Abstract A55: KRAS gene amplification is a distinct molecular subgroup of gastroesophageal adenocarcinoma that may benefit from combined RAS/RAF/MEK/ERK and PI3K/PTEN/AKT/mTOR pathway inhibition

Abstract

Abstract Background: Given the prevelance of chromosomal instability and gene amplification (amp+) as driver mechanisms in gastroesophageal adenocarcinoma (GEC), we assessed the incidence of KRAS gene amp+ and gene copy number (GCN) along with protein expression, at baseline and after molecular targeted therapy in GEC cell lines/tissues. KRAS undergoes activating mutation in various cancers. KRAS mutation is rare (<5%) in GEC. KRAS is a downstream effector of HER2 and MET tyrosine kinases, and KRAS upregulation may be a resistance mechanism to targeted anti-HER2/anti-MET therapy. We assessed KRAS amp+ in a series of 81 GEC samples and 19 GEC cell lines. We established a KRAS amp+ model to evaluate potential therapeutic strategies. Methods: 81 GEC samples and 19 GEC cell lines were assessed for KRAS GCN by fluorescence in situ hybridization (FISH) using HER2 scoring criteria. KRAS expression was assessed by immunohistochemistry (IHC) and selected reaction monitoring mass spectrometry (SRM-MS) using a novel mulit-plex (GEC-plex) assay for formalin fixed paraffin embedded (FFPE) samples. Patient samples with HER2+ status treated with trastuzuamb as well as a cohort of patients treated with anti-MET therapy were available for pre/post treatment assessment. A KRAS-amplified cell line (Cat2) was established from ascites fluid and grown in vitro and screened for KRAS mutation using deep next-gerenation sequencing to determine if KRAS mutant alleles were present. Tumorigenic activity of Cat2, and another amplified line, MKN1, were assessed using standard proliferation assays in vitro, and an intraperitoneal (IP) murine model assessing time to ascites development (TTA), with/without anti-MEK, anti-PI3K/mTOR, or combinations thereof. Cat2 and MKN1 were also treated with KRAS siRNA and compared to siRNA control in vitro. Results: Assessment of KRAS GCN revealed clustered wild type gene amp+ in 16% (13/81) of FFPE samples, while KRAS mutation was observed in 7% (6/81) (mutually exclusive). KRAS amp+ was found in newly diagnosed metastatic patients (12/55, 22%) more than locally advanced (1/26, 4%) (p=0.06). GCN correlated with expression as determined by IHC and the MS GEC-plex. Patients treated with anti-HER2 or anti-MET therapy were observed to have developed KRAS amp+ in post-treatment biopsies at progressive disease as compared to pre-treatment biopsy. In the cell line, Cat2, KRAS was amp+ (by FISH, mean copies/nucleus: 43.2) and extremely overexpressed by IHC and MS (60 fold over KRAS non-amp+ samples); KRAS was not mutated. In vitro, KRAS siRNA substantially inhibited KRAS amp+ lines versus siRNA control (>80%). We observed diminished KRAS amp+ growth/proliferation both in vitro and in vivo using combined MEK and AKT pathway inhibitors over either alone, and compared to various inhibitor controls (anti-EGFR, anti-MET, anti-HER2). Conclusions: In this series, we observed KRAS wild type gene amp+ to be present in a subset (16%) of GEC patients at diagnosis, correlating with very high protein expression. KRAS amp+ was present after treatment with trastuzumab in HER2+ patients, and also after anti-MET therapy. These data suggest that KRAS amp+ represents a molecular subset with advanced disease at diagnosis. The observation of acquired KRAS amp+ after targeted therapies may be a resistance mechanism to anti-HER and anti-MET inhibitors. Inhibition using combined MEK/AKT pathway inhibitors, and proof-of-principle siRNA, warrants further investigation for KRAS amp+ tumors. Citation Format: Les Henderson, Peng Xu, Brittany Rambo, Wei-Li Liao, Todd Hembrough, Daniel Catenacci. KRAS gene amplification is a distinct molecular subgroup of gastroesophageal adenocarcinoma that may benefit from combined RAS/RAF/MEK/ERK and PI3K/PTEN/AKT/mTOR pathway inhibition. [abstract]. In: Proceedings of the AACR Special Conference on RAS Oncogenes: From Biology to Therapy; Feb 24-27, 2014; Lake Buena Vista, FL. Philadelphia (PA): AACR; Mol Cancer Res 2014;12(12 Suppl):Abstract nr A55. doi: 10.1158/1557-3125.RASONC14-A55