CAR-T cell therapy for treatment-refractory rheumatic autoimmune diseases: a systematic review of clinical outcomes and safety profiles

Objectives Current therapeutic paradigms for treatment-refractory rheumatic diseases rely on chronic immunosuppression with variable efficacy and potential for cumulative toxicity. We systematically synthesised evidence on the clinical efficacy, safety and durability of chimeric antigen receptor T-cell (CAR-T) therapy in autoimmune rheumatic diseases. Methods We conducted a systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (PROSPERO: CRD42025641602). Five databases were searched from inception to February 2025 for primary studies investigating CAR-T therapy in autoimmune rheumatic diseases. Two reviewers independently assessed studies using the Joanna Briggs Institute tools and the Risk of Bias in Non-Randomised Studies of Interventions tool. Results 12 studies encompassed 44 patients with severe treatment-refractory disease across six rheumatic conditions. Patients had extensive prior exposure (median 5 therapies), including conventional and biological agents. Cluster of Differentiation antigen 19 (CD19)-targeted constructs predominated (83% of studies), with emerging dual-target approaches (17%). Universal clinical responses were achieved (100% of studies; individual response rates 92%–100%), with complete responses occurring within 2–16 weeks. Sustained drug-free remissions extended 6–46 months, accompanied by profound autoantibody reductions (80%–99% across disease-specific markers). Safety profiles diverged markedly from oncological experience: cytokine release syndrome (CRS) remained exclusively confined to grades 1–2, while immune effector cell-associated neurotoxicity occurred in 25% of studies, with predominantly grade 1 severity. Conclusions CAR-T therapy demonstrates substantial efficacy in treatment-refractory rheumatic diseases, achieving sustained remissions through comprehensive B-cell elimination. Importantly, safety profiles were favourable, with CRS limited to grades 1–2 severity, no grade ≥3 events and rare immune effector cell-associated neurotoxicity, representing markedly superior tolerability compared with oncological applications. PROSPERO registration number CRD42025641602.