Chronic Myelogenous Leukemia: Update on Biology and Treatment

Introduction

Chronic myelogenous leukemia (CML) accounts for 15% to 20% of leukemias in adults and occurs with an incidence of 1 to 2 cases per 100,000 population. This myeloproliferative disorder results from neoplastic transformation of hematopoietic progenitor cells and affects myeloid, monocytic, erythroid, megakaryocytic, and lymphoid lineages.

Chronic myelogenous leukemia occurs more frequently in males than in females (ratio of 1.3 to 1).[1] Incidence increases with age, and the median age at presentation is between 45 and 55 years. Up to 30% of patients with CML are 60 years or older, which is an important consideration for the selection of therapeutic strategies, such as stem-cell transplantation and treatment with interferon-alfa (Intron A, Roferon-A).

The outlook for patients with CML has improved dramatically over the last decade, thanks to refinements in allogeneic stem-cell transplantation and growing expertise in the use of interferon-alfa. This review provides a concise update of the biology of CML, as well as current therapeutic options and management strategies.

Clinical Course

Chronic myelogenous leukemia typically follows a biphasic or triphasic course. A chronic phase of variable length precedes an accelerated phase, which is often followed by a blastic phase.

Chronic Phase

Chronic-phase disease is indolent, and up to 50% of patients in this stage have no symptoms and are diagnosed by routine blood testing.[1] Among those who have symptoms, fatigue and anorexia, weight loss, abdominal fullness, left upper quadrant discomfort, early satiety, bleeding, and sweats are encountered most frequently.

In the rare patient with very high white blood cell (WBC) counts, symptoms of hyperviscosity may occur; these include visual changes from retinal hemorrhage, headaches, stupor, tinnitus, and priapism. Physical examination reveals splenomegaly in 50% of patients with CML and hepatomegaly in a lesser percentage.[2]

Marked leukocytosis, anemia, and thrombocytosis are common laboratory features at presentation.[3] Granulocytes are present in all stages of maturation. The activity of leukocyte alkaline phosphatase (LAP score) is reduced in almost all patients and can be used to distinguish CML from other myeloproliferative disorders. The bone marrow of patients with CML is usually hypercellular and may reveal reticulin fibrosis, especially with disease progression.[2]

Accelerated and Blastic Phases

Chronic myelogenous leukemia invariably transforms and becomes refractory to therapy with such agents as hydroxyurea (Hydrea) and busulfan(Drug information on busulfan) (Myleran). It then enters an accelerated phase, which is characterized by basophilia and increases in peripheral blood blast and promyelocyte counts. The definition of accelerated-phase disease is vague and relies on several generally accepted clinical and laboratory criteria (Table 1).[4]

In about 75% of patients, accelerated-phase disease is followed, after 3 to 18 months, by a blastic phase, which resembles acute leukemia and causes the death of the patient within 3 to 6 months. One-fourth of patients develop blastic-phase disease without an intervening accelerated phase.[4]

The blastic phase of CML is usually defined by the presence of extramedullary infiltrates of leukemic cells or blast counts in excess of 30% in peripheral blood or marrow. In one-third of cases, blasts are characterized by lymphoid morphology and expression of lymphoid markers, such as terminal deoxynucleotidyl transferase (TdT) or CD10 (common acute lymphoblastic leukemia antigen [CALLA]). The remaining two-thirds of patients have either the acute myeloblastic leukemia (AML) or acute undifferentiated leukemia (AUL) phenotype and form a heterogeneous group.[5]

Patients with lymphoid blastic-phase CML may respond to treatment with regimens active against acute lymphoid leukemia (ALL).[6] The complete response rates with such therapies are 50% to 60%, and median survival durations range from 9 to 12 months.

Initial Diagnostic Work-up

The essential diagnostic work-up for CML at presentation includes a complete blood count (CBC) with platelets and differential blood count, marrow aspiration for morphology (percentages of blasts and basophils) and biopsy, and cytogenetic studies to demonstrate the Philadelphia (Ph) chromosome and other markers of clonal evolution. Approximately 5% of patients present with a morphologic picture consistent with CML without documentation of the Ph chromosome by cytogenetic studies. In these cases, molecular studies, either Southern blot analysis for detection of the bcr-abl rearrangement or assays (eg, Western blot analysis to detect the bcr-abl protein) to characterize the specific fusion messenger RNA (mRNA) and protein product (p210^bcr-abl, p190b^cr-abl, or p230^bcr-abl), should be performed.

The Philadelphia Chromosome

The Ph chromosome results from a reciprocal translocation between the long arms of chromosome 9 and chromosome 22. This transposes the large 3¢ segment of the c-abl gene from chromosome 9q34 to the 5¢ part of the bcr gene on chromosome 22q11 in a head-to-tail fashion, creating a hybrid bcr-abl gene that is transcribed into a chimeric bcr-abl mRNA (Figure 1).

Initially described by Nowell and Hungerford in 1960, the Ph chromosome became the first chromosomal abnormality to be associated with a specific neoplastic disorder.[7] Translocation t(9;22)(q34;q11) can be demonstrated in more than 90% of patients with CML. It is also seen in up to 5% of children and 15% to 30% of adults with ALL and in 2% of patients with AML who showed no evidence of a preceding CML phase.[8]

Role of the bcr-abl Fusion Gene in CML Pathogenesis

The c-abl gene is a proto-oncogene that encodes a nonreceptor tyrosine kinase with a molecular mass of 145 kd (p145^c-abl) that is localized in both cytoplasm and nucleus. It consists of 11 exons (also referred to as a1 to a11) and spans 230 kilobases (kb). Exon 1 has two alternative forms, 1a and 1b. In most cases, the breakpoint in the abl gene occurs in the 5¢ part of abl exon a2, within the segment between exons 1a and 1b.[8]

Abl exons a2 to a11 are transposed into a region of the bcr gene between exons 12 and 16 (also referred to as b1 to b5) on chromosome 22, which extends over 5.8 kb and is called the major breakpoint cluster region (M-bcr). The breakpoint locations fall either 5¢ between exons b2 and b3 or 3¢ between exons b3 and b4, creating a bcr-abl fusion mRNA of 8.5 kb with either a b2a2 or b3a3 junction (Figure 1). The fusion mRNAs are translated into a 210-kd chimeric protein called p210^bcr-abl.[8]

In about 50% of adults and 80% of children with Ph-positive ALL, the breakpoint on chromosome 22 falls 5¢ of the M-bcr within a long intron segment separating alternative exon e2¢ from exon e2; this is called the minor breakpoint cluster region (m-bcr).[9] Splicing out exons e1¢ and e2¢ creates an e1a2 junction of the bcr-abl transcript and a smaller bcr-abl fusion protein of 190 kd termed p190b^cr-abl.

A third breakpoint location in the bcr gene has been identified 3¢ from the M-bcr region, between exons e19 and e20 (also referred to as c3 and c4); this breakpoint creates a fusion transcript with an e19a2 junction. The translation product is a 230-kd protein termed p230^bcr-abl. Although expression of these proteins is rare in CML, associations of the p190^bcr-abl variant with a prominent monocytic component and of p230^bcr-abl with the chronic neutrophilic leukemia variant have been described.[10,11]

Both p210^bcr-abl and p190b^cr-abl demonstrate significantly higher tyrosine phosphokinase activity than the normal c-abl protein.[8] Several other functional sequences of the bcr-abl protein assume transforming capacity by generating multiple protein-protein interactions that initiate diverse signaling pathways.[12]

Evidence for a direct link between the expression of bcr-abl fusion gene products and abnormal proliferation and malignant behavior of hematopoietic progenitor cells comes from experiments using in vitro and in vivo models of tumor development. In vitro bone marrow culture assays have shown that bcr-abl causes factor-independent and leukemogenic cell growth in hematopoietic cell lines.[13]

Several in vivo animal systems using transgenic mouse models or retrovirus-mediated gene transfer of bcr-abl into murine hematopoietic cells have demonstrated that diverse hematologic malignancies can be generated, among them a syndrome that closely resembles the chronic phase of human CML (reviewed in reference 14). These data support the hypothesis that the v-abl and bcr-abl gene transcripts are central mediators of myeloid proliferation and transformation in CML.

Detection of bcr-abl—Cytogenetic analysis demonstrates the Ph chromosome in 90% of patients with CML. Such analysis is tedious and time-consuming, allows the examination of only 20 to 25 metaphases per bone marrow sample, and misses the 5% of patients who are Ph-negative but bcr-abl-positive. Despite these shortcomings, cytogenetic analysis is the gold standard in the diagnosis of CML.

Molecular tools are important for detecting the molecular abnormalities associated with Ph and also for monitoring the course of disease during treatment. These include polymerase chain reaction (PCR), as well as Southern blot and Western blot analyses.

Quantitative reverse transcriptase–polymerase chain reaction (RT-PCR) is the method of choice for following patients with CML after stem-cell transplantation.[15] Its use for monitoring patients receiving interferon-alpha is not well-defined.

Fluorescence in situ hybridization (FISH) allows for the analysis of both metaphase and nondividing interphase cells.[16] Results of FISH studies are easily quantifiable.

Interphase fluorescence in situ hybridization (i-FISH) is performed on peripheral blood specimens and, thus, avoids the need for bone marrow aspirations. It is fast and permits the analysis of more cells than is possible with conventional cytogenetics. Interphase FISH has a false-positive incidence of up to 10%.

Hypermetaphase fluorescence in situ hybridization (h-FISH) analyzes up to 500 metaphases per sample. In contrast to i-FISH, h-FISH produces no false-positive results. However, peripheral blood samples are not suitable for analysis with h-FISH.[17]

A recently introduced FISH technique uses double-color probes for the detection of Ph-positive leukemias.[18] Despite promising results, not enough data are available to validate its impact on disease monitoring.

Cytogenetic Events in Disease Transformation

The Ph chromosome is the predominant cytogenetic abnormality during chronic phase. Clonal evolution is observed in 50% to 80% of patients during the transition from chronic to accelerated and blastic phase (Table 2). These changes may precede the hematologic and clinical manifestations of transformed CML.

Mitelman[19] described cytogenetic route changes in CML evolution. Minor changes were monosomies of chromosomes 7, 17, and Y; trisomies of chromosomes 17 and 21; and translocation t(3;21)(q26;q22). Major changes were trisomy 8, isochromosome i(17q), trisomy 19, and an extra Ph chromosome (double Ph). Trisomy 8 is most common, especially during myeloid transformation. Isochromosome i(17q) is seen almost exclusively in myeloid-type blastic phase.

In some cases, alterations in molecular mechanisms correspond to cytogenetic changes during the progression of CML. These include abnormalities of p53 (on chromosome 17p13), RB1 (13q14), c-myc (8q24), p16^INK4A (9p21), ras, and AML/EVI-1, a fusion protein resulting from translocation t(3;21)(q26;q22). The incidence of these molecular abnormalities appears to be low, however.