# Comparative Risks of Pneumonitis Amongst Immune Checkpoint Inhibitors in Patients with Lung Cancer: A Network Meta-Analysis of Randomized Clinical Trials

## TL;DR

This study compares the risk of lung inflammation (pneumonitis) among different immune checkpoint inhibitors used in lung cancer treatment.

## Contribution

The study provides a network meta-analysis comparing pneumonitis risks across multiple immune checkpoint inhibitors in lung cancer patients.

## Key findings

- Pembrolizumab and sugemalimab showed significantly higher pneumonitis risk compared to placebo.
- Nivolumab had increased pneumonitis risk but with unstable estimates.
- Combination therapies were associated with higher pneumonitis rates.

## Abstract

Background/Objectives: Despite immune checkpoint inhibitors (ICPIs)’s transformation of lung cancer treatment, pneumonitis remains a potentially serious immune-related adverse event. However, reliable data on the comparative risks of individual ICPIs remain unknown. We conducted this network meta-analysis (NMA) to, therefore, quantify and compare the exact pooled burden of pneumonitis risk across multiple ICPI analogs. Methods: We searched the following databases, PubMed, Embase, Scopus MEDLINE and Cochrane Database of Systematic Reviews, as well as gray literature on Google Scholar for eligible studies reporting on the prevalence of pneumonitis following immune check point inhibitor exposures. Pairwise and network meta-analyses were performed to estimate pooled odds ratios (ORs) for pneumonitis, using placebo as the common comparator. Sensitivity analyses assessed the impact of study quality and combination therapies. Results: A total of 29 studies enrolling 15,271 patients with non-small cell lung cancer (NSCLC) or small-cell lung cancer (SCLC) satisfied the inclusion criteria and are included in the meta-analysis. Pembrolizumab was associated with a significantly increased risk of pneumonitis compared to placebo (odds ratio [OR] = 2.67, 95% confidence interval [CI]: 1.70–4.17), with similar elevated risk observed for sugemalimab (odds ratio [OR] = 2.45, 95% confidence interval [CI]: 1.52–3.95). Nivolumab was associated with increased odds of pneumonitis, although with unstable point estimate (odds Ratio [OR] = 2.69, 95% confidence interval [CI]: 0.64–11.35). Statistical heterogeneity was low (H statistics = 1.34). Atezolizumab and ipilimumab demonstrated modest or uncertain risk. Heterogeneity was low (I2 = 12%), and results were robust to sensitivity analyses. Higher pneumonitis rates were observed in combination regimens. Conclusions: Our analysis demonstrates that pneumonitis risk varies among ICPIs, with pembrolizumab and sugemalimab showing the highest odds. Although the absolute incidence is low, the potential severity of pneumonitis warrants vigilant monitoring. These results should guide clinicians in risk stratification and treatment planning, and they should support the development of standardized reporting criteria and further comparative research.