PB2010 DETECTING MANTLE CELL LYMPHOMA MINIMAL RESIDUAL DISEASE IN AUTOLOGOUS GRAFT THROUGH NEXT‐GENERATION SEQUENCING AND THE IMPLICATION ON LONG‐TERM REMISSION

Background: Mantle cell lymphoma (MCL) is a non‐Hodgkin lymphoma subtype mostly with advanced stage and extranodal tissue involvement. High‐dose chemotherapy with autologous stem cell rescue (HDC‐SCR) is the current standard of care for eligible patients. Nevertheless, recurrence often intercedes in post‐chemotherapy course, which poses a challenge against a durable remission. Minimal residual disease (MRD) had been proved to be predictive of relapse with worse overall survival (OS). MRD monitoring therefore might change the plan of consolidation therapy including HDC‐SCR. How to improve polymerase chain reaction‐based MRD analysis for MCL are being investigated. Aims: This study employed next‐generation sequencing (NGS) to optimize MRD identification for MCL and correlates its implications on post‐HDC‐SCR outcome. Methods: In this single‐institute retrospective study, 112 patients (male: 90 vs. female: 22) were diagnosed as having MCL between 2006 and 2018. The median follow‐up duration was 62 months. The median age at diagnosis was 62 years old. The Ann Arbor staging at diagnosis was mostly stage 4 (81.1 %), and the blastoid variant comprised of 13.4 %. Using MIPI scores, patients were stratified as low (27 %), intermediate (34 %), and high (39 %) risk groups. In treatment, CHOP‐like induction chemotherapy was given to 75% of this cohort. The number of patients receiving HDC‐SCR and allogenic stem cell transplant (allo‐SCT) were 40 (35.7%) and 6 (5.4%), respectively. In recent years, bortezomib and ibrutinib were added to the treatment armamentarium either in front‐line setting or for salvage therapy, given to 41 (36.6%) and 18 (16%) patients, respectively. Further, LymphoTrackÒ Dx IGH/IGK assay coupled with Illumina MiSeq sequencer is the NGS platform used in this study to analyze V(D)J recombination in the diagnostic formalin‐fixed paraffin‐embedded samples and autografts. Results: The median OS was 95 months (MIPI low/intermediate/high risk: 147, 95, 46 months, respectively, p = 0.002). Patients who were older than 60‐year‐old at diagnosis had poorer outcome (median survival 51 vs. 147 months, p = 0.003). Blastoid variant was associated with worse survival (p < 0.001). Patients harboring complex cytogenetic changes also had graver prognosis (median survival 19 vs. 115 months, p < 0.001). In the multivariate analysis for OS, blastoid variant, higher MIPI scores, and complex cytogenetic change were independent adverse risk factors whereas HDC‐SCR had protective effect. Concerning pre‐HDC‐SCR disease status, patients in complete remission (CR) had better OS than those who achieved only partial remission (p = 0.042). Ten patients undertook NGS‐based MRD analysis for their diagnostic and autologous graft paired samples. Four (40%) patients developed relapse, and all of them did not receive rituximab maintenance therapy. Two of these four patients were positive for NGS‐based MRD in the autologous graft. Interestingly, the other two who were negative for NGS‐based MRD in the autologous graft developed recurrence from their pre‐HDC‐SCR FDG‐PET lesions. Summary/Conclusion: This study supports that HDC‐SCR confers OS benefit. Autologous graft MRD detection by NGS analysis helps prediction of relapse. Meanwhile, for patients without MRD in autologous graft, post‐HDC‐SCR maintenance therapy or consolidative radiotherapy might help curb relapse and warrants further investigation.