R-High Dose Cytarabine/Dexamethasone (R-HAD) Plus Bortezomib Is Superior to R-HAD Only in Relapsed Mantle Cell Lymphoma: A Randomized Phase 3 Trial of the European MCL Network

Background: Relapsed mantle cell lymphoma (MCL) is characterized by a mostly aggressive clinical course achieving only short term remission after immunochemotherapy. In first line, the addition of high dose cytarabine resulted in significantly prolonged time to treatment failure (TTF) in younger patients (Hermine et al, Lancet 2016), whereas in older patients the addition of bortezomib to a CHOP-like regimen led to superior PFS and overall survival (OR) rates after a prolonged follow-up (Robak, Lancet Oncology 2018). In the recently closed R-HAD trial we investigated efficacy and safety of the addition of bortezomib to a R-HAD regimen previously investigated in elderly patients (Weigert, Leukemia Lymphoma 2009).

Methods: This randomized, open-label phase 3 trial recruited in Germany and France. Patients with relapsed or refractory MCL with 1-3 prior lines not appropriate for myeloablative treatment were randomized between four 3-weekly cycles of R-HAD (Rituximab 375 mg/m2 day 1), high dose Cytarabine (2 g/m2 day 2+3), Dexamethasone (40 mg/m2 days 1-4) and the same regimen combined with Bortezomib (1.5 mg/m2 s.c. day 1 and 4). In patients >65 years the cytarabine dose was reduced to 2x 1g/m2. Patients not achieving at least a PR after 2 cycles were able to discontinue treatment at the discretion of the local physician. Randomization was stratified according to response to therapy (relapse vs. refractory), IPI, previous autologous stem cell transplantation (SCT), prior exposure to cytarabine and country. The primary trial endpoint, time to treatment failure (stable disease, progression, death from any cause) from randomization, was monitored statistically by a two-sided truncated sequential probability ratio test for the log-rank statistic using a 5% significance level. To achieve 95% power to detect a hazard ratio of 0.55 and with a truncation at 160 events, the observation of 78 events was anticipated to be needed with a maximum patient recruitment of 275 patients in up to 5.5 years. Response rates, duration of remission, overall survival, and toxicities were key secondary outcomes.

Results: From May 2012 to December 2016 a total of 128 patients were included before the recruitment was stopped prematurely due to the registration of BTK inhibitors. Median patient age was 70 years (range 41-85), 76% were male, 82% with advanced stage III/IV, and 38% and 43% with intermediate or high risk MIPI, respectively. 71% of patients had 1 prior treatment, 36% a prior cytarabine exposure and 40% a prior autologous SCT. After a median follow-up of 41 months, median TTF was 12 months in the R-HAD + Bortezomib arm and 2.6 months in the control arm (p=0.045, HR 0.68; corrected for sequential design, underrunning analysis). Accordingly, OR (63% vs 45%; p=0.049) and CR/CRu rates (42% vs 19%; p=0.006) differed significantly in favor of the Bortezomib arm, whereas no significant effect on duration of remission (median 21 months vs. 14 months, p=0.32) and overall survival (Median 31 months vs 35 months; p=0.83) was observed. Number of serious adverse events were in favor of the experimental arm (34 vs 50) with hematological toxicities (3 vs 11), Infections (9 in both arms), and fever (6 vs 3) being the most frequent side effects.

Conclusions: In this randomized trial the addition of Bortezomib to R-HAD achieved a prolonged time to treatment failure mainly driven by higher response rates. Long-term effects and relevant differences in survival or toxicities were not observed.