Efficacy of Chimeric Antigen Receptor T-Cell Therapy Is Not Impaired By Previous Bispecific Antibody Treatment in Patients with Large B-Cell Lymphoma

Abstract

Introduction: Potential T-cell exhaustion after bispecific antibody (BsAb) treatment remains an open question, raising the theoretical concern that prior BsAb exposure could affect subsequent chimeric antigen receptor (CAR) T-cell efficacy. Clinical data on CAR T-cell outcomes after prior BsAb treatment in the setting of large B-cell lymphoma (LBCL) are scarce and highly awaited to better define treatment sequencing in relapsed/refractory (R/R) patients. Methods: We conducted a retrospective, international study including R/R LBCL patients treated with CD19-targeted CAR T-cells at 15 centers between July 2018 and January 2023 who had been exposed to BsAbs prior to apheresis. Then, we identified a control cohort from patients included in the DESCAR-T Registry (n=764). We carried out a 1:1 propensity score matching (PSM) to achieve balance between cohorts; 13 baseline covariates were included in the PSM. We compared response rates, survival outcomes and toxicity after CAR T-cell therapy, according to previous BsAb exposure. Results: We identified 47 LBCL patients who received BsAb therapy prior to CAR T-cell apheresis. Median age was 65 years (range 31-82), with a male predominance (66%), and median prior lines of therapy before BsAb were 2 (IQR 2-3) (Table). The BsAbs targeted CD20/CD3 (91%) or CD22/CD3 (9%), either in monotherapy (n=41, 87%) or in combination with other agents (n=6, 13%) (lenalidomide [n=2], polatuzumab [n=1], chemotherapy [n=2], missing [n=1]). The median time on BsAb treatment was 98 days (IQR 56-160) with a best overall (complete) response rate (ORR [CRR]) of 47% (19%). Median progression-free survival (PFS) and duration of response (DOR) were 3.1 months (95% CI 2.7-4.4 months) and 8.8 months (95% CI 2.2-NR), respectively. Cytokine release syndrome (CRS) occurred in 59% of patients (grade 3 in 4%). No neurologic toxicity (NT) was reported. In 26 (55%) patients, bispecific antibody was the last line before CAR T-cell therapy. Median washout between the last BsAb dose and lymphocyte apheresis for CAR T-cell manufacturing was 43 days (range 34-103 days). In terms of the subsequent CAR T-cell therapy, patients received axi-cel (n=22, 47%), tisa-cel (n=20, 43%) or liso-cel (n=5, 11%). Most patients developed CRS after infusion (79%, 6% grade >2), with a low rate of NT (23%, 2% grade >2). Neutropenia and thrombocytopenia grade >2 were reported in 66% and 45% of patients, respectively. Regarding efficacy, ORR after infusion was 83% (CR 43%). Median PFS and overall survival (OS) were 6.6 months (95% CI 2.6-not reached [NR]) and NR (95% CI 9.0-NR), respectively. The rate of overall (complete) response was similar between patients who had previously responded (CR or PR) or not (SD or PD) to BsAb therapy (82% [41%] vs 84% [44%], respectively, p=0.64). However, all 9 patients achieving CR after BsAb achieved subsequent response to CAR T-cells. Given the wide variability in the washout between last BsAb dose and lymphocyte apheresis, we evaluated the impact of this interval on CAR T-cell efficacy. The 6-month PFS and OS of patients previously exposed to BsAb within 45 days of lymphocyte apheresis was similar to patients who had a longer washout (59% vs 47% [p=0.25] and 83% vs 67% [p=0.21], respectively). In the second part of the analysis, we matched our cohort with a control group of CAR T-cell recipients not previously exposed to BsAbs. The final analysis included 2 sets of 42 patients with comparable characteristics. Median follow-up from CAR T-cell infusion was 11.5 months for the BsAb cohort (CI 95% 6.4-19.0 months) and 18.8 months for the control cohort (CI 95% 10.9-23.0 months). Patients with previous BsAb exposure showed a similar PFS (p=0.10, Figure) and OS (p=0.08) after CAR T-cell infusion in comparison to the control group. In terms of toxicity, there were no differences in the incidence of CRS (p=0.41) or NT (p=1.0) after infusion. Conclusions: Previous exposure to bispecific antibody treatment targeting different antigens does not have a negative impact on survival outcomes after CAR T-cell therapy. Lack of response to a previous BsAb does not predict for lower response rates after CAR T-cell infusion.