Abstract 4606: Subclassification of patients with lung adenocarcinoma harboring EGFR-activating mutations by gene expression profiling

Abstract

Abstract Background: Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) have been effective in patients with lung adenocarcinoma (LUAD) harboring EGFR-activating mutations. However, the response to EGFR-TKIs in these patients is not uniform, indicating that patients with EGFR-activating mutations may be further stratified into some undefined molecular subgroups. Therefore, we aimed to identify these molecular subgroups in patients with LUAD harboring EGFR-activating mutations, by gene expression profiling (GEP) data. Methods: Between January 2014 and April 2017, 294 patients with LUAD who underwent surgery at the Shizuoka Cancer Center were enrolled with informed consent. Whole-exome sequencing (WES) and GEP were conducted with an Ion Proton system and an Agilent SurePrint G3 Human Gene Expression 8×60K v2 Microarray, respectively. Corresponding peripheral blood samples were used to identify tumor-specific genetic alterations. Tumor-specific gene expression was identified by comparing expression profiles between tumor and adjacent normal tissues. Samples with estimated tumor purity <20% were excluded from the following analysis to avoid false negatives; the remaining 251 patients (85%) were included in the analysis. EGFR-activating mutations were defined based on OncoKB database. Statistical evaluation of GEP data was conducted using principal component analysis (PCA) and orthogonal projection to latent structures discriminant analysis (OPLS-DA). Results: EGFR-activating mutations were identified in 98 patients (39%); the most frequently identified alteration was the deletion in exon 19 (47%, 46), followed by the L858R mutation (45%, 44). No significant differences in the expression profile were noted between patients with both major EGFR-activating mutations. Patients with EGFR-activating mutations were divided into two clusters (C1 and C2) by the evaluation of those expression profiles with PCA, and 10 and 88 patients were included in each cluster, respectively. As a result of OPLS-DA, the genes relevant to cell cycle progression, DNA repair and tumor immunity were selected, and were considered to have strongly contributed to the differences between two clusters. Associations between each cluster and clinicopathologic characteristics were evaluated by logistic regression analysis. Patients classified into C1 showed the significant association with poor differentiation, relapse and smoking history. Conclusions: These results demonstrate that patients with EGFR-activating mutations can be further stratified into two additional molecular subgroups by assessing the expression level of genes above mentioned. To establish this expression profiling-based stratification as predictive marker of EGFR-TKIs, further clinical study to evaluate the association with the therapeutic response to EGFR-TKIs is needed. Citation Format: Hirotsugu Kenmotsu, Masakuni Serizawa, Mitsuhiro Isaka, Hideaki Kojima, Shoji Takahashi, Akira Ono, Tateaki Naito, Haruyasu Murakami, Takeshi Nagashima, Shumpei Ohnami, Keiichi Ohshima, Kenichi Urakami, Masatoshi Kusuhara, Ken Yamaguchi, Takashi Sugino, Yasuhisa Ohde, Toshiaki Takahashi. Subclassification of patients with lung adenocarcinoma harboring EGFR-activating mutations by gene expression profiling [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4606.