P1151: bendamustine-containing regimens can be safely used as bridging in large b-cell lymphoma patients receiving chimeric antigen receptor t-cell therapy

Topic: 19. Aggressive Non-Hodgkin lymphoma - Clinical Background: Bridging treatment (BT) during Chimeric Antigen Receptor (CAR) T-cell manufacturing for patients with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) is a key step to guarantee disease control and CAR-T efficacy. Across countries, BT regimens are heterogeneous and dependent on patient status and reimbursement policies. Bendamustine-containing regimens are a potential option for BT but, given the prolonged lymphotoxic effect of bendamustine (benda), concerns have been raised regarding its impact on endogenous lymphocytes which could play a role in the efficacy of the CAR T-cell infusion. Aims: To evaluate the impact of benda-containing regimens as BT in CAR T-cell therapy recipients not previously exposed to bendamustine. Methods: We conducted a retrospective, international study including patients with R/R LBCL treated at 7 sites with commercial CAR T-cells until September 2022. We analyzed baseline characteristics, response rates and survival outcomes depending on the bridging strategy, focusing on benda-containing regimens versus other systemic combinations. Patients who had received benda before apheresis were excluded. Analysis were carried out separately according to CAR-T construct. Results: Eighty of the 439 patients with LBCL receiving CAR T-cell therapy in our cohort had been exposed to benda before apheresis and were excluded. Of the 359 remaining patients, 295 (82%) received BT after apheresis. Regarding BT-recipients, 25 (9%) received radiotherapy, 27 (9%) were managed with steroids and 243 (82%) received chemotherapy or targeted agents. Among the latter, 62 (26%) patients received a benda-containing regimen while 181 (74%) received a systemic approach based on different agents (gemcitabine [49%], cyclophosphamide [22%], etoposide [17%], lenalidomide [4%] or other [8%]). Baseline characteristics were balanced between benda- and non-benda BT groups except for a trend towards a higher proportion of axicabtagene ciloleucel (axi-cel) in the benda-BT group (p=0.054). Taking this into account, together with the different toxicity and efficacy profile of each construct, we analyzed patients separately according to infused product. Regarding the axi-cel cohort, 139 patients received a systemic BT of which 42 (30%) included benda. We did not identify significant differences between baseline characteristics in the benda and no-benda BT groups. We observed a similar toxicity profile between benda and no-benda BT recipients, including CRS any grade (90% vs 96%), CRS grade ≥3 (5% vs 4%), ICANS any grade (52% vs 56%), ICANS grade ≥3 (24% vs 42%) and non-relapse mortality (2% vs 6%). In terms of efficacy, complete response rates to CAR T-cell therapy in the benda and no-benda cohorts were 70% and 54%, respectively (p=0.13). Median PFS and OS were also similar in the benda and no-benda cohorts (23.5 mo vs 6.6 mo [p=0.33] and 29.9 mo vs 23.6 mo [p=0.38]). In the multivariate analysis, benda BT did not show a significant impact on PFS nor OS. Peak CAR T-cell expansion after infusion and Area Under the Curve between day of infusion and day 28 (AUC0-28) were also similar between the benda and no-benda BT groups (59 vs 50 cells/μL [p=0.96] and 790 vs 555 [p=0.63]). Concerning the tisagenlecleucel (tisa-cel) cohort, 104 patients received a systemic BT of which 20 (19%) included benda. We did not identify differences in the baseline characteristics between the benda and no-benda BT groups. In terms of safety and efficacy, outcomes were similar for both subgroups (Table 1). Summary/Conclusion: Bridging treatment with benda-containing regimens seems to provide similar safety and efficacy outcomes after CAR-T cell therapy than alternative systemic approaches and can be safely used in this context.Keywords: Lymphoid malignancy, CAR-T