Development and validation of a nomogram for predicting high-risk pathology in clinical stage lA left upper lobe lung adenocarcinoma.

Background: Accurate preoperative identification of high-risk pathological features in early-stage lung adenocarcinoma is critical for guiding surgical decisions and improving patient outcomes. This study aimed to develop and validate a nomogram to predict high-risk pathology in clinical stage IA lung adenocarcinoma located in the left upper lobe (LUL), an anatomical site with distinct surgical implications.

Methods: We retrospectively reviewed 545 patients with clinical stage IA LUL adenocarcinoma who underwent surgery between January 2018 and May 2022. The cohort was randomly divided into training (80%) and validation (20%) sets. Independent predictors were identified via multivariate logistic regression and further validated using LASSO regression. A nomogram was constructed and evaluated using ROC curves, calibration plots, decision curve analysis (DCA), and bootstrap resampling.

Results: High-risk pathology, defined by the presence of solid/micropapillary predominant patterns, complex glandular architecture, STAS, or LVI, was observed in 19.1% of patients. Four independent preoperative predictors were identified: elevated CEA levels, larger CT-measured tumor size, invasive histology on frozen section, and higher mean CT value. The nomogram demonstrated excellent discriminative ability, with AUCs of 0.837 in the training set and 0.865 in the validation set. Internal validation by bootstrap resampling confirmed model stability.

Conclusion: The proposed nomogram integrates routinely available clinical, radiologic, and intraoperative variables to enable individualized preoperative risk assessment for high-risk pathology in stage IA LUL adenocarcinoma. This tool may assist surgeons in tailoring surgical approaches and identifying patients who may benefit from more extensive resection or adjuvant therapy. Prospective external validation is required to confirm generalizability.