Although survival in stage III non-small cell lung cancer (NSCLC) patients receiving concurrent chemoradiotherapy is significantly prolonged, brain metastasis (BM) remains prevalent. This study aims to develop and validate a comprehensive model for predicting BM risk in stage III NSCLC patients, guiding personalized treatment. A total of 311 patients were retrospectively analyzed and randomly divided into a training cohort (n = 230) and a validation cohort (n = 81). Univariate analysis identified potential predictors, followed by multivariate analysis using stepwise AIC regression to determine independent risk factors. A nomogram model was constructed and validated with ROC curves, calibration curves, and decision curve analysis, which was used for risk stratification. Of the 311 patients, 45 (14.5%) developed BM. Key independent predictors included sex, EGFR mutation, liver metastasis, immune maintenance deficiency, neuron-specific enolase, carcinoembryonic antigen, and absolute lymphocyte count. The model demonstrated robust predictive performance, with an area under the ROC curve of 0.813 in the training cohort and 0.775 in the validation cohort, along with favorable calibration and decision curve analysis. Kaplan–Meier survival analysis showed that patients with BM had significantly shorter overall survival (43.3 vs. 75.8 months, p = 0.007). Using a nomogram-derived cutoff score of 393.79, patients were stratified into high- and low-risk groups, with the high-risk group exhibiting significantly shorter median overall survival compared to the low-risk group (p = 0.034). This validated nomogram offers a practical tool for early identification of high-risk patients with stage III NSCLC, facilitating personalized surveillance and intervention strategies to improve outcomes.
Gong et al. (Thu,) studied this question.