• Mohamed Rahouma, Mohamed Kamel, Abu Nasar, Sebron Harrison, Benjamin Lee, Jeffrey Port, Nasser Altorki, and Brendon M Stiles.
• Cardiothoracic Department, Weill Cornell Medicine, New York-Presbyterian Hospital, New York, NY; Surgical Oncology Department, National Cancer Institute, Cairo University, Cairo, Egypt.
• J. Thorac. Cardiovasc. Surg. 2020 Feb 24.

ObjectiveSurgery may be underused for stage IIIA non-small cell lung cancer. Although an argument can be made for definitive chemoradiation for N2/3 mediastinal nodal disease, the role of a nonsurgical strategy is less clear in patients with cT3N1M0 stage IIIA given a lack of randomized data. We sought to determine the outcomes of patients with cT3N1M0 by treatment type from the National Cancer Database.MethodsThe National Cancer Database (2004-2014) was queried for patients with cT3N1M0 non-small cell lung cancer, known treatment modalities, and sequence. Comparisons between groups were performed using Mann-Whitney and chi-square tests. Cox regression was performed to identify predictors of overall survival. Propensity score matching analysis was performed to compare overall survival in surgery versus definitive chemoradiation.ResultsWe identified 1937 patients undergoing surgery (1518 up-front and 419 after neoadjuvant treatment) and 1844 patients undergoing definitive chemoradiation. Among patients undergoing surgery without prior treatment, 19% were overstaged and were found to have pN0, whereas 9.6% had pN2/3. Median overall survival was 33.1 months in the surgery group (± adjuvant/neoadjuvant) versus 18 months in definitive chemoradiation. To compare outcomes in balanced groups, we propensity matched 1081 pairs of patients. Median overall survival was 31.1 months in the surgery group compared with 19.1 months in the definitive chemoradiation group (P < .001). By multivariable analysis, surgery (hazard ratio, 0.65; confidence interval, 0.59-0.73), female sex (hazard ratio, 0.88; confidence interval, 0.79-0.98), age (hazard ratio, 1.02; confidence interval, 1.01-1.03), squamous histology (hazard ratio, 1.22; confidence interval, 1.07-1.38), and Charlson score of 2 (hazard ratio, 1.31; confidence interval, 1.11-1.54) were predictors of survival.ConclusionsIn the National Cancer Database, approximately half of patients with clinical T3N1M0 were treated with definitive chemoradiation rather than surgery. This practice should be avoided in operable patients, because surgical resection is associated with better survival.Copyright © 2020. Published by Elsevier Inc.

4 keywords...

hide…