Challenges in the Frontline Treatment of Patients With Chronic Lymphocytic Leukemia
- First Online:
- Cite this article as:
- Lamanna, N. Curr Hematol Malig Rep (2010) 5: 45. doi:10.1007/s11899-009-0040-3
- 82 Downloads
Therapy for chronic lymphocytic leukemia has improved dramatically over the past 20 years. Traditional therapy with oral chlorambucil led to complete responses in less than 5% of treated patients, in marked contrast to modern regimens, which can reliably produce complete responses in over 50% of patients. This remarkable improvement is attributable to the use of purine analogue-based treatment as well as monoclonal antibodies. Novel combinations of these agents have emerged as effective new therapies for previously untreated patients. Clinical studies indicate that such combinations can induce higher response rates (including complete responses) than single-agent therapy. Those patients who achieve a complete response have superior progression-free survival compared with those who achieve only a partial response. Though not yet demonstrated in a prospective randomized trial, treatment approaches aimed at achieving high-quality responses may one day improve survival for patients with chronic lymphocytic leukemia. However, many challenges remain, such as finding less toxic and equally efficacious regimens for older patients, who remain the majority of the population with this disease.
KeywordsChronic lymphocytic leukemia Frontline treatment
Over the past 20 years, the frequency of complete response achieved with initial treatment of chronic lymphocytic leukemia (CLL) has increased dramatically from less than 5% with single-agent alkylators to more than 50% with multiagent chemoimmunotherapy. This remarkable improvement can be attributed primarily to an increase in the number and activity of therapeutic agents recently made available to treat this disease. It is hoped that the recent abundance of information about the underlying biology of leukemic lymphocytes will translate into the development of even more therapeutic strategies. Before the introduction of purine analogues, conventional therapy for CLL consisted of oral alkylating agents, chlorambucil and cyclophosphamide, with or without corticosteroids. Unfortunately, these agents produced primarily partial responses and only rarely induced complete responses [1, 2]. Furthermore, response durations were brief and patients succumbed to progressive resistant disease. The introduction of the purine analogues, fludarabine [3, 4], cladribine [5, 6], and pentostatin [7, 8], as well as the monoclonal antibodies, rituximab [9, 10] and alemtuzumab [11, 12], have renewed interest in this disease. As a result, the level of clinical investigation in CLL has increased, with an explosion of clinical trials examining combinations of two or more non-cross resistant agents. Yet despite these newer treatments, many challenges and unanswered questions still remain in treating this disease.
Indications for Therapy
In contrast to other malignancies, in which treatment of most patients begins soon after diagnosis (unless they are too frail or do not desire treatment), CLL patients are often actively “observed” for a period of time after initial diagnosis. In fact, one of the most important questions often asked in CLL is not how to treat, but when to treat. This question remains unanswered. Studies performed by the French Cooperative Group compared therapy with chlorambucil versus observation in patients with early-stage disease . In the first trial, 609 patients were randomly assigned to receive either daily chlorambucil or no treatment. In the second study by the same group, 926 patients were randomly assigned to receive either intermittent chlorambucil plus prednisone or no treatment. In these two trials, with more than 11 years and 6 years of follow-up on 1535 patients, immediate treatment with daily chlorambucil alone or intermittent chlorambucil and prednisone did not prolong survival. Based on these observations, the clinical standard is to manage patients with expectant observation until the patient develops symptomatic disease or signs of progressive bone marrow failure. Indications to initiate chemotherapy include disease-related anemia (hemoglobin <11 g/dL) or thrombocytopenia (platelets <100,000/µL) in the absence of autoimmune disease, bulky (or symptomatic) lymphadenopathy, and transformed disease . Additionally, patients with a rapid doubling of their lymphocyte counts (<3 months) when the total lymphocyte counts exceed 100,000/µL may be considered for therapy. However, these earlier studies have not addressed the impact of newer agents on treating early-stage patients, especially early-stage patients with poor prognostic markers (chromosomal aberrations such as deletion in 17p or 11q, absence of a mutated immunoglobulin variable region [IgVH], overexpression of ZAP-70 or CD38, elevated β-2 microglobulin), as these patients have a shorter overall survival [15, 16, 17, 18, 19, 20, 21]. Several ongoing clinical trials may address this issue in the advent of more modern chemoimmunotherapy regimens. The German CLL Study Group is randomizing untreated (Rai stage 0–II) patients with high risk of progression either to early treatment with combination fludarabine, cyclophosphamide, and rituximab or to observation alone. High-risk patients are defined as having two or more of the following: elevated serum thymidine kinase levels, unfavorable molecular cytogenetics (17p-/11q- deletion, trisomy 12), predicted lymphocyte doubling time less than 12 months, and unmutated IgVH gene (percentage of homology ≥ 98%). The Cancer and Leukemia Group B (CALGB) is currently randomizing untreated (RAI 0-I) patients with high risk of progression based on genetics (unmutated IgVH gene) to either initiation of treatment with fludarabine and rituximab or observation .
The introduction of the purine analogues has reinvigorated clinical research in CLL. Fludarabine currently is the most widely studied of these agents. It has been shown to induce responses in a majority of patients (although the frequency of complete responses is less impressive) . This activity led to the initiation of a large, randomized intergroup trial comparing fludarabine with the more traditional CLL therapy, chlorambucil, in untreated patients . The group treated with fludarabine had a superior overall response rate (63% versus 37%) and complete response rate (20% versus 4%) when compared with the group treated with chlorambucil. The median response duration was also longer in the fludarabine treatment group (25 months versus 14 months) than with chlorambucil treatment. Unfortunately, overall survival was not significantly different between the two arms of this study. Other trials also confirmed the significant activity of fludarabine in these patients and indicated that achieving a high-quality response (complete response, nodular response, or flow cytometric response) correlates with improved survival [25, 26].
Although fludarabine is quite active, its ability to induce complete responses in only a minority of previously untreated patients with CLL led to the investigation of combination chemotherapy. The results of early studies with corticosteroids  or chlorambucil [28, 29] combined with fludarabine were disappointing because of increased toxicity, which limited dose intensity without an obvious improvement in frequency of response.
Fludarabine ± Cyclophosphamide ± Rituximab
Investigations of combined fludarabine with cyclophosphamide (FC), however, showed significant activity and acceptable toxicity. O’Brien et al.  reported on 128 patients with CLL treated with this combination; 94 had been previously treated and 34 had not. All patients received fludarabine (30 mg/m2 daily for 3 days). The dose of cyclophosphamide was reduced from an initial dose of 500 mg/m2 per day for 3 days to ultimately a dose of 300 mg/m2 per day for 3 days secondary to severe myelosuppression. This regimen was found to have marked activity, with an overall incidence of response of 74%. Response rates were greatest for patients who were previously untreated (88%). Increased myelosuppression was noted, with an increase in infectious complications, including grade 3/4 infections in 40% of patients.
Subsequently the group at M. D. Anderson Cancer Center added rituximab to the FC regimen and reported on this FCR combination in 224 previously untreated patients with CLL (only one third with high-risk disease) [31, 32]. An overall response frequency of 95% was achieved in this population, with complete response in 64% of Rai stage III–IV patients. Grade 3/4 neutropenia was noted in 52%, with infections complicating 13% of the administered cycles. (The paper does not detail the fraction of patients who developed any infection while being treated.) Dose reductions were required in 35 patients (16%), and 58 patients (26%) were unable to complete the prescribed six courses of therapy. Not surprisingly, patients with more advanced stage and older age had a lower probability of achieving a complete response and were more likely to discontinue therapy.
Flinn et al.  reported on the combination of fludarabine and cyclophosphamide in 17 previously untreated patients with CLL. Fludarabine (20 mg/m2 per day for 5 days) was combined with cyclophosphamide (600 mg/m2 on day 1) administered with growth factor support. The overall response rate was 100%, with 47% achieving a complete response. The regimen was generally well tolerated, but infectious toxicities were important, including one case of cryptococcal pneumonia and one death from Pneumocystis carinii pneumonia (PCP). Following completion of treatment, three cases of herpes zoster and one additional case of PCP were noted.
Previously Weiss and colleagues  reported on a regimen combining fludarabine and cyclophosphamide in a sequential approach to allow administration of full-dose fludarabine followed by dose-intense consolidation with cyclophosphamide. Rituximab was subsequently added to this regimen in sequential fashion as a second consolidation after patients received high-dose cyclophosphamide [34•]. Patients received induction therapy with fludarabine (25 mg/m2 per day for 5 days), given every 4 weeks for six cycles, followed by consolidation with high-dose cyclophosphamide (3000 mg/m2) administered every 2 to 3 weeks for three cycles with filgrastim support. Patients then received rituximab (375 mg/m2 weekly) for four doses. In addition to standard morphologic evaluation, patients were also evaluated for minimal residual disease by flow cytometric analysis and by polymerase chain reaction analysis using a patient-specific “clonotypic” reaction. In 36 previously untreated patients with CLL who were given this three-drug regimen, the total response rate was 89%, and 61% of patients achieved a complete response. Sequential therapy with fludarabine, high-dose cyclophosphamide, and rituximab produced a high frequency of complete response, with improvements in response seen with each phase of therapy (36% improvement after cyclophosphamide and 25% after rituximab). Thus, the addition of rituximab appears to further enhance the quality of response in these patients.
The CALGB has used a regimen of fludarabine and rituximab (FR). Initially the regimen was evaluated with patients receiving the combination either simultaneously or sequentially . Patients who received the simultaneous regimen received 11 cycles of rituximab, compared with 4 cycles in the sequential arm. The simultaneous regimen produced a higher overall response (90% vs 77%) and complete response (47% vs 28%) but there was no difference between the two arms in progression-free survival or overall survival. Subsequently the CALGB reported that adding rituximab to fludarabine (compared with historical controls) may increase survival in patients who receive the combination therapy .
Most recently, the German CLL Study Group has presented preliminary results of a large multinational study randomly assigning patients to initiation of therapy with either FC or FCR . About 400 patients (median age, 61 years) were randomly assigned to each arm. The FCR combination was associated with a higher overall response rate (95% vs 88%), the frequency of complete response (52% versus 27%) was higher, and progression-free survival was increased at 2 years. At the time of the report, there was no difference in overall survival between the two groups and, as expected, there was an increase in all grade 3/4 hematologic and nonhematologic toxicities with the three-drug combination. There is great anticipation that perhaps with longer follow-up the addition of rituximab may yield a survival benefit in this younger cohort of patients. Those results are eagerly awaited.
Other Purine Analogues or Purine Analogue—Like Compounds
Combination therapeutic regimens using purine analogues other than fludarabine have been investigated. Robak et al.  reported the results of a randomized trial of cladribine with prednisone compared with chlorambucil and prednisone in 229 previously untreated patients with CLL. The cladribine group achieved a higher overall response rate than the chlorambucil group (87% vs 57%), as well as a higher frequency of complete response (47% vs 12%). However, increased toxicity was noted in the cladribine arm, including an increase in myelosuppression (23% vs 11%) and infection (56% vs 40%). Survival at 2 years was comparable for the two groups (78% vs 82%). Additional studies have been reported demonstrating no difference in overall response, complete response, progression-free survival, or overall survival between cladribine/cyclophosphamide and fludarabine/cyclophosphamide combinations .
Of the purine analogues active in CLL, pentostatin is the least myelosuppressive. At Memorial Sloan-Kettering Cancer Center, we have investigated a combination of pentostatin and cyclophosphamide in heavily pretreated patients with CLL . We also hypothesized that combining pentostatin with cyclophosphamide and rituximab would result in less myelosuppression and therefore less toxicity than combinations with either fludarabine or cladribine. This three-drug combination was active, with an overall response rate in 24 (75%) of 32 CLL patients, with 8 complete responses (25%), 1 nodular response (3%), and 15 partial responses (47%) . When given the three-drug regimen, 75% of fludarabine-refractory patients achieved a response, suggesting non-cross resistance. Because of these promising findings, this three-drug regimen has been examined in the upfront setting in untreated patients with CLL.
Kay and colleagues  have adapted this regimen by reducing the dose of pentostatin to 2 mg/m2 (compared with 4 mg/m2) and increasing the rituximab to eight doses instead of five. They evaluated 64 patients and noted an overall response frequency of 91%, with a 41% complete response rate. In addition, patients who were over the age of 70 years could tolerate this regimen as well as younger patients; they achieved similar responses without an increase in infectious complications [43•].
Most recently, bendamustine was approved for the upfront treatment of CLL. This agent is believed to have properties that encompass both purine analogue and alkylator-like activity. Bendamustine and chlorambucil were recently compared in previously untreated patients (<75 years of age) with CLL. Increased overall response (68% vs 31%) and complete response (31% vs 2%) were observed with bendamustine, as well as increased progression-free survival (21.6 months vs 8.3 months) . Combinations with bendamustine are currently being evaluated in the upfront setting.
Other Monoclonal Antibodies
In addition to rituximab, alemtuzumab, an anti-CD52 antibody, is currently approved for CLL patients in whom prior therapy with fludarabine has failed. In the pivotal trial of 93 patients with fludarabine-refractory disease that led to its approval, alemtuzumab was given intravenously at escalating doses up to 30 mg during the first week; subsequently, 30 mg was given thrice weekly for a total of 12 weeks . The overall response rate was 33%, with 2% achieving a complete response. Grade 3/4 infections were reported in 25 patients (27%) and five deaths were thought to be related to alemtuzumab. Cytomegalovirus reactivation occurred in 7.5% of patients. Alemtuzumab has also been studied in previously untreated patients with CLL with notable activity . It has been compared with chlorambucil , used as consolidation in patients with minimal residual disease , and evaluated in combination with both fludarabine  and rituximab . In many of these studies, consistent toxicity has been seen—most notably myelosuppression, immunosuppression, infusional reactions (subcutaneous administration is better tolerated), and opportunistic infections [51, 52]. Antibacterial and antiviral prophylaxis is recommended for all patients receiving alemtuzumab therapy. Alemtuzumab is most effective at reducing leukemia counts and bone marrow disease; it is less effective at shrinking bulky lymphadenopathy.
Immunomodulatory agents such as lenalidomide and thalidomide have recently been reported to have activity in CLL. The true mechanism of action of these drugs is not entirely understood, but their activity may be mediated by interaction with cytokines and the tumor microenvironment. Lenalidomide has been studied in several dose-escalation studies in untreated patients , as well as in untreated patients older than 65 years of age . No complete responses have been reported, but there were notable partial responses and patients with stable disease who have remained on therapy.
Challenges With Current Approaches
Recent studies of combination therapy for chronic lymphocytic leukemia (CLL)
Fludarabine vs chlorambucil
Cancer and Leukemia Group B 
Fludarabine + rituximab
M. D. Anderson Cancer Center 
Fludarabine + cyclophosphamide + rituximab
Memorial Sloan-Kettering Cancer Center [34•]
Sequential fludarabine → cyclophosphamide → rituximab
Mayo-Ohio State University 
Pentostatin + cyclophosphamide + rituximab
German CLL Study Group 
Fludarabine + cyclophosphamide vs fludarabine + cyclophosphamide + rituximab
Ongoing investigations are exploring the use of older cytotoxic agents in combination and the use of newer agents specific to this patient population, with the goal of increasing response, progression-free survival, and overall survival while maintaining acceptable toxicity in this older age group.
Clinical investigation in CLL has matured, with a growing number of opportunities to develop and investigate a variety of new therapies. Investigators can now develop regimens that produce high-quality responses in patients with CLL by combining active agents from different classes that are non—cross resistant. However, because it is uncertain whether combination therapy will offer improved overall survival in patients with CLL, whereas combination therapy clearly has been associated with increased toxicity, clinicians must be cautioned to offer some of the more aggressive combination regimens only to suitable patients, preferably in the setting of a clinical trial. It is hoped that continued efforts to develop new agents, along with clinical trials aimed at older patients and various subtypes of CLL patients, will help to address some of the issues and unanswered questions in the treatment of this disease.
Dr. Lamanna has been a consultant for Celgene and has received research support from Hospira, Celgene, and Biogen-Idec.