Factors determining the feasibility of segmentectomy for central non-small cell lung cancer and construction of a predictive model.

[OBJECTIVES] Multiple studies have demonstrated that segmentectomy is feasible and can yield long-term outcomes comparable to lobectomy for selected central non-small cell lung cancer (NSCLC). This study aimed to identify factors influencing the feasibility of segmentectomy for central NSCLC, and to develop a predictive nomogram.

[METHODS] Patients with central NSCLC ≤2 cm who underwent segmentectomy or lobectomy between 2020 and 2024 were screened. Patients were split into two groups based on their surgical date (Training set: 2020-2023 and Validation set: 2024). The least absolute shrinkage and selection operator (LASSO) regression and multivariable logistic regression analyses were performed to identify potential predictors. Qualified segmentectomy (surgical margin ≥ maximum tumor diameter) was classified as 0, while lobectomy or unqualified segmentectomy was classified as 1. A nomogram was developed based on these identified predictors.

[RESULTS] There were 303 and 105 patients in the training set and validation set, respectively. The probability of achieving sufficient margins during segmentectomy was 83.9 % in the training set and 81.3 % in the validation set. After LASSO and multivariable logistic regression analyses, four variables were retained: tumor size (OR = 3.81, P = 0.006), radiological type (solid vs. subsolid, OR = 5.39, P = 0.009), tumor-to-segmental bronchus distance (OR = 0.64, P = 0.036), and number of tumor-involved subsegments (OR = 1.54, P < 0.001). The nomogram developed from these predictors exhibited good predictive ability, with an area under the curve of 0.793 in the training set and 0.777 in the validation set.

[CONCLUSIONS] Tumor size, radiological type, tumor-to-segmental bronchus distance, and number of tumor-involved subsegments influenced the feasibility of segmentectomy for central NSCLC.