Abstract
INTRODUCTION: Cadonilimab (AK104) is a bispecific antibody that simultaneously targets programmed cell death-1 and cytotoxic T-lymphocyte antigen-4. It has received approval for the treatment of cervical cancer and gastric/gastroesophageal junction cancer. This meta-analysis aims to assess cadonilimab's safety profile. METHODS: A systematic review of electronic databases was conducted to identify clinical trials that reported cadonilimab's safety data. Immune-related adverse events (irAEs) was the primary endpoint, and treatment-related adverse events (TRAEs) were the secondary endpoints. A single-group proportion meta-analysis was conducted by R software. RESULTS: A total of 1271 patients across more than five cancer types in 11 clinical trials were included in this study. The incidence of any grade irAEs, grade ≥ 3 irAEs, irAEs leading to treatment discontinuation, and irAEs associated with mortality was 43.3% [95% confidence interval (CI), 33.3%-53.4%], 11.3% (95% CI, 9.5%-13.3%), 3.7% (95% CI, 1.5%-6.5%), and 0% (95% CI, 0%-0.4%), respectively. Hypothyroidism was the most common all-grade irAEs (13.3%, 95% CI, 8.9%-18.5%). The incidence of TRAEs was higher in the combined therapy group compared to the cadonilimab monotherapy group. CONCLUSIONS: The irAEs associated with cadonilimab are generally manageable. When combining with other anticancer agents, physicians and pharmacists should be particularly aware of the potential increase in TRAEs.