Validation of POD24 As a Robust Early Clinical End Point of Poor Survival in Mantle Cell Lymphoma from 1280 Patients on Clinical Trials

Abstract

Background The prognosis of mantle cell lymphoma (MCL) has largely improved in the past decade; however, the disease is characterized by a heterogeneous clinical course. Several retrospective studies identified early progression of disease (i.e. within two years, POD24) as a potential overall survival (OS) surrogate, but this has not been validated in cohorts of patients prospectively included in clinical trials in rituximab maintenance era. Methods We performed a pooled analysis of French patients with MCL included in six randomized clinical trials (EU-MCL younger NCT00209222, LyMA NCT00921414, LyMA101 NCT02896582, EU-MCL elderly NCT00209209, MCL-R2 NCT01865110 and RiBVD NCT01457144). Survival analysis using Landmark approach evaluated the association of POD24 status with post-event OS for all patients: starting from POD24 event or two years for patients without POD24 event. Logistic regression models were used to evaluate the association between POD24 status and (1) clinico-biological factors at diagnosis; (2) autologous stem cell transplantation at end-of-induction (ASCT) and anti-CD20 maintenance (RM) in responding patients after induction only. Results Among 1386 MCL patients, 106 censored for clinical follow-up before 24 months were excluded from the analysis, leading to 1280 patients evaluable for POD24 status: 299 with a POD24 event and 981 without. The 299 (23%) patients with a POD24 event had a post-event median OS of 9.3 months (95% CI 8.4-11.8) versus not reached in patients without POD24 event (95% CI 97.8-NR). Within the 981 non-POD24 patients, 314 presented a late relapse with a post-relapse median OS of 49.4 months (95% CI 30.4-56.8), significantly longer than OS of POD24 patients (HR=0.39; 95%CI 0.31-0.48; p<0.001). Compared to patients without a POD24 event, POD24 patients were older, had more frequently a performance status >1, elevated LDH and higher leucocytes leading to higher MIPI scores (high-risk MIPI 61% vs. 29%; p<0.001), more frequent blastoid variant (24% vs. 9%; p<0.001) and Ki67 > 30% (45% vs. 23%, p<0.001). Regarding treatment, POD24 patients had less frequently received high-dose cytarabine (21% vs. 39%, p<0.001) as well as ASCT (26% vs. 47%, p<0.001). In a final model, including baseline factors and treatment strategies, induction was not associated with risk of POD24, and only baseline variables (age, performance status, LDH, leucocytes and Ki67>30%) remained significantly associated with POD24 status. Within responding patients only (CR/CRu/PR at end-of-induction, n=1000), 150 had a POD24 event, and ASCT or RM were not significantly associated with POD24 status whereas age, LDH and Ki67>30% remained significant. Conclusion Using this large dataset of patients included in clinical trials, we confirm that POD24 can be used as a surrogate for OS in MCL. ASCT as well as RM have not a clear benefit to prevent early relapse within two years after the diagnosis in responding patients at end-of-induction.