The incidence of cytokine release syndrome and neurotoxicity of CD19 chimeric antigen receptor-T cell therapy in the patient with acute lymphoblastic leukemia and lymphoma.

Our objective was to summarize the side effect of chimeric antigen receptor (CAR)-T cell therapy in patients with acute lymphocytic leukemia (ALL) and lymphoma. Two independent reviewers extracted relevant data. A total of 35 hematologic malignancy studies with CD19 CAR-T cell were included (1412 participants). Severe cytokine release syndrome (sCRS) proportion was experienced by 18.5% (95% confidence interval [CI], 0.128-0.259; P = 0.000) of 982 patients with the National Cancer Institute/Lee/common terminology criteria for adverse events grading system. The pooled neurotoxicity proportion was 21.7% (95% CI, 0.167-0.287; P = 0.000) of 747 patients with the same grading system. For all of the 25 clinical trials with the same grading system, subgroup analysis was performed. Based on the different disease type, a pooled prevalence of 35.7% was observed with event rate (ER) of 0.358 (95% CI, 0.289-0.434; P = 0.000) for ALL in 12 clinical trials. For lymphoma, a pooled prevalence of 13% was observed with ER of 0.073 (95% CI, 0.028-0.179; P = 0.000) in eight clinical trials. It was demonstrated that the patients who were older than 18 years of age have the lower sCRS incidence of 16.1% (95% CI, 0.110-0.250; P = 0.000) compared with 28.6% of the remaining population who were younger than 18 years of age (95% CI, 0.117-0.462: P = 0.023) in our analysis. Based on the different co-stimulatory domain, the sCRS of 16.5% was observed with ER of 0.175 (95% CI, 0.090-0.312; P = 0.000) for 4-1BB. The sCRS of 22.2% was observed with ER of 0.193 (95% CI, 0.107-0.322; P = 0.000) for CD28. For both the CD28 and 4-1BB, the sCRS of 17.3% was observed with ER of 0.170 (95% CI, 0.067-0.369; P = 0.003). Sub-analysis sCRS of the impact with cell dose and specific disease indication were also demonstrated. Limitations include heterogeneity of study populations, as well as high risk of bias of included studies. These results are helpful for physicians, patients and the other stakeholders to understand the adverse events and to further promote the improvement of CAR-T cell therapy in the future.