Follicular Lymphoma, Survival, and Rituximab: Is It Time to Declare Victory?

Seemingly an unattainable goal in the 20th century, prolongation of survival has long been the Holy Grail of clinical investigation in follicular lymphoma. In fact, single institutional data over a 30-year time period from Stanford University were frequently cited as evidence for the lack of progress. All that has changed. Reports of change in follicular lymphoma survival first appeared in the form of retrospective analyses in Journal of Clinical Oncology. Swenson et al assessed U.S. population data, citing an improvement in overall survival of 1.8% per year among patients diagnosed from 1983 to 1999. They speculated that the advance was due to the sequential application of effective therapies and improved supportive care. Improvements in survival were reported in each of three eras of Southwest Oncology Group therapeutic trials conducted from 1974 to 2000, andthehighest4-yearoverall survivalswereachievedwithchemotherapy plus anti-CD20 antibody. The authors attributed the 10% gain in survival for patients treated from 1988 to 1994—the prerituximab era—to effective sequential treatment options, because there was no advantage in progression-free survival after initial therapy in this era. Similarly, significant gains in overall survival in stage IV follicular lymphoma were reported in a retrospective analysis from the M. D. Anderson Cancer Center during a 25-year period. In this report, the efficacy of both initial and subsequent treatments was credited for the progress. Recently, new data from Stanford University demonstrated longer overall survivals in patients diagnosed after 1986. Again, this progress was attributed to effective sequential treatments because no prolongation in the time to progression after initial therapy was observed. In this issue of JCO, Marcus et al describe a survival benefit for patients with advanced stage follicular lymphoma requiring therapy, based on their initial treatment. In a randomized controlled trial, 4-year overall survival rates were 83% for rituximab, cyclophosphamide, vincristine, and prednisone (R-CVP) and 77% for CVP (P .03). These data are concordant with two German phase III trials in which rituximab plus chemotherapy was tested against chemotherapy alone for remission induction. Although the German trials also incorporated interferon or high-dose therapy with autologous transplantation as consolidation, the described survival benefits are of similar magnitude. Collectively, do these results lead to the conclusion that primary therapy including rituximab prolongs overall survival and the corollary that rituximab as initial treatment is superior to the sequential use of rituximab at a later time in the disease course? Let’s review the data more carefully. It should be noted that each of the three cited European trials had early primary end points relative to the conventional progression-free survival end point. Patients who did not achieve partial response after four treatment cycles were considered treatment failures in the Marcus study. Patients achieving less than partial remission went off-study after four cycles of chemotherapy in the trial of Hiddemann et al, 7 and patients achieving less than partial remission after six cycles of chemotherapy were scored as treatment failures in the trial of Herold et al. Theoretically, survival bias could be introduced in favor of the chemotherapy-only arms if patients immediately crossed over to rituximab or initiated another highly effective second-line treatment, whereas the bias could be unfavorable in the chemotherapyonly arms if patients were managed conservatively with observation or less effective treatments. To address the question whether the initial use of rituximab confers a survival advantage over its sequential use, information on the type and timing of second-line therapy is needed. Although rituximab was approved for relapsed follicular lymphoma in the United States in 1997, it may not have been fully available for second-line use in combination in all the countries participating in the R-CVP versus CVP until after 2004, according to Marcus et al. This lack of accessibility may explain why just one third of patients treated for relapse after CVP received rituximab as secondary therapy. Unfortunately, as such, the results from this trial cannot fully address the question of sequence. The authors state that the survival differences—76% of patients treated with rituximab at first relapse compared to 64% of patients treated otherwise—did not achieve statistical significance; but this is a sizeable disparity, and there was little power to observe a significant survival difference based on the sample size. The enormity of the difficulty in demonstrating a survival difference in follicular lymphoma with a true cross-over design has been recently demonstrated by Ladetto et al. In this Italian multicenter study, highly unfavorable follicular lymphoma patients were randomly assigned to cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) plus rituximab or a high-dose sequential chemotherapy with rituximab regimen with autologous transplantation. Patients experiencing disease progression after CHOP plus rituximab crossed-over to the high-dose arm. Despite a large difference in failure-free survival (61% v 28% at 4 years; P .001) in favor of high-dose treatment, no survival difference was observed (P .7), leading the authors to conclude that the more intensive approach was better used as second-line treatment, based on the toxicity profile.