Effect of surgical intervention on survival of patients with clinical N2 non-small cell lung cancer: A veterans' affairs central cancer registry (VACCR) database analysis

Background: Optimal management of locally advanced non-small cell lung cancer (NSCLC) lacks consensus. A retrospective analysis of patient data entered in the Veterans Affairs Central Cancer Registry was conducted to evaluate these issues. Patients and Methods: Data of patients with cT1-4, cN2, and cM0 NSCLC diagnosed in the VA Health System between 1995 and 2003 were evaluated. Age, sex, race, smoking history, TNM stage, treatment, and overall survival were abstracted. Survival was compared using multivariate Cox proportional hazards regression analysis. Results: Of the 7328 patients analyzed, 7218 (98.5%) were male, 6061 (82.7%) were white, and 321 (4.4%) were never smokers. The treatment received included: none, 23.8%; chemotherapy alone, 14.3%; radiation alone, 23%; and chemoradiation (sequential or concurrent), 31.4%. Only 7.5% of patients had a surgical resection, with or without multimodality therapy. The median survival (months) of these patient groups were: surgery, 19.3; chemoradiation, 13; chemotherapy alone, 9.2; radiation alone, 7.3; and no treatment, 4 (P < 0.0001). African Americans had a significantly decreased risk of mortality compared with whites (hazard ratio 0.92; 95% confidence interval, 0.87-0.98). Conclusions: Inclusion of surgical resection as a treatment modality was associated with a better overall survival. Also, African Americans appeared to do better than whites. These hypothesis-generating findings should be useful in the ongoing pursuit of better treatment strategies for locally advanced NSCLC.