Construction of a novel tool for predicting chronic obstructive pulmonary disease mortality in lung cancer patients.

The rate of lung cancer patients with chronic obstructive pulmonary disease (COPD) is increasing. This study aimed to identify the risk factors associated with COPD mortality in lung cancer patients and to develop a practical tool to achieve an accurate prediction of COPD mortality in lung cancer patients. Patient-related data for this study were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. Risk factors for COPD mortality were identified using Cox regression analysis. The Kaplan-Meier curves were depicted to further validate associated risk factors. Variables screened by multivariate Cox regression analysis were used to construct a predictive model for the risk of COPD mortality. The performance of the model was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and its clinical applicability was evaluated using decision curve analysis (DCA). 28,194 patients were randomized in a 7:3 ratio into a training cohort (n = 19,736) and a validation cohort (n = 8458). Cox regression analysis showed age, race, sex, grade, histological type, T stage, N stage, surgery, chemotherapy, bone metastasis, and marital status as independent factors influencing COPD mortality. The area under the curve (AUC) values of the model in the training cohort were 0.886, 0.870, and 0.873, respectively. In the validation cohort, the AUC values were 0.901, 0.888, and 0.879, respectively. The calibration curves further demonstrate the reliability and stability of the model. DCA indicated that the model could achieve more net clinical benefit. A clinical model for predicting the risk of COPD mortality in lung cancer patients was further constructed, which could provide risk assessment and clinical decision-making for individualized treatment of patients.