The Differences in the Safety and Tolerability of Immune Checkpoint Inhibitors as Treatment for Non-Small Cell Lung Cancer and Melanoma: Network Meta-Analysis and Systematic Review

Abstract

Background: Immune checkpoint inhibitors (ICIs) have evolved for the treatment of solid tumors. In addition to the efficacy of ICIs for cancer, the adverse events (AEs) of ICIs are also noteworthy for gradually more extensive clinical use. Objective: To conduct a systematic review and network meta-analysis to evaluate the treatment-related AEs that occurred in clinical trials using different kinds of ICIs, to explore the differences in AEs among ICIs for treating non-small cell lung cancer (NSCLC) and melanoma, and to compare select immune-related AEs. Methods: PubMed, EMBASE, Cochrane Library, ClinicalTrials.gov, and other available sources were systematically searched for published reports up to January 1, 2019. Two reviewers independently selected reports about phase II/III randomized controlled trials to compare among ICIs and between ICIs and chemotherapy. After the bias assessment of all included trials, a Bayesian network meta-analysis was performed. The primary outcomes were any-grade and high-grade treatment-related AEs from all ICIs. The secondary outcomes were AEs in patients with NSCLC and melanoma and the presence of the select AEs pneumonitis/pneumonia and colitis. Results: Eighteen randomized controlled trials containing 11,223 patients with NSCLC or melanoma were included. A total network meta-analysis was conducted. The meta-analysis showed that atezolizumab 1,200 mg and pembrolizumab 2 mg/kg every 3 weeks were generally more tolerable than other ICIs. ICI combined with chemotherapy might suggest a higher risk of treatment-related AEs than monotherapy with a single ICI, except durvalumab and ipilimumab. In the NSCLC subgroup, pembrolizumab was associated with a higher risk of high-grade AEs than nivolumab. In addition, ICIs (nivolumab, atezolizumab, and avelumab) led to a lower risk of any/high-grade treatment-related AEs than traditional chemotherapy and ICI combination chemotherapy. However, ICIs did not present preferable safety and tolerability compared to chemotherapy in treating melanoma. Compared with chemotherapy, nivolumab, durvalumab, two ICIs, and ICI combined chemotherapy led to more pneumonitis/pneumonia. However, when treating NSCLC, different types of ICIs did not differ significantly regarding the incidence of pneumonitis/pneumonia. A combination of nivolumab and ipilimumab had the highest risk for colitis, while pembrolizumab and atezolizumab had a lower possibility than the other ICIs. Conclusion: Atezolizumab 1,200 mg and pembrolizumab 2 mg/kg every 3 weeks were ordinarily safer than other ICIs. When treating NSCLC, nivolumab had the lowest risk; when treating melanoma, pembrolizumab had the lowest toxicity.

Keywords: immune checkpoint inhibitors; melanoma; network meta-analysis; non–small cell lung cancer; treatment-related adverse events.