Clinical guidelines promote the identification of several targetable biomarkers to drive treatment decisions in advanced non-small cell lung cancer (NSCLC), but half of all patients do not have a viable biopsy. Specimens from endobronchial-ultrasound transbronchial needle aspiration (EBUS-TBNA) are an alternative source of material for the initial diagnosis of NSCLC, however their usefulness for a complete molecular characterization remains controversial. EBUS-TBNA samples were prospectively tested for several biomarkers by next-generation sequencing (NGS), nCounter, and immunohistochemistry (PD-L1). The primary objectives were to assess the sensitivity of EBUS-TBNA samples for a comprehensive molecular characterization and to compare its performance to the reference standard of biopsy samples. Seventy-two EBUS-TBNA procedures were performed, and 42 NSCLC patients were diagnosed. Among all cytological samples, 92.9% were successfully genotyped by NGS, 95.2% by nCounter, and 100% by immunohistochemistry. There were 29 paired biopsy samples; 79.3% samples had enough tumor material for genomic genotyping, and 96.6% for PD-L1 immunohistochemistry. A good concordance was found between both sources of material: 88.9% for PD-L1, 100% for NGS and nCounter. EBUS-TBNA is a feasible alternative source of material for NSCLC genotyping and allows the identification of patient candidates for personalized therapies with high concordance when compared with biopsy.

No abstract

No abstract

MET inhibitors have shown activity in non-small cell lung cancer patients (NSCLC) with MET amplification and exon 14 skipping (METΔex14). However, patient stratification is imperfect and thus response rates have varied widely. Here, we studied MET alterations in 474 advanced NSCLC patients by nCounter, an RNA-based technique, together with Next Generation Sequencing (NGS), fluorescence in situ hybridization (FISH), immunohistochemistry (IHC) and reverse transcriptase polymerase chain reaction (RT-PCR), exploring correlation with clinical benefit. Of the 474 samples analyzed, 422 (89%) yielded valid results by nCounter, which identified 13 patients (3%) with METΔex14 and 15 patients (3.5%) with very-high MET mRNA expression. These two subgroups were mutually exclusive, displayed distinct phenotypes and did not generally co-exist with other drivers. For METΔex14, 3/8 (37.5%) samples positive by nCounter tested negative by NGS. Regarding patients with very-high MET mRNA, 92% had MET amplification by FISH and/or NGS. However, FISH failed to identify three patients (30%) with very high MET RNA expression, among which one received MET tyrosine kinase inhibitor treatment deriving clinical benefit. Our results indicate that quantitative mRNA-based techniques can improve the selection of patients for MET-targeted therapies.