THE CHRONIC LEUKAEMIAS
Chronic myelogenous leukaemia (CML), also called chronic myeloid leukaemia, is a clonal myeloproliferative neoplasm of the primitive hematopoietic stem cell that is characterized by overproduction of cells of the myeloid series, resulting in marked splenomegaly and leukocytosis. Basophilia and thrombocytosis are common. A characteristic cytogenetic abnormality, the Philadelphia (Ph) chromosome, which produces the fusion oncogene BCR-ABL, is present in the bone marrow cells in more than 90% of cases. Most patients (85 to 90%) present in the chronic phase. Eventually, if poorly controlled, CML evolves into the accelerated and blastic phases.
CML constitutes one fifth of all cases of leukaemia in the United States. It is diagnosed in 1 or 2 persons per 100,000 per year and has a slight male preponderance. This incidence of 4800 to 5000 cases annually has not changed significantly in the past few decades. The incidence of CML increases with age; the median age at diagnosis is 50 to 55 years. Ph-positive BCR-ABL- positive CML is uncommon in children and adolescents. No familial association of CML has been noted; for example, the risk is not increased in monozygotic twins or in relatives of patients with CML. Because of the availability of effective therapy, the annual mortality has been reduced from 15 to 20%, before 2000, to 1 to 2% currently. Thus, the prevalence of CML is predicted to increase gradually, from 15,000 to 20,000 cases before 2000 up to 180,000 cases by 2030 in the United States.
Usually, no etiologic agent is incriminated in CML. Exposure to ionizing radiation (e.g., in survivors of the atomic bomb explosions in Japan in 1945, in those undergoing radiation treatment for ankylosing spondylitis or cervical cancer) increases the risk for CML; the peak incidence occurs 5 to 12 years after exposure and is dose related. No increase in the risk for CML has been demonstrated among individuals working in the nuclear industry. Radiologists working without adequate protection before 1940 were more likely to develop myeloid leukaemia, but no such association has been found in recent studies. Benzene exposure increases the risk for acute myelogenous leukaemia (AML) but not of CML. CML is not a frequent secondary leukaemia after treatment of other cancers with radiation, alkylating agents, or both.
The Ph chromosome abnormality, present in more than 90% of patients with typical CML, results from a balanced translocation of genetic material between the long arms of chromosomes 9 and 22: t(9;22)(q34;q11.2). The breakpoint at band q34 of chromosome 9 results in translocation of the cellular oncogene ABL1 (previously c-ABL) to a region on chromosome 22 coding for the major breakpoint cluster region (BCR). ABL1 is a homologue of v-ABL, the Abelson virus that causes leukaemia in mice. This translocation allows juxtaposition of a 5′ portion of a BCR and 3′ position of ABL; the two genetic sequences produce a new hybrid oncogene (BCR-ABL1), which codes for a novel BCR-ABLl oncoprotein with a molecular weight of 210 kD (p210 BCR-ABLI ). The p210 BCR-ABLI oncoprotein results in uncontrolled kinase activity of BCR-ABLl, which triggers the excessive proliferation and reduced apoptosis of CML cells, thereby giving CML cells a growth advantage over normal cells and suppressing normal haematopoiesis. Although in most cases 100% of the metaphases on cytogenetic analysis show BCR-ABL1, normal stem cells emerge on long-term bone marrow culture and after treatment with interferon-α (IFN-α), imatinib, and other BCR-ABLl-selective tyrosine kinase inhibitors (TKIs).
The constitutive activation of BCR-ABLI results in autophosphorylation and activation of multiple downstream pathways that affect gene transcription, apoptosis, cytoskeletal organization, cytoadhesions, and degradation of inhibitory proteins. The signal transduction pathways implicated involve RAS, mitogen-activated-protein (MAP) kinases, signal transducers and activators of transcription (STAT), phosphatidyl inositol 3-kinase (PI3K), MYC, and others. Many of these interactions are mediated through tyrosine phosphorylation and require binding of the BCR-ABLI to adapter proteins such as GRB-2, CRK, CRK-like protein (CRKL), and SCR homology-containing proteins (SHC). Although imatinib and new-generation TKIs (nilotinib, dasatinib, bosutinib, ponatinib) have been extremely successful at targeting BCR-ABLI, understanding of the pathophysiology of the downstream events of BCR-ABLI is important for the future development of agents that may target these events.
In Ph-positive acute lymphocytic leukaemia (ALL), the breakpoint in BCR is proximal, in the minor BCR, resulting in a smaller BCR gene apposing ABL1 ; the resulting fusion gene, messenger RNA, and BCR-ABLI oncoprotein (p190 BCR-ABLI ) are smaller. A third rare, “micro” BCR breakpoint distal to the major BCR produces a p230 BCR-ABLIhybrid oncoprotein, which is associated with a more indolent CML course.
What induces this molecular rearrangement is unknown. Molecular techniques that amplify detection of BCR-ABL1 have demonstrated its presence in the marrow cells of 25 to 30% of healthy volunteers and 5% of infants, but not in cord blood. Because clinical CML develops in only 1 to 2 of 100,000 individuals (i.e., 1 to 2 per 25,000 to 30,000 individuals who express BCR-ABL in their bone marrow), immune regulatory processes or additional molecular events presumably contribute to the development of CML.
BCR-ABL1 is found only in hematopoietic cells and has its origin close to the pluripotent stem cell. The Ph chromosome occurs in erythroid, myeloid, monocytic, and megakaryocytic cells; less commonly in B lymphocytes; rarely in T lymphocytes; and not at all in marrow fibroblasts. The fusion BCR-ABL1 gene and the p210 protein can be found in cases of morphologically typical CML in which no cytogenetic abnormality occurs or in which changes other than the typical t(9;22) (q34;q11.2) are identified. These patients have a survival rate and a response to therapy similar to those of patients with Ph-positive CML. Patients with atypical CML (usually older and more frequently exhibiting anaemia, thrombocytopenia, monocytosis, and dysplasia) who are Ph negative and BCR-ABL1 negative have a worse prognosis than those who are either Ph positive or Ph negative and BCR-ABL1 positive; they more closely resemble patients with myelodysplastic syndrome. Thus, three groups of patients with CML can be identified: (1) those who are positive for Ph and BCR-ABL1;(2) those who are Ph negative but BCR-ABL1 positive; and (3) those who are negative for Ph and BCR-ABL1. PDGFB (previously SIS), which codes for platelet-derived growth factor (PDGF) and is the homologue of the simian sarcoma virus, is also translocated from chromosome 22 to chromosome 9 in CML, but it is distant from the breakpoint and is not expressed.
The pathophysiologic mechanisms underlying CML resistance to TKIs is a fascinating topic that has now been replicated with other targeted therapies in other hematologic and solid tumours. Several mechanisms of resistance to TKIs have been identified; the most common are mutations in the BCR-ABLl kinase domain. More than 100 different mutations have been reported and can involve any of the important domains in the BCR-ABLl structure, including the P-loop (the area where adenosine triphosphate [ATP] binds), the activation loop, and the catalytic domain, as well as the amino acids where imatinib makes contact with BCR-ABLl. The different mutations have considerable variability with respect to resistance to imatinib and other TKIs. Some mutations are overcome by higher concentrations of imatinib than required to inhibit the wild-type form; others are completely insensitive to imatinib. Mutational analysis is useful in patients with imatinib resistance to identify those with the T3151 mutation, who do not respond to imatinib or the second-generation TKIs (dasatinib, nilotinib, bosutinib) but do respond to ponatinib therapy. Knowledge of the sensitivity of the different mutations, as determined by the IC 50 for particular agents, can help select the TKIs.
About 40 to 50% of patients diagnosed with CML do not have symptoms, and the disease is found on routine physical examinations or blood tests. In these patients, the white blood cell (WBC) count may be relatively low at diagnosis. The degree of leukocytosis correlates with tumour burden, as defined by spleen size.
The symptoms of CML, when present, are due to anaemia and splenomegaly; they include fatigue, weight loss, malaise, easy satiety, and left upper quadrant fullness or pain. Rarely, bleeding or thrombosis occurs. Other rare presentations include gouty arthritis (from elevated uric acid levels), priapism (usually with marked leukocytosis or thrombocytosis), retinal haemorrhages, and upper gastrointestinal ulceration and bleeding (from elevated histamine levels due to basophilia). Headaches, bone pain, arthralgias, pain from splenic infarction, and fever are uncommon in the chronic phase but more frequent as CML progresses. Symptoms of leukostasis, such as dyspnoea, drowsiness, loss of coordination, or confusion, which are due to leukocyte sludging in the pulmonary or cerebral vessels, are uncommon in the chronic phase despite WBC counts exceeding 50 × 10 9 cells/µL, but these symptoms appear more frequently in the accelerated or blastic phases of the disease.
Splenomegaly, the most consistent physical sign in CML, occurs in 30 to 50% of cases. Hepatomegaly is less common (10 to 20%) and usually minor. Lymphadenopathy is uncommon, as is infiltration of skin or other tissues. If present, these findings suggest Ph-negative CML or the accelerated or blastic phase of CML.
The diagnosis of typical CML is not difficult. Patients with untreated CML usually have leukocytosis ranging from 10 to 500 × 10 9 /µL. The predominant cells are neutrophils, with a left shift extending to blast cells. Basophils and eosinophils are commonly increased. Monocytes may be slightly increased in some cases that overlap with chronic myelomonocytic leukaemia (CMML). Thrombocytosis is common, whereas thrombocytopenia is rare and, if present, suggests a worse prognosis. A haemoglobin level of less than 11 g/dL is present in one third of patients. Some patients demonstrate a cyclic oscillation of the WBC count. The presence of unexplained myeloid leukocytosis with splenomegaly should lead to a bone marrow examination and cytogenetic and molecular analysis.
The bone marrow is hypercellular, with marked myeloid hyperplasia and, at times, evidence of increased reticulin or collagen fibrosis. The myeloid-erythroid ratio is 15 : 1 to 20 : 1. About 15% of patients have 5% or more blast cells in the peripheral blood or bone marrow at diagnosis.
The presence of the t(9; 22) (q34; q11.2) abnormality establishes the diagnosis of CML. If the Ph chromosome is not found in a patient with suspected CML, molecular studies for the presence of the hybrid BCR-ABL1 gene should be performed. About 25 to 30% of patients with a typical morphologic picture of CML who are Ph negative have theBCR-ABL1 rearrangement. The Ph chromosome is usually present in 100% of metaphases, often as the sole abnormality. Between 10 and 15% of patients have additional chromosomal changes (loss of the Y chromosome, trisomy 8, an additional loss of material from 22q, or double Ph). Some patients have complex chromosomal changes involving chromosome 9 or chromosome 22 (Ph variants, three-way translocations).
CML must be differentiated from leukemoid reactions, which usually produce WBC counts lower than 50 × 10 9 /µL, and toxic granulocytic vacuolation, Döhle bodies in the granulocytes, absence of basophilia, and normal or increased leukocyte alkaline phosphatase (LAP) levels (which are typically low in CML). The clinical history and physical examination generally suggest the origin of the leukemoid reaction. Corticosteroids can rarely cause extreme neutrophilia with a left shift, but this abnormality is self-limited and of short duration.
CML may be more difficult to differentiate from other myeloproliferative or myelodysplastic syndromes. Patients with myeloid metaplasia with or without myelofibrosis frequently have splenomegaly, neutrophilia, and thrombocytosis. Polycythaemia vera with associated iron deficiency, which causes normal haemoglobin and haematocrit values, can manifest with leukocytosis and thrombocytosis. Such patients usually have a normal or increased LAP score, a WBC count less than 25 × 10 9 /µL, and no Ph chromosome or BCR-ABL1 rearrangement.
The greatest diagnostic difficulty lies with patients who have splenomegaly and leukocytosis but do not have the Ph chromosome. In some, the BCR-ABL hybrid gene can be demonstrated despite a normal or atypical cytogenetic pattern. Patients who are Ph negative and BCR-ABLI negative are considered to have Ph-negative CML or CMML. Isolated megakaryocytic hyperplasia can be seen in essential thrombocythemia, with marked thrombocytosis and splenomegaly. Some patients who present with clinical characteristics of essential thrombocythemia (with marked thrombocytosis but without leukocytosis) have CML; cytogenetic and molecular studies showing the Ph chromosome, the BCR-ABLI rearrangement, or both lead to the appropriate diagnosis and treatment.
Rarely, patients have myeloid hyperplasia, which involves almost exclusively the neutrophil, eosinophil, or basophil cell lineage. These patients are described as having chronic neutrophilic, eosinophilic, or basophilic leukaemia and do not have evidence of the Ph chromosome or the BCR-ABLI gene but may have other molecular abnormalities. Most patients with chronic neutrophilic leukaemia , characterized clinically by sustained, mature neutrophilic leukocytosis, hepatosplenomegaly, and bone marrow granulocytic hyperplasia, have oncogenic mutations in the gene for colony-stimulated factor 3 receptor ( CSF3R ). Patients with chronic eosinophilic leukaemia (or clonal hypereosinophilic syndrome) have mutations in the genes for platelet-derived growth factor receptor-aα ( PDGFRA ) or PDGFRB or FGFR1; the prototypical abnormality is the FIP1L1-PDGFRA gene fusion. Peripheral blood or bone marrow can be tested for these genetic markers by reverse transcription–polymerase chain reaction (RT-PCR) or interphase/metaphase fluorescent in situ hybridization (FISH), as can the BCR-ABL1 rearrangement for CML.
Evolution to Accelerated and Blastic Phases
More than 90% of patients present with CML in the benign or chronic phase. If symptomatic at presentation, it becomes asymptomatic once the disease is controlled. Death rarely occurs during the chronic phase of CML. When poorly controlled, CML evolves into an accelerated phase, usually defined by the presence of 15% or more blasts, 30% or more blasts plus promyelocytes, 20% or more basophils, thrombocytopenia (platelets <100 × 10 9 /µL) unrelated to therapy, or cytogenetic clonal evolution. The accelerated phase can be also characterized by worsening anaemia; increasing splenomegaly or hepatomegaly; infiltration of nodes, skin, bones, or other tissues; and fever, malaise, and weight loss. In the accelerated phase, bone marrow studies may show dysplastic changes, increased percentages of blasts and basophils, myelofibrosis, and chromosomal abnormalities in addition to the Ph chromosome (clonal evolution). About 5 to 10% of patients present in the accelerated phase.
Before the era of imatinib therapy, the risk for developing accelerated or blastic phase CML was 10% per year in the first 2 years after diagnosis and 15 to 20% per year thereafter, unless therapies such as IFN-α or allogeneic hematopoietic stem cell transplantation (AHSCT) were used. With imatinib, the annual incidence of progression of CML from the chronic to the accelerated or blastic phase has been 2% in the first 10 years of observation. Before imatinib therapy, the median survival of accelerated phase CML was 18 months or less, but survival has now increased to 4 years or more. In de novo accelerated phase CML, the estimated 8-year survival rate with TKI therapy is 75%.
The blastic phase of CML is diagnosed when 30% or more blast cells are present in the bone marrow and/or peripheral blood or when extramedullary blastic disease is present. Most patients develop features of the accelerated phase before progressing to the blastic phase, but 20% of patients evolve quickly into a blastic phase without warning. Most patients in the accelerated or blastic phase have additional chromosomal abnormalities (clonal evolution) such as duplication of the Ph chromosome, trisomy of chromosome 8, or development of an isochromosome 17. The extramedullary blastic phase of CML can occur in the spleen, lymph nodes, skin, meninges (especially in the lymphoid blastic phase), bones, and other sites; extramedullary transformation is usually followed shortly by evidence of marrow involvement. Blastic phase CML is associated with a very poor median survival time of 5 months. About 25% of patients develop a lymphoid blastic phase. Ph-negative and BCR-ABL1 -negative CML often appear to overlap clinically with CMML in their behaviour, progress, and response to therapy and seem to resemble the myelodysplastic syndromes more than Ph-positive CML. A male preponderance and older age are noted; splenomegaly is common (60 to 70%).
Today, there are five TKIs that have been approved for the treatment of CML. Imatinib mesylate was approved in 2001 by the U.S. Food and Drug Administration (FDA) for CML salvage and in 2002 for CML firstline therapy. Nilotinib, a more potent selective BCR-ABL1 TKI, was approved in 2007 for CML salvage and in 2010 for first-line CML therapy. Dasatinib, a dual BCR-ABL TKI, was approved for CML salvage therapy in 2006 and for first-line therapy in 2010. In 2012, two additional TKIs were approved for CML salvage: bosutinib, a dual SRC-ABL1 inhibitor, and ponatinib, a pan BCR-ABL1 kinase inhibitor with selective potency against the resistant T315I mutation. In 2012, omacetaxine mepesuccinate, a semisynthetic cephalotaxine that inhibits protein synthesis, was also approved for the treatment of CML following failure of two or more TKIs.
First-line therapy for CML today includes imatinib, nilotinib, or dasatinib. 45Patients who demonstrate CML resistance or intolerance to treatment may be offered salvage TKI therapy with any of the other available TKIs. The choice of second-line therapy with a TKI versus allogeneic hematopoietic stem cell transplantation (AHSCT) depends on several factors: (1) the patient’s age and general condition; (2) the availability of acceptable donors (related or matched unrelated); (3) whether a patient has intolerance or CML resistance to first-line TKI therapy; (4) the emerging mutation in the resistant CML clone; (5) whether there is additional clonal evolution at the time of relapse; (6) the response to second-line therapy with the new TKI; (7) the estimated safety and success of the AHSCT; and (8) additional comorbid conditions of the patients (e.g., diabetes, pulmonary conditions, cardiac status, prior history of pancreatitis or pulmonary hypertension).
Patients who present with or develop accelerated or blastic phase should receive second-line TKIs to reduce the disease burden and should be offered AHSCT as soon as possible (the exception possibly being de novo CML accelerated phase which may respond durably to first-line TKI therapy, particularly with achievement of a complete cytogenetic response). Patients who develop intolerance to first-line TKI in chronic phase could be offered second-line TKIs as durable therapy, particularly if they achieve complete cytogenetic response. Patients who develop resistance to a first-line TKI in chronic phase are offered a second-line TKI based on their mutation analysis. A T3151 mutation in the CML clone requires therapy with ponatinib and (as of today) early consideration of AHSCT until the results of ponatinib therapy mature. Mutations involving V299L, T315A, or F317LN/I/C are sensitive to nilotinib therapy. Mutations involving Y253H, E255KN, or F359N/C/I are sensitive to dasatinib and bosutinib therapy. Patients who harbour clonal evolution in the CML cells (additional chromosomal abnormalities in the Ph positive cells) or mutations at the time of second-line therapy, or those who do not achieve a complete cytogenetic response by 1 year of second-line TKI therapy, should be considered early for AHSCT. However, if there is no clonal evolution or mutations at the time of second-line therapy, and if the patient achieves complete cytogenetic response with second-line TKI therapy, responses are durable, and TKIs can be continued until evidence of cytogenetic relapse before AHSCT is considered as third-line therapy. Older patients (e.g., 65 to 70 years or older) may forgo a curative option of AHSCT in favour of several years of good disease control. In such patients, TKI therapy with or without additional (older) agents (hydroxyurea, cytarabine, decitabine, 6-mercaptopurine) may sustain less than a complete cytogenetic response (partial, minor) or a complete hematologic response for many years with a good quality of life, and without the risk for mortality or morbidities associated with AHSCT, particularly if the donor is not optimal (unrelated, mismatched).
THERAPY OF CHRONIC MYELOID LEUKEMIA
|First line||· Imatinib 400 mg daily
· Nilotinib 300 mg twice daily
· Dasatinib 100 mg daily
|Second/third line||· Nilotinib, dasatinib, bosutinib, ponatinib
· Allogeneic stem cell transplantation
|Other||· Decitabine, pegylated interferon
· Hydroxyurea, cytarabine, decitabine
· Combinations of tyrosine kinase inhibitor−based regimens
ROLE AND TIMING OF ALLOGENEIC STEM CELL TRANSPLANTATION IN CHRONIC MYELOID LEUKEMIA
|CHRONIC MYELOID LEUKEMIA STATUS||TYROSINE KINASE INHIBITOR THERAPY||ALLOGENEIC STEM CELL TRANSPLANTATION|
|Accelerated or blastic phase||Interim therapy to minimal residual disease||As soon as possible|
|Imatinib failure in chronic phase with T3151 mutation||Ponatinib interim therapy to minimal residual disease||As soon as possible if no good response obtained|
|Imatinib failure in chronic phase; no clonal evolution, no mutations, good initial response||Long-term second-line tyrosine kinase inhibitors||Third-line post second tyrosine kinase inhibitors failure|
|Imatinib failure in chronic phase with clonal evolution or mutation or no cytogenetic response to second-line tyrosine kinase inhibitors||Interim therapy to minimal residual disease||Second line|
|Older (>65-70) post-imatinib failure in chronic phase||Long-term||May forgo allogeneic stem cell transplantation for many years of quality of life|
Recent trials with different TKIs have indicated that early and “deep” cytogenetic and molecular responses are predictive of improved progression-free and overall survival.
Since its discovery in 1999, imatinib mesylate has become standard therapy for CML. Imatinib is a 2-phenylaminopyrimidine derivative that binds to the canonical ATP lining the groove between the N and C lobes of the ABL1 kinase domain, thus blocking the phosphorylation of tyrosine residues on substrate protein. Blocking of ATP binding inactivates the ABLl kinase because it cannot transfer phosphate to its substrate. By inhibiting phosphorylation, imatinib prevents the activation of signal transduction pathways that induce the leukemic transformation processes that cause CML. Imatinib inhibits several tyrosine kinases, including p210 BCR-ABLI , p190 BCR-ABLI , v-ABL, c-ABL, c-Kit, and PDGF receptor.
In a randomized trial of 1106 patients with newly diagnosed CML, imatinib 400 mg/day orally provided significantly higher rates of major cytogenetic response (87% vs. 35%) and complete cytogenetic response (76% vs. 14%), as well as lower rates of progression (8% vs. 26%) and transformation (3% vs. 9%) after 12 months of therapy, compared with the prior standard nontransplantation therapy (a combination of IFN-α and cytosine arabinoside). The longer term follow-up results continue to demonstrate excellent outcomes with imatinib therapy; with a median follow-up of 8 years, the complete cytogenetic response rate (occurring at least once during therapy) is 83%, the estimated 8-year event-free survival rate is 81%, and the transformation-free survival rate 92%. The estimated 8-year survival rate is 85% (93% if only CML-related deaths are included). The annual rate of transformation was 1.5 to 2.8% in the first 3 years and decreased to less than 1% in the subsequent 5 years among patients who continued on imatinib therapy. Therapy with high-dose imatinib or with combinations of imatinib and other agents (e.g., peg-IFN-α2) did not show convincingly improved results compared with standard imatinib 400 mg daily.
RESULTS OF TYROSINE KINASE INHIBITOR THERAPY IN CHRONIC PHASE CHRONIC MYELOGENOUS LEUKEMIA
|THERAPY||LEUKAEMIA STATUS||COMPLETE CYTOGENETIC RESPONSE (%) (AT INDICATED YEAR OF TREATMENT)||MAJOR/COMPLETE MOLECULAR RESPONSES (%) (AT INDICATED YEAR OF TREATMENT)||SURVIVAL (%) (AT INDICATED YEAR AFTER INITIATION OF TREATMENT)|
|Imatinib||First line||65 (5)||40/20 (5)||85 (8-10)|
|Nilotinib||First line||85-87 (4)||73-76/37-40 (4)||95-97 (4)|
|Dasatinib||First line||86 (2)||74/34 (4)||93 (4)|
|Dasatinib||Salvage||50 (5)||43 (6)||71 (6)|
|Nilotinib||Salvage||44 (4)||30-40 (4)||85 (3)|
|Ponatinib||Salvage||45-65 (2)||30-50 (2)||90 (2)|
|Bosutinib||Salvage||40 (2)||30 (2)||92 (2)|
|Omacetaxine||Salvage||10 (2)||—||85-90 (2)|
Imatinib has a 5% or lower rate of serious side effects, which include nausea, vomiting, diarrhoea, skin rash, muscle cramps, bone pain, periorbital or leg edema, weight gain, and rarely, hepatic, renal, or cardiopulmonary dysfunction; most of these are manageable with dose reduction or treatment interruption. Drug-related myelosuppression occurs in 10 to 20% of patients with newly diagnosed CML and is manageable with brief treatment interruptions, dose modifications, or both, or with the administration of growth factors (erythropoietin for anaemia, granulocyte colony-stimulating factor for neutropenia). Imatinib and other TKIs may prolong the cardiac QTc interval; medications that contribute to QTc prolongation should be avoided. Hypophosphatemia associated with altered bone metabolism can occur, and the serum phosphate level should be monitored. Chromosomal abnormalities may appear in the Ph-negative diploid cells in 5 to 10% of responding patients, probably owing to unmasking of a fragile stem cell prone to the development of CML or to chromosomal instability; such changes disappear spontaneously in 70% of cases and rarely evolve into a myelodysplastic syndrome or acute myeloid leukaemia, probably as part of the natural course of CML.
Nilotinib, a selective BCR-ABLl TKI 30 times more potent than imatinib, was initially approved for the treatment of CML after imatinib failure. In CML chronic phase after imatinib failure, nilotinib 400 mg orally twice daily was associated with complete cytogenetic response rates of 40 to 50%. The major molecular response rates, defined as BCR-ABL1 transcripts, 0.1% by International Scale [IS], are 30 to 40%, and the estimated 3-year survival rate is 80%. Subsequent studies compared nilotinib to imatinib in newly diagnosed patients with CML. In a three-arm randomized study, patients received imatinib 400 mg daily, nilotinib 400 mg twice a day, or nilotinib 300 mg twice daily. With a minimum follow-up of 4 years, the two arms of nilotinib demonstrate better early results compared with imatinib. There was no difference in the estimated 4-year survival rates (94%, 97%, 93%, respectively). Nilotinib therapy was associated with lower rates of fluid retention, diarrhoea, headaches, muscle cramps, nausea and vomiting, and neutropenia. However, it was associated with higher rates of headache, rash, pruritus, and hyperglycaemia, and with a low but notable incidence of pancreatitis (<2%), ischemic heart disease (4 to 5% vs. 1% with imatinib), and peripheral arterial occlusive disease (1.4 to 1.8% vs. 0%).
Dasatinib, a dual SRC-ABL1 inhibitor, is 300 times more potent than imatinib. In chronic phase CML after imatinib failure, dasatinib therapy was associated with complete cytogenetic response rates of 45 to 60%, major molecular response rate of 43%, and estimated 6-year survival rate of 71%. A first-line study comparing dasatinib to imatinib (DASISION trial) randomized patients to receive either imatinib 400 mg daily or dasatinib 100 mg daily. With a minimum follow-up of 48 months, the incidence of complete cytogenetic response by 24 months was 86% with dasatinib and 82% with imatinib. The incidence of major molecular response was 74% vs. 60%. The rate of transformation to accelerated or blastic phase was 4.6% vs. 7%. The estimated 4-year progression-free survival rates were similar, 90%. The estimated 4-year survival rates were 93% and 92%, respectively. Dasatinib therapy was associated with lower rates of fluid retention, edema, myalgia, nausea, vomiting, and rashes. However, it was associated with higher rates of pleural effusions (about 10 to 15%) and cytopenias, particularly thrombocytopenia, as well as a low but notable incidence of pulmonary hypertension (<2 to 3%).
Bosutinib, a dual SRC-ABLl inhibitor (similar to dasatinib), is 30 to 200 times more potent than imatinib. It has minimal inhibitory activity against c-Kit and PDGF receptor and therefore is expected to produce less myelosuppression and fewer pleural effusions. In studies of patients with chronic phase CML after imatinib failure treated with bosutinib 500 mg orally daily, the major cytogenetic response rate was 53%, the complete cytogenetic response rate was 41%, and the estimated 2-year survival rate was 92%. Grade 3 to 4 toxicities included diarrhoea (8%), rashes (9%), and thrombocytopenia (5 to 10%).
Ponatinib is a pan-BCR-ABLl TKI with potent activity against native and mutated BCR ABLl kinases, including T3151. In a phase II study of patients with chronic, accelerated, or blastic (CML) or Ph-positive acute lymphocytic leukaemia, all of whom were resistant or intolerant to several TKIs, ponatinib showed high activity. In these chronic phase CML patients, the complete cytogenetic response rate was 44%, and the major molecular response rate was 30%. In the subset of patients with T3151 mutation, the major cytogenetic response rate was 70%, the complete cytogenetic response rate 66%, and the major molecular response rate 50%. Significant side effects included pancreatitis (5 to 10%), thrombocytopenia (30%), and skin rash (30%). Because of the cumulative incidence of serious thrombotic events since the approval of ponatinib, the FDA has restricted its use to patients resistant to other TKIs. Additional clinical trials assessing the safety profile and different dose schedules of ponatinib are being considered.
Omacetaxine mepesuccinate, a semisynthetic analogue of homoharringtonine, is a first-in-class cephalotaxine that acts as a protein synthesis inhibitor that induces apoptosis in leukemic cells by reducing levels of multiple oncoproteins, including BCR-ABL1. Data pooled from two phase II trials of subcutaneous omacetaxine, 1.25 mg/m 2 twice daily for 2 weeks every 4 weeks until response, then for 1 week every 4 weeks, in patients with chronic phase CML after failure of two TKIs, a major cytogenetic response was reported in 20% and complete cytogenetic response in 10%. Grade 3 to 4 side effects included cytopenias in 37 to 67% of patients, which were reversible. This led to FDA approval of omacetaxine for patients whose disease has progressed despite treatment with two TKIs.
AHSCT, a potentially curative therapy in selected patients with CML, is most effective during the chronic phase, when it is associated with a 20-year survival rate of 40 to 50%. Transplant-related mortality rates range from 5 to 50%, depending on the patient’s age, whether the donor is related or unrelated, the degree of matching, and other, less important factors such as positivity for cytomegalovirus, preparative and post-transplantation regimen, and institutional expertise. Disease-free survival rates with related allogeneic stem cell transplantation are 40 to 80% in chronic phase, 15 to 40% in accelerated phase, and 5 to 20% in blastic phase. In chronic phase CML, patients younger than 30 to 40 years have disease-free survival rates of 60 to 80%, compared with only 30 to 40% for patients older than 50 years. A major limitation of allogeneic stem cell transplantation is the availability of related donors. Human leukocyte antigen (HLA)-compatible unrelated donors can be found for 50% of patients; the median time from initiation of the donor search to transplantation is 3 to 6 months.
Nonmyeloablative preparative regimens have expanded the indications for AHSCT to older patients and have reduced transplant-related mortality and complications. Early results show acceptable degrees of engraftment, less mortality and organ damage, more persistent residual disease, and similar degrees of graft-versus-host disease. Patients whose CML recurs after AHSCT may respond to imatinib or new-generation TKIs, donor lymphocyte infusions, IFN-α, or a second AHSCT.
AHSCT can produce an estimated cure rate of 40% at 20 years. However, it is associated with a 1-year mortality rate of 5 to 40% and with morbidities such as cataracts, infertility, second cancers (5 to 10%), immune-mediated complications, and chronic graft-versus-host disease. Delaying AHSCT beyond 1 to 3 years after diagnosis may be associated with worse results and with occasional sudden blastic transformation, which may not be salvageable. The outcome of AHSCT may be even better after exposure to TKIs.
Currently, all three TKIs (imatinib, nilotinib, and dasatinib) are acceptable first-line therapies for CML. The choice of TKI may depend on patient and physician preferences and patient prior history and comorbidities (e.g., diabetes, pancreatitis, cardiopulmonary conditions, and pulmonary hypertension). Current oncology practice trends appear to increasingly favour nilotinib and dasatinib over imatinib as initial therapy because of their better early results, particularly the lower early incidence of CML transformation. However, the costs of TKIs may shift treatment paradigms in some emerging nations to using a particular TKI over others, or even to consider first-line AHSCT (total one-time procedure cost of $30 to $100,000) in situations in which patients or the national health care system cannot afford the burden of the TKI. Imatinib may become available in generic formulations in 2015. The price of generic imatinib is unknown but may be lower than that of other agents ($2000 to $10,000 per year). The choice of first-line TKI therapy may then depend on the differential pricing of generic imatinib versus dasatinib and imatinib and on the maturing long-term data (5 to 8 years) for survival, transformation-free survival, and event-free survival with the three TKIs. With an estimated 8-year survival rate of 93% with imatinib (considering only CML-related deaths) and the high efficacy of new-generation TKIs as salvage therapies, the survival benefit with dasatinib or nilotinib may or may not be apparent compared with imatinib first-line therapy, careful monitoring for cytogenetic relapse, and rapid institution of second-line TKI therapies at that time.
In several studies, the achievement of a complete cytogenetic response (Ph-positive metaphases 0%; BCR-ABLl transcripts 1%) at 12 months or later on TKI therapy was associated with significant survival benefit compared with achievement of lesser degrees of response. Therefore, achievement of complete cytogenetic response is now the primary end point of TKI therapy. The achievement of complete molecular response (nonmeasurable BCR-ABLl transcripts) offers the possibility of treatment discontinuation in the clinical trial setting. Lack of achievement of major molecular response or of complete molecular response should not be interpreted as a need to change TKI therapy or to consider AHSCT. Response assessments at earlier times on first-line TKI therapy (3 to 6 months) have shown better outcomes, with achievement of a major cytogenetic response by 3 to 6 months on imatinib therapy (Ph-positive metaphases 35%, or BCR-ABLl transcripts 10%). Although this is interpreted to mean that a change to a second TKI therapy may be considered if such an outcome is not obtained, no studies have shown that changing therapy from imatinib to second TKI for this indication has improved patient outcome. When nilotinib or dasatinib is used for first-line therapy, achievement of complete cytogenetic response by 3 to 6 months of TKI therapy has been associated with improved outcomes.
Currently, imatinib failure (requiring a change of therapy) should be strictly defined as failure to achieve a major cytogenetic response after 6 months of imatinib therapy and a complete cytogenetic response after 12 months or cytogenetic or hematologic relapse at any later time, on an optimal imatinib dose schedule (adjusting dose for significant side effects or for intolerance and checking for treatment compliance). With the use of second-generation TKIs in the first-line setting, failure of TKI therapy has been suggested to be lack of achievement of complete cytogenetic response or BCR-ABLl transcript levels of 1% by 3 to 6 months of therapy. Such patients (<10% of the denominator) have a worse event-free survival, although their survival at 3 to 5 years remains in the range of 90%, better or equivalent to what would be achievable with AHSCT. Thus, although the early surrogate response parameters at 3 to 6 months on first-line TKI therapy predict for differences in outcome, a change of therapy at that point in time has not been proved to improve longer term prognosis.
Patients with CML whose disease progresses on imatinib therapy may be treated with a newer generation TKI or with AHSCT, as discussed earlier. Patients with CML and failure on first-line dasatinib or nilotinib therapy may possibly be salvaged with ponatinib if the CML clones exhibit a T3151 mutation. If no such mutation is detected, they could be considered for other TKI therapies, AHSCT, treatment with omacetaxine, or combined-modality therapies including TKIs and older agents (hydroxyurea, cytarabine, decitabine). The choice of AHSCT as second-line versus later salvage therapy was discussed earlier.
Patients with accelerated or blastic phase CML may receive initial therapy with TKIs (newer generation TKIs like dasatinib or ponatinib are preferred over imatinib) to reduce the CML burden and may be considered for early AHSCT. Response rates with combinations of TKIs and chemotherapy are 40% in nonlymphoid blastic phase CML and 70 to 80% in lymphoid blastic phase CML. Median survival times are 6 to12 months and 12 to 24 months, respectively. The addition of TKIs to chemotherapy has improved the response rates and prolonged the median survival time in blastic phase CML.
At present, AHSCT is the only curative therapy for accelerated and blastic phase CML; overall cure rates are in the range of 15 to 40% and 5 to 20%, respectively. Patients with cytogenetic clonal evolution as the only accelerated phase criterion have a long-term event-free survival rate of about 60%. Otherwise, TKIs provide hematologic responses in 80% of patients and an estimated 4-year survival rate of 40 to 55% in accelerated phase CML, but only a 40% response rate and a median survival of 9 to 12 months in blastic phase CML. Patients in the accelerated or blastic phase should be encouraged to participate in investigational strategies to improve their prognosis. Patients with de novo CML accelerated phase have a better outcome with first-line TKI therapy than patients who evolve from chronic to accelerated phase. The estimated 6- to 8-year survival rates with TKI therapy in de novo accelerated phase CML are 60 to 80%. Such patients may continue on TKI therapy as their long-term treatment if they achieve a complete cytogenetic response on TKI therapy.
Patients with severe leukocytosis and manifestations of leukostasis may benefit from initial leukapheresis. Severe thrombocytosis uncontrolled with anti-CML measures may respond to anagrelide, thiotepa, IFN-α, 6-mercaptopurine, 6-thioguanine, hydroxyurea, and platelet pheresis. CML during pregnancy may be controlled with pheresis in the first trimester and then with hydroxyurea until delivery. Use of IFN-α during pregnancy has been reported anecdotally to be safe. An analysis of 125 babies delivered to women with CML on imatinib therapy (who discontinued imatinib once the pregnancy was known) showed most babies to be healthy. However, the study demonstrated imatinib therapy to be associated with a syndrome of ocular, skeletal, and renal abnormalities in three babies delivered. Therefore, imatinib (and presumably other TKIs, although there is little experience with them) should be discontinued immediately once pregnancy is documented, but abortion is not recommended because foetal malformations are rare. Partners of men with CML on TKI therapy who become pregnant have delivered normal babies. Splenectomy can be useful as a palliative measure in patients with massive, painful splenomegaly, hypersplenism, or thrombocytopenia.
With TKI therapy, complete cytogenetic response, major molecular response, and even complete molecular response have been achieved. These response rates improve with continued therapy and are higher with new-generation TKIs (dasatinib, nilotinib) compared with imatinib as first-line therapy. Techniques have been developed to measure these responses more accurately (rather than relying on only 20 metaphases by cytogenetic analysis), with less tedious and less painful procedures (peripheral blood rather than marrow studies), and at levels below the level of detection by routine cytogenetic evaluations. FISH studies with improved probes can measure 200 interphase cells using peripheral blood, and they have false-positive rates of less than 2 to 3%. Quantitative PCR tests usually measure the BCR-ABLl transcript levels (ratio of the abnormal message, BCR-ABLl, to a normal message, such as ABL1). BCR-ABLl transcript levels of 0.1% [IS], about a 3-log reduction of disease, have been associated with a very low risk for CML relapse on TKI therapy. This is referred to as a major molecular response. Undetectable BCR-ABL1 transcript levels, (usually <0.0032% [IS], or 4.5 log of reduction) are sometimes referred to as complete molecular response. The percentage of patients achieving complete molecular response is significantly higher with new-generation TKIs compared with imatinib.
In monitoring the response to TKI-based therapies, patients require a bone marrow analysis before treatment (to determine the percentages of blasts and basophils and clonal evolution) and FISH and quantitative PCR analyses. Quantitative PCR can be falsely negative at diagnosis in 5 to 8% of patients with unusual breakpoints and messages (e.g., b2a3 or b3a3) if proper procedures are not used. Thus, knowledge of the pretreatment BCR-ABL1 transcript levels avoids the false assumption of complete molecular response because of the false negativity. Bone marrow analysis may be useful at 6 and 12 months (to assess cytogenetic response and confirm complete cytogenetic response) and once every 1 to 3 years in patients with stable, durable complete cytogenetic responses (to look for chromosomal abnormalities in both Ph-positive and Ph-negative cells). Monitoring in patients with confirmed durable complete cytogenetic responses can be continued with either FISH or quantitative PCR studies every 6 months (or more often, such as every 3 months, if there are concerns about significant and consistent increases in BCR-ABLl transcripts levels). Some CML experts have shifted from monitoring by marrow studies to monitoring by peripheral blood studies using molecular analysis, with or without FISH studies. Among patients achieving major molecular response, molecular analysis without FISH studies is sufficient.
Resistance to imatinib therapy (discussed earlier) requires a change of therapy to other TKIs, TKI combinations with chemotherapy, or consideration of AHSCT. It is important to emphasize that many patients with apparent CML resistance to a TKI therapy may be noncompliant with the treatment. This should be discussed clearly with patients when they exhibit signs of CML progression by either molecular or FISH studies. If they are noncompliant, they may continue on the same TKI treatment with emphasis on compliance and evaluated 3 to 6 months later, before CML resistance is declared.
Among patients in complete cytogenetic response on a particular TKI, failure to achieve a major molecular response does not, at present, indicate resistance to the particular TKI or a need to change therapy. Mutational studies are recommended in patients who develop cytogenetic or hematologic resistance or relapse on a particular TKI therapy, when considering changing therapy to another TKI. The detection rate of mutations in this situation is 30 to 50%. Mutational studies are not recommended in patients in complete cytogenetic response on a particular TKI because the detection rate of mutations are then very low (<3 to 5%).
Treatment of CML with imatinib and other TKIs has revolutionized the outcome of the disease. In patients with newly diagnosed CML, imatinib therapy is associated with an estimated 8- to 10-year survival rate of 85% (93% if non-CML deaths are censored). If this favourable trend continues with longer follow-up, the median survival time in CML may exceed 25 years. The annual mortality rate of CML with TKIs in the first decade of experience has been reduced from the historical rate of 10 to 20%, down to 2% (1% if only CML deaths are counted). Many well established poor prognostic factors in CML (e.g., older age, splenomegaly, presence of marrow fibrosis, deletion of 9q) have lost much of their prognostic importance since the advent of TKI therapy. With AHSCT, cures can be expected in 40 to 80% of patients with chronic phase CML, 15 to 40% of those with accelerated phase CML, and 5 to 20% of those with blastic phase CML.
Chronic Myelomonocytic Leukaemia and Atypical Chronic Myeloid Leukaemias
Although superficially resembling CML in its clinical and morphologic presentation, CMML should be considered a separate entity because of its particular clinical, therapeutic, and prognostic aspects. CMML is a hybrid entity manifesting as a proliferation of the myeloid monocytic series and dysplasia of the erythroid-megakaryocytic series. Patients with CMML are older (median age, 65 to 70 years) than most patients with CML.
The cytogenetic findings in patients with CMML are either normal or involve an additional chromosome 8 or findings other than the Ph chromosome. Patients with CMML have RAS mutations in 40 to 60% of cases.
Patients often present with symptoms related to anaemia and thrombocytopenia (fatigue, bleeding). Other typical features include splenomegaly, leukocytosis, and monocytosis. Organ infiltration (lymph nodes, skin, liver) is less common. Basophilia and thrombocytosis are not presenting features. High-frequency mutations in the granulocyte colony-stimulating factor 3 receptor gene ( CSF3R ) in chronic neutrophilic leukaemia (CNL) and in some patients with atypical chronic myeloid leukaemia (aCML) have been identified. In addition, recurrent mutations in SETBPl(Set binding protein) have been identified in 25% of aCML patients.
AHSCT, which is the only curative modality, should be considered first-line therapy in candidate patients. Other therapies include hydroxyurea to control leukocytosis, erythropoietin to improve anaemia, azacytidine or decitabine (both approved by the FDA for the treatment of CMML), topotecan and cytarabine or other intensive anti-AML regimens for CMML transformation, splenectomy for symptomatic splenomegaly and/or hypersplenism, and investigational agents. Inhibition of Janus kinase 2 or SRC kinase signalling downstream of mutated CSF3R is being explored therapeutically.
Poor prognostic factors include the presence of anaemia (haemoglobin <10 g/dL), thrombocytopenia, and more than 5% blasts. Median survival is 12 to 18 months.
Definition and Epidemiology
Hairy cell leukaemia (HCL) is an uncommon and indolent B-cell leukaemia (1 to 2% of all leukaemias). The median age at diagnosis is 50 years, and there is a 4 : 1 male preponderance.
The cell of origin of HCL is the B lymphocyte, as documented by the demonstration of heavy and light chain immunoglobulin gene rearrangements. In a series of 47 patients, all had a BRAF V600E activating mutation. Hairy cells express CD19, CD20, CD11C, CD103, FMC7, and CD22, but not CD21, CDS, CD10, or CD23. The cells demonstrate a κ or λ light chain phenotype. The cells also express CD25 (TAC), the low-affinity interleukin-2 (IL-2) receptor, and CD103, a unique hairy cell antigen. High levels of soluble IL-2 receptor (more than five times normal) are present in the sera of almost all patients with HCL, with extremely high levels noted in many cases. Immune dysfunction is wide ranging in HCL. Monocytopenia is universal; B and T lymphocytes are decreased in number; the CD4/CD8 (helper T/suppressor T) ratio is often inverted; and skin test reactivity to recall antigens is impaired, as is antibody-dependent cellular cytotoxicity. Humoral immunity is relatively preserved, with normal immunoglobulin levels. Marrow failure in HCL may be due in part to inhibitory factors (e.g., tumour necrosis factor) produced by the leukemic infiltrate; the pancytopenia is often more marked than would be anticipated from the degree of leukemic infiltration.
Most patients present with pancytopenia and splenomegaly. Patients may also have fatigue, fever, weight loss, and infection secondary to granulocytopenia or monocytopenia. Leukocytosis is uncommon, and lymphadenopathy is rare. Anaemia is present in up to 85% of patients, whereas leukopenia and thrombocytopenia are present in 60 to 75%. The cytopenias are caused by a combination of bone marrow failure due to leukemic infiltration and hypersplenism. Patients may experience repeated infections and, rarely, a systemic vasculitis resembling polyarteritis nodosa. Although bacterial infections occur, as would be expected with neutropenia, patients with HCL have a predilection to develop tuberculosis, atypical mycobacterial infections, and fungal infections, perhaps related to the severe monocytopenia that is characteristic of this disorder.
In conjunction with the clinical features, careful examination of the peripheral blood smear may demonstrate the occasional typical cells with cytoplasmic projections, giving rise to the name hai1y cell leukaemia. The hairy cells are 10 to 15 mm in diameter, with pale blue cytoplasm, a nucleus with a loose chromatin structure, and one or two indistinct nucleoli. Bone marrow aspiration is often inadequate owing to increased deposition of reticulum, collagen, and fibrin; bone marrow biopsy is usually necessary. Bone marrow involvement is interstitial or patchy, and the infiltrate is characterized by widely spaced nuclei due to the abundant cytoplasm, giving rise to the commonly described fried-egg appearance.
Hairy cells exhibit a strong acid phosphatase (isoenzyme 5) cytochemical reaction in 95% of cases, a reaction that is resistant to the inhibitory effect of tartaric acid (TRAP). Other lymphoproliferative diseases are rarely TRAP positive. Electron microscopy clearly demonstrates the microvillar projections. Often, ribosomal-lamellar complexes, which are characteristic but not diagnostic of HCL, can be identified. The peroxidase stain is negative, and lysozyme activity is absent in hairy cells, thereby differentiating these cells from monocytes.
The differential diagnosis must distinguish HCL from non-Hodgkin lymphoma or chronic lymphocytic leukaemia (CLL), which can manifest with predominant splenomegaly and minimal lymphadenopathy. Some patients with a myelodysplastic syndrome or a chronic myeloproliferative neoplasm have splenomegaly and pancytopenia with only a few atypical cells. Patients with other diseases, such as systemic lupus erythematosus and other autoimmune diseases, B-cell and T-cell prolymphocytic leukaemias, infiltrative splenomegaly, or tuberculosis, may have splenomegaly and cytopenia, but these diagnoses can usually be made by history, physical examination, and appropriate blood and bone marrow tests. Splenomegaly, cytopenia, and nonaspirable marrow in a middle-aged man should create a high index of suspicion for HCL. Splenectomy or lymph node biopsy is sometimes necessary to establish the diagnosis in difficult cases. Cases of HCL variant manifest with higher WBC counts, are TRAP negative, have prominent nucleoli, and are only occasionally positive for antibodies against CD25. HCL variant does not respond as well to the agents that are usually effective in the management of typical HCL.
A small proportion (<5%) of patients with HCL do not require therapy. These patients have mild cytopenias, are not transfusion dependent, have no history of infections, and have a low level of marrow infiltration by hairy cells. 2-Chlorodeoxyadenosine (cladribine), an adenosine analogue that is resistant to deamination by adenosine deaminase, produces complete remission in more than 80% of HCL patients after a single course of 0.1 mg/kg/day for 7 days given by continuous intravenous infusion, and it is now the recommended first-line therapy. It can also be given at 0.14 mg/kg/day for 5 days as a short daily intravenous infusion. Remissions are durable, and patients who relapse can often attain a second remission after retreatment with cladribine. The drug is well tolerated, with a low infection rate. Despite long-lasting suppression of CD4 +lymphocyte counts, there does not appear to be an increase in late opportunistic infections or second malignancies. Partial response to purine analogs is regarded as a poor prognostic factor, and a second course of purine analog therapy is recommended if patients do not enter complete remission, with the addition of rituximab to be considered. Rituximab in combination with a purine analogue is often used in the treatment of relapsed disease.
Deoxycoformycin (pentostatin; 4 mg/m 2 weekly or every 2 weeks for up to 6 months), an adenosine deaminase inhibitor, produces complete remission in 70 to 80% of patients. The response to treatment is rapid. Toxicity includes nausea and vomiting, infection, renal and hepatic dysfunction, conjunctivitis, and photosensitivity, albeit mild in most cases.
Human leukocyte interferon (HuIFN), or recombinant interferon-α (r-IFN-α), rapidly improves granulocyte, platelet, and haemoglobin levels (within 1 to 3 months); reduces spleen size; and decreases marrow infiltration. Peripheral blood cell counts return to normal in 80% of cases, but complete remission is uncommon. In addition, when treatment is discontinued, relapse occurs within 1 to 2 years. Rituximab, the monoclonal antibody targeting CD20, also produces responses; eight weekly infusions appear to be more effective than four. Two immunotoxins can produce responses in refractory patients. LMB2 is composed of the Fc portion of the anti-TAC antibody linked to a Pseudomonas exotoxin. Moxetumomab also contains a Pseudomonas exotoxin linked to an antibody targeting CD22. The B-RAF inhibitor, vemurafenib, has successfully been used to treat a patient with HCL, and a clinical trial in relapsed HCL is underway. Splenectomy is recommended mainly for patients with splenic infarcts or massive splenomegaly.
More than 85 to 90% of patients treated with cladribine or pentostatin are expected to be alive at 10 years.
Chronic lymphocytic leukaemia (CLL) is a neoplasm characterized by the accumulation of monoclonal lymphocytes of B-cell origin. The cells accumulate in the bone marrow, lymph nodes, liver, spleen, and occasionally other organs. After decades of chemotherapy-based treatment of CLL, recent progress has turned attention to mechanism-driven therapy with targeting of the B-cell receptor signalling pathway.
CLL is the most common leukaemia (one third of all cases) in the Western world and is twice as common as CML. The disease occurs rarely in those younger than 30 years; most patients with CLL are older than 60 years. CLL increases in incidence exponentially with time; by age 80 years, the incidence rate is 20 cases per 100,000 persons per year. The male-to-female ratio is approximately 2 : 1. The incidence of CLL among Asians in Japan and China is only 10% of that in the United States and other Western countries. Intermediate incidence rates are seen in persons of Hispanic origin.
The cause of CLL is unknown. Ionizing radiation and viruses have not been associated with CLL, although hepatitis C infection has recently been associated with splenic lymphoma with villous lymphocytes (another indolent B-cell disorder). Familial clustering in CLL is more common than in other leukaemias; first-degree relatives of patients have a two- to four-fold higher risk and develop CLL at a younger age compared with the general population. Farmers have a higher incidence of CLL than do those in other occupations, raising the possibility of an etiologic role for herbicides or pesticides. Agent Orange, the defoliating agent used in Vietnam, has been associated with the development of CLL.
Leukaemia cells in CLL are homogeneous and have the appearance of normal mature lymphocytes. However, clonality can be documented by the presence of immunoglobulin gene rearrangements and the restriction to either κ or λ light chains on the cell surface. The cells express low-intensity monoclonal surface immunoglobulin (Smig; usually immunoglobulin [Ig] M ± IgD) and the pan-B-cell antigens CD19, CD20, CD23, and CD24 in almost all cases, as well as CD21 (which includes the receptor for the Epstein-Barr virus and the C3d component of complement) in more than 75% of cases. Almost all cells exhibit Ia antigen and receptors for the Fc fragment of IgG and spontaneously form rosettes with mouse erythrocytes. In addition to B cell antigens, CLL cells express CDS (Leu 1, Tl, and TlOl), a pan-T-cell antigen. Other T-cell antigens are absent. CD25 (TAC, IL-2 receptor) antigen is positive in about 25% of cases. T cells are increased in number at diagnosis, and the CD4/CD8 ratio is often inverted, owing to a relatively greater increase in CD8+ cells. The CD4/CD8 ratio declines as the disease progresses and after therapy. The T cells have a blunted response to T-cell mitogens and decreased delayed hypersensitivity reactions to recall antigens. However, these T-cell functions are impaired by factors produced by the CLL cells because purified T cells have a normal response to T-cell mitogens.
Genes mutated in CLL include TP53 (15% of patients), SF3B1 (15%), ATM (9%),MYD88 (10%), and NOTCH1 (4%). Standard cytogenetic analysis identifies abnormalities in 40 to 50% of cases of CLL, but CLL cells have low mitotic activity. By FISH, the likelihood of detecting abnormalities increases to 80%. A 13q deletion is the most common abnormality; other abnormalities include llq deletion (15 to 20%), trisomy 12 (15 to 20%), and 17p deletion (5 to 10%). The 17p deletion increases in frequency as the disease progresses, recurs after therapy, and is associated with a very poor prognosis. The llq deletion also is associated with a poorer prognosis, whereas the 13q deletion, if present as the sole abnormality, is associated with a favourable prognosis.
Most patients with CLL do not have symptoms, and the disease is diagnosed when absolute lymphocytosis is noted in the peripheral blood during evaluation for other illnesses or when the patient undergoes a routine physical examination. Symptoms such as fatigue, lethargy, loss of appetite, weight loss, and reduced exercise tolerance are nonspecific. Many patients have enlarged lymph nodes. B symptoms (fever, night sweats, weight loss) are rarely present initially, and their presence in later stages of the disease suggests transformation to large cell lymphoma (Richter transformation). The most common infections are sinopulmonary. As the disease progresses, the frequency of neutropenia, T-cell deficiency, and hypogammaglobulinemia increases, resulting in infections with gram-negative bacteria, fungi, and viruses such as herpes zoster and herpes simplex.
The major physical findings relate to infiltration of the reticuloendothelial system. Lymphadenopathy with discrete, soft, mobile lymph nodes is present in two thirds of patients at diagnosis. Later, as the lymph nodes enlarge, they can become matted. Enlargement of the liver or spleen is less common at diagnosis (approximately 10% and 40% of cases, respectively) but occurs more frequently with progression. Organ failure resulting from infiltration with CLL is uncommon. Infiltration of the central nervous system in CLL is rare, and central nervous system symptoms are more likely to be caused by opportunistic infections such as cryptococcosis or listeriosis.
CLL is characterized by absolute lymphocyte counts that typically range from 5000 to 600,000 × 10 9 /µL in the peripheral blood. Even with markedly elevated WBC counts, hyperviscosity symptoms rarely occur, probably because of the small size and pliability of the cells. Anaemia (haemoglobin <11 g/dL) is present in 15 to 20% of patients at diagnosis and thrombocytopenia (platelet count <100 × 10 9 /µL) in 10%. However, bone marrow replacement and hypersplenism, which are seen with progressive disease, increase the frequency of anaemia and thrombocytopenia. The anaemia is usually normochromic and normocytic, and the reticulocyte count is normal unless the patient has autoimmune haemolytic anaemia, which usually results from the development of a warm-reacting IgG antibody. The diagnosis of autoimmune haemolytic anaemia, which occurs in 10% of cases, is confirmed by a positive direct Coombs (DAT) test (80 to 90% of cases), reticulocytosis, a low serum haptoglobulin concentration, and an elevated unconjugated serum bilirubin level. In such patients, reactive erythroid hyperplasia as a response to the haemolysis may be masked in the bone marrow by the marked lymphocytic infiltration. Cold agglutinin haemolysis occurs rarely in CLL. Autoimmune thrombocytopenia (immune thrombocytopenic purpura) can be diagnosed in 10 to 15% of cases. The antibodies causing red cell and platelet destruction are not produced by the CLL cells, and the mechanisms for the associated autoimmune diseases are not known. Pure red cell aplasia is an additional, underappreciated cause of anaemia in CLL.
The lymphocytes in CLL are indistinguishable on light or electron microscopy from normal small B lymphocytes. On bone marrow aspiration, the proportion of lymphocytes is greater than 30% and may be up to 100%. Four patterns of lymphocyte infiltration on bone marrow biopsy occur: nodular (15%), interstitial (30%), mixed nodular and interstitial (30%), and diffuse (35%). Most early-stage cases have one of the first three patterns; diffuse histology is common in advanced-stage disease and becomes more prominent as the disease evolves. A diffuse histologic pattern confers a poor prognosis regardless of the stage of disease.
There are many diseases that can cause lymphocytosis, including pertussis, cytomegalovirus, Epstein-Barr virus mononucleosis, tuberculosis, toxoplasmosis, chronic inflammatory disorders, and autoimmune syndromes. These diseases are seldom confused with B-cell CLL, largely because the lymphocytosis in these conditions is usually less than 15 × 10 9 /µL and is not sustained. If doubt persists, immunophenotypic or molecular studies can distinguish the monoclonal lymphocytosis in CLL from the T-cell or polyclonal B-cell proliferation in the other disorders.
In individuals 62 to 80 years old, monoclonal CLL-phenotype B cells are found in about 5% of individuals with normal blood counts. In patients with greater than 4000 lymphocytes/mL, about 45% have CLL, about 40% have reactive lymphocytosis, and about 15% have monoclonal CLL-phenotype B cells that confer a 1.1% per year risk for developing CLL. The latter is a more recently recognized entity that has been termed monoclonal B-cell lymphocytosis . In patients ultimately diagnosed with CLL, B-cell clones were previously present in peripheral blood in 98% of patients, sometimes many years before diagnosis.
The more difficult differential diagnosis is distinguishing CLL from other lymphoproliferative disorders such as prolymphocytic leukaemia (PLL), splenic lymphoma with villous lymphocytes, HCL (see earlier section), the leukemic phase of mantle cell lymphoma, and Waldenström macroglobulinemia. Although certain clinical features are more common in some of these disorders (e.g., marked splenomegaly with minimal or no lymphadenopathy in PLL, splenic lymphoma, and HCL vs. extensive lymphadenopathy with or without splenomegaly in CLL), none of these clinical features is specific. The differential diagnosis therefore depends largely on histopathologic and, more specifically, immunophenotypic features.
DIFFERENTIAL DIAGNOSIS OF INDOLENT LYMPHOPROLIFERATIVE DISORDERS
|DISEASE||LYMPHADE NOPATHY (%)||SPLENOMEGAL Y(%)||CELL OF ORIGIN (BIT)||POSITIVE MARKERS|
|Smlg||CDS||CD19, CD20 (%)||Other|
|Chronic lymphocytic leukaemia (CLL)||75||50||B (20 : 1)||Weak||>90%||90||Mouse red blood cell receptors|
|Prolymphocytic leukaemia (PLL)||33||95||B (4 : 1)||Bright||T-cell PLL||75||FMC-7|
|Hairy cell leukaemia||<10||80||B (T rare)||Bright||—||>90||CD25, CD11C, CD103|
|Lymphoma (leukemic phase)||90||90||B (T rare)||Bright||Some||>90||CD10|
|Splenic lymphoma with villous lymphocytes||10||80||B||Bright||20%||>90||FMC-7, CD22|
|Waldenström macroglobulinemia||33||33||All B||Weak||Some||Many||CD38, PCA-1|
|Large granular lymphocytosis||10||10||All T||Absent||–||–||CD2, CD3, CD8|
CD2 = pan-T cell; CD3 = pan-mature T cell; CDS = pan-T cell, B-cell CLL; CD8 = T cell (suppressor cytotoxic); CD10 = early B cell; CD11C = hairy cell, activated T cell, NK cell; CD19 =early pan-B cell; CD20 = pan-B cell; CD25 = low-affinity interleukin-2 receptor; CD38 = activated B cell, thymocyte, plasma cell; FMC-7 = PLL, hairy cell leukaemia; PCA-1 = plasma cell; Smlg = monoclonal surface immunoglobulin.
PLL is an uncommon disease (incidence <5% that of CLL), and its characteristics of massive splenomegaly, minimal lymphadenopathy, and markedly elevated WBC count (often >100 × 10 9 /µL), with 10 to 90% of the cells being prolymphocytes, distinguish this disease from typical B-cell CLL. Prolymphocytes are larger cells that have a distinct nucleolus and express FMC-7. The male-to-female ratio is 4 : 1, and the median age at diagnosis is 70 years. Survival is shorter than in CLL (median, 3 years), and response to therapies usually applied in CLL is poor. A serum paraprotein, typically IgG or lgA, is present in one third of cases. The immunoglobulin on the surface of the cells is occasionally IgG or IgA, not IgM ± IgD, as in CLL. Several karyotypic abnormalities have been reported in PLL, including t (11; 14) (q13; q32). Deletions of llq3, 23, and 17p are more common in B-cell PLL than in CLL. Abnormalities in the TP53 oncogene are found in 75% of cases of B-cell PLL. One fifth of PLL cases express a T-cell phenotype.
Small lymphocytic lymphoma (SLL) shares histopathologic and immunophenotypic features with CLL, differing only in the lack of lymphocytosis in the peripheral blood. The bone marrow in SLL may or may not have more than 30% lymphocytes. LFA-1 adhesion protein is much more commonly expressed on SLL cells than on CLL cells. Other lymphomas, such as follicular and mantle cell lymphomas, occasionally manifest a leukemic phase on initial presentation. Follicular lymphoma cells are often cleaved on light microscopy, have bright staining for Smlg, and are positive for FMC-7 and CD10. Lymph node biopsy should be performed to identify these cases with greater precision. The presence of lymphoma cells in the blood in follicular lymphoma is more common with advanced disease. Follicular lymphoma can usually be identified by the presence of the translocation t(14;18) and consequentBCL2 rearrangement, both of which are rare in CLL. The WBC count in Waldenström macroglobulinemia is usually much lower than in CLL (<10 × 10 9 /µL), and many patients are leukopenic. The cells have a plasmacytoid appearance, CD38 and PCA-1 positivity, and more Smlg and cytoplasmic Ig. A serum monoclonal IgM peak is present in almost all cases of Waldenström macroglobulinemia but is uncommon in CLL.
The predominant clinical manifestation of Sézary syndrome (a CD4 + T-cell malignant disorder related to mycosis fungoides ) is chronic exfoliative erythroderma with a low number of circulating monoclonal T cells. The clinical and laboratory differentiation from CLL is not difficult.
Other T-cell malignant disorders with peripheral blood involvement are adult T-cell leukaemia-lymphoma and large granular lymphocytosis , also referred to as large granular lymphoproliferative disorder, T-cell lymphocytosis with neutropenia, or T-γ lymphocytosis syndrome. Adult T-cell leukaemia-lymphoma is associated with a retrovirus (human T-lymphotropic virus I) and is common in Japan and the Caribbean. It frequently manifests with lytic bone lesions and hypercalcemia. In T-cell large granular lymphoproliferative disorders, the absolute lymphocyte count is usually low (<5 × 10 9 /µL). The disease is defined by clonal amplification of either CD3 +cytotoxic T-lymphocytes or CD3 − natural killer (NK) cells. These patients often have splenomegaly, neutropenia, and rheumatoid arthritis−like symptoms. A subset, called NK-cell large granular lymphocytosis, has an NK-cell phenotype (CD16 − ) and no molecular evidence of T-cell receptor rearrangement. The lymphocytes in large granular lymphoproliferative disorder have abundant cytoplasm with azurophilic granules. Most patients have a benign course, although repeated infections can occur.
The natural history of CLL is heterogeneous, with survival times ranging from 2 to 20 years after diagnosis. Either of two validated clinical staging systems can be used. The Rai staging system (1975) defines five stages and is most frequently used in the United States, whereas the Binet system (1981) defines three stages and is most frequently used in Europe. Patients with anaemia and thrombocytopenia (Rai III and IV, Binet C) have the worst prognosis; patients with lymphocytosis alone (Rai 0, some Binet A patients) have an excellent prognosis. A group of patients with a lymphocyte count of less than 30 × 10 9 /µL, haemoglobin greater than 13 g/dL, platelet count greater than 100 × 10 9 /µL, fewer than three involved node areas, and lymphocyte doubling time greater than 12 months has been described as having “smouldering” CLL, with survival equal to that of an age- and sex-matched control population. Patients tend to progress through stages, with many patients developing more sites of involvement over time and eventually experience marrow failure; however, anaemia and thrombocytopenia can develop abruptly, even without antibody-mediated destruction or markedly increased tumour burden.
RAI AND BINET STAGING SYSTEMS IN CHRONIC LYMPHOCYTIC LEUKEMIA
|STAGE||LYMPHOCYTOSIS||LYMPHADENOPATHY||HEPATOMEGALY OR SPLENOMEGALY||HEMOGLOBIN (g/dL)||PLATELETS ×10 3 /mL|
|RAI STAGING SYSTEM|
|BINET STAGING SYSTEM|
|A||+||±||± (<3 lymphatic groups * positive)||≥10||≥100|
|B||+||±||± (≥3 lymphatic groups * positive)||≥10||≥100|
|C||+||±||±||<10 †||<100 †|
The three lymphatic groups are (1) cervical, axillary, and inguinal nodes; (2) liver; and (3) spleen. Each group is considered one group whether unilateral or bilateral.
† The criterion is haemoglobin <10 g/dL and/or platelets <100 × 10 3 /mL.
Other adverse factors include a diffuse pattern of lymphocytic infiltration observed on bone marrow biopsy; molecular abnormalities, including deletion of llq or 17p; advanced age; male sex; elevated serum levels of thymidine kinase, β 2 -microglobulin, and soluble CD23; rapid lymphocyte doubling time; an increased proportion of large or atypical lymphocytes in the peripheral blood; and lack of somatic mutation of the VH gene within the B-CLL cell or the presence of the ZAP-70 protein or CD38 on the CLL cell surface.
The major therapeutic questions are when to treat and which therapeutic regimen to use. Recent progress has transformed the treatment of CLL, initially from chemotherapy to chemoimmunotherapy and now to mechanism-driven drugs that target the B-cell receptor signalling pathway. Patients with CLL are usually older, and the prognosis of the disease is variable (with some early-stage cases being stable for 10 to 20 years). Treatment of early-stage CLL (Rai 0, Binet A) is delayed until the disease progresses. In randomized trials, early treatment with alkylating agents did not prolong survival and was associated with a heightened risk for developing second malignant tumours. Treatment of Rai stages III and IV (Binet stage C) is recommended at the time of diagnosis because of the morbidities associated with cytopenias and the poor survival time of these patients. Treatment of intermediate-stage disease (Rai I and II, Binet B) is recommended if symptomatic disease (fever, sweats, weight loss, severe fatigue) or massive lymphadenopathy, with or without hepatosplenomegaly, is present.
Fludarabine monophosphate (25 mg/m 2 /day for 5 days every 4 weeks), an adenosine analogue, is approved by the FDA for treatment of relapsed CLL; it produces an overall response rate of 50 to 60%. The dose-limiting toxicity is myelosuppression. In a large randomized trial of initial therapy for CLL, fludarabine was compared with chlorambucil, the historical standard therapy; it resulted in higher overall and complete remission rates, longer duration of remission, and improved response rates on crossover. Ten-year follow-up showed a survival benefit in the fludarabine arm. Cladribine is used primarily in Europe, where it appears to have efficacy similar to fludarabine. Pentostatin has not been as widely studied in CLL as in HCL.
The combination of fludarabine and cyclophosphamide (FC) was a logical attempt to improve on the efficacy of fludarabine alone by combining it with the other most active agent in this disease, an alkylating agent. The FC combination has been compared with fludarabine alone in three randomized trials. All these trials consistently showed a higher complete response rate, higher overall response rate, and longer progression-free survival with the FC combination.
After therapy, many patients remain stable for months to years before progressive disease indicates the need for further treatment. The goal of treatment is to achieve complete response.
DEFINITION OF REMISSION IN CHRONIC LYMPHOCYTIC LEUKEMIA: THE INTERNATIONAL WORKSHOP IN CHRONIC LYMPHOCYTIC LEUKEMIA−NATIONAL CANCER INSTITUTE WORKING GROUP CRITERIA
|CRITERION||COMPLETE REMISSION||PARTIAL REMISSION|
|None ≥1.5 cm
|≥1500/mL or ≥50% increase from baseline
>100,000/µL or ≥50% increase from baseline
11 g/dL or >50% increase from baseline
|Bone marrow||<30%, no B-lymphoid nodules||50% reduction in infiltrate or B-lymphoid nodules|
Bendamustine is a potent alkylating agent that has some structural similarity to nucleoside analogues, but preclinical data suggest that it does not function as a nucleoside analogue. Bendamustine was recently approved by the FDA for the treatment of CLL based on a randomized trial comparing this agent to chlorambucil as initial therapy for patients with CLL. Complete and overall response rates were higher with bendamustine, and progression-free survival was longer. The main side effect is myelosuppression.
Rituximab, a monoclonal antibody targeting the CD20 antigen, is associated with response rates of about 50% when given at the standard dose (375 mg/m 2 /week for 4 weeks) as initial therapy for CLL and significantly lower rates when used in the salvage setting; complete responses are rare in either situation. The major benefit of this antibody appears to be its use in combination with chemotherapy. Fludarabine combined with rituximab produces better responses than those seen historically with fludarabine alone. In addition, progression-free and overall survival rates are better than those seen in the historical cohort. A three-drug regimen of fludarabine, cyclophosphamide, and rituximab (FCR) appears to produce the best and most durable complete remission rates when used as first-line therapy. A randomized trial compared the activity of FC to that of FCR. The complete response rate and the overall response rate were significantly higher with FCR, and progression-free and overall survival rates were significantly longer than with chemotherapy alone. A trial of previously untreated CLL patients with coexisting conditions tested the efficacy of chlorambucil alone versus chlorambucil plus rituximab versus chlorambucil plus obinutuzumab, a glycoengineered anti-CD20 monoclonal antibody. Combining an anti-CD20 antibody with chemotherapy improved outcomes, and obinutuzumab was found to be superior to rituximab when each was combined with chlorambucil.
Alemtuzumab (Campath-lH; 30 mg intravenously three times a week for 4 to 12 weeks), a monoclonal antibody that binds to the CD52 antigen, was originally approved for the treatment of fludarabine-refractory CLL. One third of such patients can achieve remission. Recently, alemtuzumab was compared with chlorambucil as initial therapy for symptomatic patients with CLL. Alemtuzumab produced higher complete and overall response rates and longer progression-free survival. In the first-line treatment of high-risk CLL, the addition of subcutaneous alemtuzumab to oral FC chemotherapy (fludarabine plus cyclophosphamide) improved progression-free survival and overall survival (the latter only in patients younger than 75 years) compared with FC chemotherapy alone. Alemtuzumab was withdrawn from the market in the United States and Europe in 2012 but is available free of charge from the company by compassionate request.
Ofatumumab is a humanized monoclonal antibody that binds to CD20, but to a different epitope than rituximab. This drug was recently approved by the FDA for the treatment of fludarabine- and alemtuzumab-refractory CLL. In the pivotal trial, the drug was given intravenously for 8 weeks and then monthly for 4 months. The overall response rate in this highly refractory population was 58%; the median progression-free survival was 6 months, with a median overall survival of 13.7 months. As with other monoclonal antibodies, the predominant side effects are infusion reactions, which tend to be more common with the initial doses.
A new category of targeted agents, called B-cell receptor signalling pathway inhibitors, is providing significant increased efficacy in CLL. Several inhibitors are being tested, targeting different kinases in the B-cell receptor pathway. Recent data have demonstrated that B-cell receptor signalling inhibitors are very active, particularly for the treatment of relapsed CLL. Ibrutinib is a first-in-class, oral covalent inhibitor of Bruton tyrosine kinase, an essential enzyme in B-cell receptor signalling, homing, and adhesion. It has been granted accelerated approval by the FDA for use in CLL. A randomized trial compared the effect of ibrutinib with that of the anti-CD20 antibody ofatumumab, both used alone, in the treatment of patients with relapsed or refractory CLL or small lymphocytic lymphoma. Ibrutinib significantly improved both progression-free and overall survival, compared with ofatumumab. Idelalisib is an inhibitor of phosphatidylinositol 3-kinase δ (PI3Kδ). Signalling through the B-cell receptor is mediated in part by activation of PI3Kδ. The δ isoform of PI3K is highly expressed in lymphoid cells, and it is the most critical isoform involved in the malignant phenotype of CLL. In a randomized, phase 3 study, patients with relapsed CLL who had clinically significant coexisting medical conditions that made them less able to undergo standard chemotherapy were randomized to receive eight intravenous infusions of rituximab plus either idelalisib 150 mg orally twice daily or placebo. The combination of idelalisib and rituximab, compared with rituximab alone, significantly improved response rate, progression-free survival, and overall survival.
Chimeric antigen receptors (CARs) are fusion proteins that combine antigen moieties and costimulatory T-cell receptors to redirect T cells toward malignant cells. All CARs targeting CLL have thus far directed the T cells through recognition of CD19. 14 Complete remissions, including absence of residual disease by sensitive testing, has been produced in refractory patients. Cytokine release syndrome, which can be life-threatening, is the initial toxicity. (Cytokine release syndrome, a rare phenomenon, also occurs in graft-versus-host disease after transplantation, severe infections, hemophagocytic lymphohistiocytosis or macrophage activation syndrome, and monoclonal antibody therapy; cytokines trigger an acute systemic inflammatory response that leads to endothelial and organ damage, microvascular leakage, and heart failure.) Late toxicity includes the need for immunoglobulin replacement due to simultaneous eradication of normal B cells. CARs targeting other antigens in CLL are being investigated.
Autologous stem cell transplantation has no proven benefit in terms of survival or longterm disease control in CLL. Data on allogeneic stem cell transplantation are limited to young patients with refractory disease, in whom a long-term control rate of 40 to 55% has been reported. Nonmyeloablative stem cell transplantation, which works mainly by its graft-versus leukaemia effect, has been used in older patients with CLL with some success.
In CLL, radiation therapy is used palliatively to shrink unsightly or painfully enlarged nodal masses or an enlarged spleen.
Autoimmune haemolytic anaemia and immune-mediated thrombocytopenia do not correlate closely with the activity of CLL. Prednisone (60 to 100 mg/day) is indicated as treatment for autoimmune haemolytic anaemia and for some cases of immune-mediated thrombocytopenia in CLL. If there is no response in 3 to 4 weeks, the treatment has failed, and the dose should be tapered over 1 to 2 weeks. If a response is obtained, the dose is reduced by 25% each week over 4 weeks. Patients for whom corticosteroids fail often respond to low-dose oral cyclosporine 100 mg three times a day. Other therapeutic options include splenectomy, intravenous immunoglobulin, rituximab, and alemtuzumab. Intravenous immunoglobulin (400 mg/kg every 3 to 4 weeks) significantly decreases the incidence of infections in patients with recurrent infections and hypogammaglobulinemia. However, the cost of this therapy is substantial.
Approximately one third of patients who present with early-stage CLL never require therapy and have the same survival as age-matched controls. Frequent characteristics of such patients include WBC less than 30 × 10 9 /L, haemoglobin greater than 13 g/dL, nondiffuse pattern on bone marrow biopsy, and slow lymphocyte doubling time.
Factors associated with shorter time to treatment failure as well as either poorer response to at least chemotherapy-based regimes, or shorter remission durations, include an unmutated IgVH gene, presence of 17p or llq deletions, presence of ZAP70 and CD38, and mutations in TP53, SF3Bl, ATM, and NOTCH . A poor response to therapy is an adverse factor in all phases of the disease. As CLL progresses, the development of a prolymphocytic transformation (10% of cases) or transformation to large cell lymphoma (Richter transformation) portends a median survival time of less than 6 months. Other factors that may suggest transformation are the development of B symptoms (fevers, night sweats, weight loss), a markedly elevated lactate dehydrogenase level, or fluorodeoxyglucose-avid disease on positron emission tomography. A high incidence of second malignant tumours (10 to 20% of patients) either precedes or follows the diagnosis of CLL; the roles of therapy versus impaired immune surveillance as causative factors are unclear. Skin cancer, including melanoma, as well as colorectal and lung cancers particularly are common. CLL tends to develop in older people; in indolent cases, death occurs from other intercurrent illnesses seen in this age group. Almost all patients younger than 60 years and those with progressive disease die as a result of CLL, primarily from infections. Gram-positive organisms usually cause nonfatal infections early in CLL, but most deaths due to infection are associated with gram-negative bacterial or fungal infections. Infection with other opportunistic organisms, such as Mycobacterium tuberculosis, herpes virus, and Pseudomonas jiroveci , may also be fatal.