Efficacy and safety of chimeric antigen receptor T-cell therapy in primary central nervous system lymphoma: a systematic review and meta-analysis.

BACKGROUND: Evidence for chimeric antigen receptor T-cell (CAR-T) therapy in primary central nervous system lymphoma (PCNSL) remains limited and heterogeneous, with prior syntheses often combining primary and secondary CNS lymphoma. METHODS: We systematically searched PubMed, Scopus, and Web of Science through February 2026. The protocol was not registered in PROSPERO. Twenty-three reports were included: 15 studies in the quantitative synthesis (194 patients) and 8 reports in the qualitative synthesis (9 patients). RESULTS: The pooled overall response rate was 68% (95% CI, 59-76; I2=17%), including a complete response rate of 57% (95% CI, 49-65; I2=4%) and a partial response rate of 16% (95% CI, 11-22; I2=0%). Pooled overall survival rates were 79% at 6 months and 60% at 12 months; progression-free survival rates were 52% and 42%, respectively. Safety outcomes showed pooled rates of 72% for any-grade cytokine release syndrome (CRS) and 12% for grade ≥3 CRS, while any-grade and grade ≥3 immune effector cell-associated neurotoxicity syndrome (ICANS) occurred in 46% and 16%, respectively. Treatment-related mortality was 5% (95% CI, 2-15; I2= 0%). Sensitivity analyses yielded comparable results. CONCLUSIONS: CAR-T therapy demonstrates promising efficacy with manageable toxicity in PCNSL. Although early disease control is encouraging, long-term durability remains limited.