Risk factors and nomogram prediction model for checkpoint inhibitor-related pneumonitis in patients with advanced...

Immune checkpoint inhibitors have improved outcomes in advanced non-small cell lung cancer (NSCLC) but are associated with immune-related adverse events such as pneumonitis. This study aimed to identify risk factors for checkpoint inhibitor-related pneumonitis (CIP) and to develop a predictive nomogram for individualized risk assessment in advanced NSCLC patients.

This retrospective study included consecutive patients with advanced NSCLC treated with ICIs between January 2021 and December 2024. All patients who developed CIP were included as cases (n = 96), and non-CIP controls were selected from the source population using propensity score matching (n = 191). CIP was diagnosed using a standardized adjudication process with systematic exclusion of infectious, malignant, radiation-related, and cardiogenic etiologies. Multivariate logistic regression was performed to identify independent predictors, and a nomogram was constructed. Model performance was evaluated using receiver operating characteristic analysis, calibration curves, bootstrap internal validation, and decision curve analysis.

Disease duration (OR 1.66, 95% CI 1.20-2.31), smoking history (OR 3.32, 95% CI 1.34-8.26), prior chest radiotherapy (OR 2.75, 95% CI 1.09-6.92), and baseline Hamilton Anxiety Rating Scale score (OR 1.12 per point, 95% CI 1.04-1.21) were independent predictors of CIP. The nomogram demonstrated good discrimination (AUC 0.819, 95% CI 0.752-0.891) and calibration, with a bootstrap-corrected C-index of 0.751. Subgroup analyses showed consistent associations across immune checkpoint inhibitor types and treatment lines.

The developed nomogram, incorporating key clinical and psychological predictors, offers a practical tool for individualized risk assessment of CIP in advanced NSCLC patients, potentially guiding early intervention and improving immunotherapy safety.