Background: By prolonging overall survival and reducing disease recurrence rates, immune checkpoint inhibitors (ICIs) are an emerging adjuvant therapy option for patients with resectable malignant tumors. However, the safety profile (deaths and adverse events [AEs]) of adjuvant ICIs has not been fully described.Methods: We searched the literature for phase III randomized clinical trials that compared PD-1, PD-L1, and CTLA-4 inhibitors in solid malignant tumors. Incidences of death, discontinuation, AEs of any cause, treatment-related adverse events (TRAEs), and immune-related adverse events (IRAEs) were extracted for the network meta-analysis. Network meta-analyses with low incidence and poor convergence are reported as incidences with 95% confidence intervals (95% CIs).Results: Ten randomized clinical trials that included 9243 patients who received ICI adjuvant therapy were eligible. In total, 21 deaths due to TRAEs were recorded, with an overall incidence of 0.40% (95% CI: 0.26-0.61). The treatment-related mortality rates for ipilimumab (0.76%, 95% CI: 0.31-1.55) and atezolizumab (0.56%, 95% CI: 0.18-1.31) were higher than for pembrolizumab (0.24%, 95% CI: 0.10-0.56) and nivolumab (0.30%, 95% CI: 0.08-0.77). The most frequent causes of death were associated with the gastrointestinal (0.10%, 95% CI: 0.04-0.24) and pulmonary (0.08%, 95% CI: 0.03-0.21) systems. Compared with the control arm, we found that nivolumab (odds ratio [OR]: 2.73, 95% CI: 0.49-15.85) and atezolizumab (OR: 12.43, 95% CI: 2.42-78.48) caused the fewest grade ≥3 TRAEs and IRAEs. Commonly reported IRAEs of special interest were analyzed, and two agents were found to have IRAEs with incidences >10%, i.e.,