Chemoimmunotherapy
Based on the above background studies, it is clear that adding cyclophosphamide to fludarabine in the upfront setting improves OR, CR and PFS in CLL. Kay and colleagues recently examined whether the addition of an alkylating agent to other purine nucleoside analogs similarly improves outcomes [
10]. The Mayo Clinic and Ohio State retrospectively compared results of a phase II study of pentostatin and rituximab (PR) to previously published results using pentostatin, cyclophosphamide and rituximab (PCR) [
8]. The pentostatin dose was increased to 4 mg/m
2 in the PR regimen, but demographics of patients in both studies were similar. OR and CR rates were similar for PR (79%, 30%) and PCR (91%, 41%), but median PFS was significantly shorter for PR (12 months) compared to PCR (31 months). These results supported previous findings, described above, that the addition of cyclophosphamide to fludarabine improves OR, CR and PFS (Table
1).
While excellent results were obtained with FR and FCR in phase II studies, the benefit of adding rituximab to a purine nucleoside analog has not been examined in a prospective randomized trial. Therefore, the GCLLSG CLL8 study randomized 817 previously untreated patients to fludarabine 25 mg/m
2 days 1–3 and cyclophosphamide 250 mg/m
2 days 1–3, with or without rituximab 500 mg/m
2 day 1 (375 mg/m
2 day 1 of cycle 1) [
11]. OR, CR and median PFS favored FCR (93%, 45%, 43 months) over FC (85%, 23%, 32 months), although 2-year OS (91% vs. 88%) was similar (Table
1). The MDACC FCR trial previously showed that the ability to achieve ≤ 1% residual CLL by two-color flow cytometry significantly affected relapse free survival (RFS) and OS [
7]. Five of 138 patients (4%) whose post-FCR bone marrow flow cytometry had ≤ 1% residual CLL relapsed, in contrast to 17 of 62 patients (27%) with > 1% residual CLL. The importance of eradicating minimal residual disease (MRD) was confirmed by the GCLLSG CLL8 trial, which demonstrated that median PFS depended upon the ability to eradicate MRD in the peripheral blood, with PFS increasing from 15 months (MRD ≥ 10
-2) to 34 months (10
-4 ≥ MRD > 10
-2) to not reached (MRD < 10
-4) with increasing eradication of MRD [
12]. Furthermore, 67% of patients receiving FCR achieved MRD < 10
-4, compared to only 34% of FC patients, thus accounting for the improved PFS with FCR. In addition to reaffirming the importance of eradicating MRD, the GCLLSG CLL8 study demonstrated that MRD can be determined from peripheral blood and that bone marrow biopsies are not necessary to gain MRD information.
While regimens such as FR, FCR and PCR have achieved excellent results in academic centers, there are limited data on whether such regimens are able to achieve similar results in the community setting. Therefore, US Oncology performed a multicenter, community-based trial that randomized 184 patients (80% previously untreated, 20% relapsed) to PCR or FCR, using the Memorial Sloan Kettering PCR regimen (pentostatin dose 4 mg/m
2) given for 8 cycles and the Johns Hopkins FCR regimen (fludarabine 20 mg/m
2 days 1–5, cyclophosphamide 600 mg/m
2 day 1) [
13]. The primary endpoint, incidence of grade 3–4 infections, was similar for PCR (34%) and FCR (31%). Only 50% of patients in both arms completed therapy, resulting in surprisingly low OR and CR rates for PCR (45%, 7%) and FCR (58%, 17%), as summarized in Table
1. While the study was not powered to show a statistically significant difference between FCR and PCR, these findings indicated that results from academic centers may not necessarily be reproducible in the community. Specifically, a reduced ability to complete planned treatment may reduce the clinical efficacy of these regimens outside academic centers.
Because of the ability of the anti-CD52 alemtuzumab (Campath-1H) to eradicate MRD and its activity in high-risk CLL patients, the MDACC added alemtuzumab to FCR, to determine if the efficacy of FCR could be improved [
14]. The CFAR regimen consists of fludarabine 25 mg/m
2 on days 2–4, cyclophosphamide 250 mg/m
2 on days 2–4, rituximab 375 mg/m
2 (cycle 1) or 500 mg/m
2 (cycles 2–6) on day 2, and alemtuzumab 30 mg IV on days 1, 3 and 5 every 28 days for up to 6 cycles. Patients receive pegfilgrastrim, as well as prophylaxis for
pneumocystis carinii pneumonia (PCP) and cytomegalovirus (CMV). A phase II study in 78 relapsed CLL patients had attained OR 65% (CR 24%); median PFS was 27 months for the 19 CR patients and 10 months for the 32 PR patients. Given these promising results, the MDACC is conducting a phase II study of CFAR in previously untreated patients [
15]. Results in the first 48 patients with high-risk features revealed OR and CR of 94% and 69%, respectively, with OR 77% and CR 54% in 13 patients with del(17p13). Grade 3–4 neutropenia and thrombocytopenia were observed in 71% and 42% of patients, respectively, and 6% and 27% of patients developed major and minor infections, respectively. While the results do not appear to be superior to FCR at first glance, it must be noted that CFAR is being tested in higher risk patients and that PFS data are not yet available.
Novel agents
Bendamustine hydrochloride is a novel alkylating agent which contains a benzimidazole ring and is only partially cross-resistant with other alkylating agents
in vitro. The FDA approved bendamustine for the treatment of CLL in 2008 based upon a multi-center, European phase III study which randomized 319 previously untreated CLL patients to oral chlorambucil 0.8 mg/kg on day 1 and 15 every 28 days or bendamustine 100 mg/m
2 IV on day 1 and 2 every 28 days [
16]. Treatment in both arms was given for up to 6 cycles or until disease progression. The results of this study were updated at the 2008 ASH meeting. OR, CR and median PFS favored bendamustine (67%, 32%, 21.5 months) over chlorambucil (30%, 2%, 8.3 months), although bendamustine caused greater hematologic toxicity (40% vs. 19%), especially grade 3–4 neutropenia (23% vs. 9%) [
17]. The GLLSG is currently conducting a randomized trial in previously untreated patients to compare the combination of bendamustine and rituximab (BR) to the MDACC regimen of FCR, which has achieved the highest CR rate of any upfront CLL regimen to date.
Lenalidomide is an immunomodulatory drug that is a more potent analog of thalidomide. Lenalidomide has shown activity in relapsed CLL patients including activity in patients with high-risk cytogenetic features. The Roswell Park Cancer Insititute conducted a phase II study which administered lenalidomide 25 mg daily orally on days 1–21 every 28 days to 45 patients with relapsed CLL [
18]. Results showed OR and CR rates of 47% and 9%, and responses were observed in fludarabine refractory patients and those with poor-risk 11q22 and 17p13 deletions. The MDACC chose a different dosing strategy and administered lenalidomide by continuous low dose at 10 mg orally daily, with a 5 mg dose escalation every 28 days up to a maximum dose of 25 mg daily, to 44 patients with relapsed CLL with 10 mg being the median delivered dose [
19]. OR and CR rates of 32% and 7% were observed, and the OR rate was 31% in patients with high-risk cytogenetic abnormalities. However, tumor lysis and tumor flare reactions are potentially serious toxicities of lenalidomide, and the optimal dosing schedule with respect to safety and efficacy are undefined.
Two studies of lenalidomide in previously untreated patients were presented at the 2008 ASH meeting. Chen et al. summarized results of a phase I study in 25 Canadian patients [
20]. Due to grade 5 sepsis and grade 3–4 tumor lysis, the dose was decreased from 25 mg to 2.5 mg and then escalated to 10 mg daily for 21 days every 28 days. Toxicity included fatigue (78%), tumor flare (78%), rash (48%) and grade 3–4 neutropenia (43%). OR and CR were 65% and 0%, respectively. The MDACC presented a study in 43 elderly patients age 65 years or older [
19]. Lenalidomide was given continuously, and 5–10 mg daily was the median delivered dose. Grade 3–4 myelosuppression and tumor flare were observed in 26% and 44% of patients, respectively. OR and CR were 54% and 0%, respectively. While lenalidomide is clearly active in CLL, the absence of CR in previously untreated patients was disappointing.
Finally, James et al. presented a phase II study giving high dose methylprednisolone 1000 mg/m
2 day 1–3 every four weeks and weekly rituximab (total dose 4500–6750 mg/m
2) to 28 previously untreated patients [
21]. OR and CR were 96% and 32%, respectively. Patients with less prominent splenomegaly and lower beta-2-microglobulin levels were more likely to respond.
Relapsed CLL
Despite advances in first line therapy for CLL, patients invariably relapse and often acquire high risk chromosomal abnormalities such as del(11q22) and del(17p13), which result in resistance to therapy [
22,
23]. Patients who have fludarabine refractory disease have a median survival of less than one year. Thus, new agents are needed for the treatment of relapsed CLL, particularly for those patients with high-risk cytogenetic features. It is important to emphasize that the same NCI and IWCLL criteria for initiating therapy in previously untreated patients also apply for patients with relapsed CLL [
1,
2]. Asymptomatic relapsed patients should be followed expectantly and should receive treatment only upon development of cytopenias or symptoms.
Stilgenbauer et al. presented final results of the GCLLSG CLL2H study which administered subcutaneous alemtuzumab to 103 relapsed patients, many of whom had high-risk features [
24]. Infusion toxicity was minimal, but grade 3–4 anemia (56%), thrombocytopenia (57%), anemia (49%), cytomegalovirus reactivation (8%) and non-CMV infection (29%) were significant toxicities. Seventy-five patients died; 56% died of progressive CLL, and 31% died of infection. OR (34%), CR (4%) and median PFS (7.7 months) were similar to the results achieved by IV alemtuzumab in the pivotal CAM211 study [
25]. While alemtuzumab was an effective therapy, the outcome for high-risk patients remained poor without the use of allogeneic stem cell transplantation.
The GCLLSG presented a phase II trial administering bendamustine 70 mg/m
2 on day 1–2 and rituximab 500 mg/m
2 on day 1 to 81 relapsed CLL patients [
26]. OR and CR were 77% and 15%, respectively. Twelve of 13 patients (92%) with del(11q22), 4/9 patients (44%) with del(17p13), and 29/39 patients (74%) with unmutated IgVH responded, suggesting that bendamustine may be active in high-risk relapsed CLL. However, further studies are needed to determine whether bendamustine can be considered as a treatment for relapsed CLL patients with poor-risk features.
Flavopiridol is a synthetic flavone which broadly inhibits cyclin dependent kinases (CDK); down-regulates expression of anti-apoptotic proteins such as Mcl-1 and X-linked inactivator of apoptosis (XIAP); decreases phosphorylation and transcriptional activity of RNA polymerase II, resulting in decreased gene transcription; and induces apoptosis distally to p53 by activating caspase 3 in primary CLL cells. Lin et al. presented combined phase I/II results of flavopiridol (alvocidib) in 116 relapsed patients treated at Ohio State, 70% of whom were fludarabine-refractory [
27]. OR in this high-risk population was 47%. Furthermore, 19/39 del(17p13) patients (49%), 28/47 del(11q22) patients (60%) and 22/52 complex karyotype patients (42%) responded, demonstrating the activity of flavopiridol in poor-risk groups with limited therapeutic options. Forty-one of 85 patients (48%) with bulky lymphadenopathy > 5 cm responded. Median PFS in responders was 10–12 months across all risk groups. Based upon these promising results, a registration study in 165 relapsed CLL patients is ongoing in the United States and Europe.
Several monoclonal anti-CD20 antibodies which have been engineered to improve antibody-dependent cytotoxicity or complement fixation are under clinical investigation. Of this new generation of anti-CD20 antibodies, ofatumumab has generated the most mature clinical data. Ofatumumab (HuMax-CD20) is a fully humanized, high-affinity monoclonal antibody whose epitope on the CD20 molecule of B cells is distinct from that of rituximab. Ofatumumab has higher affinity for CD20 than rituximab, activates complement-dependent cytotoxicity more effectively, and is superior to rituximab in killing B-cell lines with low CD20 expression. An initial phase I/II study in 33 patients with relapsed CLL giving weekly therapy for 4 week showed a 50% OR [
28]. At the 2008 ASH meeting, Osterberg et al. presented a pivotal phase II study of ofatumumab in relapsed patients refractory to both fludarabine and alemtuzumab (DR, n = 59) or with bulky lymphadenopathy refractory to fludarabine (BFR, n = 79) [
29]. OR, time to next therapy, and OS were similar for the DR (51%, 9.0 months, 13.7 months) and BFR groups (44%, 7.9 months, 15.4 months). Based on these results, ofatumumab has been submitted for FDA approval.
Investigational Agents
Several exciting novel therapeutic agents are under study, although a complete description of these drugs is beyond the scope of this review. ABT-263, a small molecule inhibitor of Bcl-2, has shown clinical activity as a single agent in CLL, and studies of this drug in combination with other agents are ongoing. Based on the prior results with flavopiridol, several new CDK inhibitors are under clinical study. The MDACC has studied SNS-032, an inhibitor of CDK 2, 7 and 9. SCH 727965 is a CDK inhibitor which appears to have a better therapeutic window than flavopiridol, and this drug is under study in relapsed CLL as well as non-Hodgkin's lymphoma and multiple myeloma. The CD37 small molecule inhibitor (SMIP), Tru-016, is under investigation in relapsed CLL, and initial results of this study will be presented at the 2009 ASCO meeting. Similarly, preclinical results of an ongoing phase I study of CAL-101, an orally available inhibitor of the phosphatidylinositol-3-kinase (PI-3-K) delta isoform, will be presented at the ASCO meeting. Other novel agents are under preclinical and clinical study, but space precludes a discussion of these agents in their review.