Identifying the Appropriate FISH Criteria for Defining MET Copy Number-Driven Lung Adenocarcinoma through Oncogene Overlap Analysis

Introduction: Mesenchymal-epithelial transition factor gene (MET) gene copy number gain may be a predictive biomarker for mesenchymal-epithelial transition factor (MET) inhibition in lung cancer, but the most appropriate method and criteria for defining MET positivity are uncertain. Methods: MET copy number was assessed by fluorescence in situ hybridization in lung adenocarcinoma. Positivity criteria included mean MET per cell values greater than 5 (low [>= 5 to <6], intermediate [>= 6 to <7], and high [>= 7]) and mean MET-to-chromosome 7 centromere ratios (MET/CEP7) of at least 1.8 (low [>= 1.8 to <= 2.2], intermediate [>2.2 to <5], and high [>= 5]). Associated clinical and molecular characteristics were captured. Results: Of 686 cases, 99 (14%) had a mean MET per cell value of 5 or greater, 52 of 1164 (4.5%) had a MET/CEP7 ratio of 1.8 or higher. Other oncogenic drivers (in EGFR, KRAS, anaplastic lymphoma receptor tyrosine kinase gene [ALK], erb-b2 receptor tyrosine kinase 2 gene [ERBB2], BRAF, NRAS, ROS1, or ret proto-oncogene [RET]) were detectable in 56% of the group with a mean MET per cell value of 5 or higher and 47% of the group with a MET/CEP7 ratio of 1.8 or higher, suggesting that many MET-positive cases are not truly MET addicted. The rates of concomitant drivers in the groups of patients in the low, indeterminate, and high categories of mean MET per cell were 32 of 52 (62%), 12 of 19 (63%), and 11 of 27 (41%) (p = 0.2), and the rates of concomitant drivers in the low, intermediate, and high categories of MET/CEP7 ratios were 15 of 29 (52%), 9 of 18 (50%), and 0 of 4 (0%), respectively (p = 0.04). A MET/CEP7 ratio of 1.8 or higher in the absence of other oncogenes was associated with a higher rate of adrenal metastases (p = 0.03) but not with never smoking status. Conclusions: A fluorescence in situ hybridization MET/CEP7 ratio of 5 or higher defined a MET-positive group with no oncogenic overlap. As this method and criteria are also associated with the highest response rate to MET inhibition, it represents the clearest definition of a MET copy number gain-addicted state. However, a MET-associated phenotype may also exist across cases with a MET/CEP7 of 1.8 or higher when no other oncogene overlap occurs. (C) 2016 International Association for the Study of Lung Cancer. Published by Elsevier Inc. All rights reserved.