CAR T-cell therapy in secondary CNS lymphomas: A real-world study from the descar-t registry

Abstract Introduction Lymphomas with central nervous system (CNS) involvement were excluded from major clinical trials on anti-CD19 CAR T-cells due to concerns about potential unacceptable neurotoxicity. Recently, several studies have shown promising results of this therapy in primary or secondary CNS lymphomas, with high response rates and neurotoxicity rates similar to those observed in systemic lymphomas. However, these studies included few patients, had short follow-up periods, and the medium-term efficacy data varied greatly between studies. The aim of this study was to assess the efficacy and toxicity of CAR T-cell therapy in a large real-world cohort of secondary CNS lymphomas and to compare the results with those of a cohort of systemic lymphomas. Methods Since 2019, the DESCAR-T registry (NCT04328298) has included patients with hematologic malignancies who had an indication for CAR T-cell therapy established by the multidisciplinary tumor board (MTB) of a French CAR T-cell accredited center. We selected from this registry patients with an initial diagnosis of B-cell lymphoma (excluding Burkitt and primary CNS lymphomas) and with CNS involvement at the time of the MTB or at the time of lymphodepletion (with or without systemic involvement) who received CAR T-cells. Patients who received CAR T-cells as part of a clinical trial or patients who received another treatment out of progression after CAR T-cell infusion were excluded. We compared this population with patients from the registry who were treated with CAR T-cells for systemic diffuse large B-cell lymphoma (DLBCL) without CNS involvement, using a propensity score (PS) method. Results The CNS cohort included 195 patients with a median age of 65 years (min-max: 24-82), the majority of whom were males (66%) and had low ECOG performance status (65% with ECOG 0–1) at the time of CAR T-cell infusion. At the time of the MTB, the histological diagnosis was mainly DLBCL (77%), with few transformed indolent lymphomas (13%) and other high-grade B-cell lymphomas (10%). 47% of patients had received at least 2 prior lines of therapy and 90% of patients received bridging therapy. At the time of CAR T-cell infusion, 109/195 (56%) patients were in complete (CR) or partial response (PR), and 84 (43%) had stable (SD) or progressive disease (PD) (missing data: 1%). One hundred thirty-eight (71%) patients received axi-cel, 26 (13%) tisa-cel, and 31 (16%) liso-cel. The median follow-up after CAR T-cell infusion was 18 months (95% CI: 13-19). The best response after CAR T-cell infusion was CR in 120/181 (66%) patients and PR in 21 (12%) patients. Progression-free survival (PFS) rates at 12 and 18 months were 46% (95% CI: 37-54) and 37% (95% CI: 27-47), respectively. Only six relapses occurred beyond 12 months after CAR T-cell therapy. Overall survival (OS) rates at 12 and 18 months were 66% (95% CI: 57-73) and 53% (95% CI: 42-63), respectively. A responder tumor status (CR/PR) before CAR T-cell infusion and female sex were significantly associated with prolonged PFS and OS in multivariate analysis. 153/189 (81%) patients experienced cytokine release syndrome (CRS), including 9 (5%) with grade ≥3 and 82/189 (43%) patients developed neurotoxicity, of which 37 (20%) were grade ≥3. The comparison between the CNS cohort and the systemic cohort was based on 1614 patients (138 CNS vs 1476 systemic lymphomas) with complete data. After PS-weighting, there was no significant imbalance between the two cohorts, and no significant difference was observed (CNS vs systemic) in terms of CRS incidence (84% vs 87%, p=0.43), neurotoxicity incidence (45% vs 43%, p=0.68), including grade ≥3 neurotoxicity (16% vs 13%, p=0.28), and overall response rate (77% vs 82%, p=0.20). PFS (p=0.13) and OS (p=0.28) were also not significantly different between the two populations. Conclusion This study, which is the largest real-world cohort of patients, confirms the effectiveness of CAR T-cell therapy in secondary CNS lymphomas. The medium-term efficacy results were encouraging, with a 12-month PFS close to 50% and a low relapse rate beyond 12 months. Efficacy and toxicity data, particularly with respect to neurotoxicity, were consistent with those observed in systemic DLBCL. Further studies are needed to support these results over the long term and to determine the optimal place of CAR T-cells within the therapeutic options of CNS lymphomas.