Six versus eight doses of rituximab in combination with six cycles of CHOP chemotherapy in patients with previously untreated diffuse large b-cell lymphoma

Abstract

Abstract BACKGROUND Six cycles of R-CHOP (rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone) are the standard first-line treatment for diffuse large-B-cell lymphoma (DLBCL). Past studies showed that adding rituximab to CHOP not only improved survival outcome but also reduced central nervous system (CNS) relapse (Boehme et al. Blood. 2009, Villa et al. Ann Oncol. 2010). However, the optimal dose of rituximab has not been fully investigated. Results from a subset analysis from the PETAL trial (Hüttmann et al. Ann Hematol 2019) showed no superiority of six cycles of R-CHOP plus two doses of rituximab (R8-CHOP6) over six cycles of R-CHOP (R6-CHOP6) in event free survival and overall survival (OS), although the population analyzed in the study was limited to patients who were sensitive to first two courses of chemotherapy. To our knowledge, there have been no prior reports addressing the effect of increasing rituximab on the incidence of CNS relapse. We aim to compare progression free survival (PFS), OS, and cumulative incidence of CNS relapse between patients receiving R6-CHOP and R8-CHOP in an observational cohort of patients with newly diagnosed CD20-positive DLBCL. METHODS We performed a retrospective cohort review of patients with newly diagnosed CD20-positive DLBCL receiving R6-CHOP6 or R8-CHOP6 as first-line therapy at two institutions in Japan between May, 2006 and July, 2022. Patients with CNS involvement at the time of diagnosis were excluded from the study cohort. PFS and OS were estimated with the use of the Kaplan-Meier method starting from the date of diagnosis. Differences between groups were assessed by log-rank test. A Cox proportional-hazards regression model that included IPI was used to perform an adjusted analysis. To evaluate the heterogeneity of additional rituximab effect on OS, subgroup analyses according to variables based on disease characteristics at baseline were done. A Cox model was used for each subgroup. The presence of interaction was tested by including an interaction term between the treatment effect and the subgroup covariate of interest. The cumulative incidence of CNS relapse was estimated using the method proposed by Gray. Death without CNS relapse was defined as a competing event and comparisons between groups were done using the Fine and Gray model, adjusted for CNS IPI. RESULTS A total of 453 patients were analyzed, of whom 294 patients received R6-CHOP6 and 159 patients received R8-CHOP6. Median age was 70 years (range, 19-93). The median follow-up time was 52 months (range, 5-198). Baseline patient characteristics in the treatment groups slightly differed in Eastern Cooperative Oncology Group (ECOG) performance-status score ≥ 2 (R6-CHOP6 vs. R8-CHOP6; 26.1% vs. 15.1%) and LDH greater than upper limit of normal (R6-CHOP6 vs. R8-CHOP6; 65.6% vs. 50.3%). The complete response rates were similar between the groups (R6-CHOP6 vs. R8-CHOP6; 91.8% vs. 89.8%). Five-year PFS was 68.6% (95% CI, 62.1-74.2) in the R6-CHOP6 group and 67.8% (95% CI, 59.7-74.6) in the R8-CHOP6 group (HR, 0.96; 95% CI, 0.68-1.36; p=0.838, IPI-adjusted HR, 1.05; 95% CI, 0.74-1.50; p=0.506). Five-year OS was 78.6% (95% CI, 72.3-83.6) in the R6-CHOP6 group and 81.1% (95% CI, 73.8-86.6) in the R8-CHOP6 group (HR, 0.92; 95% CI, 0.90-1.41; p=0.705, IPI-adjusted HR, 1.00; 95% CI, 0.65-1.54; p=0.994). Five-year cumulative incidence of CNS relapse was 7.5% (95% CI, 4.8-11.8) in the R6-CHOP6 group and 6.6% (95% CI, 3.6-12.1) in the R8-CHOP6 group (HR, 0.79; 95% CI, 0.37-1.71; p=0.533, CNS-IPI-adjusted HR, 0.95; 95% CI, 0.44-2.08; p=0.902). R8-CHOP6 was not associated with better OS than R6-CHOP6 in subgroup analyses stratified according to key covariates, including high-risk features. CONCLUSIONS Our populatiion-based cohort study showed that additional two rituximab applications after six cycles of R-CHOP did not improve outcome including CNS relapse. Six cycles of rituximab seem to be optimal in the frontline treatment combined with CHOP for patients with DLBCL.