Outcomes of CD20xCD3 bispecific antibody therapy as holding or bridging therapy prior to chimeric antigen receptor T-cell infusion in aggressive B-cell lymphomas

Abstract

Abstract Introduction CD19-directed chimeric antigen receptor T-cell (CAR-T) therapy is approved as second-line treatment for aggressive B-cell lymphoma (aBCL), while CD20xCD3 bispecific antibodies (BsAbs) are approved in the third-line setting. Although targeting distinct B-cell antigens, both strategies redirect cytotoxic T cells via different mechanisms: CAR-T uses autologous engineered T cells, while BsAbs rely on antibody-based T-cell engagement (TCE). As both modalities become more accessible, optimal sequencing remains unclear. Outcomes of patients receiving BsAbs as bridging (BT) or holding therapy (HT) prior to intended CAR-T have not been well described. Methods We retrospectively identified adult patients with aBCL who received BsAb as standard of care at Moffitt Cancer Center between June 1st, 2023, and May 1st, 2025, using our institutional bispecific therapy database. Demographic and clinical variables were abstracted from the electronic medical record. Clinical intent of BsAb use was determined by detailed chart review. Descriptive statistics were used to summarize patient characteristics and clinical outcomes, with a specific focus on those treated with BsAbs as bridging prior to CAR-T. Result Out of 81 patients treated with BsAb for aBCL, 37 patients (46%) had no prior CAR-T. Among these, 16 patients (43%) received BsAb as BT/HT prior to intended CAR-T; 8 (22%) due to aggressive disease with frequent hospitalizations precluding CAR-T consultation; 8 (22%) by patient preference; 4 (11%) due to comorbidities; and 1 (3%) due to CAR-T manufacture failure. Among the 16 patients who received BsAb as BT/HT prior to CAR-T, all but one completed step-up dosing. Median age was 71 years (range, 22–86); 63% were female and 94% were White. Central nervous system (CNS) involvement was present in 4 (15%), double/triple hit lymphoma in 4 (15%), and transformed disease in 3 (19%). Median prior lines of therapy was 2 (range, 1–5). Eastern Cooperative Oncology Group (ECOG) performance status was ≥2 in 7 of 13 patients with available data (54%). Glofitamab was used in 12 patients (75%) and epcoritamab in 4 (25%). Of the 16 patients, three (18%) died before CAR-T due to rapid disease progression, and 2 (13%) declined CAR-T due to ongoing BiAbs response. Eleven (69%) proceeded to CAR-T with axicabtagene ciloleucel (n=8), lisocabtagene maraleucel (n=2), or brexucabtagene autoleucel (n=1). Median number of BsAb cycles prior to CAR-T was 2 (range, 1–4). One patient with mantle cell lymphoma was unevaluable for response. Among the 10 evaluable patients, 1 patient achieved a complete response (CR) and 5 (50%) achieved a partial response (PR) to BsAb and maintained ongoing complete response after CAR-T at a median follow-up of 8 months (range, 2–14). Of the 4 (40%) patients with progressive disease on BsAb, 2 progressed after CAR-T and 2 had a short-lived response lasting around 3 months. CAR-T related toxicities included: cytokine release syndrome (CRS) occurred in 9 (82%), all grade 1–2, and immune effector cell-associated neurotoxicity syndrome (ICANS) occurred in 7 (64%), including grade 1–2 in 2 (18%) and grade 3–4 in 5 (45%). Conclusion In this real-world cohort, BsAb as a BT/HT prior to CAR-T was feasible and associated with meaningful activity. Response to BiAbs was associated with ongoing CR to CAR-T at a median follow-up of 8 months. Higher grade ICANS may be related to higher tumor burden, immune priming from BiAbs or CNS lymphoma involvement. This suggests that BsAb responsiveness may reflect an immunologically permissive tumor microenvironment predictive of efficacy, irrespective of the TCE platform. These findings support further study of BsAb bridging as a functional biomarker of CAR-T responsiveness.