Genome-Wide Analysis Reveals That Cmet Is A Driver Of Ras Signaling And Implicates Ras Activation Status As A Predictor For Response To Cmet Inhibitors

Proceedings: AACR 101st Annual Meeting 2010‐‐ Apr 17‐21, 2010; Washington, DC In order to facilitate the development of HGF/ c-MET inhibitors, we undertook a genome-wide analysis to identify a “c-MET activation signature” that could be used to identify tumors that are potentially dependent on HGF / c-MET signaling and therefore more responsive to HGF/ c-MET inhibition. We first developed a c-MET co-expression signature by identifying genes that correlate with the mRNA expression of c-MET across internal panels of kidney, lung, and colorectal tumors. Many genes correlated with the expression of c-MET, including the epidermal growth factor receptor (EGFR). In order to assess the biology of genes co-expressed with cMET, we performed pathway enrichment analysis to identify pathways or pre-existing signatures that were enriched in this cMET co-expression signature more than we would expect by chance. Our previously reported RAS signature (Loboda et al, 2009 AACR annual meeting) was the top pathway enrichment, indicating a link between genes involved in RAS signaling and cMET expression across tumors. To further assess this relationship, we tested the correlation between c-MET and our RAS signature across independent cohorts of breast and lung tumors, including primary and metastatic samples. A strong correlation between c-MET expression and our RAS signature was present in all cases and was conserved in metastatic tumors. We then assessed the relationship between c-MET and RAS activation using reverse-phase protein arrays measuring approximately 70 proteins across a panel of 89 lung cancer cell lines. Four proteins were identified that significantly correlated positively with the RAS signature: total ERBB4, pMEK, pERK, and pMET. The finding that pMET correlated with RAS signature in addition to pMEK and pERK supports the connection between c-MET signaling and RAS activation, as this relationship is observed on the phospho-protein as well as mRNA levels. Taken together, these data suggest that cMET can serve as a driver of RAS signaling. To test this, we performed a xenograft study in which GTL-16 xenografts (MET amplified gastric cancer cell line) were treated with a novel small molecule inhibitor of c-MET (MK-8033). Animals were treated with vehicle or with a single dose of 10, 33, 66, 100, or 200 mpk MK-8033. At each dose, we collected samples for profiling at 2, 8, and 12 hours post-dose. MK-8033 treatment caused a dose-and time-dependent inhibition of the RAS signature at all three time points, and the level of RAS signature inhibition was related to efficacy. These data support the hypothesis that cMET is a driver of RAS signaling in some tumor contexts and suggest that basal levels of RAS pathway activation may be an important determinant of response to HGF / c-MET inhibition in the clinic. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 107.