Very long-term follow-up of rituximab maintenance in young patients with mantle cell lymphoma included in the LYMA trial, a LYSA study.

7508 Background: The phase III LYMA trial demonstrated the efficacy (OS, PFS, EFS) and safety of rituximab maintenance (RM) post autologous stem cell transplant (ASCT) in first line for young patients with mantle cell lymphoma (MCL) (Le Gouill et al, NEJM). Herein, we present the first long-term analysis of the LYMA trial. Methods: 299 patients were included in LYMA trial of whom 240 were randomized to receive 3 years of RM (n = 120) or no RM (observation; n = 120). Primary endpoint was 3 years EFS. We updated the outcome of all patients enrolled in this trial. Treatment at relapse and cause of death were collected. Results: At the time of the present analysis, the median follow-up is 7 years (0.2-10). The median EFS and PFS were still statistically superior in favor of RM (not reached (NR), versus 5.8 years, p < .0001 for EFS and NR versus 6.1 years for PFS in RM and observation arm respectively). The 7 years estimated EFS and PFS were 76.2% (95% CI 67.4-82.9%) versus 46% (95% CI 36.6-54.9%) and 78.5% (95% CI 69.9-85.0%) versus 47.4% (95% CI 37.9-56.3%), for RM and observation respectively, p < .0001). In the RM arm, 22 patients had died (18.3%) versus 33 (27.5%) in the observation arm with a 7-year OS estimate of 83.2% (95% CI: 74.7.7%-89.0%) and 72.2% (95% CI 62.9%-79.5%) in RM and observation arm, respectively (p = 0.087). Causes of death were not significantly distinct between the 2 groups, lymphoma being the leading cause in both (50% in both arm), with less than 10% of infectious related death. The end of RM was not associated with an increase in the relapse incidence, since in the RM arm, 17/21 (81%) of the progression occurred during the theoretical 3 years of RM and only 19% after the end of RM. On the contrary, in the observation arm, 32/53 (60%) of the progression/relapses occurred during the theoretical 3 years of maintenance part, and 40% after. Salvage treatments were not different between the RM and observation arm, including anti CD20 and BTK inhibitors administered in 75% versus 79% and 10% versus 15% of the relapsing patients in the RM and observation arms, respectively. Post relapse OS was significantly impacted by previous RM, with a median OS-2 of 1.1 years (95%CI 0.7-2.1) for the 21 relapsing patients in the RM arm versus 4.6 years (95%CI 2-NA) for the 53 relapsing patients in the observation arm. Given the timing of relapses in RM arm, this reflects the aggressiveness of early disease progression during RM. Conclusions: The benefice in favor of RM in term of EFS (the primary objective), PFS remains after 7 years of FU, but not in OS. The end of RM was not associated with an increase in relapse rate. RM was not associated with an increase in infectious related mortality, making of this strategy a safe standard of care with long term FU. Patients relapsing during RM have a particularly poor outcome and represent an unmet medical need. Clinical trial information: NCT00921414 .