Treatment-Related Adverse Events of Chimeric Antigen Receptor T-Cell (CAR T) in Clinical Trials: A Systematic Review and Meta-Analysis

Simple Summary Successful treatment of hematological malignancies with chimeric antigen receptors T (CAR-T) cells has led to much enthusiasm for the wide clinical usage and development of novel CAR-T therapies. However, it also challenges physicians and investigators to recognize and deal with treatment-associated toxicities. We conducted a systematic review and meta-analysis from 84 eligible study and a total of 2592 patients to identify the comprehensive incidences and severity of CRS and neurological symptoms (NS) as well as the potential differences in AEs across a variety of cancer types, CAR-T targets, and other factors, thereby offering a significant implication on its future application and research. Abstract Chimeric antigen receptors T (CAR-T) cell therapy of cancer is a rapidly evolving field. It has been shown to be remarkably effective in cases of hematological malignancies, and its approval by the FDA has significantly increased the enthusiasm for wide clinical usage and development of novel CAR-T therapies. However, it has also challenged physicians and investigators to recognize and deal with treatment-associated toxicities. A total of 2592 patients were included from 84 eligible studies that were systematically searched and reviewed from the databases of PubMed, de, the American Society of Hematology and the Cochrane Library. The meta-analysis and subgroup analysis by a Bayesian logistic regression model were used to evaluate the incidences of therapy-related toxicities such as cytokine release syndrome (CRS) and neurological symptoms (NS), and the differences between different targets and cancer types were analyzed. The pooled all-grade CRS rate and grade ≥ 3 CRS rate was 77% and 29%, respectively, with a significantly higher incidence in the hematologic malignancies (all-grade: 81%; grade ≥ 3: 29%) than in solid tumors (all-grade: 37%; grade ≥ 3: 19%). The pooled estimate NS rate from the individual studies were 40% for all-grade and 28% for grade ≥ 3. It was also higher in the hematologic subgroup than in the solid tumors group. The subgroup analysis by cancer type showed that higher incidences of grade ≥ 3 CRS were observed in anti-CD19 CAR-T therapy for ALL and NHL, anti-BCMA CAR-T for MM, and anti-CEA CAR-T for solid tumors, which were between 24–36%, while higher incidences of grade ≥ 3 NS were mainly observed in CD19-ALL/NHL (23–37%) and BCMA-MM (12%). Importantly, subgroup analysis on anti-CD19 CAR-T studies showed that young patients (vs. adult patients), allologous T cell origin (vs. autologous origin), gamma retrovirus vector, and higher doses of CAR-T cells were associated with high-grade CRS. On the other hand, the patients with NHL (vs ALL), administered with higher dose of CAR-T, and adult patients (vs. young patients) had an increased incidence of grade ≥ 3 NS events. This study offers a comprehensive summary of treatment-related toxicity and will guide future clinical trials and therapeutic designs investigating CAR T cell therapy.