P2.17-25 Genomic Insight into Stage I Pulmonary Adenocarcinoma for Recurrence by Targeted Next-Generation Sequencing Analysis

Despite surgical resection, stage I pulmonary adenocarcinoma has a recurrence rate of 18.4%–25%. Prognostic markers of localized lung adenocarcinoma remain undefined. Targeted next-generation sequencing (NGS) is supposed to be useful to identify recurrence-related genes. We aimed to identify genes and co-occurring mutations that might predict recurrence after surgical resection of stage I pulmonary adenocarcinoma through targeted NGS. Tissues from 202 patients who had complete resection of stage I pulmonary adenocarcinoma at Konkuk University Medical Center, from 2005 to 2017 (median follow-up: 48 months), were analyzed by targeted NGS using a panel of the 170 cancer-related genes (KF1 panel by Macrogen Inc. Korea). We compared RFS (relapse-free survival) according to frequent genetic alterations using the Kaplan-Meier method and compared their combinations to identify recurrence-related genes by multivariate analysis. The most frequent genetic alterations were EGFR (56.9%), TP53 (37.7%), KRAS (13.4%), and PIK3CA (10.4%) in surgically resected stage I pulmonary adenocarcinoma. In the tumors with single gene mutation, EGFR mutation was a good prognostic factor and TP53 mutation was a poor prognostic factor for recurrence. (EGFR mutation, HR 0.26, 95% CI 0.07-0.95, p=0.024 vs TP53 mutation, HR 3.39, 95% CI 1.17-9.8, p=0.02). In the tumors with two gene mutations, similar tendency has been observed but there was no statistical significance. Mutation of CTNNB1 was significantly related to recurrence in multivariate analysis (HR3.76, P=0.003) and this finding suggests the CTNNB1 as a molecular biomarker for recurrence by NGS method. TP53 and KRAS combination mutations tended to be associated with shorter RFS (HR=1.89, p=0.08). Recurrence was not associated with the number of the co-occurring mutations but with their specific combinations such as the combination of CTNNB1 and EGFR. The CTNNB1 mutation is associated with shorter RFS as compared with wild-type CTNNB1 in resected stage I adenocarcinoma. Larger datasets are required to validate whether the CTNNB1 gene is an independent predictive factor of early stage adenocarcinoma recurrence. NGS can be useful tool to predict the risk of recurrence and select the patient to need the adjuvant chemotherapy in stage I adenocarcinoma.