The Washington Manual of Oncology, 3 Ed.

Chronic Leukemias

Rizwan Romee • Todd A. Fehniger

 I. INTRODUCTION. Chronic leukemias are malignancies of the myeloid or lymphoid hematopoietic lineages that have historically been characterized as having an indolent course when compared with their acute counterparts. While the indolent nature of these diseases results in a relatively long median survival as compared with other cancers, chronic leukemias have not typically been considered curable, except in some cases following allogeneic hematopoietic cell transplantation (HCT). Here we review clinical features and current treatment approaches to the most common chronic leukemias, chronic myelogenous leukemia (CML) and chronic lymphocytic leukemia (CLL).

 CML is the most frequent chronic leukemia of myeloid derivation and is categorized as a clonal myeloproliferative disorder within the World Health Organization (WHO) classification. This disease is characterized by peripheral blood leukocytosis resulting from an expansion of normally differentiated myeloid cells and typically presents incidentally on routine laboratory testing. CML is notable as the first leukemia identified with a causative clonal chromosomal rearrangement, t(9;22), or the Philadelphia chromosome (Ph), which juxtaposes the ABL tyrosine kinase next to the break point cluster (BCR) region, yielding the BCR-ABL fusion protein. The management of CML has been revolutionized over the past 15 years through targeted inhibition of BCR-ABL by the oral tyrosine kinase inhibitor imatinib mesylate (Gleevec), which has provided the first proof-of-principle for small molecule targeted therapy of cancer. CML therapy continues to rapidly evolve with the advent of new generations of small molecular inhibitors of BCR-ABL, and current challenges include defining the optimal approach to manage CML patients who have developed resistance or are unable to tolerate these agents.

 CLL is the most common lymphoid leukemia, and is combined with small lymphocytic lymphoma (SLL) as a mature B-cell neoplasm in the WHO classification. Advances in our understanding of the pathophysiology of CLL have yielded an updated view on the natural history, genomic causes, and important prognostic factors in this disease. Improvements in the initial therapy of CLL include combining chemotherapeutic agents with monoclonal antibodies, and expanded treatment options now exist for patients with disease refractory to purine analogs such as fludarabine, including kinase inhibitors such as idelalisib (PI3Kδ inhibitor) and ibrutinib (BTK inhibitor). Other new agents have preliminary evidence of activity, and the use of immunotherapy appears particularly promising.

1. CHRONIC MYELOGENOUS LEUKEMIA (CML)
2. Epidemiology. CML accounts for 14% of all leukemias and 20% of adult leukemias, with an annual incidence of 1.6 cases per 100,000 adults. Since the advent of imatinib, the annual mortality has decreased to 1% to 2%. The median age at presentation is 67, and incidence increases with age. The etiology is unclear; no correlation with monozygotic twins, geography, ethnicity, or economic status has been observed. However, a significantly higher incidence of CML has been noted in survivors of the atomic disasters at Nagasaki and Hiroshima, in radiologists, and in patients treated with radiation to the spine for ankylosing spondylitis.
3. Pathogenesis. Historically, CML was the first disease in which a specific chromosomal abnormality was linked to the pathogenesis of the disease: the foreshortened chromosome 22, named the Philadelphia (Ph) chromosome. Subsequently, the BCR-ABL fusion gene resulting from the common t(9;22) translocation has been noted in 90% to 95% of patients with CML. This fusion of the BCR (breakpoint cluster region) serine kinase with the human homologue ABL1 of the Abelson murine leukemia virus oncogene results in constitutive tyrosine kinase activity of ABL and thereby dysregulated activity of multiple signal-transduction pathways controlling cell proliferation and apoptosis. BCR-ABL may also play a direct role in signals leading to independence from external growth signaling, cell adhesion modulation, and DNA repair. CML patients who truly lack BCR-ABL gene fusion are called atypical CML (aCML) and account for <5% of CML cases. Recently, mutations in G-CSF receptor gene called colony-stimulating factor 3 (CSF3R) were found in 40% of aCML patients and mutations in set binding protein (SETBP1) in 25% of aCML patients. Interestingly, some aCML cases harbor mutations in both of these genes. For this chapter, however, we will focus only on BCR-ABL positive CML.
4. Clinical and laboratory features. In most patients, CML is diagnosed incidentally. Symptomatic constellations typically result from concurrent anemia and splenomegaly: fatigue, early satiety, and sensation of abdominal fullness, but may also include weight loss, bleeding, or bruising in advanced disease. Leukocytosis with a myeloid shift is universal. In contrast to cases of acute leukemia, in which an arrest in maturation is the rule, granulocytes at all stages of maturation are observed on the peripheral smear. Anemia and thrombocytosis are common, although basophilia (more than 7%) occurs in only 10% to 15% of patients. Leukocyte alkaline phosphatase (LAP) activity is usually reduced, but can be increased with infections, stress, on achievement of remission, or on progression to blast phase (BP). The diagnosis is confirmed by the detection of the Ph chromosome t(9;22)(q34.1;q11.21). In around 5% of patients, a BCR-ABL fusion can be detected without classic Ph chromosomal cytogenetics, and rarely translocations can involve three or more chromosomes. The bone marrow is typically hypercellular and devoid of fat. All stages of myeloid differentiation are present and megakaryocytes may be increased, suggesting that chronic-phase CML is a disease of discordant maturation, where a delay in myeloid maturation results in increased myeloid cell mass.
5. Natural history. The natural history of CML is a triphasic process: a chronic phase, an accelerated phase, and a blast phase. Most patients present in chronic phase, characterized by an asymptomatic accumulation of differentiated myeloid cells in the bone marrow, spleen, and peripheral blood. Without therapy, CML patients almost invariably progress from chronic phase to accelerated phase and ultimately into blast phase, though some patients in chronic phase evolve directly into blast phase with no intervening accelerated phase.

 In the 2 years after initial diagnosis of CML, 5% to 15% of untreated patients will enter blast crisis. In subsequent years, the annual rate of progression increases to 20% to 25%, with progression commonly occurring between 3 and 6 years after diagnosis.

 The definition of accelerated phase CML relies on several clinical and laboratory features and is characterized by increasing arrest of maturation. Current WHO criteria include at least one of the following: 10% to 19% blasts in peripheral blood or bone marrow, equal to or more than 20% peripheral basophils, persistent thrombocytopenia less than 100,000/μL unrelated to therapy, persistent thrombocytosis more than 1000,000/μL and unresponsive to therapy, increasing spleen size and increasing white blood cell (WBC) count unresponsive to therapy, or cytogenetic evidence of clonal evolution. Once either accelerated phase or blast crisis occurs, the success of any therapy declines dramatically.

 The current WHO criteria for diagnosis of blast phase (also called blast crisis) include at least one of the following: equal to or higher than 20% blasts in the peripheral blood or bone marrow, large foci or clusters of blasts in the bone marrow biopsy, or extramedullary disease.

 Several prognostic models (Sokal score, Hasford, MD Anderson Cancer Center staging system) have been developed to stratify patients into groups with different average survival using variables such as age, spleen size, platelet count, percentage of peripheral blood blast count, hematocrit, cytogenetic clonal evolution, and gender. Although these scoring systems were developed before imatinib, post hoc analysis of the IRIS (International Randomized Study of Interferon vs. STI-571) study provided an initial validation of the Sokal score in this imatinib-treated population. Other predictors of outcome derived from the IRIS study include response to imatinib at 3, 12, and 18 months. EUTOS scoring, which was specifically developed to predict response in CML patients undergoing initial treatment with imatinib, is relatively simple and relies only on spleen size and circulating percentage of basophils. It better predicts complete cytogenetic remission at 18 months after initiation of imatinib therapy, which is an important predictor of outcomes in CML patients. However, it remains to be seen whether EUTOS scoring predicts survival/response in CML patients who are being treated with newer tyrosine kinase inhibitors.

1. Treatment of chronic myeloid leukemia: tyrosine kinase inhibitors
2. Imatinib mesylate (gleevec). Imatinib is a targeted tyrosine kinase inhibitor (TKI), which antagonizes the activity of the ABL tyrosine kinase as well as c-Kit, and platelet-derived growth factors α and β. At nanomolar concentrations, imatinib binds to the adenosine triphosphate (ATP)-binding pocket of the BCR-ABL fusion protein while in the inactive conformation, resulting in competitive inhibition. This nearly completely abolishes autophosphorylation of BCR-ABL, inactivates dysregulated downstream signaling through multiple pathways including JAK-STAT, PI3K, RAS, AKT, and ERK, thereby specifically inhibiting the growth of BCR-ABL positive bone marrow progenitor cells.

 The current practice is to initiate imatinib therapy at a dose of 400 mg once daily, which can be titrated up to 600 mg once daily dose in case of disease progression, lack of hematologic response by 3 months, lack of cytogenetic response after 6 to 12 months, or in case of a loss of previous response at 400 mg dose. In the phase III IRIS trial, an initial dose of 400 mg every day followed by escalation to 400 mg twice daily dosing if needed resulted in 98% complete hematologic response (CHR) and 87% complete cytogenetic response (CCyR) rates at 60 months with an estimated 5-year survival of 90%. However, higher dosing schedules of 600 mg and 800 mg have not been associated with improved survival outcomes.

 Side effects of imatinib mesylate are generally mild, but include hematologic suppression (neutropenia, thrombocytopenia, and anemia), constitutional symptoms (diarrhea, edema, and rash) and rare organ damage (transaminitis, hypophosphatemia, and potential cardiotoxicity). These can usually be managed with growth factors or dose reduction, but occasionally require discontinuation, either briefly or permanently.

2. Imatinib resistance. Resistance to imatinib has been noted in 2% to 4% of patients annually for the first 3 years of imatinib therapy and may decrease thereafter. Mechanisms proposed include the acquisition of point mutations in the BCR-ABL kinase SH1 domain, overexpression of BCR-ABL, activation of BCR-ABL–independent pathways including SRC kinases, increased imatinib efflux through the multidrug resistance (MDR) pump and progressively abnormal cytogenetics. Of these, point mutations in the SH1 kinase domain likely play the most prominent role, and more than 50 distinct mutations have been documented in 42% to 90% of resistant cases. ATP-binding loop (P-loop) and T3151 mutations are particularly more common in advanced phase CML patients. Mutations have also been found de novo in untreated chronic-phase patients, suggesting they may exist before treatment and are slowly selected out during therapy. As rates of progression decline over time, imatinib therapy is not currently thought to induce new mutations. These mutations act by either decreasing the affinity for imatinib binding in the ATP-binding pocket or shifting the kinetics of BCR-ABL to prefer the active conformation, to which imatinib will not bind.

 Imatinib resistance can be overcome either with increasing doses or a second-generation tyrosine kinase inhibitor. Second-generation tyrosine kinase inhibitors are effective in most of the mutations with the exception of T315I, which imparts a high degree of resistance to all currently available TKI with the exception of ponatinib. Mutational analysis is therefore critical in determining clinical course after resistance is noted.

3. Second generation tyrosine kinase inhibitors. Several more potent tyrosine kinase inhibitors have been developed since the initial introduction of imatinib.
4. Dasatinib. Dasatinib is a potent inhibitor of ABL tyrosine kinase but also inhibits SRC family kinases, c-KIT, EPHA2, and platelet-derived growth factor receptor β (PDGFRβ). It is active against most of imatinib-resistant mutational forms of ABL1 except for T315I and F317V. First-line therapy for chronic phase CML patients on standard 100 mg daily dose of dasatinib is associated with faster and deeper response rates compared with imatinib. However, so far, no survival advantage has been demonstrated with the use of dasatinib over imatinib as a first-line therapy in chronic phase CML. Dasatinib has also shown excellent response rates as a second-line agent in chronic phase CML patients otherwise intolerant or resistant (except for patients with the above-mentioned mutations) to imatinib. Overall, dasatinib is well tolerated with easily manageable cytopenias and diarrhea. Pleural effusion is a relatively common side effect and tends to be more common in patients with accelerated phase CML, prior cardiac history, hypertension, and those receiving higher doses of dasatinib (70 mg twice a day vs. 100 mg once daily).
5. Nilotinib. Nilotinib is a highly potent inhibitor of ABL tyrosine kinase and also inhibits other tyrosine kinases including c-KIT and PDGFR, but unlike dasatinib, it has no activity against SRC family kinases. Similarly to dasatinib, nilotinib is active against most of the mutations in ABL1 but has no activity against T315I mutation. At a twice daily dose of 300 mg, patients achieve faster and deeper responses compared with imatinib, although without survival improvement. In addition to the common adverse effects including nausea, vomiting, diarrhea, and myelosuppression, nilotinib has also been associated with significant QTc prolongation in some patients, and thus the black box warning in its labeling from FDA. Because of this QTc prolongation, it is important to correct any electrolyte abnormalities prior to its use, and then also to monitor them periodically.
6. Bosutinib. Bosutinib has activity against BCR-ABL and SRC family kinases, but minimal activity against c-KIT and PDGFR. In addition, it has activity against most of the imatinib-resistant mutations except for T315I and V299L. Compared with imatinib, use of bosutinib at a standard daily dose of 500mg in chronic phase CML patients leads to faster and somewhat deeper responses; however, as with the use of other second-generation TKI’s, no survival advantage has been demonstrated. Overall, bosutinib has a favorable toxicity profile with only minimal effect on QTc. Diarrhea is the most common nonhematologic side effect of this medication; the other relatively common nonhematologic side effects are elevation of ALT, increased lipase, hyperglycemia, and electrolyte abnormalities, which are often manageable without needing its discontinuation or dose modification. Currently, bosutinib is approved as a second-line therapy in patients intolerant and/or resistant to prior TKIs (imatinib, dasatinib, and nilotinib).
7. Ponatinib. Ponatinib is a potent orally active tyrosine kinase inhibitor with activity against a wide range of tyrosine kinases including T315I mutant forms of the ABL1 tyrosine kinase. In a phase II using a daily dose of 45mg, patients with chronic phase CML with resistance (including patients with T315I mutations) or intolerance to prior TKI therapy, ponatinib induced a major cytogenetic response rate of 56%, a complete cytogenetic response of 46%, and a major molecular response rate of 34%. Furthermore, the estimated progression-free survival and overall survival at 12 months were 80% and 94%, respectively. Presence of T315I mutations, young age, prior exposure to fewer TKIs, and shorter duration of leukemia were associated with higher response rates. The most common adverse effects reported with the use of ponatinib were thrombocytopenia, neutropenia, rash, abdominal pain, and fluid retention, all relatively easily manageable (N Engl J Med 2013;369:1). However serious vascular complications including thromboembolic events (arterial and venous) have been reported in ≥25% of the patients treated with ponatinib, leading to its withdrawal from the market in late 2013. Nevertheless, the drug was reinstated in the spring of 2014 with narrower indications. Ponatinib is currently approved at an oral daily dose of 45 mg for adult CML patients harboring the T315I mutation and for patients where all other TKIs have failed. In addition to thromboembolic events, ponatinib also carries a black box warning for heart failure and liver toxicity. Patients being treated with ponatinib need to be closely monitored for thromboembolic events, and despite lack of evidence for its efficacy, many clinicians now concomitantly use aspirin for the prophylaxis of these thromboembolic events.
8. Initial workup, therapy, and disease monitoring in chronic phase cml. In addition to the routine labs, including CBC with differential and comprehensive metabolic panel, CML patients should undergo bone marrow aspirate and biopsy with conventional cytogenetic analysis (karyotyping) to identify Philadelphia chromosome. In a small minority of patients with variant or cryptic translocations where Philadelphia chromosome is not readily detected by conventional karyotyping, fluorescence in-situ hybridization (FISH) can be used to identify BCR-ABL1 gene fusion in these patients. Peripheral blood (or marrow at diagnosis) quantitative PCR (qPCR) is done at baseline (prior to treatment initiation), then every 3 months to monitor the response once treatment is initiated.

 Current recommendations for treatment initiation include either imatinib 400 mg daily, nilotinib 300 mg twice daily, or dasatinib 100 mg daily. Imatinib remains a reasonable first-line agent, although patients with intermediate- or high-risk Sokal or Hasford scores should be considered for either nilotinib or dasatinib. Bosutinib is currently approved for use as a second-line TKI after patients have failed the first line TKI’s.

 CML patients initiated on TKI therapy are assessed for hematologic, cytogenetic, and molecular responses. Complete hematologic remission (CHR) is defined as normalization of peripheral blood counts (no immature cells, <5% basophils on differential, WBC count <10 × 10⁹/L, and platelet count <450 × 10⁹/L). CHR also includes absence of palpable splenomegaly on physical exam. Cytogenetic response is defined as complete cytogenetic response (CCyR) by the absence, major by 1% to 35%, minor by 36% to 95%, and no response by >95% Philadelphia chromosome metaphases on bone marrow banding analysis of at least 20 metaphases. Molecular response is defined as major molecular response (MMR) when qPCR for BCR-ABL1 transcripts in the peripheral blood are ≤0.1% on the International Scale (IS), which is equivalent to the ≥3 log reduction from the standardized baseline. Complete molecular remission (CMR) denotes undetectable BCR-ABL1 transcripts by qPCR; however, as the sensitivity of the qPCR has been steadily increasing, it is becoming clear that CMR is a misleading term as very low burden of malignant clone may still be found in patients with otherwise complete CMR, and therefore it has been proposed not to use this term any more.

 Increased risk of progression to accelerated and blastic phases has been demonstrated if the initial TKI therapy does not achieve specific clinical goals, which currently include complete hematologic response with normal peripheral counts and BCR-ABL transcripts by quantitative polymerase chain reaction (qPCR) of ≤10% (International Scale, IS), and/or ≤35% Ph chromosome-positive bone marrow cells at 3 months, BCR-ABL transcripts by qPCR of ≤1% (IS), and/or 0% Ph chromosome-positive bone marrow at 6 months, BCR-ABL transcripts by qPCR of ≤0.1 (IS) at 12 months after initiation of the TKI therapy. Failure to reach any of these goals warrants close follow-up, change of therapy, and ABL tyrosine kinase domain mutation analysis.

 After initiation of therapy, patients should undergo weekly complete blood counts (CBC) with differential until complete hematologic remission. Afterwards, follow-up includes peripheral blood BCR-ABL qPCR with CBCs and chemistries every 3 months for 3 years, and if patient continues to maintain major molecular response, then every 3 to 6 months afterwards. Bone marrow biopsy with cytogenetics is performed at the time of the diagnosis and repeated at 3 and 6 months if qPCR for BCR-ABL transcripts is not available. Bone marrow biopsy with cytogenetics is again performed at 12 months if the patient is still not in MMR and then again at 18 months if the previous marrow at 12 months did not demonstrate a CCyR. Peripheral blood FISH can replace qPCR for BCR-ABL transcripts to monitor disease progression after patients in centers where qPCR is not readily available. Any signs of disease progression such as change in blood counts and/or rising BCR-ABL1 transcripts should be quickly reevaluated with a bone marrow biopsy with cytogenetics as well as mutational analysis of BCR-ABL.

5. Treatment duration. Treatment length continues to be defined, but we recommend indefinite tyrosine kinase inhibition. Recent data suggest tyrosine kinase inhibitors are not cytotoxic to early, quiescent BCR-ABL positive precursor cells. Several small recent studies have shown the feasibility of safely discontinuing TKI therapy in patients in MMR and low Sokal risk scores. This is currently an evolving field and at this time discontinuation or TKI drug holidays of TKI therapy should be considered only in the context of a clinical trial.
6. Conventional chemotherapy. Until 1980, hydroxyurea and busulfan were the two most effective anti-CML agents. Both offer mild hematologic control associated with myelosuppression but without affecting the uniform transformation to the acute phase of the disease. Subsequently, interferon α used alone or in combination with cytarabine has demonstrated improved response over chemotherapy with major cytogenetic responses in 40% to 50% of patients, with up to 80% of these patients achieving a durable response resulting in a 10-year survival of 75%. However, interferon α therapy is complicated by significant side effects including flu-like symptoms, anorexia, weight loss, depression, autoimmune (AI) disorders, thrombocytopenia, alopecia, rashes, and neuropathies, resulting in discontinuation in approximately a fifth of patients. Given the superior response to tyrosine kinase inhibitors and their relatively benign side effect profile, conventional chemotherapy and interferon have fallen from common use in CML.
7. Treatment of chronic myelogenous leukemia: transplant options
8. Allogeneic hematopoietic cell transplantation. Allogeneic HCT from either related or unrelated donors remains the only known curative therapy for CML. Transplantation from a matched-sibling donor during the chronic phase is associated with a 10-year survival of 50% to 70%. Results of transplantation from unrelated donors are somewhat less impressive, but are improving with better matching strategies and supportive care. The objective of allogeneic HCT is cure of CML by eradication of the leukemic clone with myeloablative chemoradiotherapy, and restoration of hematopoiesis by transplantation of normal donor-derived stem cells. In addition, the donor-derived allogeneic immune cells confer an important graft-versus-leukemia (GVL) effect, which acts to prevent recurrence of disease. GVL has been closely associated with the presence of graft versus host disease (GVHD). GVHD does not develop in patients receiving transplants from identical twin donors. These patients have at least twice the risk of relapse of CML compared with transplant recipients from HLA-identical siblings.

 The best results occur when patients are transplanted while in chronic phase with 5-year survival in the 50% to 60% range. However long-term survival after allogeneic HCT in accelerated phase is only 20% to 40%, whereas survival after transplants performed in blastic phase further declines to approximately 20%. The highest rates of survival for related and unrelated-donor transplantation in the preimatinib era are documented in patients transplanted within 1 year of their diagnosis where long-term survival in chronic phase approaches the 60% to 70% range.

 Use of allogeneic HCT is also limited by histocompatible donor availability as only one-third of patients will have a HLA-matched sibling, and only around 50% of the patients are able to locate a suitable unrelated donor (70% for whites and 15% for African Americans and other minorities in the United States). However, the recent advances in alternate donor allogeneic HCT such as haploidentical and cord blood transplantation have further broadened the availability of the transplant modality to patients who otherwise lack HLA-matched related and unrelated donors. Age older than 50 to 60 years has been found to constitute a significant hurdle for transplant success, especially in the unrelated-donor transplant setting, although this is improved with modified intensity and nonmyeloablative transplant.

 Most CML patients transplanted in the chronic phase are cured of their disease, although transplant-related morbidity and mortality remain a significant problem. The cumulative incidence of severe GVHD is approximately 20% to 35% in matched-sibling transplantation and 40% to 55% in recipients of transplants from unrelated donors. Infection is a major cause of nonrelapse mortality in allogeneic transplantation. GVHD and immunosuppression are predisposing factors for infectious complications.

2. Allogeneic bone marrow transplantation in the TKI era. The effect of pretransplant tyrosine kinase inhibitor therapy, which will delay transplant and may thereby increase its risks, is still under study. Initial retrospective studies comparing imatinib-treated transplant patients with historical controls show no difference in rates of engraftment and acute or chronic GVHD, suggesting delays due to therapy do not adversely affect the transplant itself. Upfront consideration for allogeneic HCT in the TKI era is suggested for patients who have T315I or other pan-TKI resistant mutations, patients who are intolerant of all currently available TKIs and who present with blast phase.

 High-risk patients, who present with cytogenetic changes beyond Ph chromosome, who demonstrate increasing cytogenetic complexity, who do not meet standard goals of therapy at 3, 6, 12, or 18 months, or who have rising qPCR levels of BCR-ABL should be evaluated for human leukocyte antigen (HLA)–matched siblings and potential unrelated donors. Further progressive disease despite appropriate tyrosine kinase inhibition warrants transplant consideration as early as possible.

 Relapse following transplant has been successfully treated with both donor lymphocyte infusion and tyrosine kinase inhibition. Mutational analysis of ABL kinase domain may help guide appropriate therapy choice.

1. Accelerated and blast-phase chronic myelogenous leukemia. Despite profound advances in the treatment of chronic-phase CML, outcomes of patients with accelerated and blast phase continue to be suboptimal. BRC-ABLmutational analysis should be performed with early consideration for transplant or clinical trial. However, because most patients now developing accelerated phase have been previously treated with imatinib, mutations in and overexpression of BCR-ABL are to be expected. Simple dose escalation of imatinib is rarely sufficient therapy. Patients diagnosed with AP CML are typically treated with second-generation TKIs. Patients with the T315I mutation or intolerant to other TKIs are treated with ponatinib. Omacetaxine is another option for patients harboring this mutation as it has shown some activity in this setting. Allogeneic HCT is reserved as an option for patients who do not achieve deep response to the TKI therapy.

 Blast phase is characterized by cytogenetic evolution in approximately 70% of patients. The most common chromosomal abnormalities are trisomy 8 in 30% to 40% of patients, additional Ph chromosome in 20% to 30%, and isochromosome 17 in 15% to 20%. Corresponding mutations in p53 are also seen in 20% to 30% of patients, amplification of c-myc in 20%, and, less commonly, mutations and deletions of ras, Rb, or p16. As with de novo acute myelogenous leukemia (AML), complex cytogenetics are associated with decreased response rates and survival.

 Treatment of blast-phase CML remains a challenge, and is dictated by hematologic features. Myeloid features are seen in 50% of patients, lymphoid in 25%, and undifferentiated in 25%. Patients with myeloid blast crisis are treated with a second-generation TKI (alone or in combination with AML type of chemotherapy) followed by allogeneic HCT. However, even after allogeneic HCT, the survival continues to be around 20% to 30%.

 Patients with lymphoid blast crisis are typically treated with Hyper CVAD or other regimens including vincristine and prednisone plus a TKI. This approach has been shown to induce acceptable response rates with a CHR of around 80% and CCR in the 50% to 60% range. When followed by allogeneic HCT, prolonged survival has been demonstrated in these patients (93 months in one series from MD Anderson Cancer Center).

 III. CHRONIC LYMPHOCYTIC LEUKEMIA (CLL)

6. Epidemiology. CLL is the most common form of leukemia in adults, accounting for approximately 30% of adult leukemias in the United States. Approximately 16,000 new cases are diagnosed annually, and 4,600 deaths are attributed to CLL each year in the United States. According to the Surveillance, Epidemiology, and End Results (SEER) cancer database, from 2007 to 2011, the median age at presentation was 71 years, and only 14% of patients were less than 60 years old at the time of diagnosis. The age-adjusted incidence for CLL was 6.0 per 100,000 men and 3.1 per 100,000 women per year, with a 2:1 male:female ratio. There are no clear environmental or occupational risk factors that predispose to CLL, and patients who are exposed to radiation do not appear to have an increased frequency of CLL. Interestingly, the incidence of CLL is much lower (10% of that of Western countries) in Asian countries such as China and Japan, which is attributed to genetic rather than environmental factors. CLL (and other malignancies) occur at a higher than predicted frequency among first-degree relatives of patients with CLL (relative risk of approximately 1.5 to 7.5), the highest familial risk of all hematologic malignancies, suggesting a subset of patients have inherited risk factors. Studies of familial cohorts with CLL are ongoing, and the future identification of genes involved in familial CLL may provide insights into the pathogenesis of CLL. Of note, monoclonal B lymphocytosis (MBL) with a CLL phenotype was detected in 3.5% of normal healthy control subjects, and there was a significant increase in the detection of such cells in family members of CLL patients (13.5%). It is now generally accepted that MBL precedes the development of CLL, but only a minority of individuals with MBL eventually develop the disease.
7. Pathogenesis. CLL is a clonal lymphoproliferative disorder characterized by the accumulation of neoplastic, functionally incompetent B lymphocytes in the blood, bone marrow, lymph nodes, spleen, or other organs. After encountering antigen, a normal B-cell enters the germinal center and proliferates, where the B-cell receptor genes undergo somatic hypermutation, which allows for B-cell receptor affinity maturation and selection of B-cell clones with high affinity for the antigen. Despite their uniform morphologic appearance and immunophenotype, there appears to be significant heterogeneity in CLL patients with regard to the mutational status of the immunoglobulin (Ig) heavy-chain variable region (IgVH), which usually indicates whether the B cell has experienced somatic hypermutation in the germinal center. In CLL, approximately half of patients have a mutated IgVH (M-IgVH) indicative of a post-germinal center B cell, whereas other patients have an un-mutated IgVH (UM-IgVH), a finding that has prognostic significance. The precise normal counterpart(s) of CLL cells during B-cell development has not been definitely identified; however, the CLL immunophenotype is similar to that of mature, antigen-experienced, activated B cells. Recent gene expression array experiments indicate that both M-IgVH and UM-IgVH CLL cells resemble memory B cells more than any other identified normal B-cell subset. In addition, overexpression of the src tyrosine kinase ZAP-70 in CLL, which is normally expressed in T and natural killer (NK) cells and not B cells, is strongly correlated with UM-IgVH subset. Further advances in defining the relationship between CLL cells and normal B-cell development may yield novel therapeutic targets in CLL.

 Unlike many hematologic malignancies, CLL cells do not contain balanced chromosomal translocations, detectable using traditional cytogenetic techniques. However, in fluorescence in situ hybridization (FISH) technology on nondividing cells (i.e., interphase cytogenetics) has identified recurrent chromosomal abnormalities in approximately 80% of CLL cases. The most common cytogenetic abnormalities in CLL are del(13q14), del(11q), trisomy 12, and del(17p), which influence prognosis (Table 28-1). The gene(s) involved in 13q14 deletion have not been definitively identified; however, the retinoblastoma (Rb) gene maps close to this region. Two micro-RNAs (miR-15, miR-16) have been mapped to the 13q locus and are also potential candidates for mediating this locus effect in CLL. Del(11q22-23) encompasses the ataxia telangiectasia mutated (ATM) gene locus, and mutations in the ATM gene have been observed in CLL, suggesting that ATM is the target of this deletion. Similarly, del(17p) encompasses the p53 tumor suppressor gene, and point mutations or deletions of p53 are also present in CLL patients with poor prognosis, suggesting p53 is the target gene in del(17p) CLL patients. The gene(s) important for trisomy 12 effects have not been identified. Approximately 95% of CLLs have increased expression of the antiapoptotic Bcl-2 oncogene, and 70% have expression levels equivalent to follicular lymphoma cells harboring t(14;18). CLL cells utilize other mechanisms to increase Bcl-2 expression, and do not typically contain the classic t(14;18) present in follicular lymphoma. Recent studies using cancer genome sequencing have defined recurrent somatic mutations in CLL patients. These genes affect common pathways including DNA damage and cell-cycle control (ATM, p53), Notch signaling (NOTCH1, FBXW7), RNA splicing (SF3B1, DDX3X), and cytokine/toll like receptor signaling (MYD88 DDX3X, MAPK1). Of these, p53 mutations affect about 10%, NOTCH1 mutations affect about 10%, and SF3B1 mutations affect 10% to 15% of CLL patients, and all confer an independent poor prognosis. Further, substantial clonal heterogeneity exists in CLL patients, and ongoing research is studying how CLL subclonal architecture affects, and is in turn influenced by, anti-CLL therapy.

1. Clinical presentation, laboratory features, and diagnosis. Patients with CLL may have a wide range of symptoms, signs, and laboratory abnormalities at the time of initial diagnosis. Many patients are asymptomatic, and a routine CBC reveals a lymphocytosis, whereas fewer patients present with extreme fatigue or B symptoms including fevers, night sweats, unintentional weight loss equal to or less than 10% of body weight. Other presentations include painless lymphadenopathy, anemia, thrombocytopenia, and infections. Physical examination findings are normal in 20% to 30% of patients, but may include lymphadenopathy, splenomegaly, and hepatomegaly in approximately half the number of patients. Laboratory findings uniformly include a lymphocytosis (greater than 5,000/μL), and may include anemia, thrombocytopenia, elevated lactose dehydrogenase (LDH) levels, elevated B2M levels, positive Coombs’ test, polyclonal increase in γ-globulin levels, or hypogammaglobulinemia. The peripheral blood smear typically shows numerous small, mature appearing lymphocytes with clumped chromatin and no nucleolus, with “smudge” cells present as crush artifacts of fragile CLL cells. Bone marrow biopsy shows infiltration with small lymphocytes in nodular, interstitial, or diffuse pattern. The histopathologic lymph node findings in CLL/SLL consist of diffuse effacement of the nodal architecture by small, mature appearing lymphocytes with a low mitotic rate, and few (less than 10%) larger prolymphocytes. Peripheral blood, bone marrow, or lymph node flow cytometry reveals a characteristic immunophenotype (Table 28-2). The essential diagnostic criteria for CLL identified by the CLL international working group (IWG) include an absolute monoclonal B lymphocytosis of more than 5,000/μL with a typical morphology, and commonly, the bone marrow is infiltrated with small lymphocytes accounting for more than 30% of nucleated cells, and a typical immunophenotype (CD5+, CD23+, CD10−, CD19+, CD20^+dim, CyclinD1−, CD43±). In addition, the following tests may be useful under certain circumstances: molecular genetic analysis to detect antigen receptor rearrangements, interphase FISH for 17p−, 11q−, 13q−, +12, and determination of CD38 and/or ZAP-70 expression. The differential diagnosis of CLL includes other indolent B-cell lymphomas (mantle cell, follicular, lymphoplasmacytic), hairy cell leukemia, large granular lymphocytic (LGL) leukemia, prolymphocytic leukemia (PLL), and adult T-cell leukemia/lymphoma. Once the diagnosis is made, initial workup should include physical examination, performance status, assessment of B symptoms, CBC with differential count, LDH, comprehensive metabolic panel, and in certain circumstances quantitative Igs, reticulocyte count, direct Coombs’ test, computed tomography (CT) scans of the chest/abdomen/pelvis, B2M, and uric acid.

4. Staging and prognosis. The Rai and Binet clinical staging systems were described in the 1970s and provided prognostic information on survival based upon physical examination findings and blood counts (Table 28-3). These clinical staging systems have been extensively validated and are widely used in clinical practice to provide estimates of survival to patients. However, the clinical course of early stage patients is heterogeneous: some patients do not require therapy for many years, whereas others have rapid progression and poor responses to therapy. Further, these staging systems provide little information to help predict the clinical outcome of early stage patients or the response to therapy. Additional laboratory parameters have been identified as markers of tumor burden and independent poor prognostic factors including an elevated LDH, a lymphocyte doubling time of less than 12 months, and diffuse bone marrow infiltration pattern. Serum proteins have also been linked to poor prognosis, including elevated levels of thymidine kinase (TK), soluble CD23 (sCD23), and B2-microglobulin (B2M). Notably, patients with an elevated B2M level have a shorter survival, and worse responses to traditional chemotherapy approaches. A summary of prognostic factors in CLL is provided in Table 28-4.

CLL, chronic lymphocytic leukemia; sIg, surface immunoglobulin.

 The mutational status of the IgVH locus is an important genetic parameter for prognostication in CLL patients, with the mutated IgVH (M-IgVH) associated with slow progression and long survival, whereas the un-mutated IgVH (UM-IgVH) is associated with an unfavorable course and rapid progression. Surrogate markers of UM-IgVH and poor prognosis include flow cytometry-detected expression of CD38 and ZAP-70 by CLL cells. Recent studies have suggested that more than 20% ZAP-70 expression as assessed by flow cytometry is associated with a median survival of less than 5 years, compared with 10 years in patients with less than 20% expression, and is now widely available. This was superior in predicting outcomes in CLL patients compared with IgVH mutational status testing.

BM, bone marrow; ATM, ataxia telangiectasia mutated; LDH, lactose dehydrogenase; B2M, beta 2 microglobulin.

 Interphase cytogenetics has identified both favorable (13q−) with a longer treatment-free interval and overall survival, and unfavorable (17p−, 11q−) with a short treatment-free interval and overall survival, subsets of CLL patients (Table 28-2). Consistent with this, subsets of patients with mutations in p53 (17p) or ATM (11q) have a poorer prognosis. When molecular parameters consisting of IgVH mutational status, 17p−, and 11q− were included in a multivariate analysis, the clinical stage was not identified as an independent prognostic factor. Four distinct molecular prognostic groups (17p−, 11q−, UM-IgVH, M-IgVH) have been identified, and have provided a framework for risk-adapted therapeutic strategies. Laboratory testing by interphase cytogenetics for genetic abnormalities of clinical significance in CLL patients is now widely available. Although these advances in genetic and molecular prognostic factors are promising, the decision to treat patients with CLL based upon them requires validation in prospective, randomized clinical trials. More recently, recurrent somatic mutations have been identified in CLL that have been integrated into cytogenetics-based risk scores. For example, a retrospective study of previously untreated CLL patients identified high risk (p53 or BIRC3 abnormalities), intermediate risk (NOTCH1 or SF3B1 mutations and/or 11q deletion), low risk (trisomy 12 or no cytogenetic abnormality), and very low risk (13q− as an isolated abnormality). These findings will require confirmation in prospective clinical trials, and assessment in relapsed/refractory patients.

1. Complications associated with chronic lymphocytic leukemia
2. Richter’s transformation. Richter’s syndrome (RS), originally described as the development of an aggressive NHL in patients with CLL/SLL, is now commonly used to describe transformation into any aggressive malignancy including diffuse large B-cell lymphoma (DLBCL), or, less commonly, PLL, Hodgkin’s lymphoma (HL), lymphoblastic lymphoma, hairy cell leukemia, or aggressive T-cell NHL. RS occurs in 2% to 8% of CLL patients, and results are conflicting on whether purine analog therapy increases the risk for transformation. In about 80% of patients, the malignant clone in RS develops through transformation of the original CLL clone (molecularly related to original CLL), while in about 20% of patients, they arise as an independent neoplasm. This distinction is clinically important, since patients with clonally unrelated RS/DLBLC have a prognosis similar to de novo DLBCL and frontline treatment with R-CHOP is adequate. RS is suspected clinically in patients with CLL/SLL that have a rapidly enlarging lymph node group, rapidly progressive splenomegaly or hepatomegaly, elevated LDH and B2M, new B symptoms (fever, night sweats, weight loss), or a sudden decline in performance status. Patients with suspected RS should have a tissue biopsy to confirm the diagnosis. Treatment of clonally related RS usually involves aggressive chemotherapy combinations utilized in NHL (e.g., R-CHOP) with response rates of 5% to 43%, and median survival between 5 and 8 months. For responding patients, an allogeneic SCT in CR1 is preferred, and is typically offered to those patients who respond to initial chemotherapy and who are good transplant candidates with an HLA-matched donor. For patients not responding to R-CHOP, other salvage regimens such as OFAR, RDHAP, or RICE are possibilities. Overall, clonally related RS patients have a poor prognosis, and novel treatment strategies are needed.
3. Autoimmune complications. AI phenomena are common in CLL, occur more frequently in advanced-stage patients and those with UM-IgVH, and include autoimmune hemolytic anemia (AIHA), autoimmune thrombocytopenia (ITP), AI neutropenia, and pure red cell aplasia (PRCA) (Table 28-5). In CLL patients with isolated anemia, laboratory evaluation for hemolysis should be performed, including direct Coombs’ (direct anti-globulin) testing, LDH, haptoglobin, indirect bilirubin, and reticulocyte count. It is important to note that these tests may not provide consistent findings of hemolysis in CLL patients, as an elevated LDH may be due to other causes in CLL, and a low reticulocyte count could be due to poor bone marrow responses when infiltrated with CLL. In addition, AI complications may be triggered in CLL patients at the time of treatment with fludarabine. Treatment of AIHA in CLL patients is similar to the steroid approach in non-CLL patients, with a typical course of prednisone 1 mg/kg/day until a response is achieved, followed by a prolonged, 2- to 3-month, taper. If the AI complication is not responsive to steroids, other treatments include intravenous immunoglobulin (IVIG), cyclosporine, splenectomy, and rituximab. In addition, treatment of the underlying CLL may improve the AI cytopenias.

GI, gastrointestinal.

3. Infectious complications. The immune deficiency associated with CLL is multifactorial, and includes hypogammaglobulinemia, T and NK cell dysfunction, and decreased phagocytic function. Infectious complications are frequent and respond to appropriate antimicrobial therapy. For CLL patients with hypogammaglobulinemia (IgG <500 mg/dL) and recurrent serious infections, treatment with IVIG 400 mg/kg IV every 3 to 4 weeks (goal IgG trough approximately 500 mg/dL) reduces serious bacterial infection rates without clear effects on overall survival. Patients treated with fludarabine or alemtuzumab develop therapy-related T-cell immune defects, and are at a significantly increased risk for cytomegalovirus (CMV) reactivation, pneumocystis, varicella zoster, herpes viruses, listeria, and other opportunistic infections. Prophylaxis against pneumocystis, herpes simplex virus (HSV), and varicella zoster virus (VZV), as well as monitoring for CMV reactivation should be considered when treating CLL patients with these agents.
4. Decision to initiate treatment. Treatment of CLL has historically been palliative as chemotherapeutic regimens have not impacted overall survival to date, and therefore the diagnosis of CLL does not mandate the need for therapy. Indications for treatment include (a) eligibility for treatment on a clinical trial, (b) advanced clinical stage (Rai III/IV), (c) AI cytopenias, (d) recurrent infections, (e) B symptoms, (f) threatened compromise of organ function, (g) cytopenias, (h) bulky disease, (i) rapid progression of disease, (j) histologic transformation, or (k) patient preference for immediate treatment. Observation until an indication for treatment arises is not currently thought to impact CLL patients’ overall survival or response to therapy when initiated. As therapies for CLL become more effective and strategies to monitor for minimal residual disease develop, this paradigm of initial observation may change in the future.
5. Initial treatment of chronic lymphocytic leukemia. The goal of CLL chemotherapy remains palliative; however, these objectives are currently being reevaluated in clinical trials as chemotherapy and monoclonal antibody combinations, kinase inhibitors, and more recently immunotherapy approaches produce high complete remission rates and in some cases elimination of disease based on sensitive molecular and flow cytometric techniques. Accepted first-line therapy options include clinical trials (preferred as standard therapy is not curative), radiation therapy (especially for stage I SLL), alkylating agents (e.g., chlorambucil or cyclophosphamide), purine analogs (fludarabine), or combinations such as fludarabine plus rituximab (FR), fludarabine plus cyclophosphamide plus rituximab (FCR) (see Appendix for treatment regimens). Randomized clinical trials have established higher complete remission rates and PFS intervals for fludarabine over alkylating agents and FC over fludarabine alone; however, no difference in overall survival has been demonstrated. A US intergroup study is comparing upfront FCR, FR, and FR followed by lenalidomide consolidation, and may provide randomized clinical trial data to prioritize these regimens when results are reported. In addition, a European study compared FCR with BR in the upfront setting, and preliminary results suggest equal efficacy and lower toxicity with BR in patients >65 years of age. Treatment duration is clinically aimed at palliating the inciting cause for therapy, and is typically 4 to 8 cycles. Clinical studies suggest a longer duration of remission for patients that achieve a complete remission (CR). Ongoing studies tracking cytogenetic and molecular minimal residual disease are seeking to clarify how MRD impacts relapse, and whether targeting MRD following initial therapy is clinically beneficial. Before initiating treatment, consideration should be given to tumor lysis syndrome prophylaxis, especially in patients with very high lymphocyte counts or bulky disease. In addition, patients treated with purine analogs should receive prophylaxis against pneumocystis and varicella zoster. Moreover, patients treated with anti-CD20 mAbs are at risk for hepatitis B and C reactivation, and testing for these viral infections is recommended prior to starting therapy, and prophylactic entecavir or other appropriate antiviral drug may be indicated. Very rarely, anti-CD20 mAbs may cause progressive multifocal leucoencephalopathy (PML), and this complication should be considered in patients that develop cognitive or neurologic symptoms. We currently utilize FCR as our standard frontline therapy for younger, fit CLL patients, and BR or obinutuzumab/ofatumumab plus chlorambucil for older patients. Patients harboring a 17p deletion or p53 mutation are at considerable risk for not responding or having a very brief remission to standard fludarabine-based therapy, and should be considered for clinical trials. While FCR and FR can induce remissions and represent appropriate therapy for 17p deleted CLL at this time, alternative approaches include alemtuzumab plus rituximab, ibrutinib, idelalisib plus rituximab, high dose methylprednisolone plus rituximab, or obinutuzumab plus chlorambucil. For patients with 11q-deletion, alkylator-based therapy has been shown to improve prognosis. Patients who are younger or have minimal comorbidities may be treated with FCR, BR, PCR, or obinutuzumab plus chlorambucil, while patients aged >70 and comorbidities are better candidates for obinutuzumab or rituximab plus chlorambucil or BR. Clinical research defining the optimal treatment approach for CLL patients continues at a rapid pace, and as new data emerge, these treatment recommendations will also evolve.
6. Treatment of relapsed and relapsed chronic lymphocytic leukemia. Treatment of patients with CLL that have relapsed after at least 1 prior therapy has changed markedly over the past several years. Treatment options now include ibrutinib (Bruton’s tyrosine kinase inhibitor), idelalisib (PI3Kγ inhibitor), ofatumumab (2nd generation anti-CD20 mAb) and obinutuzumab (ADCC optimized anti-CD20 mAb), and alemtuzumab (anti-CD52 mAb). Patients who have a long remission after initial therapy with fludarabine–rituximab-based regimens may still be treated in a similar manner as untreated patients, with retreatment with first-line therapies. However, new second-line options provide excellent alternatives with lower incidences of long-term complications and myelosuppression. It remains unclear which second-line therapy is the best initial choice, since this has not been clarified in randomized trials in light of very recent approvals of a number of novel agents. In a randomized study comparing ibrutinib to ofatumumab, the former was associated with a response rate of approximately 70% and a 2-year PFS of 75%. Both PFS and OS were improved compared with ofatumumab (N Engl J Med 2014;371:213). Typical adverse prognostic factors such as 17p and 11q deletion did not affect the ORR or PFS. Importantly, ibrutinib can induce an isolate lymphocytosis in CLL patients that occurs in the first weeks of therapy and may persist for weeks to months, that does not signify disease progression. An important uncommon (5%), but poorly understood toxicity of ibrutinib is bleeding events, which do not have a clear mechanistic explanation at this time. Idelalisib represents a PI3K-γ inhibitor with activity in relapsed CLL and is now approved for use in combination with rituximab. In relapsed CLL patients that were not deemed appropriate for cytotoxic agents, idelalisib plus rituximab resulted in an ORR of 81% (0% CR), and improved PFS and OS compared with a rituximab plus placebo control arm (N Engl J Med 2014;370:997). Encouragingly, subgroup analysis demonstrated no difference in outcomes in patients with 17p deletion, p53 mutation, or IGVH un-mutated patients. Similar to ibrutinib, a transient lymphocytosis was also observed, which appeared to be dampened by concurrent rituximab therapy. Anti-CD20-mAb options include rituximab, ofatumumab, and obinutuzumab, although combinations with a targeted agent or chemotherapy are commonly utilized. Other treatment strategies for relapsed/refractory CLL patients include retreatment with various combinations of fludarabine, cyclophosphamide, and rituximab, which yield overall response rates of 29% to 59% in this patient population, but also have significant toxicities, especially in older patients previously treated with similar agents. High-dose steroids (e.g., methylprednisolone) in combination with anti-CD20 mAbs, have also provided responses in refractory CLL patients, with response rates up to 77% including patients with p53 or 17p− genetic abnormalities. Alemtuzumab has shown activity in fludarabine-refractory CLL; however, because of its high risk of serious infections, it is reserved for multiply relapsed CLL not responding to other agents, the presence of a 17p deletion, or if the patient transforms to more aggressive prolymphocytic leukemia. Life-threatening infections can occur in patients treated with high dose methylprednisolone or alemtuzumab, and they necessitate routine prophylaxis against infections. While the long-term remission duration with ibrutinib and other novel agents remains unclear, most patients are expected to relapse and require sequential anti-CLL therapy. Clinical trials are also addressing moving novel agents up-front, and combining various agents to optimize rate, depth, and duration of a response.

 Myeloablative allogeneic SCT has historically been used in a limited manner in CLL patients, primarily on account of their advanced age. Small single-center and registry studies of selected CLL patients undergoing myeloablative SCT have shown treatment-related mortality rates of 24% to 46%, PFS of 26% to 62%, and overall survival of 31% to 76% at 3 to 10 years of projected follow-up. Recent studies utilizing reduced-intensity or nonmyeloablative allogeneic SCT have shown promising results, with treatment-related mortalities of 15% to 22%, PFS of 52% to 67%, and overall survival of 60% to 80% with a short 2-year projected follow-up. In general, results are superior in younger patients, those with little comorbidity, and those with chemosensitive disease before transplantation. The optimal conditioning regimens, patient age eligible for transplantation, and salvage chemotherapeutic regimens remain under investigation.

 Despite the recent FDA drug approvals described above, numerous new drugs are being evaluated in early clinical trials for patients with relapsed and refractory CLL, and enrollment on a clinical trial should be considered routinely for these patients. Lenalidomide is a thalidomide analog with multiple potential mechanisms of action that has shown considerable single-agent clinical activity in multiple hematologic diseases, including multiple myeloma and myelodysplastic syndrome. Two phase II clinical trials have investigated lenalidomide doses of 25 mg daily for days 1 to 21 of a 28-day cycle in relapsed CLL patients, and have shown excellent tolerability and overall response rates of 32% to 65% and a complete response rate of 5% to 9%. Notably, there was a tumor flare reaction noted following treatment that should not be interpreted as rapid disease progression. More recent studies have combined lenalidomide with rituximab with ORR of 66% (12% CR) with similar response rates in patients with 17p deletion, and this combination is an option for relapsed CLL. Flavopiridol is a synthetic flavone with early clinical activity in a single-institution study that demonstrated overall response rates of 43% (many of which were durable for more than 12 months) as well as responses in patients with high-risk genetic features. The most notable toxicity reported was tumor lysis syndrome, which should be monitored for closely when treating CLL patients with this agent. Clinical trials evaluating lenalidomide and flavopiridol alone and with other agents are ongoing.

 In summary, depending on patient characteristics, appropriate treatment options for relapsed CLL include clinical trials, ibrutinib, idelalisib, alemtuzumab, ofatumumab, obinutuzumab, combination fludarabine- or bendamustine-based chemotherapeutic regimens, high-dose steroids, lenalidomide, and, in appropriate candidates, allogeneic SCT. Finally, autologous T cells engineered to express an anti-CD19 chimeric antigen receptors (CAR) appear to have preliminary activity in small numbers of refractory CLL patients. Since these T cells have been shown to persist long-term in the patient, this approach provides a potential for long-term anti-CLL immunity without the toxicity of allogeneic HSCT, and will certainly be the focus of future CLL studies. Additional emerging immunotherapy agents under investigation include bispecific T cell engagers (BiTEs) such as blinatumomab, that simultaneously engage tumor cells (via anti-CD19 scFV) and T cells (via anti-CD3 scFv). Immune checkpoint blockade, for example with anti-PD-1/PD-L1/L2 blocking agents, are also being explored.

SUGGESTED READINGS

Chronic Myeloid Leukemia

Baccarani M, Deininger MW, Rosti G, et al. European Leukemia net recommendations for the management of chronic myeloid leukemia: 2013. Blood 2013;122:872–884.

Bacigalupo A. Stem cell transplantation for chronic myeloid leukemia (CML): current indications and perspectives. Curr Cancer Drug Targets. 2013;13:775–778.

Deininger M, Schleuning M, Greinix H, et al. The effect of prior exposure to imatinib on transplant-related mortality. Haematologica 2006;91:452–459.

Druker BJ, Guilhot F, O’Brien SG, et al. Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006;355:2408–2417.

Melo JV, Ross DM. Minimal residual disease and discontinuation of therapy in chronic myeloid leukemia: can we aim at a cure? Hematology Am Soc Hematol Educ Program 2011:136–142.

Osborn M, Hughes T. Managing imatinib resistance in chronic myeloid leukemia. Curr Opin Hematol 2010;17:97–103.

Chronic Lymphocytic Leukemia

Brown JR. The treatment of relapsed refractory chronic lymphocytic leukemia. ASH Education Program, Hematology. Am Soc Hematol Educ Program 2011:110–118.

Byrd JC, Brown JR, O’Brien S, et al. Ibrutinib versus ofatumumab in previously treated chronic lymphocytic leukemia. N Engl J Med 2014;371:213–223.

Byrd JC, Furman RR, Coutre SE, et al. Targeting ibrutinib in relapsed chronic lymphocytic leukemia. N Engl J Med 2014;369:32–42.

Byrd JC, Jones JJ, Woyach JA, et al. Entering the era of targeted therapy for CLL: impact on the practicing clinician. J Clin Oncol 2014;32:3039–3047.

ESMO Guidelines consensus conference on malignant lymphoma 2011 part 1: DLBCL, FL, and CLL. Ann Oncol 2013;24:561–576.

Furman RR, Sharman JP, Coutre SE, et al. Idelalisib and rituximab in relapse chronic lymphocytic leukemia. N Engl J Med 2014;370:997–1007.

Gaindano G, Foa R, Dalla-Favera R. Molecular pathogenesis of chronic lymphocytic leukemia. J Clin Invest 2012;122:3432–3438.

Hallek M. CLL: 2013 updated on diagnosis, risk stratification and treatment. Am J Hematol 2013;88:804–816.

Jones JA, Byrd JC. How will B-cell-receptor-targeted therapies change future CLL therapy? Blood 2014;123:1455–1460.