CELL INTELLIGENCE · 肿瘤细胞治疗研究

BACKGROUND: Chimeric antigen receptor T (CAR-T) cell therapy provides breakthrough efficacy for relapsed/refractory hematological malignancies, yet the role of cytokine release syndrome (CRS) as an efficacy predictor remains controversial and lacks large-scale validation. This pooled-analysis aimed to systematically evaluate the incidence of CAR-T-related adverse events (TRAEs) and their correlation with therapeutic efficacy. METHODS: We searched four major databases up to December 10, 2025, and included 222 prospective trials enrolling 7538 patients. Pooled analyses were performed using a random effects model, and Pearson correlation was used to assess the association between TRAE incidences and key efficacy endpoints. RESULTS: The pooled rates of all grade and grade 3 CRS were 78.07% and 7.67%, with objective response rate (ORR) 83.26% and complete response rate (CRR) 67.46%. CRS only showed a weak correlation with short term remission, while specific TRAEs had moderate to strong correlations with efficacy (r=0.6099-0.9677, P<0.05). CONCLUSION: CAR-T therapy exhibits remarkable efficacy and manageable toxicity in hematological malignancies. CRS is not a reliable efficacy marker; grade 3 dyspnea/hypotension/tremor and all grade encephalopathy/fatigue are potential early efficacy predictors.