Abstract
Objective：Given the poor prognosis of relapsed/refractory (R/R) large B-cell lymphoma (LBCL), we compared the efficacy and safety of sequential CD19/CD22 chimeric antigen receptor T (CAR-T) cell therapy following autologous stem cell transplantation (ASCT) versus CAR-T monotherapy to identify more effective treatment strategies.
Methods：We conducted a single-arm, prospective study to evaluate sequential CD19/CD22 CAR-T cell therapy following ASCT. Concurrently, patients who declined ASCT and received CD19/CD22 CAR-T monotherapy during the same period were included as a contemporaneous control cohort. Group allocation was based solely on patient preference, allowing for a comparative analysis of treatment efficacy and safety between the two approaches (NCT03196830).
Results：Between February 2020 to August 2023, 50 patients were included in the final analysis, comprising 23 in the ASCT+CAR-T group and 27 in the CAR-T alone group. As of April 2025, the median follow-up duration was 28.4 months. In ASCT+CAR-T group, the complete response (CR) rate in the best response evaluation (17/23) was as high as 73.9% (95%CI 51.6%-89.8%). Moreover, the ASCT+CAR-T group demonstrated superior clinical outcomes when compared to CAR-T monotherapy. A total of 49 patients were response-evaluable. The best response results revealed a striking advantage for ASCT+CAR-T group, with adjusted CR rates of 76.7% (95%CI 56.2%-92.4%) versus 23.9% (95%CI 8.3%-42.0%) (p=0.003) and overall response rates (ORR) of 100% (95%CI 100%-100%) versus 61.9% (95%CI 39.5%-84.2%) compared to CAR-T monotherapy (p=0.001). In addition, these robust response benefits also translated into superior long-term outcomes, with significantly improved adjusted 2-year event-free survival (EFS) (50.7%, 95%CI 31.1%-70.3% vs 23.3%, 95%CI 6.4%-40.2%; p=0.014) and adjusted 2-year overall survival (OS) (72.2%, 95%CI 53.0-91.5% vs 33.9%, 95%CI 14.0-53.9%; p=0.008) in the ASCT+CAR-T group. These clinical benefits persisted in sensitivity analyses excluding patients with CNS involvement (n=45), with adjusted 2-year EFS 54.1% (95%CI 33.3%-74.9%) versus 24.5% (95%CI 7.3%-41.8%) in ASCT+CAR-T and CAR-T group, respectively (p=0.013), and 2-year OS 76.9% (95%CI 55.4%-98.3%) versus 35.0% (95%CI 18.4%-51.7%) , respectively (p=0.005).
The most common adverse event (AE) in both groups was hematologic toxicity. All patients experienced grade ≥3 neutropenia, with significantly higher incidence of grade ≥3 thrombocytopenia (p=0.011) and a trend toward increased grade ≥3 anemia (p=0.056) in the ASCT+CAR-T group, which were attributed to the myeloablative conditioning regimen. Notably, all 23 patients in the ASCT+CAR-T group achieved successful hematopoietic reconstitution following transplantation. The median time to neutrophil and platelet engraftment was 9 days (range, 8-13) and 12 days (range, 9-30), respectively. The analysis revealed that the incidence of severe (grade ≥3) cytokine release syndrome (CRS) did not differ significantly between groups (13.0% vs 7.4%, p=0.651). Importantly, no fatal CRS events occurred in either group. The safety profile was further characterized by a single fatal immune effector cell-associated neurotoxicity syndrome (ICANS) case in the CAR-T group, which did not occur in the ASCT+CAR-T group.
Conclusions: Sequential CD19/CD22 CAR-T cell therapy following ASCT represents a clinically significant advancement for R/R LBCL, demonstrating unprecedented survival outcomes with a manageable safety profile. This approach may serve as a potential new standard of care for eligible patients. Further validation through larger, multicenter randomized controlled trials with extended follow-up is warranted to confirm its long-term efficacy and safety.
