Outcomes of patients with Relapsed/Refractory aggressive B-cell lymphoma receiving bispecific antibody therapy as monotherapy or in combination with targeted therapy

Abstract

Abstract Introduction Bispecific antibodies (BsAbs) have added to the treatment landscape for relapsed/refractory (R/R) large B-cell lymphoma (LBCL), achieving complete response (CR) rates of ~ 40% in pivotal trials as monotherapy. BsAbs in combination with other agents that target distinct components of malignant B-cell biology are being explored to improve response and survival. We present real-world data comparing clinical characteristics and outcomes for BsAb monotherapy (B-M) vs BsAbs in combination with targeted therapy (B+T) in patients (pts) with R/R LBCL. Methods We developed a multicenter cohort of pts with R/R LBCL treated with B-M vs B+T across 15 US institutions from 2022-2025, excluding those treated on clinical trials. Differences in clinical and treatment characteristics were investigated using Pearson's Chi-squared test, Fisher's exact test, and Wilcoxon rank sum test. Progression-free survival (PFS) and overall survival (OS) were estimated using Kaplan-Meier curves. Variables of clinical significance in univariate analyses (UVA) were included in multivariable regression analyses (MVA) using Cox Proportional Hazards Models to examine impact on PFS and OS. Results A total of 157 pts received BsAbs: 137 (87%) received B-M and 20 (13%) received B+T. For B+T pts, 9 (45%) were treated with lenalidomide, 6 (30%) with polatuzumab, 2 (10%) with a BTK inhibitor, and 3 (15%) with other targeted agents. For B-M vs B+T cohorts: there was no difference in median age (66 vs 66 years, p=0.8), male sex (62 vs 56%, p=0.6), Caucasian race (74 vs 70%, p=0.5), de novo DLBCL (68 vs 55%) or transformed follicular lymphoma (15 vs 15%; p=0.3), double hit status (DHL, 16 vs 6%, p=0.5), primary refractory disease with frontline therapy (PRD, 36 vs 45%, p=0.3), prior bendamustine (17 vs 21%, p=0.7), or prior CART (60 vs 80%, p=0.08). At time of BsAb initiation, rates of elevated LDH (63 vs 65%), extranodal disease (76 vs 89%), and bulky disease (36 vs 25%) were similar for each cohort (p ≥0.3). The B+T cohort included more pts with progression/relapse within 6 months of CART infusion (75 vs 44%, p=0.009). For B-M vs B+T: median time from diagnosis to BsAb infusion was 20 vs 13 mo (p=0.03) and median line of therapy was 4 for both (p=0.5). Glofitamab was more commonly used for B-M (n=83, 61%) and just as often as epcoritamab for B+T (n=8 for each, 40%; p=0.2). Median cycles of BsAb therapy were 4 in each cohort (p=0.8). Rates of any grade or ≥ grade 3 cytokine release syndrome and neurotoxicity were comparable across both groups (p>0.9), with no toxicity related deaths. For B-M vs B+T: median follow-up was 10.8 vs 17.1 mo. CR rates were 37 vs 31% (p=0.14). Although relapse rates (53 vs 65%, p=0.3) were similar, median duration of CR was not reached (7.4, NR) vs 3.2 mo (2.3, NR; p=0.009). Median PFS (mPFS) was 3.4 vs 4.9 mo (p=0.7), and median OS (mOS) was 10.1 vs 10.4 mo (p=0.6). From time of B-M vs B+T relapse, with next line therapy, mPFS was 1.6 vs 3.3 mo (p=0.4) and mOS was 2.6 vs 5.1 mo (p=0.6). On UVA, age >60, age >80, PRD, DHL, prior bendamustine or CART exposure, BsAb treatment within 3 mo of CART, LDH, bulky disease, and absolute lymphocyte count (ALC) ≤0.3 at time of BsAb did not impact PFS or OS in either B-M or B+T cohorts (p≥0.2). On MVA, choice of B-M vs B+T had no impact on PFS (p=0.6) or OS (p=0.4). ALC ≤0.3 negatively impacted PFS [HR=2.75, (95% CI 1.38-5.48), p=0.006] but not OS (p=0.2). Elevated LDH at time of BsAb was independently associated with worse PFS [HR=3.36, (95% CI 1.64-6.86), p<0.001] and OS [HR=3.39, (95% CI 1.37-8.43), p=0.005]. PRD and BsAb within 3 mo of CART trended towards negative impact on PFS (p<0.07) but not OS (p=0.2). Age >60 years, DHL, and bulky disease had no impact on PFS and OS in pts treated with BsAb +/- T. ConclusionsIn real world practice, there is a tendency to consider B+T rather than B-M in pts with quicker relapses after CART. B-M vs B+T may yield similar response rates and survival. However, the optimal combination partner for BsAbs remains unclear and requires further study. Low ALC and elevated LDH negatively impacted PFS with BsAb +/- T suggesting a need to balance disease burden and lymphotoxicity of treatments prior to the use of BsAbs. Elevated LDH also negatively impacted OS with BsAb +/- T and may serve as a key prognostic marker for treatment with BsAbs. In pts who relapse post-BsAb +/- T, survival is dismal underscoring the need for novel treatments in this population.