Commentary on Abstracts #3025, #3502, #2519, and #1447

Rituximab as a single agent induces responses in about 32% of patients with aggressive B-cell NHL, although these responses tended to be partial and brief (Coiffier et al: Blood 92:1927-1932, 1998). Nevertheless, because of its favorable toxicity profile, the antibody was rapidly incorporated into combination chemotherapy regimens, most often with CHOP, the standard program for this histology. Vose and coworkers (J Clin Oncol 19:389-397, 2001) published the results of a multicenter phase II trial in which the antibody was administered on day 1 of each 21-day cycle, with CHOP beginning on day 3, in contrast to the schedule reported by Czuczman et al (J Clin Oncol 17:268-276, 1999).

Patients primarily had diffuse large B-cell NHL, but other histologies were seen as well. Of 33 patients evaluable for response, the overall response rate was 98%, including 61% complete remissions, which is better than would be expected from CHOP alone (Fisher et al: N Engl J Med 328:1002-1006, 1993). With a median follow-up of 24 months, the median response duration was 18+ months, with no evidence of progression in the complete responders or in seven of the partial responders. Thirteen patients were bcl-2 positive prior to therapy, and 11 of those became negative. There was no additional toxicity associated with the antibody aside from the expected infusion reactions.

In the first phase III trial, Coiffier et al of the GELA (Group d’Etude des Lymphomes de l’Adulte) group (abstract #3025) randomized 399 patients between the ages of 60 and 80 years with diffuse large B-cell NHL to either CHOP or CHOP plus rituximab (R-CHOP). Rituximab was delivered on day 1 of each of the eight cycles. In the final analysis, the CR/CRu (unconfirmed CR) rate (Cheson et al: Blood 96:3671-3674, 2000), the 24-month projected progression-free survival, and the overall survival were longer with the combined modality approach. However, at 3 years, the difference between the two curves is only about 10% and decreasing. Therefore, longer follow-up is clearly required to determine if the differences will persist.

The addition of rituximab did not add to the immediate toxicity of CHOP, with the exception of grade 3/4 antibody-related infusion reactions in 9% of patients. However, there were more late deaths (9 vs 2) in the R-CHOP arm from cardiac disease, infection, and cachexia. The relationship between these events and therapy is not clear. It is important to note that none of the patients who failed CHOP were treated with rituximab because it was not yet available for aggressive B-cell lymphomas in the GELA countries. Therefore, it is not clear whether R-CHOP as initial therapy would be superior to CHOP followed by a rituximab-containing regimen at the time of relapse. The increased cure rate with the addition of rituximab was also cost-effective (abstract #3586, Coiffier et al).

A US Intergroup trial recently completed randomization of more than 630 patients onto a similar comparison of CHOP vs R-CHOP in patients over the age of 60 years, but also with a secondary randomization following response to observation or further antibody therapy. Hopefully, the results of the US study will confirm the French observations, leading to a change in our approach to these patients. An ongoing international trial is currently accruing younger patients with diffuse large B-cell NHL to CHOP or R-CHOP; however, publication of the GELA data may, unfortunately, compromise accrual to that important study (Coiffier et al: N Engl J Med 346:235-242, 2002).

A few notes of caution are warranted. First, these R-CHOP data should not be extrapolated to other histologies, such as follicular and low-grade NHL, until the results of randomized studies in those biologically different diseases are available. Second, it may not be whether R-CHOP is better, but R-CHOP that becomes the next question, because the schedules of administration of the antibody and chemotherapy vary among studies and may affect outcome (Czuczman et al: J Clin Oncol 17:268-276, 1999; abstract #3502; Vose et al: J Clin Oncol 19:389-397, 2001; Coiffier et al: N Engl J Med 346:235-242, 2002). Czuczman et al (J Clin Oncol 17:268-276, 1999; abstract #2519) first published a syncopated schedule, while Vose delivered the antibody on day 1 with CHOP on day 3 for six courses. SWOG treated patients with six cycles of CHOP followed by rituximab (abstract #3502), and the GELA group gave the antibody on day 1 of each of eight cycles. In the US Intergroup study, six to eight cycles of CHOP were administered, with a secondary randomization to antibody or observation.

Promising results have also been reported with a variety of other chemotherapy/rituximab combinations in aggressive NHL. Wilson et al (abstract #1447) added the antibody to their promising EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin HCl) infusional chemotherapy regimen (Wilson et al: J Clin Oncol 11:1573-1582, 1993). In the low-risk International Prognostic Index (IPI) patients with diffuse large B-cell NHL, the complete remission rate was 100%, with a response rate of 71% in the poor IPI risk groups. The regimen was extremely well tolerated. In the investigators’ previous experience with dose-adjusted EPOCH without rituximab, they made the observation that the outcome of patients whose tumor cells were found to overexpress bcl-2 had a significantly inferior outcome, presumably reflecting the multidrug resistance resulting from that oncogene. Surprisingly, in abstract #1447, the addition of rituximab abrogated the effect of bcl-2 overexpression and the outcomes of the two groups were comparable.

If all of the subsequent phase III trials comparing chemotherapy with or without rituximab are consistent, then it will be clear that rituximab enhances the efficacy of chemotherapy and that a shift in the therapeutic paradigm is warranted. However, if there is discrepancy among the outcomes of the various studies, it will not be clear if the differences are related to the specific drugs, their schedules of administration, the patient population, or other factors. Clinical research will then be handcuffed to years spent comparing one chemotherapy/rituximab regimen with another instead of building on those results to develop more effective strategies.