R-DHA-Oxaliplatin before Autologous Stem Cell Transplantation Prolongs PFS and OS As Compared to R-DHA-Carboplatin and R-DHA-Cisplatin in Patients with Mantle Cell Lymphoma, a Subgroup Analysis of the LyMa Trial

BACKGROUND. The LyMA (ClinicalTrials.gov, NCT00921414) trial is a prospective international randomized phase III trial that investigated Rituximab maintenance (RM) after autologous-stem cell transplantation (ASCT) in young previously untreated mantle cell lymphoma (MCL) patients (Le Gouill et al., in press, 2017). Patients were included at diagnosis (<66y; stage >I) and induction immuno-chemotherapy consisted of 4 courses of R-DHAP-21 (Rituximab, Dexamethasone, High-dose cytarabine, Platinum salt) followed by ASCT consolidation. The conditioning regimen for ASCT was R-BEAM. Patients in response after ASCT were randomized (1:1) between RM or observation. 299 patients (<66y) were included of whom 240 were randomly assigned to RM or observation after ASCT. Median follow-up from randomization after ASCT was 50.2 (46.4-54.2) months. The final analysis demonstrates that RM is superior to observation with respect of EFS, PFS and OS. For induction with R-DHAP, local investigators were free to choose between cisplatin (continuous infusion cisplatin 100 mg/m² over 24 h on day 1; DHA-Cis), carboplatin (with AUC = 5 mg/ml.min ; target Area Under the concentration vs. time Curve using the Calvert formula ; DHA-Ca) or oxaliplatin (130 mg/m² on day 1 ; DHA-Ox).

METHODS. In the present analysis, we investigated the prognostic impact on PFS and OS of the nature of platinum salt used in the R-DHAP regimen. The intention-to-treat (ITT) analysis includes all patients who have received at least one course of R-DHAP (n=298; data missing in one patient) and per protocol analysis (PP) includes all patients who have received the same platinum salt during induction phase (n=227).

RESULTS. Cisplatin, Carboplatin and oxaliplatin were used at course 1 in 184, 76 and 38 patients, respectively. Patients' characteristics at diagnosis (including MIPI) and randomization arm (RM or observation) were well balanced between the 3 groups. Twenty-seven (15%) and 38 (21%) out of 184 patients who received cisplatin from course 1 switched to carboplatin and oxaliplatin because of toxicity, respectively. In contrast, only one patient with DHA-Ca switched to DHACis and one patient with R-DHAOx switched to R-DHACa. All cases of renal failure during induction (n=8) occurred in the R-DHACis group. PFS and OS in both ITT and PP were identical in the R-DHACis and R-DHACa groups while there was a trend in favor of R-DHAOx in both PFS (p=0.092 in ITT and 0.0675 in PP) and OS (p= 0.013 in ITT and p= 0.0859 in PP). Because patients' outcome was identical with R-DHACa and R-DHACis, both groups were combined in one group (R-DHACa/DHACis) and then compared to R-DHAOx. In ITT, the 4y PFS and OS in the R-DHAOx vs R-DHACis/Ca were 86.5% (95%CI; 70.5 to 94.1) vs 65% (95%CI; 58.8 to 70.6) and 92% (95%CI; 77.1 to 97.3) vs 75.9% (95%CI; 70.1 to 80.7), respectively. In ITT, patients with R-DHAOx have a better PFS (HR=0.438; 95%CI, 0.204-0.944; p=0.035) and OS (HR=0.373; 95%CI, 0.136-1.023; p=0.0554) as compared to patients with R-DHACa/DHACis . Similar results were reached in the PP analysis: PFS (HR=0.414; 95%CI, 0.191-0.899; p=0.0257) and OS (HR=0.347; 95%CI, 0.126-0.959; p=0.0412). In multivariate analysis (with MIPI), R-DHAOx was independent prognostic marker for PFS and OS.

Conclusion. The present analysis shows for the first time that R-DHAOx is a better induction regimen than R-DHACa and R-DHACis in MCL patients and strongly suggests that platinium salt are critical (and not only high dose AraC) in the R-DHAP regimen. Our finding may also have a major impact in other forms of lymphoma where R-DHAP is frequently used as a standard salvage therapy.