A Model to Predict the Use of Surgical Resection for Advanced-Stage Non-Small Cell Lung Cancer Patients

Abstract

For advanced-stage non-small cell lung cancer, chemotherapy and chemoradiotherapy are the primary treatments. Although surgical intervention in these patients is associated with improved survival, the effect of selection bias is poorly defined. Our objective was to characterize selection bias and identify potential surgical candidates by constructing a Surgical Selection Score (SSS). Patients with clinical stage IIIA, IIIB, or IV non-small cell lung cancer were identified in the National Cancer Data Base from 1998 to 2012. Logistic regression was used to develop the SSS based on clinical characteristics. Estimated area under the receiver operating characteristic curve was used to assess discrimination performance of the SSS. Kaplan-Meier analysis was used to compare patients with similar SSSs. We identified 300,572 patients with stage IIIA, IIIB, or IV non-small cell lung cancer without missing data; 6% (18,701) underwent surgical intervention. The surgical cohort was 57% stage IIIA (n= 10,650), 19% stage IIIB (n= 3,483), and 24% stage IV (n= 4,568). The areas under the receiver operating characteristic curve from the best-fit logistic regression model in the training and validation sets were not significantly different, at 0.83 (95% confidence interval, 0.82 to 0.83) and 0.83 (95% confidence interval, 0.82 to 0.83). The range of SSS is 43 to 1,141. As expected, SSS was a good predictor of survival. Within each quartile of SSS, patients in the surgical group had significantly longer survival than nonsurgical patients (p < 0.001). A prediction model for selection of patients for surgical intervention was created. Once validated and prospectively tested, this model may be used to identify patients who may benefit from surgical intervention.