Genetic alterations of driver genes as independent prognostic factors for disease-free survival in patients with resected non-small cell lung cancer

Abstract

ObjectivesThis study assessed the associations between the molecular signatures and clinical information in non-small cell lung cancer (NSCLC) patients with postoperative disease-free survival (p-dfs) to identify novel prognostic factors, focusing on associations with driver gene alterations. Materials and methodsBetween February 2014 and September 2015, 242 patients with NSCLC, including 192 patients with adenocarcinoma (Ad) and 50 patients with squamous cell carcinoma (Sq), underwent surgery and were enrolled in this study. Surgically resected tissues were subjected to whole exome sequencing. Mt detected in 138 cancer-related genes were evaluated as driver mutations. A multivariate analysis using the multi-state model was used to establish the associations between co-variables and p-dfs. ResultsPostoperative recurrence (p-rec) was observed in 49 (20.2%) and 19 (7.9%) patients with Ad and Sq, respectively. The median (range) follow-up period for all the censored cases was 2.5 (2.0–3.5) years. The characteristics of the patients with postoperative recurrence were as follows: median age (range), 71 (50–87) years; male, 38 (56%); smoker, 51 (75%); p-stage (I/II/III), 30 (44%)/19 (28%)/19 (28%); histological type (Ad/Sq), 49 (72%)/19 (28%); adjuvant chemotherapy (yes/no), 30 (44%)/38 (56%); and driver gene alteration (presence/absence), 65 (96%)/3 (4%). In univariate analyses, age (<70/≧70 years), smoking history (yes/no), p-stage (I, II/III), histological type (Ad/Sq), and driver mutation (presence/absence) were favorable prognostic factors (P = .017, P = .048, P = .0002, P = .006, P = .029, respectively). A multivariate analysis also revealed a significant association between the driver mutation status and p-dfs (P = .046; odds ratio [OR], 2.86; 95% confidence interval [CI], 1.02–8.08), when adjusted according to histological type (P = .10), smoking status (P = .09), gender (P = .51), age (P = .008) and p-stage (P = .00003). ConclusionThe driver mutation status may be an independent prognostic factor of p-dfs in NSCLC.