We read with great interest the article by Byrd et al1 comparing, in a phase 2 randomized study, sequential versus concurrent fludarabine plus rituximab in untreated B-cell chronic lymphocytic leukemia (B-CLL). The sequential scheme consisted of 6 monthly courses of fludarabine (25 mg/m2 per day for 5 days) followed by rituximab for 4 weeks (375 mg/m2 per week) in responding patients. The concurrent scheme consisted of the same dose of fludarabine associated on day 1 of each course to rituximab (375 mg/m2 per day). In the first cycle, rituximab was administered on days 1 and 4. Patients with stable disease, at least, were treated with weekly rituximab for 4 doses (375 mg/m2 per week). Interestingly, the authors found that the complete remission (CR) rate improved with concurrent combination (47% versus 28%). The toxicity profile was similar in the 2 patient groups, with more infusion-related symptoms in the concurrent schedule.
In their discussion, the authors proposed the hypothesis that rituximab induced a modulation of antiapoptotic proteins (XIAP and Mcl-1) during the concurrent scheme, thus explaining its greater activity.
The low efficacy of rituximab as a single agent in B-CLL reported in the initial study2 was mainly attributed to both low expression of CD20 on leukemic cells and low drug concentration in serum. A third potential mechanism of resistance to rituximab could be the presence of circulating CD20 molecules interfering with rituximab activity.3
Based on these considerations, 2 recent studies4,5 testing higher doses of rituximab as a single agent in pretreated and untreated B-CLL patients disclosed a dose-response effect.
The protocol applied in the Byrd et al study,1 however, provided different doses of rituximab in the 2 treatment arms: in the sequential group it was administered at a total dose of 1500 mg/m2, whereas in the concurrent group a total dose of 4125 mg/m2 was administered. The authors did not discuss this point, which could partially explain the results obtained with these 2 different schedule treatments. In other words, the higher dose of rituximab, other than a synergistic mechanism, could be hypothesized in the concurrent arm. If this hypothesis is correct, a dose-finding study of rituximab associated to a fixed dose of fludarabine could provide useful data in this interesting field.
Response: Rituximab and fludarabine combination therapy for CLL: potential future directions of clinical investigation
Dr Castagna et al in their letter to the editor identify another important variable that exists between arm 1 and arm 2 of Cancer and Leukemia Group B 9712 (CALGB 9712)1 that could explain the discordant complete response rate observed. The authors suggest that a dose-finding study of rituximab in combination with fixed doses of fludarabine might provide important information for future studies in chronic lymphocytic leukemia (CLL). We are appreciative of the well thought-out comments of Castagna et al and agree that identifying the effective dose and schedule of any agent is of great importance to its successful application in the clinic. Information from other trials, however, suggests that further dose escalation of rituximab will not result in bone marrow clearance. Treatment failure in both arms of CALGB 9712 was most often due to failure to clear residual bone marrow CLL disease. In addition, 2 dose-finding studies of single agent rituximab in previously treated CLL patients2,3 failed to produce bone marrow complete remission (CR), and standard non-Hodgkin lymphoma (NHL) doses of rituximab in untreated CLL patients4 cleared the bone marrow in only 9% of patients. We, therefore, would offer an alternative focus for future investigation with rituximab. Specifically, efforts should be focused on understanding potential factors such as stromal cell interaction that disrupt apoptosis,5 CLL cell density, antibody penetration, and decreased access to complement and effector cells within the bone marrow that might prevent efficient clearance of tumor cells in the majority of patients. Such studies will facilitate development of alternative combination strategies to improve bone marrow responses with this promising combination therapy.
Correspondence: John C. Byrd, Department of Hematology-Oncology, The Ohio State University, B302A Starling Loving Hall, 320 W 10th Ave, Columbus, OH 43210-1240; e-mail: byrd-3@medctr.osu.edu
