P1.08-066 Prognostic Factors of Post-Recurrence Survival in Patients with Completely Resected Stage III-N2 Non-Small Cell Lung Cancer

Abstract

Post-recurrence survival (PRS) after curative resection has been considered a multifactorial process dependent on clinicopathological, biological, and treatment modality in non-small cell lung cancer (NSCLC). The aim of this study is to investigate the prognostic factors for PRS in patients with completely resected stage III-N2 NSCLC. Two hundred forty-five patients who had complete resection for pathologic stage III-N2 NSCLC between 2003 and 2014 were enrolled. First, a number of clinicopathological factors were evaluated to find prognostic factors for recurrence by Cox proportional hazards models. Second, the following additional data were evaluated: presence of recurrent symptom, recurrence patterns, treatment modality, use of targeted agents, and recurrence-free interval. The prognostic effects of these factors were analyzed for PRS. One hundred twenty-four patients experienced recurrence during a median follow-up period of 36.3 months. Univariate analysis showed that vascular invasion, lymphatic invasion, tumor size, number of positive lymph nodes (LNs), and multistation N2 were poor prognostic factors for recurrence. Lymphatic invasion, tumor size, and number of positive LNs were even worse independent prognostic factors for recurrence by multivariate analysis. Of 124 recurred patients, 21 patients (17%) were symptomatic at the time of initial recurrence, and the remaining 103 patients (83%) were asymptomatic. In these asymptomatic patients, recurrence was detected by tumor markers in 3, computed tomography (CT) in 80, or positron emission tomography-CT (PET/CT) in 20 patients. The mean recurrence-free interval was 14.0 months (≤ 12 months in 72, > 12 months in 52 patients). The patterns of recurrence were presented as loco-regional recurrence in 37 (30%), distant metastasis in 33 patients (27%), and both in 54 patients (43%). The types of initial treatment included surgery in 15 (12%), chemotherapy in 68 (55%), radiotherapy in 19 (15%), and chemo-radiation in 16 patients (13%). The median duration of PRS was 30.5 (1-109) months and the 2-year and 5-year of PRS were 54% and 23%, respectively. Univariate analysis identified no symptom of recurrence, only LN metastasis without distant organ metastasis, treatment modality, and a longer recurrence-free interval as good prognostic factors, while no symptom and a longer recurrence-free interval were independent prognostic factors for PRS in a multivariate analysis. No symptom at the time of recurrence and a longer recurrence-free interval were significant predictors of better PRS in patients that have underwent complete resection of stage III-N2 NSCLC.