A CT-Based and Serum-Derived Nomogram for Preoperative Prediction of Lymphovascular Invasion in Stage Ⅰ Lung Adenocarcinoma

Abstract Background Lymphovascular invasion (LVI) is a crucial prognostic factor in stage I lung adenocarcinoma (LUAD), yet it can only be confirmed postoperatively. This study aimed to develop a preoperative predictive model for LVI by integrating hematological, clinical, and CT-based radiological features. Methods We retrospectively analyzed 800 patients with stage I LUAD who underwent curative resection. Least Absolute Shrinkage and Selection Operator (LASSO) regression and multivariate logistic regression were employed to identify independent predictors. A nomogram was constructed and internally validated via bootstrap resampling. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis (DCA). Results Six independent predictors were identified: neuron-specific enolase (NSE), tumor diameter, consolidation tumor ratio (CTR) > 0.5, spiculation, vacuolar sign, and vessel convergence sign. The nomogram demonstrated good discrimination, with AUCs of 0.841 (95% CI: 0.807–0.875) in the training cohort and 0.786 (95% CI: 0.704–0.868) in the validation cohort. Calibration curves showed excellent agreement between predicted and observed probabilities, and DCA confirmed clinical net benefit across a wide threshold range (5%–80%). Conclusion This simple nomogram incorporating routine CT features and serum NSE offers a robust, noninvasive tool for preoperative individualized prediction of LVI in stage I LUAD, potentially aiding in risk stratification and treatment planning.