldwide standards of MDS literatures and provide much more relevant prognostic groups.

#### **3. Diagnosis of acute leukemia in under-resourced laboratories**

FAB classification (Table 2) of acute leukemia should be applied in under-resourced laboratories where the only available routine techniques for diagnosis are morphology and special stains (Abdulsalam, 2010). The scheme in (Figure 1) may be used as a general guideline for the diagnosis of acute leukemia in under-resourced laboratories; however, it should be modified to optimally fit into the locally available techniques.

The practical application of WHO classification for acute leukemia (Tables 3 and 4) requires both diagnosis and risk stratification. The diagnosis can be based on morphology, special

Uni- or bi-cytopenia

Anemia and <1% blasts

Bi- or pan-cytopenia and <1% blasts

Bi- or pan-cytopenia and <1% blasts

Uni-, bi- or pancytopenia and <1% blasts

Anemia with upper normal or increase platelet count

\*\* If Auer rods or pseudo-Chediak Higashi inclusions are present then MDS RAEB-2 is diagnosed even

normal platelet count or even thrombocytosis, bone marrow megakaryocytes of normal overall size but with a relatively small mono- or bi-lobed nucleus and a very good response to a therapeutic trial of lenalidomide) (Kelaidi et al, 2008) and MDS-Unclassified (which can be diagnosed on follow up when the disease persists or even progress). This classification, unlike the late FAB classification, would keep with the newest 20% blasts cut-off point to diagnose acute leukemia, coping with the worldwide standards of MDS literatures and

FAB classification (Table 2) of acute leukemia should be applied in under-resourced laboratories where the only available routine techniques for diagnosis are morphology and special stains (Abdulsalam, 2010). The scheme in (Figure 1) may be used as a general guideline for the diagnosis of acute leukemia in under-resourced laboratories; however, it

The practical application of WHO classification for acute leukemia (Tables 3 and 4) requires both diagnosis and risk stratification. The diagnosis can be based on morphology, special

\*\*\* If blasts > 5% then it is classified as RAEB, although still lenalidomide treatment should be tried.

\* Dysplasia is considered significant only if it is present in >10% of the cells.

Table 1. WHO classification of MDS, 2008 (Vardiman, et al, 2009)

when the peripheral blood and bone marrow blasts are <5% and <10% respectively.

**3. Diagnosis of acute leukemia in under-resourced laboratories** 

should be modified to optimally fit into the locally available techniques.

blasts-1 (RAEB-1) 1-4% blasts 5-9% blasts and no Auer rods

blasts-2 (RAEB-2) 5-19% blasts 10-19% blasts, or <19% blasts plus

Refractory cytopenia with unilineage dysplasia: - Refractory anemia (RA) - Refractory neutropenia - Refractory thrombocytopenia

Refractory anemia with ring sideroblasts (RARS)

Refractory cytopenia with multilineage dysplasia

Refractory cytopenia with multilineage dysplasia and ring sideroblasts

Refractory anemia with excess

Refractory anemia with excess

MDS-Unclassified

syndrome

provide much more relevant prognostic groups.

Isolated 5q-

**Type Peripheral blood Bone marrow aspirate** 

and <1% blasts Unilineage dysplasia\* with <5% blasts

Erythroid dysplasia only with <5% blasts and >15% ring sideroblasts

Bi- or tri-dysplasia with <5% blasts and <15% ring sideroblasts

Bi- or tri-dysplasia with <5% blasts and >15% ring sideroblasts

Auer rods\*\*

Cytogenetic diagnosis of MDS with uni-, bi- or tri-lineage dysplasia in <10% of the cells and <5% blasts

Isolated del(5q) on cytogenetic study, prominent large megakaryocytes with hypolobated nuclei and <5% blasts\*\*\*

stains and immunophenotyping (preferably using flowcytometry, or if it is not available then one can rely on either immunocytochemistry with/without immunohistochemistry). Laboratory risk stratification relies on cytogenetics and multiplex conventional PCR.

Later on, a follow up for minimal residual disease can be performed using the same genetic abnormality found at diagnosis, i.e., cytogenetic remission, or more accurately, real time quantitative (RQ) PCR for quantization of the characteristic translocation that was found positive (but without quantization using conventional "qualitative" PCR) at diagnosis.



Diagnosis of Acute Leukemia in Under-Resourced Laboratories 23

inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB–MYH11

inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1–EVI1

Provisional entity: AML with mutated NPM1 Provisional entity: AML with mutated CEBPA



t(15;17)(q22;q12); PML–RARA t(9;11)(p22;q23); MLLT3–MLL t(6;9)(p23;q34); DEK–NUP214

t(1;22)(p13;q13); RBM15–MKL1




Table 3. WHO classification of AML, 2008 (Vardiman, et al, 2009)


t(9;22)(q34;q11.2) and BCR–ABL1

Table 4. WHO classification of ALL, 2008 (Vardiman, et al, 2009)

t(12;21)(p13;q22) and ETV6–RUNX1 hyperdiploidy (> 50 chromosomes) hypodiploidy (< 46 chromosomes) t(5;14)(q31;q32) and IL3–IGH t(1;19)(q23;p13.3) and TCF3–PBX1


translocation involving 11q23 and MLL rearrangement







t(8;21)(q22;q22); RUNX1–RUNX1T1


Follow up for minimal residual disease using a multi-color flowcytometry can be adequately performed (Thorn et al, 2011) but is usually more demanding than the genetic techniques, and therefore, it may not be the best choice in a resource-poor laboratory.

Fig. 1. Options for diagnosis of acute leukemia in resource-poor laboratories: FAB-based classification serving a clear therapeutic target.

#### **3.1 Basis of diagnosis of acute leukemia**

In the WHO classification of acute leukemia (Jaffe, et al, 2001) the diagnosis is based on an arbitrary cut-off point of 20% blasts as a percentage of bone marrow total or non-erythroid cells or as a percentage of peripheral blood cells. This exact percent is also applied nowadays in under-resourced laboratories were the FAB classification should be used (Bain, 2010a).

This 20% myeloblasts cut-off point seems to be universally accepted and for the time being it represents the best known tool for defining acute leukemia. However, the word "arbitrary" may still precede it and this may be attributed to (Abdulsalam, 2011):

1. This precise percent does not represent some specific biological event in the continuum of increasing blast count, but it is merely, to the best to our current knowledge, a cut-off point that permits a relatively clear classification and therapeutic aim. However, the fact that some high risk MDS patients are being treated actively with only 10% bone marrow blasts should be remembered.


Follow up for minimal residual disease using a multi-color flowcytometry can be adequately performed (Thorn et al, 2011) but is usually more demanding than the genetic techniques,

> **Acute leukemia**  ≥ 20 blast cells of total or nonerythroid bone marrow cells

> > Clinical features, Romanowsky and special stains morphology

**ALL** Romanowsky stain: **L1**, **L2** and **L3** SBB: Negative

PAS: **cALL**

**L1:** Start chemotherapy

**L2:** Therapeutic trial

Clinical, radiological and IHC

> **T-ALL**: cCD3

CD1a and sCD3

**TdT:** positive in **L2** and negative in **leukemic phase of lymphoma** including **L3**, which is positive for sIg

**Undifferentiated acute leukemia**: Therapeutic trial for ALL

**B-ALL**: CD79a

CD10 and cIg

**Adult**

CD20 IHC

BCR-ABL1 oncogene: PCR or FISH

Fig. 1. Options for diagnosis of acute leukemia in resource-poor laboratories: FAB-based

In the WHO classification of acute leukemia (Jaffe, et al, 2001) the diagnosis is based on an arbitrary cut-off point of 20% blasts as a percentage of bone marrow total or non-erythroid cells or as a percentage of peripheral blood cells. This exact percent is also applied nowadays in under-resourced laboratories were the FAB classification should be used (Bain,

This 20% myeloblasts cut-off point seems to be universally accepted and for the time being it represents the best known tool for defining acute leukemia. However, the word "arbitrary"

1. This precise percent does not represent some specific biological event in the continuum of increasing blast count, but it is merely, to the best to our current knowledge, a cut-off point that permits a relatively clear classification and therapeutic aim. However, the fact that some high risk MDS patients are being treated actively with only 10% bone

may still precede it and this may be attributed to (Abdulsalam, 2011):

and therefore, it may not be the best choice in a resource-poor laboratory.

**AML** 

classification serving a clear therapeutic target.

marrow blasts should be remembered.

**3.1 Basis of diagnosis of acute leukemia** 

2010a).

Romanowsky stain: **M2**, **M3**, **M4**, **M5b**, **M5c** & **M6**

**M3v:** Therapeutic trial

SBB: **M1, M3v**

NSE: **M5a** Anti-MPO: **M0** CD41: **M7**


t(8;21)(q22;q22); RUNX1–RUNX1T1 inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB–MYH11 t(15;17)(q22;q12); PML–RARA t(9;11)(p22;q23); MLLT3–MLL t(6;9)(p23;q34); DEK–NUP214 inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1–EVI1 t(1;22)(p13;q13); RBM15–MKL1 Provisional entity: AML with mutated NPM1 Provisional entity: AML with mutated CEBPA


Table 3. WHO classification of AML, 2008 (Vardiman, et al, 2009)

	- B lymphoblastic leukemia/lymphoma, not otherwise specified
	- B lymphoblastic leukemia/lymphoma with *recurrent genetic abnormalities* including

t(9;22)(q34;q11.2) and BCR–ABL1 translocation involving 11q23 and MLL rearrangement t(12;21)(p13;q22) and ETV6–RUNX1 hyperdiploidy (> 50 chromosomes) hypodiploidy (< 46 chromosomes) t(5;14)(q31;q32) and IL3–IGH t(1;19)(q23;p13.3) and TCF3–PBX1 --- T lymphoblastic leukemia/lymphoma

Table 4. WHO classification of ALL, 2008 (Vardiman, et al, 2009)

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 25

Image 1. Pseudo-Chédiak-Higashi inclusions together with atypical "thick" Auer rods in

At diagnosis, acute leukemia should, in most of the cases, be clinically manifested within the last month with non-specific features like lethargy and fatigue or more commonly with specific features related to organ infiltration including bone marrow which results in anemiarelated features, infections and bleeding. Other organ infiltration may refer not only to broad suspicion of acute leukemia but more likely to lineage assignment or even to a specific diagnosis, e.g., hepatosplenomegaly, lymphadenopathy, CNS symptoms and testicular involvement are in favor of ALL, severe bone pain in lower extremities would refer to B-ALL, thymic mediastinal mass and pleural effusion to T-ALL, bleeding tendency with overt coagulation tests defect can refer to AML-M3 and its variant and gum hypertrophy, skin

Acute leukemia in most of the cases would present with one, or more often more abnormalities of the CBC, including anemia, leucocytosis (or less common leucopenia) and thrombocytopenia. Even when the WBC count is within normal limits the anemia, thrombocytopenia and WBC

Clinical features combined with CBC should be very sensitive in directing acute leukemia

Should be the backbone for the diagnosis of acute leukemia when there is leucocytosis or when leucocyte count is within reference range, as in most of the cases it provides a specific

involvement and hepatosplenomegaly in M4 and, more commonly, in M5.

flags (in most automated cell counters) would raise fair enough suspicion.

AML.

**3.2 Clinical features** 

**3.3 Complete blood count (CBC)** 

**3.4 Peripheral blood smear** 

cases to be studied by peripheral blood smear.


Laboratory diagnosis of acute leukemia in modern hematology practice is increasingly relying on objective techniques to detect a specific ultrastructural or genetic abnormality. Therefore, the era of 20% blasts to diagnose acute leukemia may not stand the time any longer than that of the old FAB group 30% blasts lower threshold. However, at least in the present and the near future the morphology will remain the initial diagnostic test of acute leukemia and the abovementioned blast threshold will still be useful as a tool for classification (Abdulsalam, 2011).

The presence or absence of myeloblasts has a central role in diagnosis of AML and suspicion of ALL respectively. The blasts are divided into agranular (type I) and granular (type II and III) blasts based on Romanowsky stain morphology. However, when using SBB stain many of the "apparently" agranular blasts turn to be granular. Pathognomonic signs of AML that can be seen with Romanowsky stain and more frequently with SBB stain include SBBpositive granules, Auer rods, atypical "thick" Auer rods, pseudo-Chediak-Higashi inclusions (Abdulsalam et al, 2011a) (Image 1) and rectangular crystalline structures (Merino & Esteve, 2005) .

Image 1. Pseudo-Chédiak-Higashi inclusions together with atypical "thick" Auer rods in AML.

#### **3.2 Clinical features**

24 Acute Leukemia – The Scientist's Perspective and Challenge

2. The significant difference in the cut-off point of blast percent between peripheral blood and bone marrow is well established in MDS as the two groups RAEB-1 (blast count less than 5% in peripheral blood and 5-9% in bone marrow) and RAEB-2 (blast count 5- 9% in peripheral blood and 10-19% in bone marrow). In acute leukemia no such

3. The morphological finding of pathological "clonal" blast, type II that contains Auer rods, Pseudo-Chédiak-Higashi (Abdulsalam & Sabeeh, 2011) (Image 1) or other specific inclusions that are not seen in reactive marrow, is referring to the diagnosis of RAEB-2 or AML, here again the arbitrary cut-off point of 20% blasts will decide the specific

4. The original FAB classification was based for many years on the arbitrary cut-off point of 30% bone marrow blasts and previously some patients with 20-29% blasts remained stable over months without chemotherapy. However, this major percent change was driven by the survival studies which revealed that patients with 20-29% myeloblasts have a similar

survival pattern as those with 30% and more in the bone marrow (Jaffe et al, 2001). 5. Although myeloblasts recognition criteria as agranular and granular blasts achieved a reasonable consensus, there are minor discordances in their definitions and in practice it may be a matter of convention (subjective method) to discriminate it from the continuum of cells, as in deciding whether this cell is a blast type III or a promyelocyte. 6. The "blasts" refer to myeloblasts, lymphoblasts, monoblasts, promonocytes and

8. Cases with blast cells less than 20% may still be diagnosed as acute leukemia if they present with certain recurrent cytogenetic abnormalities as in AML M4 with inv(16) or

9. The utilization of 20% lower blast threshold is not really an issue in ALL because most patients at diagnosis already have more than 50% blasts. Moreover, a patient with normal or reduced peripheral blood count and bone marrow lymphoblasts about or slightly above 20% would usually be classified as lymphoblastic lymphoma rather than ALL. A 25% cut-off point has been suggested to arbitrarily differentiate between the

Laboratory diagnosis of acute leukemia in modern hematology practice is increasingly relying on objective techniques to detect a specific ultrastructural or genetic abnormality. Therefore, the era of 20% blasts to diagnose acute leukemia may not stand the time any longer than that of the old FAB group 30% blasts lower threshold. However, at least in the present and the near future the morphology will remain the initial diagnostic test of acute leukemia and the abovementioned blast threshold will still be useful as a tool for

The presence or absence of myeloblasts has a central role in diagnosis of AML and suspicion of ALL respectively. The blasts are divided into agranular (type I) and granular (type II and III) blasts based on Romanowsky stain morphology. However, when using SBB stain many of the "apparently" agranular blasts turn to be granular. Pathognomonic signs of AML that can be seen with Romanowsky stain and more frequently with SBB stain include SBBpositive granules, Auer rods, atypical "thick" Auer rods, pseudo-Chediak-Higashi inclusions (Abdulsalam et al, 2011a) (Image 1) and rectangular crystalline structures (Merino

7. Diagnosis of AML-M3 and its variant is not related to the blast percent.

10. In AML-M0 and M1 the 20% blasts cut-off point is also of no use in practice.

t(16;16)(p13;q22) and AML M2 with t(8;21) (Jaffe et al, 2001).

discrimination is available.

diagnosis.

megakaryoblasts.

two conditions.

classification (Abdulsalam, 2011).

& Esteve, 2005) .

At diagnosis, acute leukemia should, in most of the cases, be clinically manifested within the last month with non-specific features like lethargy and fatigue or more commonly with specific features related to organ infiltration including bone marrow which results in anemiarelated features, infections and bleeding. Other organ infiltration may refer not only to broad suspicion of acute leukemia but more likely to lineage assignment or even to a specific diagnosis, e.g., hepatosplenomegaly, lymphadenopathy, CNS symptoms and testicular involvement are in favor of ALL, severe bone pain in lower extremities would refer to B-ALL, thymic mediastinal mass and pleural effusion to T-ALL, bleeding tendency with overt coagulation tests defect can refer to AML-M3 and its variant and gum hypertrophy, skin involvement and hepatosplenomegaly in M4 and, more commonly, in M5.

#### **3.3 Complete blood count (CBC)**

Acute leukemia in most of the cases would present with one, or more often more abnormalities of the CBC, including anemia, leucocytosis (or less common leucopenia) and thrombocytopenia. Even when the WBC count is within normal limits the anemia, thrombocytopenia and WBC flags (in most automated cell counters) would raise fair enough suspicion.

Clinical features combined with CBC should be very sensitive in directing acute leukemia cases to be studied by peripheral blood smear.

#### **3.4 Peripheral blood smear**

Should be the backbone for the diagnosis of acute leukemia when there is leucocytosis or when leucocyte count is within reference range, as in most of the cases it provides a specific

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 27

absence of dysmyelopoiesis does not affect the diagnosis. Dysmegakaryopoiesis, especially in the form of micromegakaryocytes is in favour of diagnosis of AML with a preceding

Is indicated when failed to obtain an adequate marrow aspirate, which may result from improper aspiration technique, presence of fibrosis (especially in ALL and AML-M7; in both conditions there may be a leucoerythroblastic anemia and tear drop poikilocytes in peripheral blood), aleukemic or subleukemic peripheral blood and bone marrow aspirate smears due to heavily packed marrow (especially in ALL) or presence of hypoplastic acute leukemia (especially in AML). It is also indicated when there is an intention to apply routine immunophenotyping (although this can be adequately applied on clotted marrow sections, see paragraph 3.5). It can be stated that the bone marrow biopsy is not an essential investigation in

acute leukemia diagnosis when obtaining an adequate marrow aspirate (Bain, 2001b).

leucopenia or pancytopenia suggest the need for bone marrow biopsy.

The presence of peripheral blood leucocyte count above or within the upper normal count can be used as an indicator that a bone marrow biopsy would not be essential; in contrast,

With Romanowsky stain morphology AML- M2, M3, M4, M5b, M5c and M6 can be

By adding few special stains such as Sudan black B, and not myeloperoxidase (MPO) as SBB has a little more sensitivity in detecting myeloblasts which is crucial for diagnosis of AML, plus a non-specific esterase (NSE) stain as ANAE it becomes possible to recognize AML-M1

Rare types of AML like M5c require a higher degree of morphology experience, in which malignant cells appearance is reminiscent of tissue histiocytes (Image 2) (Abdulsalam &

The AML cases that cannot be distinguished by Romanowsky and special stains morphology are M0 and M7, for which the presence of myeloid dysplasia (abnormal nuclear morphology and cytoplasm hypogranularity using a Romanowsky stain or absence of SBB stained granules from maturing myeloid cells and neutrophils) (Image 7) in the former and the blasts' cytoplasmic blebs and bone marrow fibrosis in the latter may give a hint for the probable diagnosis, however there is still the need for more positive diagnostic technique and as the flow cytometry immunophenotyping may not be available then the use of a limited number of CD markers study by ICC/IHC is the option, these include mainly anti-

When resources are limited then it is for the best to concentrate on cytoplasmic ICC/IHC

There is a small proportion of cases that would be only certainly subclassified after the response to treatment as in rare forms of AML-M3v in spite that SBB stain is usually of help

Consideration of clinical as well as hematological features permits a strong presumptive diagnosis of ALL (Bain, 2010). ALL-L3 diagnosis (which should be referred to as the

myelodysplastic syndrome.

**3.6 Bone marrow biopsy** 

**3.7 Acute Myeloid Leukemia (AML)** 

and most cases of AML-M5a respectively (Bain, 2006).

myeloperoxidase for M0 and CD41 for M7.

**3.8 Acute Lymphoblastic Leukemia (ALL)** 

in this form (Images 5 and 6).

CD markers with the highest lineage sensitivity and specificity.

Risk stratification of AML is based on genetic studies.

recognized readily.

Sabeeh, 2009a).

diagnosis (ALL-L3 and AML-M3), a provisional diagnosis (ALL-L1 and AML M2, M3 variant, M4, M5b, M5c and M6) or at least a limited differential diagnosis (ALL-L2 and AML-M0, M1, M5a and M7).

When blasts are numerous in peripheral blood then special stains like SBB can be applied directly to it, this can be very useful for emergency diagnosis of AML-M3 variant especially within short time like before the weekend (Abdulsalam & Sabeeh, 2010). In cases with leucopenia, although frank blast cells may not be easily found, still there should be at least a clue to the diagnosis (including anemia, thrombocytopenia and myeloid dysplasia).

The presence of nucleated red cells and myeloid dysplasia mainly in the form of Pelger-Huet neutrophils should be investigated as the former can refer to AML-M6 and the later can suggest an AML with myelodysplasia. A bone marrow study should follow including aspirate and biopsy (when there is peripheral blood pancytopenia, and dry tap, hypocellular, diluted or difficult aspirate).

#### **3.5 Bone marrow aspirate (BMA)**

The diagnosis of acute leukemia in many instances is evident from the peripheral blood film; however, the bone marrow aspirate examination is always essential for confirmation of diagnosis, classification and application of special techniques including cytochemical stains, genetic studies, Immunocytochemistry (ICC) and immunohistochemistry (IHC).

IHC staining can be applied using the clotted marrow aspirate as a regular tissue sample after fixation and without decalcification (Bain, 2001a), to avoid bone marrow biopsy when there is no need for this procedure apart from the intention to apply immunophenotyping; however, it should be noted that the only use of clotted aspirate is for immunophenotyping, i.e., it should not be used as a regular morphology sample or any other application.

Apart from AML-M3 and its variant, the provisional and final diagnosis of the subtype of acute leukemia should not be issued before a proper BMA is performed. The classification of AML FAB groups is based on the percentages of blasts, maturing myeloid series (promyelocytes to neutrophils), monocytic series and erythroblasts from the total nucleated marrow cells. Also in some occasions many vital signs may be seen only in marrow aspirate like few Faggot cells in AML-M3 variant and even Auer rods. This phenomenon is aggravated when there is a peripheral leucopenia.

Diagnosis of acute leukemia is based on the presence of at least 20% blasts of total nucleated marrow cells, this condition may not be applicable especially in some AML cases, then 20% blasts limit should be obtained from the non-erythroid non-lymphoid marrow cells, otherwise the case would be labelled as MDS (Table 1). The reason for setting two lower thresholds is to simplify the morphology counts in practice, where in the first type of threshold the hematologist needs only to count the blast cells from all the nucleated cells in the field, this is much easier in practice but it would certainly require much higher blast threshold (which is available in almost all cases of ALL and many AML patients) than what would be required in the second form, which would be much more effort demanding and time consuming as one has to exclude erythroblasts, lymphocytes, plasma cells, macrophages and mast cells from the count.

Diagnosis of AML with myelodysplasia can only be confirmed by studying the bone marrow aspirate morphology with trilineage dysplasia. Dyserythropoiesis alone can be seen in many malignancies and is multifactorial, dysmyelopoiesis is supportive to the diagnosis of AML; however in cases where neutrophils and maturing myeloid cells are few then absence of dysmyelopoiesis does not affect the diagnosis. Dysmegakaryopoiesis, especially in the form of micromegakaryocytes is in favour of diagnosis of AML with a preceding myelodysplastic syndrome.

#### **3.6 Bone marrow biopsy**

26 Acute Leukemia – The Scientist's Perspective and Challenge

diagnosis (ALL-L3 and AML-M3), a provisional diagnosis (ALL-L1 and AML M2, M3 variant, M4, M5b, M5c and M6) or at least a limited differential diagnosis (ALL-L2 and

When blasts are numerous in peripheral blood then special stains like SBB can be applied directly to it, this can be very useful for emergency diagnosis of AML-M3 variant especially within short time like before the weekend (Abdulsalam & Sabeeh, 2010). In cases with leucopenia, although frank blast cells may not be easily found, still there should be at least a

The presence of nucleated red cells and myeloid dysplasia mainly in the form of Pelger-Huet neutrophils should be investigated as the former can refer to AML-M6 and the later can suggest an AML with myelodysplasia. A bone marrow study should follow including aspirate and biopsy (when there is peripheral blood pancytopenia, and dry tap,

The diagnosis of acute leukemia in many instances is evident from the peripheral blood film; however, the bone marrow aspirate examination is always essential for confirmation of diagnosis, classification and application of special techniques including cytochemical stains,

IHC staining can be applied using the clotted marrow aspirate as a regular tissue sample after fixation and without decalcification (Bain, 2001a), to avoid bone marrow biopsy when there is no need for this procedure apart from the intention to apply immunophenotyping; however, it should be noted that the only use of clotted aspirate is for immunophenotyping,

Apart from AML-M3 and its variant, the provisional and final diagnosis of the subtype of acute leukemia should not be issued before a proper BMA is performed. The classification of AML FAB groups is based on the percentages of blasts, maturing myeloid series (promyelocytes to neutrophils), monocytic series and erythroblasts from the total nucleated marrow cells. Also in some occasions many vital signs may be seen only in marrow aspirate like few Faggot cells in AML-M3 variant and even Auer rods. This phenomenon is

Diagnosis of acute leukemia is based on the presence of at least 20% blasts of total nucleated marrow cells, this condition may not be applicable especially in some AML cases, then 20% blasts limit should be obtained from the non-erythroid non-lymphoid marrow cells, otherwise the case would be labelled as MDS (Table 1). The reason for setting two lower thresholds is to simplify the morphology counts in practice, where in the first type of threshold the hematologist needs only to count the blast cells from all the nucleated cells in the field, this is much easier in practice but it would certainly require much higher blast threshold (which is available in almost all cases of ALL and many AML patients) than what would be required in the second form, which would be much more effort demanding and time consuming as one has to exclude erythroblasts, lymphocytes, plasma cells,

Diagnosis of AML with myelodysplasia can only be confirmed by studying the bone marrow aspirate morphology with trilineage dysplasia. Dyserythropoiesis alone can be seen in many malignancies and is multifactorial, dysmyelopoiesis is supportive to the diagnosis of AML; however in cases where neutrophils and maturing myeloid cells are few then

genetic studies, Immunocytochemistry (ICC) and immunohistochemistry (IHC).

i.e., it should not be used as a regular morphology sample or any other application.

clue to the diagnosis (including anemia, thrombocytopenia and myeloid dysplasia).

AML-M0, M1, M5a and M7).

hypocellular, diluted or difficult aspirate).

aggravated when there is a peripheral leucopenia.

macrophages and mast cells from the count.

**3.5 Bone marrow aspirate (BMA)** 

Is indicated when failed to obtain an adequate marrow aspirate, which may result from improper aspiration technique, presence of fibrosis (especially in ALL and AML-M7; in both conditions there may be a leucoerythroblastic anemia and tear drop poikilocytes in peripheral blood), aleukemic or subleukemic peripheral blood and bone marrow aspirate smears due to heavily packed marrow (especially in ALL) or presence of hypoplastic acute leukemia (especially in AML). It is also indicated when there is an intention to apply routine immunophenotyping (although this can be adequately applied on clotted marrow sections, see paragraph 3.5). It can be stated that the bone marrow biopsy is not an essential investigation in acute leukemia diagnosis when obtaining an adequate marrow aspirate (Bain, 2001b).

The presence of peripheral blood leucocyte count above or within the upper normal count can be used as an indicator that a bone marrow biopsy would not be essential; in contrast, leucopenia or pancytopenia suggest the need for bone marrow biopsy.

#### **3.7 Acute Myeloid Leukemia (AML)**

With Romanowsky stain morphology AML- M2, M3, M4, M5b, M5c and M6 can be recognized readily.

By adding few special stains such as Sudan black B, and not myeloperoxidase (MPO) as SBB has a little more sensitivity in detecting myeloblasts which is crucial for diagnosis of AML, plus a non-specific esterase (NSE) stain as ANAE it becomes possible to recognize AML-M1 and most cases of AML-M5a respectively (Bain, 2006).

Rare types of AML like M5c require a higher degree of morphology experience, in which malignant cells appearance is reminiscent of tissue histiocytes (Image 2) (Abdulsalam & Sabeeh, 2009a).

The AML cases that cannot be distinguished by Romanowsky and special stains morphology are M0 and M7, for which the presence of myeloid dysplasia (abnormal nuclear morphology and cytoplasm hypogranularity using a Romanowsky stain or absence of SBB stained granules from maturing myeloid cells and neutrophils) (Image 7) in the former and the blasts' cytoplasmic blebs and bone marrow fibrosis in the latter may give a hint for the probable diagnosis, however there is still the need for more positive diagnostic technique and as the flow cytometry immunophenotyping may not be available then the use of a limited number of CD markers study by ICC/IHC is the option, these include mainly antimyeloperoxidase for M0 and CD41 for M7.

When resources are limited then it is for the best to concentrate on cytoplasmic ICC/IHC CD markers with the highest lineage sensitivity and specificity.

There is a small proportion of cases that would be only certainly subclassified after the response to treatment as in rare forms of AML-M3v in spite that SBB stain is usually of help in this form (Images 5 and 6).

Risk stratification of AML is based on genetic studies.

#### **3.8 Acute Lymphoblastic Leukemia (ALL)**

Consideration of clinical as well as hematological features permits a strong presumptive diagnosis of ALL (Bain, 2010). ALL-L3 diagnosis (which should be referred to as the

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 29

Image 3**.** Composite photograph of a patient with AML-M1 who had a 95% PAS-positive

The diagnosis of this acute leukemia requires a simultaneous application of several myeloid and lymphoid CD markers, or at least a request for the main lymphoid markers (CD3 and

In resource-poor laboratories a step-by-step algorithm is followed in order to use the least possible resources, therefore, the identification of a mixed acute leukemia can be missed, as when some clinical, morphological, cytochemical or immunological markers refer to one diagnosis then the other lines of investigations are usually skipped to save expenses. However, misdiagnosis of this rare type of acute leukemia to only one of its components

In a resource-poor laboratory these types of acute leukemia can be identified only if it happened to show some characteristic features using one of the essential techniques including, e.g., AML-M5c characteristic peripheral blood and bone marrow morphology, while others like biphenotypic acute leukemia may be misdiagnosed to only one of its components as described in paragraph 3.9. Natural killer-cell leukemia can be confused initially with reactive lymphocytosis as it results in CD3 negative and its characteristic CD56

For diagnosis of AML, especially M1 and M5, the addition of Sudan black B (SBB) and a non-specific esterase stain as α-naphthyl acetate esterase is respectively essential. While for ALL a negative result (0-2%, these rare SBB positively stained blasts represent remnant normal myeloblasts) with SBB staining is crucial to support the diagnosis. The addition of PAS stain would not add a lot to support the diagnosis of ALL as it can, at least occasionally, be equally positive in AML; however, a positive PAS stained vacuolated blasts can be useful

blasts (left image) and SBB positive blasts (right image).

may, in some cases, not adversely affect the patient.

CD79a) after finding a SBB positive result (Matutes et al, 1997).

marker is not usually tested for in an under-resourced laboratory.

**3.9 Biphenotypic acute leukemia** 

**3.10 Rare types of acute leukemia** 

**3.11 Special stains** 

Image 2. Blast cells appearance in AML-M5c.

leukemic phase of Burkitt's lymphoma as it arises from mature B-cells) would be obvious by morphology alone and it is convenient to rely on morphological diagnosis of ALL-L1 and start treatment. Also if a patient with an acute leukemia showing heterogeneous blasts that has no morphological markers of myeloid differentiation, negative staining with SBB with unavailability of further differentiating procedures then it may be treated initially as ALL-L2, as statistically speaking it would be much more possible than AML-M0. The negative result in staining with SBB is very helpful, while the addition of the special stain PAS would improve the chances of the correct diagnosis of common ALL. However, a case with positive staining results for both SBB and PAS is an acute myeloid leukemia (Image 3).

Clinical features as bone pain and radiological sign of mediastinal mass may presumptively aid in differentiating between B- and T-ALL, however, using ICC/IHC antibodies including CD79a for B lineage and CD3 for T lineage are necessary. After setting the diagnosis of B-ALL in adults then ICC/IHC CD20 typing and PCR or FISH for BCR-ABL1 fusion gene would affect the treatment options.

In children (neonate up to 15 years) there is some reluctance for BCR-ABL1 testing due to its low frequency, only about 3%. However, it may be prudent to test for this transcript in children who have some lymphoblasts with large azurophilic granules (represent approximately 10% of cases) (Jaffe, et al, 2001) as this, beside cutting short additional costs, can offer a safer limit.

Rare cases of ALL-L2 that are confused with leukemic phase of large cell lymphoma can be differentiated through the use of TdT immunohistochemistry typing on bone marrow biopsy slide, which would be positive in ALL but not in lymphoma.

Risk stratification of ALL is based on immunophenotypic and genetic studies.

leukemic phase of Burkitt's lymphoma as it arises from mature B-cells) would be obvious by morphology alone and it is convenient to rely on morphological diagnosis of ALL-L1 and start treatment. Also if a patient with an acute leukemia showing heterogeneous blasts that has no morphological markers of myeloid differentiation, negative staining with SBB with unavailability of further differentiating procedures then it may be treated initially as ALL-L2, as statistically speaking it would be much more possible than AML-M0. The negative result in staining with SBB is very helpful, while the addition of the special stain PAS would improve the chances of the correct diagnosis of common ALL. However, a case with positive staining results for both SBB and PAS is an acute myeloid

Clinical features as bone pain and radiological sign of mediastinal mass may presumptively aid in differentiating between B- and T-ALL, however, using ICC/IHC antibodies including CD79a for B lineage and CD3 for T lineage are necessary. After setting the diagnosis of B-ALL in adults then ICC/IHC CD20 typing and PCR or FISH for BCR-ABL1 fusion gene

In children (neonate up to 15 years) there is some reluctance for BCR-ABL1 testing due to its low frequency, only about 3%. However, it may be prudent to test for this transcript in children who have some lymphoblasts with large azurophilic granules (represent approximately 10% of cases) (Jaffe, et al, 2001) as this, beside cutting short additional costs,

Rare cases of ALL-L2 that are confused with leukemic phase of large cell lymphoma can be differentiated through the use of TdT immunohistochemistry typing on bone marrow

biopsy slide, which would be positive in ALL but not in lymphoma.

Risk stratification of ALL is based on immunophenotypic and genetic studies.

Image 2. Blast cells appearance in AML-M5c.

leukemia (Image 3).

can offer a safer limit.

would affect the treatment options.

Image 3**.** Composite photograph of a patient with AML-M1 who had a 95% PAS-positive blasts (left image) and SBB positive blasts (right image).

#### **3.9 Biphenotypic acute leukemia**

The diagnosis of this acute leukemia requires a simultaneous application of several myeloid and lymphoid CD markers, or at least a request for the main lymphoid markers (CD3 and CD79a) after finding a SBB positive result (Matutes et al, 1997).

In resource-poor laboratories a step-by-step algorithm is followed in order to use the least possible resources, therefore, the identification of a mixed acute leukemia can be missed, as when some clinical, morphological, cytochemical or immunological markers refer to one diagnosis then the other lines of investigations are usually skipped to save expenses. However, misdiagnosis of this rare type of acute leukemia to only one of its components may, in some cases, not adversely affect the patient.

#### **3.10 Rare types of acute leukemia**

In a resource-poor laboratory these types of acute leukemia can be identified only if it happened to show some characteristic features using one of the essential techniques including, e.g., AML-M5c characteristic peripheral blood and bone marrow morphology, while others like biphenotypic acute leukemia may be misdiagnosed to only one of its components as described in paragraph 3.9. Natural killer-cell leukemia can be confused initially with reactive lymphocytosis as it results in CD3 negative and its characteristic CD56 marker is not usually tested for in an under-resourced laboratory.

#### **3.11 Special stains**

For diagnosis of AML, especially M1 and M5, the addition of Sudan black B (SBB) and a non-specific esterase stain as α-naphthyl acetate esterase is respectively essential. While for ALL a negative result (0-2%, these rare SBB positively stained blasts represent remnant normal myeloblasts) with SBB staining is crucial to support the diagnosis. The addition of PAS stain would not add a lot to support the diagnosis of ALL as it can, at least occasionally, be equally positive in AML; however, a positive PAS stained vacuolated blasts can be useful

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 31

iv. The presence of 3% or more SBB stain positive blasts would characteristically refer to

vi. Speedy and firm enough diagnosis of AML-M3 variant cases (Image 6) to start ATRA

viii. The stain can be easily applied to peripheral blood as well as bone marrow aspirate

Image 5. Peripheral blood film of a patient with AML-M3 on ATRA treatment showing a

The literatures are always referring to directly counting the blasts from the SBB stain slide, which is the best technique if the blasts can be easily recognized, but in practice and especially with AML-M1 this is not always feasible due to the nature of the stain which renders many blasts indistinguishable from other less immature cells. Therefore, there should be a second best technique to count the percent of smear positive SBB blasts, because

dysplastic neutrophil that contains an Auer rod (Abdulsalam and Sabeeh, 2009b)

it is not always possible to differentiate all the blasts directly from the SBB slide.

**3.11.1.1 How to count the percentage of SBB positively stained blasts** 

v. Increased SBB stain positivity at diagnosis is associated with better prognosis.

treatment in the same day (Abdulsalam & Sabeeh, 2010). vii. Demonstration of myeloid series dysplasia (Image 7) (Bain, 2010b).

(Wong, 2010) (Image 5).

confirm the nature of the blasts.

smears.

the diagnosis of AML-M1 rather than ALL.

is usually a part of response to treatment as it appears in more maturing myeloid cells

Although it is now about 30 years since first reporting that in very rare cases even ALL blasts may show SBB positivity (Tricota, et al, 1982); however characters like being of less intensity than the control (remnant normal cells of the myeloid series), nongranular and diffuse reaction help to indicate that these are not myeloblasts. Also lymphoblasts would universally stain negative with MPO which can then be used to

to refer to cALL in 98% of cases (Bain, 2010a), here again CD10 ICC/IHC staining would be more meaningful.

#### **3.11.1 Sudan Black B (SBB) stain**

It is one of the few, but very useful, cytochemical stains to choose in a resource-poor laboratory. Care should always be paid for counting blasts with the right black color, intensely stained granules. The appealing characters that entail the use of SBB stain are:


Image 4. Composite photograph of the peripheral blood film of a patient with AML-M2, the blasts showed unusual nuclear lobulation, these blasts contained SBB positive granules and Auer rods (Abdulsalam et al, 2011b)

In all AML subtypes, except for AML-M3, the presence of even one blast cell with Auer rod would refer to failure to achieve remission and indicate the need for a second induction chemotherapy course, while in AML-M3 finding an Auer rod after induction

to refer to cALL in 98% of cases (Bain, 2010a), here again CD10 ICC/IHC staining would be

It is one of the few, but very useful, cytochemical stains to choose in a resource-poor laboratory. Care should always be paid for counting blasts with the right black color, intensely stained granules. The appealing characters that entail the use of SBB stain are: i. The reaction and non-reaction with SBB stain are both significant, as the former refer practically to AML and the latter supports the diagnosis of ALL or AML-M0. ii. The intensity of a positive reaction with SBB in general parallels myeloperoxidase activity. Generally local experience would decide which stain to choose. However, SBB gives a slightly more intense reaction and sensitivity than myeloperoxidase staining in the detection of myeloblasts and is safer than the older technique of MPO staining

iii. Better demonstration of Auer rods by using SBB stain than any of the usual Romanowsky stains. This would be of utmost benefit to identify all MDS cases with Auer rods, to differentiate some AML from ALL cases (Image 4) and also to follow up

Image 4. Composite photograph of the peripheral blood film of a patient with AML-M2, the blasts showed unusual nuclear lobulation, these blasts contained SBB positive granules and

In all AML subtypes, except for AML-M3, the presence of even one blast cell with Auer rod would refer to failure to achieve remission and indicate the need for a second induction chemotherapy course, while in AML-M3 finding an Auer rod after induction

AML cases for morphological remission after induction chemotherapy.

more meaningful.

**3.11.1 Sudan Black B (SBB) stain** 

Auer rods (Abdulsalam et al, 2011b)

(using carcinogenic benzidine or its derivatives).

is usually a part of response to treatment as it appears in more maturing myeloid cells (Wong, 2010) (Image 5).


Image 5. Peripheral blood film of a patient with AML-M3 on ATRA treatment showing a dysplastic neutrophil that contains an Auer rod (Abdulsalam and Sabeeh, 2009b)

#### **3.11.1.1 How to count the percentage of SBB positively stained blasts**

The literatures are always referring to directly counting the blasts from the SBB stain slide, which is the best technique if the blasts can be easily recognized, but in practice and especially with AML-M1 this is not always feasible due to the nature of the stain which renders many blasts indistinguishable from other less immature cells. Therefore, there should be a second best technique to count the percent of smear positive SBB blasts, because it is not always possible to differentiate all the blasts directly from the SBB slide.

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 33

Image 7. Dysplastic metamyelocyte (bottom right) that is completely agranular with SBB

It adds a minor support to diagnosis of ALL as a similar reaction can be seen, although less frequently in AML. Although the pattern of reaction was considered important in some literatures (Lewis et al, 2006) to differentiate between ALL (with clear cytoplasm between the positive granules) and AML (with cytoplasmic smudge positivity between the positive granules) but in practice relying on such a difference is very difficult, therefore, PAS use is

The other late advantage of PAS was to refer to possible cases of c-ALL; again this use has

Refers to identification of antigens within or on the surface of cells for the purpose of

In AML patients it is essential for proper diagnosis of M0 and M7. In ALL it is essential for diagnosis and risk stratification including T and B lineage assignment (as there is no reliable morphological features to differentiate between them), and subclassification into pro-,

A suggested list of ICC and IHC CD markers that should be available for diagnosis of acute leukemia can include: CD3 for T-ALL (CD7 is more sensitive than CD3, it almost reach 100% sensitivity for T-ALL but is not specific as it is also positive in 20% of AML cases; however, CD7 is still an excellent substitute for T-lineage assignment in rare cases of CD3 negative T-ALL where with the proper clinical and radiological features, Romanowsky and special stains morphology and other CD markers, as negative anti-MPO, then the diagnosis of ALL

considered non-essential and should be replaced by CD3 and CD79a ICC/IHC.

lineage assignment. It is not a proof for clonality instead of the genetic study.

common (c), pre- and mature B-ALL and early, cortical and mature T-ALL.

staining. {Same patient in Image 4}

**3.11.3 Periodic Acid Schiff (PAS)** 

been superseded by CD10.

**3.12 Immunophenotyping** 

is evident) and CD79a for B-ALL.

In the author's hematology laboratory practice the following procedure is applied by first utilizing the Leishman stain slide for counting cells into 3 categories as fractions from all the total marrow cells: 1st the blast cells; 2nd the maturing myeloid cells, which would be all assumed to stain positive, although some may actually be negative as a feature of myelodysplasia but nevertheless in calculations this would provide a higher safety threshold to avoid inappropriately classifying a case as AML and 3rd category for lymphocytes and nucleated red cells, which would be negatively stained. Then from the SBB stain slide count all the SBB positive cells and deducing the relative percentage of the SBB positive blasts.

Image 6. Composite photograph showing Leishman staining (*left*) and SBB, cytochemical, staining (*right*) diagnosis of the variant form of AML-M3 (Abdulsalam & Sabeeh, 2010).

#### **3.11.2 NSE**

Including α-naphthyl acetate esterase (ANAE), or preferably, α-naphthyl butyrate esterase (ANBE) which is more specific than the acetate stain for the monocytic lineage, either stain is required to confirm the morphological diagnosis of AML-M4, M5a and M5b.

In some occasions there might be a differential diagnosis between M3v and M5b, in which cases it is best to avoid discrimination between them based only on NSE as it may be positive in both leukemias, instead a strong reaction with SBB in M3v can be used.

In the author's hematology laboratory practice the following procedure is applied by first utilizing the Leishman stain slide for counting cells into 3 categories as fractions from all the total marrow cells: 1st the blast cells; 2nd the maturing myeloid cells, which would be all assumed to stain positive, although some may actually be negative as a feature of myelodysplasia but nevertheless in calculations this would provide a higher safety threshold to avoid inappropriately classifying a case as AML and 3rd category for lymphocytes and nucleated red cells, which would be negatively stained. Then from the SBB stain slide count all the SBB positive cells and deducing the relative percentage of the SBB

Image 6. Composite photograph showing Leishman staining (*left*) and SBB,

is required to confirm the morphological diagnosis of AML-M4, M5a and M5b.

positive in both leukemias, instead a strong reaction with SBB in M3v can be used.

cytochemical, staining (*right*) diagnosis of the variant form of AML-M3 (Abdulsalam &

Including α-naphthyl acetate esterase (ANAE), or preferably, α-naphthyl butyrate esterase (ANBE) which is more specific than the acetate stain for the monocytic lineage, either stain

In some occasions there might be a differential diagnosis between M3v and M5b, in which cases it is best to avoid discrimination between them based only on NSE as it may be

positive blasts.

Sabeeh, 2010).

**3.11.2 NSE** 

Image 7. Dysplastic metamyelocyte (bottom right) that is completely agranular with SBB staining. {Same patient in Image 4}

#### **3.11.3 Periodic Acid Schiff (PAS)**

It adds a minor support to diagnosis of ALL as a similar reaction can be seen, although less frequently in AML. Although the pattern of reaction was considered important in some literatures (Lewis et al, 2006) to differentiate between ALL (with clear cytoplasm between the positive granules) and AML (with cytoplasmic smudge positivity between the positive granules) but in practice relying on such a difference is very difficult, therefore, PAS use is considered non-essential and should be replaced by CD3 and CD79a ICC/IHC.

The other late advantage of PAS was to refer to possible cases of c-ALL; again this use has been superseded by CD10.

#### **3.12 Immunophenotyping**

Refers to identification of antigens within or on the surface of cells for the purpose of lineage assignment. It is not a proof for clonality instead of the genetic study.

In AML patients it is essential for proper diagnosis of M0 and M7. In ALL it is essential for diagnosis and risk stratification including T and B lineage assignment (as there is no reliable morphological features to differentiate between them), and subclassification into pro-, common (c), pre- and mature B-ALL and early, cortical and mature T-ALL.

A suggested list of ICC and IHC CD markers that should be available for diagnosis of acute leukemia can include: CD3 for T-ALL (CD7 is more sensitive than CD3, it almost reach 100% sensitivity for T-ALL but is not specific as it is also positive in 20% of AML cases; however, CD7 is still an excellent substitute for T-lineage assignment in rare cases of CD3 negative T-ALL where with the proper clinical and radiological features, Romanowsky and special stains morphology and other CD markers, as negative anti-MPO, then the diagnosis of ALL is evident) and CD79a for B-ALL.

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 35

Fig. 8. Composite photograph of the same AML-M1 patient in (Image 2) showing CD 3 negative blasts (top image) with 28% "uncounted" positive small lymphocytes, and CD20

Including mainly cytogenetics, InSitu Hybridization (ISH) and PCR. All of these techniques have advantages and limitations and the choice in acute leukemia should be based on at

negative blasts (bottom image) with 5% "uncounted" positive small lymphocytes.

**3.13 Genetic studies** 

When B-ALL diagnosis is confirmed then CD20 would help to decide for anti-CD20 (Rituximab) treatment option.

 TdT would help to differentiate ALL-L2 (where it is positive) from leukemic phase of lymphoma (where it is negative).

For B-ALL CD10 is negative in pro-B-ALL and positive in common-ALL (c-ALL) which confers better prognosis. In pre-B-ALL cytoplasmic immunoglobulin (cIg) is positive while surface Ig (sIg) is positive only in mature B-ALL (ALL-L3 or leukemic phase of Burkitt's lymphoma, in which case TdT is negative).

Only cytoplasmic CD3 (cCD3) is positive in early T-ALL, cCD3 and CD1a are both positive in thymic or cortical T-ALL which confers better prognosis, while in mature T-ALL surface CD3 (sCD3) is positive and CD1a is negative.

Anti-myeloperoxidase, CD117 or CD33 can be used for AML-M0 and CD41 for AML-M7.

The availability of CD45 can be useful in rare occasions to ensure the hemopoietic nature of a poorly differentiated malignancy (Bain et al, 2002).

#### **3.12.1 Flowcytometry**

The newer multicolor (detecting many CD markers in/on the same single malignant cell) and multiparametric (a character comparative to that of automated blood counters studying characters like cell size and granularity) flowcytometer is one of the ultimate routine techniques in diagnosis of AML and ALL, primary risk stratification of ALL and follow up for MRD. When there is leucocytosis due to leukemic blasts then the flowcytometry study can be done on peripheral blood, otherwise, a bone marrow aspirate is the specimen of choice. However, the current price of the flowcytometer, cost of operating kits and maintenance make it unsuitable for laboratories with small budget.

#### **3.12.2 Immunocytochemistry (ICC)**

This technique should be consistently used in resource-poor laboratories for lineage and sub-lineage assignment of acute leukemia. It is applied on the bone marrow aspirate smear or, less conveniently on the peripheral blood after removal of plasma or on buffy coat (only if the blast percent is high). Sample spread can be done on a regular glass-slide (it is not essential to use a positively charged slide as in IHC) and after fixation in alcohol, ICC can be applied directly or after storage.

In acute leukemia the results of ICC (Image 8) can be interpreted in much more logical sense than IHC as the remaining normal or reactive cells can express some diagnosis-unrelated but confusing CD marker that in the aspirate can be easily detected to appear only for nonblast cells.

#### **3.12.3 Immunohistochemistry (IHC)**

Can be used as a substitute for ICC as the second best test for immunophenotyping of acute leukemia in the resource-poor laboratories if the blast percent is high, and if the results are unequivocally positive or negative or when the bone marrow aspirate is inadequate, otherwise when the blast percent is low or when the IHC positive result is in the borderline zone (20-30%) the judgment on the result of IHC can be difficult. For paraffin embedded IHC the designation between marker-positive blasts or reactive cells can be very difficult and in almost all conditions the total positivity per all marrow cells is expressed.

When B-ALL diagnosis is confirmed then CD20 would help to decide for anti-CD20

TdT would help to differentiate ALL-L2 (where it is positive) from leukemic phase of

For B-ALL CD10 is negative in pro-B-ALL and positive in common-ALL (c-ALL) which confers better prognosis. In pre-B-ALL cytoplasmic immunoglobulin (cIg) is positive while surface Ig (sIg) is positive only in mature B-ALL (ALL-L3 or leukemic phase of Burkitt's

Only cytoplasmic CD3 (cCD3) is positive in early T-ALL, cCD3 and CD1a are both positive in thymic or cortical T-ALL which confers better prognosis, while in mature T-ALL surface

Anti-myeloperoxidase, CD117 or CD33 can be used for AML-M0 and CD41 for AML-M7. The availability of CD45 can be useful in rare occasions to ensure the hemopoietic nature of

The newer multicolor (detecting many CD markers in/on the same single malignant cell) and multiparametric (a character comparative to that of automated blood counters studying characters like cell size and granularity) flowcytometer is one of the ultimate routine techniques in diagnosis of AML and ALL, primary risk stratification of ALL and follow up for MRD. When there is leucocytosis due to leukemic blasts then the flowcytometry study can be done on peripheral blood, otherwise, a bone marrow aspirate is the specimen of choice. However, the current price of the flowcytometer, cost of operating kits and

This technique should be consistently used in resource-poor laboratories for lineage and sub-lineage assignment of acute leukemia. It is applied on the bone marrow aspirate smear or, less conveniently on the peripheral blood after removal of plasma or on buffy coat (only if the blast percent is high). Sample spread can be done on a regular glass-slide (it is not essential to use a positively charged slide as in IHC) and after fixation in alcohol, ICC can be

In acute leukemia the results of ICC (Image 8) can be interpreted in much more logical sense than IHC as the remaining normal or reactive cells can express some diagnosis-unrelated but confusing CD marker that in the aspirate can be easily detected to appear only for non-

Can be used as a substitute for ICC as the second best test for immunophenotyping of acute leukemia in the resource-poor laboratories if the blast percent is high, and if the results are unequivocally positive or negative or when the bone marrow aspirate is inadequate, otherwise when the blast percent is low or when the IHC positive result is in the borderline zone (20-30%) the judgment on the result of IHC can be difficult. For paraffin embedded IHC the designation between marker-positive blasts or reactive cells can be very difficult

and in almost all conditions the total positivity per all marrow cells is expressed.

(Rituximab) treatment option.

**3.12.1 Flowcytometry** 

lymphoma (where it is negative).

lymphoma, in which case TdT is negative).

CD3 (sCD3) is positive and CD1a is negative.

**3.12.2 Immunocytochemistry (ICC)** 

applied directly or after storage.

**3.12.3 Immunohistochemistry (IHC)** 

blast cells.

a poorly differentiated malignancy (Bain et al, 2002).

maintenance make it unsuitable for laboratories with small budget.

Fig. 8. Composite photograph of the same AML-M1 patient in (Image 2) showing CD 3 negative blasts (top image) with 28% "uncounted" positive small lymphocytes, and CD20 negative blasts (bottom image) with 5% "uncounted" positive small lymphocytes.

#### **3.13 Genetic studies**

Including mainly cytogenetics, InSitu Hybridization (ISH) and PCR. All of these techniques have advantages and limitations and the choice in acute leukemia should be based on at

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 37

Image 9. Peripheral blood film showed 94% abnormal granulated promyelocytes. In the absence of any specialized tests, the diagnosis was made from cytological features. There was a dramatic response to ATRA, confirming the morphological diagnosis of AML-M3v

Looking for cerebrospinal fluid (CSF) involvement with acute leukemia is advised in all patients with ALL, while for patients with AML it is only indicated for patients with

Durable remission in acute leukemia is based on clinical and morphological evidences. Clinical remission includes absence of symptoms and signs of leukemia. Complete blood count consistent with remission would include absence of severe anemia, neutrophil count more than 1 ×109/l and platelet count more than 100 ×109/l (Bain, 2010). Morphological remission of acute leukemia from peripheral blood involves absence of blasts, immature myeloid cells and nucleated red cells. Bone marrow aspirate consistent with morphological remission would include blast cells being less than 5% with absence of Auer rods. The presence of even one Auer rod on SBB stain would refer to failure to

Flowcytometry, cytogenetics or molecular genetics may be used to validate a morphological

remission if any of these techniques were already utilized at diagnosis.

(Abdulsalam & Nafila, 2009).

**3.15 Lumbar puncture** 

neurological symptoms.

**3.16 Remission** 

achieve remission.

least two different techniques that would give complementary information especially for risk stratification and follow up for MRD.

Older techniques, like Feulgen stain for quantization of DNA contents should be avoided even in under-resourced laboratories because these techniques are non-standard and confusing.

#### **3.13.1 Cytogenetics**

It should be routinely applied for every suspected case of acute leukemia. Cytogenetics would represent to genetic studies what a blood smear represents to hematology, i.e., study of morphology of chromosomes and blood cells respectively. However, it has major limitations as the procedure-inherent failure rate and inability to detect small size aberrations or cryptic translocations.

Cytogenetics, beside a molecular study, is essential for the application of the WHO classification of acute leukemia and its risk stratification.

#### **3.13.2 InSitu Hybridization (ISH)**

Including Flourescent InSitu Hybridization (FISH) and Chromogenic InSitu Hybridization (CISH). Each technique has its advantages and limitations. FISH would represent a molecular genetic study plus demonstration of some chromosomal morphology. While CISH would represent a molecular genetic study plus demonstration of tissue morphology. In the author's opinion both techniques are not ideal for diagnosis, risk stratification and follow up of acute leukemia in an under-resourced laboratory.

#### **3.13.3 Polymerase Chain Reaction (PCR)**

Using only one detection kit multiplex RT-PCR assay is an effective, sensitive, accurate and cost-effective one-step multiple molecular re-arrangements diagnostic and risk-stratification tool. It is a complementary technique to conventional cytogenetics for risk stratification of acute leukemia and it provides a platform for the later on possibility of RQ-PCR detection of minimal residual disease (MRD) as multiplex RT-PCR is a qualitative procedure and is not used by itself as a mean for detection of MRD. For ALL, ETV6-RUNX1 and TCF3-PBX1 (both confer good prognosis), and MLL-MLLt2 and BCR-ABL1 (both confer poor prognosis) (Cerveira at al, 2000 and Shai, 2010); and for AML, FLT3 and MLL (both confer poor prognosis), and NPM1 and CEBPA (both confer good prognosis) (Strehl et al, 2001) paired primers are useful options (Salto-Tellez et al, 2003).

#### **3.14 Chemotherapeutic trial for acute leukemia**

A chemotherapeutic trial for those who cannot afford to seek a more precise diagnosis with genetic study and lineage specification abroad is a realistic option, as the response to treatment could be a very useful confirmation of the provisional diagnosis. The two examples already the author had faced are AML-M3v diagnosed provisionally only by morphology but with a dramatic response to ATRA trial, confirming the diagnosis (Image 9) and a few cases of morphologically undifferentiated acute leukemia in which the induction therapy for ALL is tried first (using vincristine and prednisolone only). If the patient responds, then a diagnosis of ALL can be deduced; if not, the regimen should be shifted to chemotherapy of AML (Abdulsalam, 2009).

least two different techniques that would give complementary information especially for

Older techniques, like Feulgen stain for quantization of DNA contents should be avoided even in under-resourced laboratories because these techniques are non-standard and

It should be routinely applied for every suspected case of acute leukemia. Cytogenetics would represent to genetic studies what a blood smear represents to hematology, i.e., study of morphology of chromosomes and blood cells respectively. However, it has major limitations as the procedure-inherent failure rate and inability to detect small size

Cytogenetics, beside a molecular study, is essential for the application of the WHO

Including Flourescent InSitu Hybridization (FISH) and Chromogenic InSitu Hybridization (CISH). Each technique has its advantages and limitations. FISH would represent a molecular genetic study plus demonstration of some chromosomal morphology. While CISH would represent a molecular genetic study plus demonstration of tissue morphology. In the author's opinion both techniques are not ideal for diagnosis, risk stratification and

Using only one detection kit multiplex RT-PCR assay is an effective, sensitive, accurate and cost-effective one-step multiple molecular re-arrangements diagnostic and risk-stratification tool. It is a complementary technique to conventional cytogenetics for risk stratification of acute leukemia and it provides a platform for the later on possibility of RQ-PCR detection of minimal residual disease (MRD) as multiplex RT-PCR is a qualitative procedure and is not used by itself as a mean for detection of MRD. For ALL, ETV6-RUNX1 and TCF3-PBX1 (both confer good prognosis), and MLL-MLLt2 and BCR-ABL1 (both confer poor prognosis) (Cerveira at al, 2000 and Shai, 2010); and for AML, FLT3 and MLL (both confer poor prognosis), and NPM1 and CEBPA (both confer good prognosis) (Strehl et al, 2001) paired

A chemotherapeutic trial for those who cannot afford to seek a more precise diagnosis with genetic study and lineage specification abroad is a realistic option, as the response to treatment could be a very useful confirmation of the provisional diagnosis. The two examples already the author had faced are AML-M3v diagnosed provisionally only by morphology but with a dramatic response to ATRA trial, confirming the diagnosis (Image 9) and a few cases of morphologically undifferentiated acute leukemia in which the induction therapy for ALL is tried first (using vincristine and prednisolone only). If the patient responds, then a diagnosis of ALL can be deduced; if not, the regimen should be shifted to

risk stratification and follow up for MRD.

aberrations or cryptic translocations.

**3.13.2 InSitu Hybridization (ISH)** 

**3.13.3 Polymerase Chain Reaction (PCR)** 

primers are useful options (Salto-Tellez et al, 2003).

**3.14 Chemotherapeutic trial for acute leukemia** 

chemotherapy of AML (Abdulsalam, 2009).

classification of acute leukemia and its risk stratification.

follow up of acute leukemia in an under-resourced laboratory.

confusing.

**3.13.1 Cytogenetics** 

Image 9. Peripheral blood film showed 94% abnormal granulated promyelocytes. In the absence of any specialized tests, the diagnosis was made from cytological features. There was a dramatic response to ATRA, confirming the morphological diagnosis of AML-M3v (Abdulsalam & Nafila, 2009).

#### **3.15 Lumbar puncture**

Looking for cerebrospinal fluid (CSF) involvement with acute leukemia is advised in all patients with ALL, while for patients with AML it is only indicated for patients with neurological symptoms.

#### **3.16 Remission**

Durable remission in acute leukemia is based on clinical and morphological evidences. Clinical remission includes absence of symptoms and signs of leukemia. Complete blood count consistent with remission would include absence of severe anemia, neutrophil count more than 1 ×109/l and platelet count more than 100 ×109/l (Bain, 2010). Morphological remission of acute leukemia from peripheral blood involves absence of blasts, immature myeloid cells and nucleated red cells. Bone marrow aspirate consistent with morphological remission would include blast cells being less than 5% with absence of Auer rods. The presence of even one Auer rod on SBB stain would refer to failure to achieve remission.

Flowcytometry, cytogenetics or molecular genetics may be used to validate a morphological remission if any of these techniques were already utilized at diagnosis.

Diagnosis of Acute Leukemia in Under-Resourced Laboratories 39

Abdulsalam, A., Sabeeh, N., Bain, B. 2011b. Myeloblasts with unusual morphology.

Bain, B. 2010b. Neutrophil dysplasia demonstrated on Sudan black B staining. *American* 

Bain, B., Barnett, D., Linch, D., Matutes, E., Reilly, J.T. 2002. Revised guideline on

Cerveira, N., Ferreira, S., Do ria, S., Veiga, I., Ferreira, F., Mariz, J., Marques, M., Castedo,

Jaffe, E., Harris, N., Stein, H., Variman, J. 2001. *World Health Organization classification of tumors, Tumors of Haemopoietic and Lymphoid tissues*. IARC press, Lyon. Kelaidi, C., Eclache, V., Fenaux, P. 2008. The role of lenalidomide in the management of myelodysplasia with del 5q. *British Journal of Haematology*, 140: 267-278. Lewis, M., Bain, B., Bates, I. 2006. Dacie and Lewis practical haematology 10th ed. Churchill

Matutes, E., Morilla, R., Farahat, N., Carbonell, F., Swansbury, J., Dyer, M., Catovsky, D. 1997. Definition of acute biphenotypic leukemia. *Haematologica*, 82: 64-66. Merino, A., Esteve, J. 2005. Acute myeloid leukaemia with peculiar blast cell inclusions and

Salto-Tellez, M., Shelat, S., Benoit, B., Rennert, H., Carroll, M., Leonard, D., Nowell, P., Bagg,

Shai, I. 2010. Application of genomics for risk stratification of childhood acute

Strehl, S., Konig, M., Mann, G., Haas, O. 2001. Multiplex reverse transcriptase–polymerase

Thörn, I., Forestier, E., Botling, J., Thuresson, B., Wasslavik, C., Björklund, E., Aihong, L.,

Tricota, G., Broeckaert, A., Van Hoof, A., Verwilgdhen RL. 1982. Sudan Black B

A. 2003. Multiplex RT-PCR for the Detection of Leukemia-Associated

lymphoblastic leukaemia: from bench to bedside? *British Journal of Haematology*,

chain reaction screening in childhood acute myeloblastic leukemia. *Blood*, 97 (3):

Eleonor L.E., Malec, M., Grönlund, E., Torikka, K., Heldrup, J., Abrahamsson, J., Behrendtz, M., Söderhäll, S., Jacobsson, S., Olofsson, T., Porwit, A., Lönnerholm, G., Rosenquist, R., Sundström, C. 2011. Minimal residual disease assessment in childhood acute lymphoblastic leukaemia: a Swedish multi-centre study comparing real-time polymerase chain reaction and multicolour flow cytometry. *British Journal* 

positivity in acute lymphoblastic leukaemia. *British Journal of Haematology*, 51,

immunophenotyping in acute leukaemias and chronic lymphoproliferative

S. 2000. Detection of prognostic significant translocations in childhood acute lymphoblastic leukaemia by one-step multiplex reverse transcription polymerase

Bain, B. 2001a. Bone marrow aspiration. *Journal of Clinical Pathology*, 54: 657-663. Bain, B. 2001b. Bone marrow trephine biopsy. *Journal of Clinical Pathology*, 54: 737-742. Bain, B. 2006. *Blood cell, a practical guide*. 4th ed. Blackwell publishing, Singapore.

Bain, B. 2010a. *Leukaemia diagnosis* 4th ed. Wiley-Blackwell, Singapore.

disorders. Clinical and Laboratory Haematology, 24: 1-13.

chain reaction. *British Journal of Haematology*, 109: 638-640.

pseudo-eosinophilia. *British Journal of Haematology*, 131, 286.

Translocations. *Journal of Molecular Diagnostics*, 5 (4): 231-236.

*American Journal of Hematology*, 86 (6): 499.

*Journal of Hematology*, 85, 9, 707.

Livingstone, Germany.

151: 119-131.

*of Haematology*, 152 (6): 743-753.

805-808.

615-621.

#### **3.17 Minimal residual disease (MRD)**

Detection of MRD entails the availability of either RQ-PCR or multi-color flowcytometry. Both techniques may not be routinely feasible for a laboratory with poor-resources.

#### **4. Minimal technical requirements for application of WHO classification of acute leukemia**

There should be at least a routine availability of CBC (manual or, preferably, automated), peripheral blood and bone marrow aspirate smears (and in some occasions bone marrow biopsy), SBB and a NSE stains, immunocytochemistry with/without immunohistochemistry (including at least CD3, CD79a, anti-MPO and CD41), cytogenetics and conventional PCR for the multiplex primers already mentioned in paragraph 3.13.3.

#### **5. Conclusion**

In hematology laboratories where the diagnostic resources are limited, it is essential to establish local guidelines that are practical in developing cost-effective diagnostic protocols for conditions for which the treatment is available, plus leaving the door wide open for future improvements, as to the introduction of newer techniques to the already available procedures once a newer therapeutic agent with certain lineage assignment demands has been introduced.

#### **6. Acknowledgement**

I would like to thank Dr. Nafila Sabeeh, laboratory hematologist at Al-Yarmouk Teaching hospital, for her invaluable notes.

#### **7. References**


Detection of MRD entails the availability of either RQ-PCR or multi-color flowcytometry.

There should be at least a routine availability of CBC (manual or, preferably, automated), peripheral blood and bone marrow aspirate smears (and in some occasions bone marrow biopsy), SBB and a NSE stains, immunocytochemistry with/without immunohistochemistry (including at least CD3, CD79a, anti-MPO and CD41), cytogenetics and conventional PCR

In hematology laboratories where the diagnostic resources are limited, it is essential to establish local guidelines that are practical in developing cost-effective diagnostic protocols for conditions for which the treatment is available, plus leaving the door wide open for future improvements, as to the introduction of newer techniques to the already available procedures once a newer therapeutic agent with certain lineage assignment demands has

I would like to thank Dr. Nafila Sabeeh, laboratory hematologist at Al-Yarmouk Teaching

Abdulsalam, A. 2009. Chemotherapeutic trial for acute leukemia in Iraq. *Turkish Journal of* 

Abdulsalam, A. 2010. Laboratory diagnosis of acute leukemia in Iraq, the available options.

Abdulsalam, A. 2011. "Arbitrary" criterion for the diagnosis of acute leukemia. *Turkish* 

Abdulsalam, A., Sabeeh, N. 2009a. Acute myeloid leukemia with histiocytic differentiation.

Abdulsalam, A., Sabeeh, N. 2009b. Auer rod in a neutrophil following ATRA treatment of

Abdulsalam, A., Sabeeh, N. 2010. Cytological/cytochemical diagnosis of the variant form of

Abdulsalam, A., Sabeeh, N., Bain, B. 2011a. Pseudo-Chédiak-Higashi inclusions together

acute promyelocytic leukemia. *Slide atlas, BloodMed, British Society of Haematology*.

acute promyelocytic leukaemia. *Slide atlas, BloodMed, British Society of Haematology*.

with Auer rods in acute myeloid leukemia. *American Journal of Hematology*, 86 (7):

Both techniques may not be routinely feasible for a laboratory with poor-resources.

for the multiplex primers already mentioned in paragraph 3.13.3.

**4. Minimal technical requirements for application of WHO classification of** 

**3.17 Minimal residual disease (MRD)** 

**acute leukemia** 

**5. Conclusion** 

been introduced.

**7. References** 

**6. Acknowledgement** 

hospital, for her invaluable notes.

*Hematology*, 264, 216.

Wiley-Blackwell.

Wiley-Blackwell.

602.

*Turkish Journal of Hematology*, 27, 320-321.

*American Society of Hematology image bank*, 9, 80.

*Journal of Hematology*, 28 (2): 149-150.


**Part 2** 

**Molecular Mechanisms and Markers** 


## **Part 2**

### **Molecular Mechanisms and Markers**

40 Acute Leukemia – The Scientist's Perspective and Challenge

Vardiman, J., Thiele, J., Arber, D., Brunning, R., Borowitz, M., Porwit, A., Harris, N., Le

leukemia: rationale and important changes. *Blood,* 114, 937-951. Wong, F. 2010. All-*trans*-retinoic acid therapy. *British Journal of Haematology*, 149: 309.

Beau, M., Lindberg, E., Tefferi, A., Bloomfield, C. 2009. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute

**3** 

*The Netherlands* 

**The PI3K/PKB Signaling Module in** 

Roel Polak1 and Miranda Buitenhuis1,2

*1Department of Hematology, Erasmus MC, Rotterdam,* 

**Normal and Malignant Hematopoiesis** 

*2Erasmus MC Stem Cell Institute for Regenerative Medicine, Erasmus MC, Rotterdam,* 

Hematopoiesis is a complex series of events resulting in the formation of mature blood cells. This process is regulated by cytokines at various levels, including self-renewal, proliferation, and differentiation. Upon binding of cytokines to their cognate receptors, the activity of intracellular signal transduction pathways is regulated, leading to modulation of gene expression. Although our appreciation of the transcriptional regulators of hematopoiesis has developed considerably, until recently, the roles of specific intracellular signal transduction pathways were largely unknown. An important mediator of cytokine signaling implicated in regulation of hematopoiesis is the Phosphatidylinositol-3-Kinase (PI3K) / Protein Kinase

The PI3K family consists of three distinct subclasses of which, to date, only the class I isoforms have been implicated in regulation of hematopoiesis. Four distinct catalytic class I isoforms have been identified; p110, p110, p110and p110(reviewed by Vanhaesebroeck et al., 2001) These isoforms are predominantly activated by protein tyrosine kinases and form heterodimers with a group a regulatory adapter molecules, including p85, p85, p50 p55, p55and p101(reviewed by Vanhaesebroeck et al., 2001). The most important substrate for these Class I PI3Ks is phosphatidylinositol 4,5 bisphosphate (PI(4,5)P2) which can be phosphorylated at the D3 position of the inositol ring upon extracellular stimulation, resulting in the formation of phosphatidylinositol 3,4,5 trisphosphate (PI(3,4,5)P3) (reviewed by Hawkins et al., 2006). PI(3,4,5)P3 subsequently serves as an anchor for pleckstrin homology (PH) domain-containing proteins, such as Protein Kinase B (PKB/ c-akt) (Burgering & Coffer, 1995). Activation of PKB requires phosphorylation on both Thr308, in the activation loop, by phosphoinositide-dependent kinase 1 (PDK1) and Ser473, within the carboxyl-terminal hydrophobic motif, by the MTORC2 complex that consists of multiple proteins, including Mammalian Target of Rapamycin (mTOR) and Rictor (Sarbassov et al.,

PKB itself subsequently regulates the activity of multiple downstream effectors, including the serine/threonine kinase Glycogen Synthase Kinase-3 (GSK-3) (Cross et al., 1995), members of the FoxO subfamily of forkhead transcription factors FoxO1, FoxO3, and FoxO4 (Brunet et al., 1999; Kops et al., 1999) and the serine/threonine kinase mammalian target of rapamycin (mTOR) as part of the MTORC1 complex, which also includes the regulatory

**1. Introduction** 

2005).

B (PKB/c-Akt) signaling module (Figure 1).

### **The PI3K/PKB Signaling Module in Normal and Malignant Hematopoiesis**

Roel Polak1 and Miranda Buitenhuis1,2 *1Department of Hematology, Erasmus MC, Rotterdam, 2Erasmus MC Stem Cell Institute for Regenerative Medicine, Erasmus MC, Rotterdam, The Netherlands* 

#### **1. Introduction**

Hematopoiesis is a complex series of events resulting in the formation of mature blood cells. This process is regulated by cytokines at various levels, including self-renewal, proliferation, and differentiation. Upon binding of cytokines to their cognate receptors, the activity of intracellular signal transduction pathways is regulated, leading to modulation of gene expression. Although our appreciation of the transcriptional regulators of hematopoiesis has developed considerably, until recently, the roles of specific intracellular signal transduction pathways were largely unknown. An important mediator of cytokine signaling implicated in regulation of hematopoiesis is the Phosphatidylinositol-3-Kinase (PI3K) / Protein Kinase B (PKB/c-Akt) signaling module (Figure 1).

The PI3K family consists of three distinct subclasses of which, to date, only the class I isoforms have been implicated in regulation of hematopoiesis. Four distinct catalytic class I isoforms have been identified; p110, p110, p110and p110(reviewed by Vanhaesebroeck et al., 2001) These isoforms are predominantly activated by protein tyrosine kinases and form heterodimers with a group a regulatory adapter molecules, including p85, p85, p50 p55, p55and p101(reviewed by Vanhaesebroeck et al., 2001). The most important substrate for these Class I PI3Ks is phosphatidylinositol 4,5 bisphosphate (PI(4,5)P2) which can be phosphorylated at the D3 position of the inositol ring upon extracellular stimulation, resulting in the formation of phosphatidylinositol 3,4,5 trisphosphate (PI(3,4,5)P3) (reviewed by Hawkins et al., 2006). PI(3,4,5)P3 subsequently serves as an anchor for pleckstrin homology (PH) domain-containing proteins, such as Protein Kinase B (PKB/ c-akt) (Burgering & Coffer, 1995). Activation of PKB requires phosphorylation on both Thr308, in the activation loop, by phosphoinositide-dependent kinase 1 (PDK1) and Ser473, within the carboxyl-terminal hydrophobic motif, by the MTORC2 complex that consists of multiple proteins, including Mammalian Target of Rapamycin (mTOR) and Rictor (Sarbassov et al., 2005).

PKB itself subsequently regulates the activity of multiple downstream effectors, including the serine/threonine kinase Glycogen Synthase Kinase-3 (GSK-3) (Cross et al., 1995), members of the FoxO subfamily of forkhead transcription factors FoxO1, FoxO3, and FoxO4 (Brunet et al., 1999; Kops et al., 1999) and the serine/threonine kinase mammalian target of rapamycin (mTOR) as part of the MTORC1 complex, which also includes the regulatory

The PI3K/PKB Signaling Module in Normal and Malignant Hematopoiesis 45

dephosphorylate PIP3 resulting in the formation of PI(4,5)P2 (Maehama & Dixon, 1998). Although both PTEN and SHIP1 act on the main product of PI3K activity, PIP3, the products generated are distinct. PI(3,4)P2 and PI(4,5)P2 both act as discrete second messengers activating distinct downstream events (Dowler et al., 2000; Golub & Caroni, 2005) indicating that the activation of SHIP1 and PTEN not only inhibit PI3K activity, but also can re-route

The role of PI3K class I isoforms was initially examined utilizing knockout mice deficient for one or multiple regulatory or catalytic subunits. Combined deletion of p85, p55 and p50 resulted in a complete block in B cell development (Fruman et al., 2000). Similarly, introduction of a mutated, catalytically inactive p110 (p110D910A) in the normal p110 locus also resulted in a block in early B cell development while T cell development was unaffected (Jou et al., 2002; Okkenhaug et al., 2002). These results indicate that PI3K activity is essential for normal B lymphocyte development. Pharmacological inhibition of PI3K activity in human umbilical cord blood derived CD34+ hematopoietic stem and progenitor cells revealed that inhibition of the activity of PI3K is sufficient to completely abrogate both proliferation and differentiation during *ex vivo* eosinophil and neutrophil development eventually leading to cell death (Buitenhuis et al., 2008). Conditional deletion of either PTEN or SHIP1 in adult HSCs resulting in activation of the PI3K pathway not only reduced the level of B-lymphocytes but also enhanced the level of myeloid cells (Helgason et al., 1998; Liu et al., 1999; Zhang et al., 2006). In addition, these mice developed a myeloproliferative disorder that progressed to leukemia (Helgason et al., 1998; Liu et al., 1999; Zhang et al., 2006). Furthermore, enhanced levels of megakaryocyte progenitors have been observed in SHIP1 deficient mice (Perez et al., 2008). In PTEN heterozygote (+/-) SHIP null (-/-) mice, a more severe myeloproliferative phenotype, displayed by reduced erythrocyte and platelet numbers and enhanced white blood cell counts including elevated levels of neutrophils and monocytes in the peripheral blood, could be observed (Moody et al., 2004). Interestingly, PI3K appears not only to be involved in lineage development, but is also required for stem cell maintenance. In PTEN and SHIP1 deficient mice, an initial expansion of HSCs could be observed which was followed by a depletion of long-term repopulating HSCs (Damen et al., 1996; Helgason et al., 2003). Recently, a shorter SHIP1 isoform (s-SHIP1), which is transcribed from an internal promoter in the SHIP1 gene, has also been implicated in positive regulation of lymphocyte development during hematopoiesis. (Nguyen et al., 2011). Its role in regulation of HSCs and long-term hematopoiesis remains to be investigated (Nguyen et al., 2011). A third negative regulator of the PI3K/PKB signaling module is Inositol 1,3,4,5-tetrakiphosphate (Ins(1,3,4,5)P4), which is generated from Inositol 1,4,5 triphosphate (Ins(1,4,5)P3) by Inositol triphosphate 3-kinase B (InsP3KB). It has been shown that Ins(1,3,4,5)P4 can bind to the PIP3-specific PH domains and competes for binding to those PH domains with PIP3 (Jia et al., 2007). In the bone marrow of mice deficient for InsP3KB, an acceleration of proliferation of the granulocyte macrophage progenitor has been observed resulting in higher levels of GMPs and mature neutrophils (Jia et al., 2008). In addition, although B lymphocytes could still be observed in InsP3KB deficient mice, mature CD4+ and CD8+ T lymphocytes were almost completely absent (Pouillon et al., 2003). Although InsP3KB is also involved in regulation of other pathways, the enhanced PKB phosphorylation in these mice (Jia et al., 2008) suggest that the observed phenotype is at

the signal transduction pathways activated by PI-lipid second messengers.

**2. PI3K/PKB signaling and normal hematopoiesis** 

**2.1 PI3K** 

associated protein of mTOR (Raptor). In contrast to GSK-3 and the FoxO transcription factors that are inhibitory phosphorylated by PKB, activation of mTOR is positively regulated (Nave et al., 1999; Inoki et al., 2002). It has been demonstrated that PKB can inhibit the GTPase activating protein Tuberous sclerosis protein 2 (TSC2)/TSC1 complex, resulting in accumulation of GTP-bound Rheb and subsequent activation of mTOR (Inoki et al., 2002).

Fig. 1**.** *Schematic representation of the PI3K/PKB signaling module.* Activation of PI3K by receptor stimulation results in the production of PtdIns(3,4,5)P3 at the plasma membrane. PKB subsequently translocates to the plasma membrane where it is phopshorylated by PDK1 and the mTORC2 complex. Upon phosphorylation, PKB is released into the cytoplasm where it can both inhibitory phosphorylate multiple substrates, including FoxO transcription factors and GSK-3 and induce the activity of other substrates such as mTOR as part of the mTORC1 complex. Negative regulators of the PI3K/PKB signaling module include PTEN, SHIP1 and Ins(1,3,4,5)P4.

While cytokines and growth factors positively induce PI3K activity, its activity can also be inhibited by SH2-containing inositol-5'-phosphatase 1 (SHIP1) (Damen et al., 1996), a protein predominantly expressed in hematopoietic cells (Liu et al., 1998), that hydrolyzes PIP3 to generate PI(3,4)P2 (Damen et al., 1996). Similarly, Phosphate and Tensin Homologue (PTEN) (Maehama & Dixon, 1998), a ubiquitously expressed tumor suppressor protein, can dephosphorylate PIP3 resulting in the formation of PI(4,5)P2 (Maehama & Dixon, 1998). Although both PTEN and SHIP1 act on the main product of PI3K activity, PIP3, the products generated are distinct. PI(3,4)P2 and PI(4,5)P2 both act as discrete second messengers activating distinct downstream events (Dowler et al., 2000; Golub & Caroni, 2005) indicating that the activation of SHIP1 and PTEN not only inhibit PI3K activity, but also can re-route the signal transduction pathways activated by PI-lipid second messengers.

#### **2. PI3K/PKB signaling and normal hematopoiesis**

#### **2.1 PI3K**

44 Acute Leukemia – The Scientist's Perspective and Challenge

associated protein of mTOR (Raptor). In contrast to GSK-3 and the FoxO transcription factors that are inhibitory phosphorylated by PKB, activation of mTOR is positively regulated (Nave et al., 1999; Inoki et al., 2002). It has been demonstrated that PKB can inhibit the GTPase activating protein Tuberous sclerosis protein 2 (TSC2)/TSC1 complex, resulting in accumulation of GTP-bound Rheb and subsequent activation of mTOR (Inoki et al., 2002).

Fig. 1**.** *Schematic representation of the PI3K/PKB signaling module.* Activation of PI3K by receptor stimulation results in the production of PtdIns(3,4,5)P3 at the plasma membrane. PKB subsequently translocates to the plasma membrane where it is phopshorylated by PDK1 and the mTORC2 complex. Upon phosphorylation, PKB is released into the

include PTEN, SHIP1 and Ins(1,3,4,5)P4.

cytoplasm where it can both inhibitory phosphorylate multiple substrates, including FoxO transcription factors and GSK-3 and induce the activity of other substrates such as mTOR as part of the mTORC1 complex. Negative regulators of the PI3K/PKB signaling module

While cytokines and growth factors positively induce PI3K activity, its activity can also be inhibited by SH2-containing inositol-5'-phosphatase 1 (SHIP1) (Damen et al., 1996), a protein predominantly expressed in hematopoietic cells (Liu et al., 1998), that hydrolyzes PIP3 to generate PI(3,4)P2 (Damen et al., 1996). Similarly, Phosphate and Tensin Homologue (PTEN) (Maehama & Dixon, 1998), a ubiquitously expressed tumor suppressor protein, can The role of PI3K class I isoforms was initially examined utilizing knockout mice deficient for one or multiple regulatory or catalytic subunits. Combined deletion of p85, p55 and p50 resulted in a complete block in B cell development (Fruman et al., 2000). Similarly, introduction of a mutated, catalytically inactive p110 (p110D910A) in the normal p110 locus also resulted in a block in early B cell development while T cell development was unaffected (Jou et al., 2002; Okkenhaug et al., 2002). These results indicate that PI3K activity is essential for normal B lymphocyte development. Pharmacological inhibition of PI3K activity in human umbilical cord blood derived CD34+ hematopoietic stem and progenitor cells revealed that inhibition of the activity of PI3K is sufficient to completely abrogate both proliferation and differentiation during *ex vivo* eosinophil and neutrophil development eventually leading to cell death (Buitenhuis et al., 2008). Conditional deletion of either PTEN or SHIP1 in adult HSCs resulting in activation of the PI3K pathway not only reduced the level of B-lymphocytes but also enhanced the level of myeloid cells (Helgason et al., 1998; Liu et al., 1999; Zhang et al., 2006). In addition, these mice developed a myeloproliferative disorder that progressed to leukemia (Helgason et al., 1998; Liu et al., 1999; Zhang et al., 2006). Furthermore, enhanced levels of megakaryocyte progenitors have been observed in SHIP1 deficient mice (Perez et al., 2008). In PTEN heterozygote (+/-) SHIP null (-/-) mice, a more severe myeloproliferative phenotype, displayed by reduced erythrocyte and platelet numbers and enhanced white blood cell counts including elevated levels of neutrophils and monocytes in the peripheral blood, could be observed (Moody et al., 2004). Interestingly, PI3K appears not only to be involved in lineage development, but is also required for stem cell maintenance. In PTEN and SHIP1 deficient mice, an initial expansion of HSCs could be observed which was followed by a depletion of long-term repopulating HSCs (Damen et al., 1996; Helgason et al., 2003). Recently, a shorter SHIP1 isoform (s-SHIP1), which is transcribed from an internal promoter in the SHIP1 gene, has also been implicated in positive regulation of lymphocyte development during hematopoiesis. (Nguyen et al., 2011). Its role in regulation of HSCs and long-term hematopoiesis remains to be investigated (Nguyen et al., 2011). A third negative regulator of the PI3K/PKB signaling module is Inositol 1,3,4,5-tetrakiphosphate (Ins(1,3,4,5)P4), which is generated from Inositol 1,4,5 triphosphate (Ins(1,4,5)P3) by Inositol triphosphate 3-kinase B (InsP3KB). It has been shown that Ins(1,3,4,5)P4 can bind to the PIP3-specific PH domains and competes for binding to those PH domains with PIP3 (Jia et al., 2007). In the bone marrow of mice deficient for InsP3KB, an acceleration of proliferation of the granulocyte macrophage progenitor has been observed resulting in higher levels of GMPs and mature neutrophils (Jia et al., 2008). In addition, although B lymphocytes could still be observed in InsP3KB deficient mice, mature CD4+ and CD8+ T lymphocytes were almost completely absent (Pouillon et al., 2003). Although InsP3KB is also involved in regulation of other pathways, the enhanced PKB phosphorylation in these mice (Jia et al., 2008) suggest that the observed phenotype is at

The PI3K/PKB Signaling Module in Normal and Malignant Hematopoiesis 47

(Miyamoto et al., 2007) and neutrophilia developed upon myelosuppressive stress conditions (Miyamoto et al., 2007). In contrast to FoxO3 deficient mice in which neutrophilia only occurred after myelosuppression while aging, conditional deletion of FoxO1, 3, and 4 in the adult hematopoietic system, was sufficient to increase the levels of myeloid cells and decrease the number of peripheral blood lymphocytes under normal conditions. In time, these mice developed leukocytosis characterized by a relative neutrophilia and lymphopenia (Tothova et al., 2007). In addition, an initial expansion of HSCs has been observed in these mice which correlated with an HSC-specific up-regulation of Cyclin D2 and down-regulation of Cyclin G2, p130/Rb, p27, and p21 (Tothova et al., 2007). Furthermore, a defective long-term repopulating capacity of bone marrow cells was observed, which could be explained by the reduction in HSC numbers that followed the initial expansion (Tothova et al., 2007). Although deletion of FoxO3 alone was not sufficient to improve myeloid development, ectopic expression of a constitutively active, non-phosphorylatable, FoxO3 mutant in mouse hematopoietic progenitors did result in a decrease in the formation of both myeloid and erythroid colonies (Engstrom et al., 2003), suggesting that FoxO3 does plays an important role in lineage

Modulation of the activity of the PI3K signaling pathway has been observed to alter the level of reactive oxygen species (ROS). While ROS levels are reduced in PKB/ deficient mice (Juntilla et al., 2010), increased levels have been observed in mice deficient for FoxO (Miyamoto et al., 2007). Increasing ROS levels in PKB/ deficient mice was sufficient to rescue differentiation defects, but not impaired long-term hematopoiesis (Juntilla et al., 2010). Restoring the ROS levels in FoxO deficient mice by i*n vivo* treatment with an antioxidative agent N-acetyl-L-cysteine was sufficient to abrogate the enhanced levels of proliferation and apoptosis in FoxO deficient HSCs and to restore the reduced colony forming ability of these cells (Tothova et al., 2007). These studies demonstrate that correct regulation of ROS by FoxO transcription factors is essential for normal hematopoiesis. Recent findings have demonstrated that correct regulation of the activity of GSK-3, another downstream effector of PKB, is also essential for maintenance of hematopoietic stem cell homeostasis. A reduction in long-term, but not short-term repopulating HSCs has, for example, been observed in GSK3 deficient mice (Huang et al., 2009). In addition, disruption of GSK-3 activity in mice with a pharmacological inhibitor or shRNAs has been shown to transiently induce expansion of both hematopoietic stem and progenitor cells followed by exhaustion of long-term repopulation HSCs (Trowbridge et al., 2006; Huang et al., 2009). In addition, since GSK-3 has been demonstrated to inhibit mTOR activity by phosphorylation and activation of TSC1/2 (Inoki et al., 2006) and the level of phosphorylated S6 was enhanced in cells with reduced GSK-3 levels, mice were treated with rapamycin. Rapamycin induced the number of LSK cells when GSK3 was depleted, but not in un-manipulated cells, suggesting that mTOR is an important effector of GSK-3 in regulation of HSC numbers (Huang et al., 2009) In addition to the observed expansion of HSCs in mice treated with a GSK-3 inhibitor, the recovery of neutrophil and megakaryocyte numbers after transplantation was accelerated in these mice, resulting in improved survival of the recipients (Trowbridge et al., 2006). In addition, *ex vivo* experiments revealed that GSK-3 can enhance eosinophil differentiation and inhibit neutrophil development (Buitenhuis et al., 2008). C/EBP, a key regulator of hematopoiesis, has been demonstrated to be an important mediator of PKB/GSK-3 signaling in regulation of granulocyte development (Buitenhuis et

development.

al., 2008).

least partially due to activation of the PI3K/PKB signaling module. Taken together, these studies suggest that correct temporal regulation of PI3K activity is critical for both HSC maintenance and regulation of lineage development.

#### **2.2 PKB**

PKB, an important effector of PI3K signaling, has been demonstrated to play an important role in regulation of cell survival and proliferation in a variety of systems (reviewed by Manning & Cantley, 2007). Three highly homologous PKB isoforms have been described to be expressed in mammalian cells; PKB, PKB, and PKB. Analysis of HSCs derived from PKB/PKB double-knockout mice revealed that PKB plays an important role in maintenance of long-term repopulating HSCs. These PKB/PKB double-deficient HSCs were found to persist in the G0 phase of the cell cycle, suggesting that the long-term functional defects observed in these mice were caused by enhanced quiescence (Juntilla et al., 2010). In contrast, loss of only one of the isoforms only minimally affected HSCs (Juntilla et al., 2010). In addition, analysis of mice deficient for both PKB and PKB revealed that the generation of marginal zone and B1 B cells and the survival of mature follicular B cells highly depend on the combined expression of PKB and PKB. Again no significant differences could be observed in mice deficient for the single isoforms (Calamito et al., 2010). In addition, ectopic expression of constitutively active PKB in mouse HSCs conversely resulted in transient expansion and increased cycling of HSCs, followed by apoptosis and expansion of immature progenitors in BM and spleen, which was also associated with impaired engraftment (Kharas et al., 2010), again demonstrating the importance of PKB in HSC maintenance. Utilizing an *ex vivo* human granulocyte differentiation system and a mouse transplantation model, it has recently been demonstrated that PKB not only plays a role in expansion of hematopoietic progenitors, but also has an important function in regulation of cell fate decisions during hematopoietic lineage commitment (Buitenhuis et al., 2008). High PKB activity was found to promote neutrophil and monocyte development and to inhibit B lymphocyte development, while conversely reduction of PKB activity is required to induce optimal eosinophil differentiation (Buitenhuis et al., 2008). In addition, PKB plays an important role in regulation of proliferation and survival of dendritic cell (DC) progenitors, but not maturation (van de Laar et al., 2010). Transplantion of mouse bone marrow cells ectopically expressing constitutively active PKB was sufficient to induce a myeloproliferative disease in most mice, characterized by extramedullary hematopoiesis in liver and spleen. In the majority of those mice, lymphoblastic thymic T cell lymphoma could also be observed. In addition, an undifferentiated AML developed in those mice that did not develop a myeloproliferative disease (Kharas et al., 2010).

#### **2.3 Downstream effectors of PKB**

To understand the molecular mechanisms underlying PKB mediated regulation of hematopoiesis, the roles of its downstream effectors in hematopoiesis have been investigated. FoxO transcription factors are known to play an important role in regulation of proliferation and survival of various cell types (reviewed by Birkenkamp & Coffer, 2003). Although proliferation and differentiation of hematopoietic progenitors appears not to be affected in FoxO3 deficient mice, competitive repopulation experiments revealed that deletion of FoxO3 is sufficient to impair long-term reconstitution (Miyamoto et al., 2007). In addition, in aging mice, the frequency of HSCs was increased compared to wild type littermate controls

least partially due to activation of the PI3K/PKB signaling module. Taken together, these studies suggest that correct temporal regulation of PI3K activity is critical for both HSC

PKB, an important effector of PI3K signaling, has been demonstrated to play an important role in regulation of cell survival and proliferation in a variety of systems (reviewed by Manning & Cantley, 2007). Three highly homologous PKB isoforms have been described to be expressed in mammalian cells; PKB, PKB, and PKB. Analysis of HSCs derived from PKB/PKB double-knockout mice revealed that PKB plays an important role in maintenance of long-term repopulating HSCs. These PKB/PKB double-deficient HSCs were found to persist in the G0 phase of the cell cycle, suggesting that the long-term functional defects observed in these mice were caused by enhanced quiescence (Juntilla et al., 2010). In contrast, loss of only one of the isoforms only minimally affected HSCs (Juntilla et al., 2010). In addition, analysis of mice deficient for both PKB and PKB revealed that the generation of marginal zone and B1 B cells and the survival of mature follicular B cells highly depend on the combined expression of PKB and PKB. Again no significant differences could be observed in mice deficient for the single isoforms (Calamito et al., 2010). In addition, ectopic expression of constitutively active PKB in mouse HSCs conversely resulted in transient expansion and increased cycling of HSCs, followed by apoptosis and expansion of immature progenitors in BM and spleen, which was also associated with impaired engraftment (Kharas et al., 2010), again demonstrating the importance of PKB in HSC maintenance. Utilizing an *ex vivo* human granulocyte differentiation system and a mouse transplantation model, it has recently been demonstrated that PKB not only plays a role in expansion of hematopoietic progenitors, but also has an important function in regulation of cell fate decisions during hematopoietic lineage commitment (Buitenhuis et al., 2008). High PKB activity was found to promote neutrophil and monocyte development and to inhibit B lymphocyte development, while conversely reduction of PKB activity is required to induce optimal eosinophil differentiation (Buitenhuis et al., 2008). In addition, PKB plays an important role in regulation of proliferation and survival of dendritic cell (DC) progenitors, but not maturation (van de Laar et al., 2010). Transplantion of mouse bone marrow cells ectopically expressing constitutively active PKB was sufficient to induce a myeloproliferative disease in most mice, characterized by extramedullary hematopoiesis in liver and spleen. In the majority of those mice, lymphoblastic thymic T cell lymphoma could also be observed. In addition, an undifferentiated AML developed in those mice that did not

To understand the molecular mechanisms underlying PKB mediated regulation of hematopoiesis, the roles of its downstream effectors in hematopoiesis have been investigated. FoxO transcription factors are known to play an important role in regulation of proliferation and survival of various cell types (reviewed by Birkenkamp & Coffer, 2003). Although proliferation and differentiation of hematopoietic progenitors appears not to be affected in FoxO3 deficient mice, competitive repopulation experiments revealed that deletion of FoxO3 is sufficient to impair long-term reconstitution (Miyamoto et al., 2007). In addition, in aging mice, the frequency of HSCs was increased compared to wild type littermate controls

maintenance and regulation of lineage development.

develop a myeloproliferative disease (Kharas et al., 2010).

**2.3 Downstream effectors of PKB** 

**2.2 PKB** 

(Miyamoto et al., 2007) and neutrophilia developed upon myelosuppressive stress conditions (Miyamoto et al., 2007). In contrast to FoxO3 deficient mice in which neutrophilia only occurred after myelosuppression while aging, conditional deletion of FoxO1, 3, and 4 in the adult hematopoietic system, was sufficient to increase the levels of myeloid cells and decrease the number of peripheral blood lymphocytes under normal conditions. In time, these mice developed leukocytosis characterized by a relative neutrophilia and lymphopenia (Tothova et al., 2007). In addition, an initial expansion of HSCs has been observed in these mice which correlated with an HSC-specific up-regulation of Cyclin D2 and down-regulation of Cyclin G2, p130/Rb, p27, and p21 (Tothova et al., 2007). Furthermore, a defective long-term repopulating capacity of bone marrow cells was observed, which could be explained by the reduction in HSC numbers that followed the initial expansion (Tothova et al., 2007). Although deletion of FoxO3 alone was not sufficient to improve myeloid development, ectopic expression of a constitutively active, non-phosphorylatable, FoxO3 mutant in mouse hematopoietic progenitors did result in a decrease in the formation of both myeloid and erythroid colonies (Engstrom et al., 2003), suggesting that FoxO3 does plays an important role in lineage development.

Modulation of the activity of the PI3K signaling pathway has been observed to alter the level of reactive oxygen species (ROS). While ROS levels are reduced in PKB/ deficient mice (Juntilla et al., 2010), increased levels have been observed in mice deficient for FoxO (Miyamoto et al., 2007). Increasing ROS levels in PKB/ deficient mice was sufficient to rescue differentiation defects, but not impaired long-term hematopoiesis (Juntilla et al., 2010). Restoring the ROS levels in FoxO deficient mice by i*n vivo* treatment with an antioxidative agent N-acetyl-L-cysteine was sufficient to abrogate the enhanced levels of proliferation and apoptosis in FoxO deficient HSCs and to restore the reduced colony forming ability of these cells (Tothova et al., 2007). These studies demonstrate that correct regulation of ROS by FoxO transcription factors is essential for normal hematopoiesis.

Recent findings have demonstrated that correct regulation of the activity of GSK-3, another downstream effector of PKB, is also essential for maintenance of hematopoietic stem cell homeostasis. A reduction in long-term, but not short-term repopulating HSCs has, for example, been observed in GSK3 deficient mice (Huang et al., 2009). In addition, disruption of GSK-3 activity in mice with a pharmacological inhibitor or shRNAs has been shown to transiently induce expansion of both hematopoietic stem and progenitor cells followed by exhaustion of long-term repopulation HSCs (Trowbridge et al., 2006; Huang et al., 2009). In addition, since GSK-3 has been demonstrated to inhibit mTOR activity by phosphorylation and activation of TSC1/2 (Inoki et al., 2006) and the level of phosphorylated S6 was enhanced in cells with reduced GSK-3 levels, mice were treated with rapamycin. Rapamycin induced the number of LSK cells when GSK3 was depleted, but not in un-manipulated cells, suggesting that mTOR is an important effector of GSK-3 in regulation of HSC numbers (Huang et al., 2009) In addition to the observed expansion of HSCs in mice treated with a GSK-3 inhibitor, the recovery of neutrophil and megakaryocyte numbers after transplantation was accelerated in these mice, resulting in improved survival of the recipients (Trowbridge et al., 2006). In addition, *ex vivo* experiments revealed that GSK-3 can enhance eosinophil differentiation and inhibit neutrophil development (Buitenhuis et al., 2008). C/EBP, a key regulator of hematopoiesis, has been demonstrated to be an important mediator of PKB/GSK-3 signaling in regulation of granulocyte development (Buitenhuis et al., 2008).

The PI3K/PKB Signaling Module in Normal and Malignant Hematopoiesis 49

mutations in p110, have been detected in a wide variety of human solid tumors (Ligresti et al., 2009). The most common mutations in p110 are located in the kinase domain (H1047R) and in the helical domain (E545A) (Lee et al., 2005). The E545A mutation has also been detected in acute, but not further specified, leukemia, albeit in a very low percentage (1/88) (Lee et al., 2005). In a series of 44 pediatric T-ALL patients, activating mutations in the catalytic subunit of PI3K (PIK3CA) have been observed in 2 patients, while in frame insertions/deletions have been detected in the PI3K regulatory subunit PIK3R1 in two other patients (Gutierrez et al., 2009). Transplantation of mice with bone marrow cells ectopically expressing mutated p110 resulted in the development of a leukemia-like disease within 5 weeks after transplantation (Horn et al., 2008), suggesting that mutations in p110 would be sufficient to induce leukemia. However, since mutations in PI3K appear to be very rare, it is unlikely that these mutations would be a major cause of leukemic development. Alternatively, the constitutive activation of PI3K observed in many patients with leukemia could also be caused by either aberrant expression or activation of modulators of PI3K

Reduced expression of PTEN has, for example, been observed in different types of leukemia (Xu et al., 2003; Nyakern et al., 2006). Both homozygous and heterozygous deletion of PTEN as well as non-synonymous sequence alterations in exon 7 have been detected in approximately 15% and 25% of T-ALL patients, respectively (Gutierrez et al., 2009). In contrast, analysis of both leukemic cell lines and primary AML blasts indicate that PTEN mutations are rare in AML (Aggerholm et al., 2000; Liu et al., 2000). In addition to mutations in PTEN itself, aberrant PTEN expression may also be caused by mutations in its upstream regulators. Both enhanced casein kinase 2 (CK2) expression/activity and enhanced ROS levels appear, for example, to correlate with decreased PTEN phosphatase activity in T-ALL cells (Silva et al., 2008). Both CK2 inhibitors and ROS scavengers were sufficient to restore PTEN activity and impaired PI3K/PKB signaling in those T-ALL cells, demonstrating that aberrant CK2 and ROS levels may affect PI3K signaling in leukemia (Silva et al., 2008). Another important, negative regulator of PI3K activity that has been demonstrated to play a critical role in hematopoiesis is SHIP1. Analysis of primary T-ALL cells revealed that full length SHIP1 expression is often low or undetectable. However, when using an antibody against the C terminal domain of SHIP1, low molecular weight proteins can frequently be observed. These low molecular weight proteins are thought to be the result of mutation induced alternative splicing (Lo et al., 2009). In addition, in leukemic cells from an AML patient, a mutation in the phosphatase domain of SHIP1 has also been detected which results in reduced catalytic activity and enhanced PKB phosphorylation (Luo et al., 2003).

For an overview of all known mutations affecting PI3K/PKB signaling, see table 1.

Constitutive activation of PKB has been demonstrated in a significant fraction of AML patients (Min et al., 2003; Xu et al., 2003; Zhao et al., 2004; Grandage et al., 2005; Gallay et al., 2009). Until recently, no PKB mutations were found in patients with leukemia. However, an activating mutation in the pleckstrin homology domain of PKB (E17K) has recently been detected in solid tumors (Carpten et al., 2007). Transplantation of mice with bone marrow cells ectopically expressing this E17K mutation was sufficient to induce leukemia, ten weeks after transplantation (Carpten et al., 2007). Although this particular mutation has been observed in different types of cancer, it appears to be rare in leukemic patients. Thus far, this

activity, including PTEN and SHIP1.

**3.1.2 PKB** 

A third, important mediator of PI3K/PKB signaling is mTOR. Conditional deletion of TSC1 in mice, resulting in activation of mTOR, has been demonstrated to enhance the percentage of cycling HSCs and to reduce the self-renewal capacity of HSCs in serial transplantation assays (Chen et al., 2008). In addition, a reduction in the number of granulocytes and lymphocytes has been observed in those mice (Chen et al., 2008). As described above, activation of the PI3K signaling pathway by conditional deletion of PTEN in adult murine HSCs resulted in an initial expansion followed by exhaustion of LT-HSCs. Inhibition of mTOR in murine HSCs deficient for PTEN with Rapamycin was sufficient to revert this phenotype, again suggesting that mTORC1 signaling plays an important role in proliferation of HSCs (Yilmaz et al., 2006). A role for mTOR in progenitor expansion has been demonstrated utilizing an *ex vivo* human granulocyte differentiation system (Geest et al., 2009). In contrast to inhibition of PKB activity which not only affects progenitor expansion but also alters lineage development (Buitenhuis et al., 2008), inhibition of mTOR activity with Rapamycin only reduced the expansion of hematopoietic progenitors, during both eosinophil and neutrophil differentiation, without altering levels of apoptosis or maturation (Geest et al., 2009). Similarly, inhibition of mTOR reduced the number of interstitial DCs and Langerhans cells in *in vitro* experiments (van de Laar et al., 2010). In contrast to granulocyte development, treatment with rapamycin appears not only to affect proliferation during megakaryocyte (MK) development, but also appears to delay the generation of pro-platelet MKs (Raslova et al., 2006). Similar to FOXO transcription factors, TSC1 also appears to be involved in regulation of ROS levels in HSCs. Elevated levels of ROS have been observed in TSC1 deficient mice. *In vivo* treatment of those mice with a ROS antagonist restored HSC numbers and function (Chen et al., 2008), suggesting that TSC1 regulates HSC numbers at least in part via ROS. In addition to GSK3, the activity of C/EBP also appears to be regulated by mTOR, albeit in a different manner. It has recently been shown that the ratio of wild type C/EBP (C/EBPp42) and truncated C/EBPp30, which is generated by alternative translation initiation, is decreased by mTOR, resulting in high levels of the smaller p30 C/EBP isoform (Fu et al., 2010) that inhibits trans-activation of C/EBP target genes in a dominant-negative manner (Pabst et al., 2001) and binds to the promoters of a unique set of target genes to suppress their transcription (Wang et al., 2007).

#### **3. PI3K/PKB signaling and malignant hematopoiesis**

#### **3.1 Deregulated PI3K/PKB signaling in malignant hematopoiesis**

The above described studies clearly demonstrate that the PI3K/PKB signaling module plays a critical role in regulation of hematopoiesis. Since constitutive activation of PI3K and/or its downstream effectors has been observed in a high percentage of patients with hematological malignancies, it is likely that the development of leukemia may at least in part depend on aberrant regulation of this signaling module.

#### **3.1.1 PI3K**

Constitutive activation of class I PI3K isoforms has been observed in a high percentage of patients with acute leukemia (Kubota et al., 2004; Silva et al., 2008; Billottet et al., 2009; Zhao, 2010). In contrast to the expression of p110, and which is only up-regulated in leukemic blasts of some patients, p110expression appears to be consistently up-regulated in cells from patients with either AML or APL (Sujobert et al., 2005; Billottet et al., 2009). Activating mutations in p110, have been detected in a wide variety of human solid tumors (Ligresti et al., 2009). The most common mutations in p110 are located in the kinase domain (H1047R) and in the helical domain (E545A) (Lee et al., 2005). The E545A mutation has also been detected in acute, but not further specified, leukemia, albeit in a very low percentage (1/88) (Lee et al., 2005). In a series of 44 pediatric T-ALL patients, activating mutations in the catalytic subunit of PI3K (PIK3CA) have been observed in 2 patients, while in frame insertions/deletions have been detected in the PI3K regulatory subunit PIK3R1 in two other patients (Gutierrez et al., 2009). Transplantation of mice with bone marrow cells ectopically expressing mutated p110 resulted in the development of a leukemia-like disease within 5 weeks after transplantation (Horn et al., 2008), suggesting that mutations in p110 would be sufficient to induce leukemia. However, since mutations in PI3K appear to be very rare, it is unlikely that these mutations would be a major cause of leukemic development. Alternatively, the constitutive activation of PI3K observed in many patients with leukemia could also be caused by either aberrant expression or activation of modulators of PI3K activity, including PTEN and SHIP1.

Reduced expression of PTEN has, for example, been observed in different types of leukemia (Xu et al., 2003; Nyakern et al., 2006). Both homozygous and heterozygous deletion of PTEN as well as non-synonymous sequence alterations in exon 7 have been detected in approximately 15% and 25% of T-ALL patients, respectively (Gutierrez et al., 2009). In contrast, analysis of both leukemic cell lines and primary AML blasts indicate that PTEN mutations are rare in AML (Aggerholm et al., 2000; Liu et al., 2000). In addition to mutations in PTEN itself, aberrant PTEN expression may also be caused by mutations in its upstream regulators. Both enhanced casein kinase 2 (CK2) expression/activity and enhanced ROS levels appear, for example, to correlate with decreased PTEN phosphatase activity in T-ALL cells (Silva et al., 2008). Both CK2 inhibitors and ROS scavengers were sufficient to restore PTEN activity and impaired PI3K/PKB signaling in those T-ALL cells, demonstrating that aberrant CK2 and ROS levels may affect PI3K signaling in leukemia (Silva et al., 2008). Another important, negative regulator of PI3K activity that has been demonstrated to play a critical role in hematopoiesis is SHIP1. Analysis of primary T-ALL cells revealed that full length SHIP1 expression is often low or undetectable. However, when using an antibody against the C terminal domain of SHIP1, low molecular weight proteins can frequently be observed. These low molecular weight proteins are thought to be the result of mutation induced alternative splicing (Lo et al., 2009). In addition, in leukemic cells from an AML patient, a mutation in the phosphatase domain of SHIP1 has also been detected which results in reduced catalytic activity and enhanced PKB phosphorylation (Luo et al., 2003). For an overview of all known mutations affecting PI3K/PKB signaling, see table 1.

#### **3.1.2 PKB**

48 Acute Leukemia – The Scientist's Perspective and Challenge

A third, important mediator of PI3K/PKB signaling is mTOR. Conditional deletion of TSC1 in mice, resulting in activation of mTOR, has been demonstrated to enhance the percentage of cycling HSCs and to reduce the self-renewal capacity of HSCs in serial transplantation assays (Chen et al., 2008). In addition, a reduction in the number of granulocytes and lymphocytes has been observed in those mice (Chen et al., 2008). As described above, activation of the PI3K signaling pathway by conditional deletion of PTEN in adult murine HSCs resulted in an initial expansion followed by exhaustion of LT-HSCs. Inhibition of mTOR in murine HSCs deficient for PTEN with Rapamycin was sufficient to revert this phenotype, again suggesting that mTORC1 signaling plays an important role in proliferation of HSCs (Yilmaz et al., 2006). A role for mTOR in progenitor expansion has been demonstrated utilizing an *ex vivo* human granulocyte differentiation system (Geest et al., 2009). In contrast to inhibition of PKB activity which not only affects progenitor expansion but also alters lineage development (Buitenhuis et al., 2008), inhibition of mTOR activity with Rapamycin only reduced the expansion of hematopoietic progenitors, during both eosinophil and neutrophil differentiation, without altering levels of apoptosis or maturation (Geest et al., 2009). Similarly, inhibition of mTOR reduced the number of interstitial DCs and Langerhans cells in *in vitro* experiments (van de Laar et al., 2010). In contrast to granulocyte development, treatment with rapamycin appears not only to affect proliferation during megakaryocyte (MK) development, but also appears to delay the generation of pro-platelet MKs (Raslova et al., 2006). Similar to FOXO transcription factors, TSC1 also appears to be involved in regulation of ROS levels in HSCs. Elevated levels of ROS have been observed in TSC1 deficient mice. *In vivo* treatment of those mice with a ROS antagonist restored HSC numbers and function (Chen et al., 2008), suggesting that TSC1 regulates HSC numbers at least in part via ROS. In addition to GSK3, the activity of C/EBP also appears to be regulated by mTOR, albeit in a different manner. It has recently been shown that the ratio of wild type C/EBP (C/EBPp42) and truncated C/EBPp30, which is generated by alternative translation initiation, is decreased by mTOR, resulting in high levels of the smaller p30 C/EBP isoform (Fu et al., 2010) that inhibits trans-activation of C/EBP target genes in a dominant-negative manner (Pabst et al., 2001) and binds to the promoters of a unique set of target genes to suppress their transcription (Wang et al., 2007).

**3. PI3K/PKB signaling and malignant hematopoiesis** 

aberrant regulation of this signaling module.

**3.1.1 PI3K** 

**3.1 Deregulated PI3K/PKB signaling in malignant hematopoiesis** 

The above described studies clearly demonstrate that the PI3K/PKB signaling module plays a critical role in regulation of hematopoiesis. Since constitutive activation of PI3K and/or its downstream effectors has been observed in a high percentage of patients with hematological malignancies, it is likely that the development of leukemia may at least in part depend on

Constitutive activation of class I PI3K isoforms has been observed in a high percentage of patients with acute leukemia (Kubota et al., 2004; Silva et al., 2008; Billottet et al., 2009; Zhao, 2010). In contrast to the expression of p110, and which is only up-regulated in leukemic blasts of some patients, p110expression appears to be consistently up-regulated in cells from patients with either AML or APL (Sujobert et al., 2005; Billottet et al., 2009). Activating Constitutive activation of PKB has been demonstrated in a significant fraction of AML patients (Min et al., 2003; Xu et al., 2003; Zhao et al., 2004; Grandage et al., 2005; Gallay et al., 2009). Until recently, no PKB mutations were found in patients with leukemia. However, an activating mutation in the pleckstrin homology domain of PKB (E17K) has recently been detected in solid tumors (Carpten et al., 2007). Transplantation of mice with bone marrow cells ectopically expressing this E17K mutation was sufficient to induce leukemia, ten weeks after transplantation (Carpten et al., 2007). Although this particular mutation has been observed in different types of cancer, it appears to be rare in leukemic patients. Thus far, this

The PI3K/PKB Signaling Module in Normal and Malignant Hematopoiesis 51

PI3K signaling. Constitutive activation of FMS-like tyrosine kinase 3 (FLT3), by internal tandem duplication (Flt3-ITD) (Brandts et al., 2005) and mutation in c-Kit (Ning et al., 2001) have, for example, been demonstrated to induce PKB activity. This induction of PKB activity appears to be essential for the survival and proliferation of cells expressing FLT3-ITD (Brandts et al., 2005) or mutated c-Kit (Hashimoto et al., 2003; Cammenga et al., 2005; Horn et al., 2008). In addition to these tyrosine kinase receptors, the activity of the PI3K/PKB pathway can also be enhanced by several fusion proteins, including Bcr-Abl, which can be detected in virtually all patients with CML (Ben-Neriah et al., 1986) and in patients with ALL (Clark et al., 1988). It has been demonstrated that the PI3K/PKB signal transduction pathway plays an important role in Bcr-abl mediated leukemic transformation (Varticovski et al., 1991; Skorski et al., 1997; Hirano et al., 2009). Other potential regulators of PI3K often mutated in leukemia include Ras (Rodriguez-Viciana et al., 1994; reviewed by Schubbert et al., 2007; Gutierrez et al., 2009) Evi1 (Yoshimi et al., 2011) and PP2A. In AML patients, decreased PP2A activity has, for example, been reported to correlate with enhanced levels of PKB phosphorylation on Thr308 (Gallay et al., 2009). In addition, restoration of PP2A

activity also resulted in a reduction of PKB phosphorylation (Cristobal et al., 2011).

As described above, the PI3K/PKB signaling module appears to be aberrantly regulated in a large fraction of patients with leukemia. Recent evidence suggests that the level of PI3K/PKB activation in leukemic blasts could be used to predict the survival rate of patients. Comparison of pediatric T-ALL patients with either no mutations in PTEN, monoallelic mutations or bi-allelic mutations revealed that the survival rate of patients positively correlates with the level of PTEN (Jotta et al., 2010). Similar observations were made in a different cohort of pediatric T-ALL patients, in which PTEN deletions correlated with early treatment failure in T-ALL (Gutierrez et al., 2009). These studies suggest that constitutive activation of PI3K and its downstream effectors reduces the survival rate of ALL patients. To determine whether the level of mTOR activity similarly correlates with reduced survival of ALL patients, mice were transplanted with blasts from pediatric de novo B cell progenitor ALL patients. In those experiments, a rapid induction of leukemia correlated with enhanced mTOR activity in the leukemic blasts (Meyer et al., 2011). In addition to ALL, constitutive activation of PI3K, as measured by enhanced FoxO3 expression or phosphorylation, is also considered to be an independent adverse prognostic factor in AML patients (Santamaria et al., 2009; Kornblau et al., 2010). In addition, a reduced survival rate has also been observed in AML patients displaying enhanced levels of phosphorylated, and therefore inactive, PTEN (Cheong et al., 2003) and phosphorylated PKB on Serine 473 (Kornblau et al., 2006) and Threonine 308 (Gallay et al., 2009). In contrast, Tamburini *et al.* suggest that PI3K activity, as was determined by analysis of the level of phosphorylation of PKB on Ser473, positively correlates with the survival of AML patients (Tamburini et al., 2007). Although the short-term survival rate (within 12 months) appeared to be slightly lower in the group displaying high PKB phosphorylation compared to the group with low levels of phosphorylated PKB, both the long-term survival and relapse free survival were significantly enhanced (Tamburini et al., 2007). Except for this last study, all other studies suggest that enhanced PI3K/PKB activity correlates with reduced survival rate in both ALL and AML patients. The molecular mechanisms underlying this reduced prognosis are, thus far, incompletely understood. However, it has been demonstrated that AML blasts

**3.2 Prognosis of acute leukemia with activated PI3K/PKB signaling** 


mutation has only been detected in one pediatric T-ALL patient (Gutierrez et al., 2009). To date, no other mutations in PKB have been described.
