Prediction of high-risk factor in early-stage lung cancer: micropapillary adenocarcinoma.

[OBJECTIVE] To identify independent risk factors for micropapillary adenocarcinoma (MPA) in lung adenocarcinoma, a subtype often associated with poor prognosis, in order to support preoperative risk assessment and treatment planning.

[METHODS] Patients diagnosed with lung adenocarcinoma were classified into the micropapillary adenocarcinoma (MPA) group or the non-MPA group based on pathological findings. Clinical, pathological, and imaging data were collected for both groups. Patients were followed up through outpatient visits or telephone interviews. The primary outcome was disease-free survival (DFS). Independent risk factors for MPA were identified using univariate and multivariate logistic regression analyses. A preoperative predictive model and corresponding nomogram were developed. The receiver operating characteristic (ROC) curve was used to determine the optimal cutoff value and the Hosmer-Lemeshow test and bootstrap - corrected calibration curves were used to evaluate the model's predictive performance.

[RESULTS] The recurrence rate was significantly higher in the MPA group compared to the non-MPA group. Independent risk factors for MPA included smoking history, CT-diameter (≥22.5 mm), partially-solid nodule (PSN), solid nodule (SN), lobulation, spiculation, presence of vacuoles, and volume doubling time (VDT) (≤310 days). Guided by the multivariable analysis, we constructed a clinically oriented nomogram to provide an intuitive and practical tool for predicting MPA. The nomogram demonstrated excellent discrimination for predicting MPA, with an area under the ROC curve (AUC) of 0.923.

[CONCLUSION] MPA predicts worse prognosis. The developed nomogram provides improved stratification of patients and supports informed clinical management.