Risk Factors and Nomogram for Checkpoint Inhibitor Pneumonitis in Advanced NSCLC

A retrospective study of 287 advanced non-small cell lung cancer patients (96 CIP cases, 191 matched controls) treated with immune checkpoint inhibitors between 2021 and 2024. It identifies independent risk factors for checkpoint inhibitor-related pneumonitis and presents a predictive nomogram with an AUC of 0.819.

Use Cases

• Validate the predictive nomogram for CIP risk using the identified independent predictors like disease duration and smoking history.
• Analyze the association between baseline Hamilton Anxiety Rating Scale scores and CIP incidence in a multivariate logistic regression framework.
• Assess the performance of the predictive model via its bootstrap-corrected C-index of 0.751 and decision curve analysis.
• Investigate the impact of prior chest radiotherapy on CIP risk across different immune checkpoint inhibitor types and treatment lines.

Strengths

• Includes 287 patient records (96 cases, 191 controls) for model development and validation.
• The nomogram model demonstrates good discrimination with an AUC of 0.819 and a bootstrap-corrected C-index of 0.751.
• Risk factors were identified through a standardized adjudication process and multivariate logistic regression.

Limitations

• The dataset is small (17.5 KB), likely containing only summary statistics and model parameters, not raw patient-level data.
• As a retrospective study, findings may be subject to selection bias and unmeasured confounding variables.
• The model's performance is based on internal validation only; external validation is required for generalizability.

Provenance

Source

Ya-Fang Chen via figshare.

Collection Method

Retrospective study of consecutive patients with advanced NSCLC treated with ICIs, using propensity score matching for control selection.

Time Range

January 2021 to December 2024.

Freshness

Data covers patients treated up to December 2024, with the dataset last updated in March 2026.