O.04: Profiling of MET-Amplified Non-Small Cell Lung Cancer (NSCLC), Correlation to cMET Protein Expression/MET Exon 14 Skipping

Abstract

Recent genomic studies of molecular subtypes of NSCLC revealed a plethora of data surrounding its underlying biology. Knowledge of EGFR-driven NSCLC has changed the way one molecular subtype is being managed. While MET-driven NSCLC has been documented, a lot remains unknown about this molecular subtype of NSCLC, in contrast to EGFR- and KRAS-mutated molecular subtypes. The purpose of this study is to investigate the genomics of MET-amplified NSCLC to further understand its tumor biology, and to indicate potential for clinical trials. In total, 3,540 NSCLC specimens underwent multiplatform testing at a CAP/ISO/CLIA-certified laboratory (Caris Life Sciences). Analysis included protein expression by immunohistochemistry (IHC), gene amplification by in situ hybridization (ISH), fusions by ArcherDx FusionPlex assay, and copy number variance (CNV)/mutation by next generation sequencing (NGS). The laboratory validated CISH assay, utilized a gene copy number > 5 to assess gene amplification. In this cohort, 3.1% (109/3,540) patients were MET-amplified. In MET-amplified tumors (n=109), protein expression by IHC showed no ALK expression (0/36), positive EGFR at 47.5% (38/80), and positive PD-L1 at 56.0% (36/64, using 5% cut off value with SP142 antibody), with a sub-group showing a PD-L1 positive rate (50% cut off) of 68.8% (11/16) using the recently FDA approved companion diagnostics kit (22C3 antibody). For gene translocation by FISH, ALK was 2.5% (2/79) and ROS1 was 0.0% (0/97). For mutational analysis, BRAF mutations were 6.5% (6/92), EGFR was 23.1% (21/91), ERBB2 was 1.1% (1/92), and KRAS was 13.0% (12/92). ForMET-amplified NSCLC specimens, cMET protein expression was detected in 92.5% (74/80), MET amplification by CNV was 66.7% (4/6), MET mutation rates were 5.4% (5/92), and MET fusion was found in 20.0% (1/5). Evaluation of the 3,540 specimens revealed 17 MET exon 14 skipping mutations with only two showing amplification (11.8%, 2/17). Our analysis reveals potentially targetable aberrations in MET-driven NSCLC. High PD-L1 rates in this group indicate potential for targeting with PD-1blockade therapy and potential for combined immune blockade/MET inhibition therapy. Differences in cMET alterations indicate the need to analyze patient specimens for MET using different technologies and to find the optimal threshold for MET. Further studies of this molecular subtype of NSCLC are urgently needed to elucidate the optimal therapy for these patients.