A chest CT based model predicts lymph node metastasis in non-small cell lung cancer

Accurate preoperative prediction of lymph node metastasis (LNM) is critical for staging and treatment planning in non-small cell lung cancer (NSCLC). This study aimed to develop and validate a chest CT–based model for predicting LNM. This retrospective study included 362 patients with pathologically confirmed NSCLC treated between January 2019 and December 2025. Patients were randomly assigned to a training cohort ( n = 252) and a validation cohort ( n = 110). Clinical and CT imaging features were systematically collected. Univariable and multivariable logistic regression analyses were performed to identify independent predictors of LNM. A nomogram was constructed to estimate individualized risk. Model performance was evaluated using receiver operating characteristic (ROC) analysis, calibration curves with bootstrap resampling, and decision curve analysis (DCA) to assess discrimination, calibration, and clinical utility. Tumor maximum diameter (TMD), carcinoembryonic antigen (CEA) level, lymph node short-axis diameter (LNSD), and tumor location were identified as independent predictors of LNM and were incorporated into the nomogram. The model demonstrated good discrimination, with AUCs of 0.814 and 0.861 in the training and validation cohorts, respectively. Calibration curves showed good agreement between predicted and observed outcomes. In addition, the model provided a higher net clinical benefit than treat-all or treat-none strategies across a range of threshold probabilities. In the validation cohort, the sensitivity, specificity, and accuracy were 76.4%, 85.5%, and 82.0%, respectively. The nomogram may serve as an adjunctive tool for preoperative risk stratification of LNM in NSCLC patients, with external validation in multicenter cohorts required to establish its clinical applicability.