CT-based nomogram for predicting EGFR mutation status in ground-glass nodules of lung adenocarcinoma.

Purpose: This study aimed to establish a nomogram based on computed tomography (CT) imaging characteristics to predict epidermal growth factor receptor (EGFR) mutation status in patients with ground-glass nodules (GGNs), thereby aiding medication decision-making. Materials and methods: In total, 935 patients diagnosed with GGNs were enrolled. Patients undergoing surgery from August 2019 to December 2023 (n=709) comprised the training cohort, whereas those treated between January 2024 and March 2025 (n=226) constituted the validation cohort. Clinical parameters and radiological features were recorded for all participants. The training group underwent univariate and multivariate logistic regression analyses to identify significant predictive variables, subsequently facilitating the construction of a nomogram prediction model. The model's discrimination, calibration, and clinical applicability were validated in both patient cohorts. Results: Multivariate logistic regression analysis revealed maximum nodule diameter, consolidation-to-tumor ratio (CTR), mean CT values, presence of air bronchogram signs, and vascular convergence signs as independent predictors of EGFR mutations. The resulting nomogram demonstrated robust predictive capability, achieving an area under the curve (AUC) of 0.87 (95% CI: 0.85-0.90) in the training group and 0.87 (95% CI: 0.82-0.92) in the validation group. Bootstrap internal validation yielded an AUC of 0.89, confirming strong model discrimination. Calibration plots and decision curve analysis further supported the model had a good calibration degree and clinical practicability across both groups. Conclusion: The nomogram integrating maximum diameter, CTR, mean CT value, air bronchogram signs, and vascular convergence signs effectively predicts EGFR mutation status in GGNs, offering a valuable tool for clinical guidance and patient management strategies.