Outcomes of patients (Pts) with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) progressing after CD3xCD20 bispecific antibody (BsAb) therapy: A new unmet need

Abstract Introduction BsAb are an emerging off-the-shelf immunotherapy for R/R LBCL. Epcoritamab and glofitamab have received accelerated FDA approval for use in pts with LBCL after ≥2 prior lines of therapy. Despite promising results, approximately 70% of pts treated with BsAb experience progression (PD) within 12 months. Currently, there is no established standard of care for subsequent treatment, and optimal post-BsAb therapy remains unknown. To address this knowledge gap and inform clinical decision-making, we reviewed our experience in pts with R/R LBCL progressing after epcoritamab or glofitamab.Methods Electronic medical records of pts with LBCL treated with commercial epcoritamab or glofitamab at our center from May (epcoritamab) or June (glofitamab) 2023 to July 2025 were reviewed. We excluded pts who completed BsAb without PD or were still receiving BsAb, those who received BsAb as bridging to CAR-T cell therapy (CAR-T) or consolidative stem cell transplantation, or who discontinued BsAb due to toxicity. Responses were assessed per the Lugano 2014 criteria. Progression-free survival (PFS) was calculated from the first day of post-BsAb treatment to PD or death. Overall survival (OS) was calculated with Kaplan-Meier methods starting from documented progression post-BsAb.Results A total of 32 pts with LBCL treated with BsAb experienced PD and represent the study population. The median age at BsAb initiation was 71 years (interquartile range [IQR] 59–75); 72% were older than 60 years and 37% were female. ECOG performance status was 0–1 in 21 pts (65%) and ≥2 in 11 (35%). At the time of BsAb initiation, the median IPI score was 3 (IQR 2–5). Ann Arbor stage III/IV disease was present in 28 (90%) pts; >1 extranodal site in 15 (47%). Cell of origin was non-GCB in 17 (57%) and GCB in 13 (43%), unknown in 2; 2 of 26 evaluable pts (8%) had double-hit lymphoma and 7/32 (22%) had a previous history of indolent lymphoma. Pts had received a median of 3 prior lines of therapy before BsAb (IQR 2–5). Primary refractory disease was documented in 34% of pts. Twenty-one pts (65%) had received CAR-T and, of these, 18/21 (86%) had relapsed within 6 months. Epcoritamab and glofitamab were administered in 20 (63%) and 12 (37%) pts, respectively, with a median treatment duration of 1.5 months (IQR 0.7–3.3). The best overall response rate (ORR) to BsAb was 25%, comprising 2 complete responses (CR) and 6 partial responses (PR). Before BsAb, all 32 pts had CD20 expression. However, following BsAb failure, CD20 was lost in 17 of 21 evaluable pts (81%). Twenty of 32 pts (63%) received subsequent therapy. The remaining 12 (37%) did not receive additional treatment due to deteriorating clinical status or death. Subsequent treatment types included clinical trial (n=5, 25%), chemotherapy (n=5, 25%), loncastuximab tesirine (lonca-T, n=4, 20%), tafasitamab ± lenalidomide (n=2, 10%), first-time CAR-T (n=2, 10%), polatuzumab-based therapy (n=1, 5%), radiation (n=1, 5%), and a second BsAb regimen of glofitamab-GemOx (n=1, 5%). Among 17 evaluable pts, 3 (17%) achieved a CR following lonca-T (n=2) and 1 ater CAR-T. One additional pt achieved a PR following CAR-T, for an ORR of 23%. Median follow-up from post-BsAb progression was 8.8 months (95% CI, 6.7–NE). Among treated pts, median PFS was 1.5 months (95% CI, 0.95–NE). PFS was 0.74 months (95% CI, 0.69–NE) for pts receiving chemotherapy vs. 5.7 months (95% CI, 3.2–NE) for those receiving lonca-T (p = 0.03). Median OS from post-BsAb progression was 3 months (95% CI, 2.0–6.1). In pts who received subsequent therapy, median OS was 6 months (95% CI, 2.3–NE), compared to 0.75 months in those who did not. OS rate at 6 months post-BsAb progression was 22% (95% CI, 11–39).Conclusion We provide the first early report of post-BsAb outcomes in pts with R/R LBCL. The prognosis of this population is dismal, with only two-thirds of the pts able to receive further treatment. Among non-CAR-T treatments, clinical responses were observed only with lonca-T. The rate of CD20 loss post-BsAb (81%) was significantly higher than previously reported, emphasizing the need for CD20-independent strategies. Pts with R/R LBCL after BsAb therapy represent a new, urgent unmet medical need. Given our limited sample size, these findings warrant validation in a larger study.