Review Article: Taming the Neuroinflammatory Cost of CAR T Therapy

Chimeric antigen receptor (CAR) T cell therapy has transformed the treatment of refractory haematological malignancies, yet its success remains limited by cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome. Although earlier intervention with tocilizumab and corticosteroids has reduced the incidence of severe toxicity, these strategies act after inflammatory signaling is already established and do not reliably prevent isolated or delayed neurological complications. This has shifted the field from reactive management towards prevention at the level of the cellular product itself. Manufacturing conditions, cellular compositions and differentiation state can shape the inflammatory setpoint of CAR T cells, but attempts to preserve less-differentiated, highly proliferative populations may also increase expansion-driven toxicity. Construct engineering therefore aims not simply to weaken CAR activity, but to recalibrate antigen recognition, signaling strength and cytokine output while preserving durable tumor control. Multimodal reversible CAR engineering approaches via drug-gated receptors, safety switches, transient surface shielding, and targeted CAR degradation, further recast CAR T cells as tuneable therapeutic systems rather than fixed products. However, most evidence remains preclinical or derived from small, heterogeneous studies. Progress will depend on prospective trials that evaluate efficacy and neurotoxicity together, using harmonised product analytics, neurological endpoints and longitudinal biomarker sampling to define clinically meaningful safety/activity trade-offs.