PB1787 OUTCOMES OF PATIENTS WITH PRIMARY MEDIASTINAL B‐CELL LYMPHOMA AFTER UPFRONT R‐DAEPOCH ARE COMPARABLE TO R‐VACOPB REGIMEN: A SINGLE CENTER EXPERIENCE WITH RETROSPECTIVE HISTORICAL COHORT COMPARISON.

Background: Primary mediastinal B‐cell lymphoma (PMBCL) is a unique entity of non‐Hodgkin lymphoma with distinct clinical and biological features. Several different chemoimmunotherapy approaches have been studied, mostly in retrospective case series, without reaching a consensus on the optimal upfront treatment. A phase 2 study, published by Dunleavy et. al. (NEJM 368, 1408‐1416, 2013), demonstrated excellent survival outcomes after R‐DAEPOCH without consolidative radiotherapy in 51 patients with PMBCL. Based on these promising data we adopted R‐DAEPOCH regimen as an upfront therapy in newly diagnosed PMBCL patients. Aims: The aim of the study was to evaluate the outcomes and toxicity of newly diagnosed PMBCL patients treated with R‐DAEPOCH vs R‐VACOPB, an ambulatory dose‐dense third generation regimen previously used in our center. Methods: We thoroughly reviewed the electronic medical records of newly diagnosed PMBCL patients treated with R‐DAEPOCH in our center between 1.2016 and 12.2017 in comparison with prior historical cohort of patients treated with R‐VACOPB between 3.2003 and 12.2009. We documented demographics, clinical characteristics and toxicity and analyzed the survival outcomes. Comparison between categorical variables was done using Chi‐square test, survival analysis with Kaplan‐Meier method and comparison between survivals with Log‐Rank test. Results: Forty nine patients were included in the study. Eighteen patients were treated with R‐DAEPOCH and thirty one patients were treated with R‐VACOPB. Median follow‐up was two years (range 0.3‐4.4) in the R‐DAEPOCH group and 10.2 years (range 0.4‐14.4) in the R‐VACOPB group. There were no statistically significant differences between the two groups in terms of gender, stage I/II at diagnosis, B symptoms, ECOG performance status, LDH levels, aaIPI category, bone marrow involvement, disease bulk, SVC syndrome, pericardial effusion, pleural effusion, and extra‐nodal involvement. Median age was 32 years (range 23‐43) in the R‐DAEPOCH and 31 years (range 17‐48) in the R‐VACOPB group. Neutropenic fever (n = 36) was observed in 56% of the R‐DAEPOCH patients compared with 20% in R‐VACOPB (p = 0.036). Most neutropenic fever events in the R‐DAEPOCH arm occurred after the fifth cycle (range 3‐6). Median last escalation was in the third cycle (range 1‐5). None of the baseline patient characteristics predicted PFS or OS. No statistically significant differences were found in PFS and OS between the two groups. Median PFS was not reached in both groups (p = 0.31). Two year PFS was 76% in the R‐DAEPOCH and 86% in the R‐VACOPB group. Median OS was not reached in both groups. Summary/Conclusion: With patient and disease characteristics comparable to historical cohort, outcomes of patients treated with R‐DAEPOCH did not differ in terms of PFS and OS from patients treated with R‐VACOPB. R‐DAEPOCH was associated with higher toxicity. These data may challenge the concept of routinely using R‐DAEPOCH in PMBCL.