Efficacy of DNA versus RNA NGS-based Methods in MET Exon 14 skipping mutation detection

9036 Background: Exon 14 skipping mutations in the mesenchymal-epithelial transition ( MET) gene are reported in 2-5% of lung adenocarcinomas and are mutually exclusive of other driver mutations. Small-molecule MET tyrosine kinase inhibitors, capmatinib and tepotinib, showed durable responses in previously treated and treatment-naïve patients harboring MET-exon-14 skipping mutations. Studies suggest that for detection of MET-ex14 mutations, DNA-based assays alone may be sub-optimal when compared to RNA-based NGS assays. We compared the performance of DNA and RNA-based assays for detection of MET-ex14 variants. Methods: We examined NGS-based profiling data of lung adenocarcinomas (or when this diagnosis could not be excluded) to identify MET-ex14 mutations missed by DNA but identified by RNA analysis. The carcinomas were profiled by a DNA-based NGS panel that targets MET exons 2, 14, 16, 18 and 19. Cases without driver mutations were reflexed to an NGS-based RNA fusion panel (Archer’s Anchored Multiplex PCR). Results: Over a 21-month period, MET-ex14 skipping events were detected in 16/644 (2.5%) lung carcinomas by DNA profiling. RNA analysis on driver-negative cases identified 9 additional MET-ex14 mutations. All 16 MET-ex14 DNA variants occurred at or around the intron 14 splice donor site, as the assay did not include the intron 13 splice acceptor site. Clinical characteristics of the MET positive cohort include a male to female ratio of 0.8:1.0, an average age of 76.5 years and 52% non-smoker status. All tumors were adenocarcinomas (including one with a < 10% spindle/pleomorphic component) with the exception of 3 adenosquamous carcinomas and 1 squamous cell carcinoma. Conclusions: DNA based NGS-panels can potentially miss MET-ex14 skipping events in lung carcinomas, when the primers do not target both 3′ splice site of intron 13, and the 5′ splice site of intron 14. A reflex work flow interrogating RNA fusions can potentially capture such events. The clinical and molecular characterization of the variants detected only by RNA NGS assays warrants further exploration.