Abstract 2147: Correlating RAS oncogenic allele dependence with drug sensitivity

Abstract

Abstract KRAS is a small G-protein that is a highly important member of the EGFR pathway. KRAS is frequently mutated in human cancers and predicts a shorter survival for patients with lung and colorectal cancers. Although any mutation at codon 12 or 13 is potentially oncogenic, Cysteine-12, Aspartate-12, Valine-12, and Aspartate 13 mutations predominate in human cancers, and the various oncogenic alleles correlate with distinct clinical outcomes. There is even some indication that Glycine 12 mutations may benefit from EGFR targeted therapy, whereas codon 12 mutations are contraindicated. We hypothesize that these clinical observations may be explained by subtle differences in effector pathway activation. It has been reported that G12C and G12V mutations preferentially activate the Ral-GDS pathway, while cell lines with G12D mutations elevate the PI3′ kinase and MAPK pathways. However, it is difficult to correlate clinical response to differences in biochemical signaling amidst a myriad of potentially confounding genetic alterations present across a panel of cell lines or patient tumor samples. In order to normalize the contributions of cell line heterogeneity, we tested pairs of isogenic cell lines to isolate KRAS allele specific effects, and to identify compounds that selectively inhibit RAS dependent growth in an allele specific manner. To do this, we used an engineered mouse embryonic fibroblast cell line - H-Ras−/−, N-Ras−/− and K-Raslox/lox, which expresses a transgene cre-recombinase - estrogen receptor fusion. Treatment with 4OHT results in excision of the endogenous K-Ras alleles, loss of KRAS expression, suppression of the MAPK pathway and G1 arrest. Reconstitution with any RAS isoform (K- H- or N-Ras), including any of the KRAS oncogenic alleles using a lentiviral vector system, allows the cells to re-enter the cell cycle and proliferate. The doubling time and basal pERK and pAKT levels are comparable across all isogenic uniclonal cell lines regardless of the RAS allele introduced. This system takes advantage of its dependency on a single RAS mutant to help screen for effector pathway dependence or synthetic lethal interactions unilaterally relevant to one mutant KRAS allele. Majority of compounds tested were unbiased towards the cell lines, but a few classes of drug showed selectivity. Cells expressing wild-type RAS were more sensitive to RTK inhibitors compared with cells expressing a mutant KRAS allele. Farnesyl Transferase Inhibitors (FTIs) such as Tipifarnib (R115777/Zarnestra) and Lonafarnib (SCH66336/Sarasar) were more potent against the HRAS WT dependent cell lines as compared to the KRAS dependent lines, validating the escape mechanism by which K-Ras and N-Ras get geranylgeranylated in the presence of FTIs. This strategy provides us with tools to delineate the subtle allele specific sensitivity differences which may translate to meaningful biology and may be used for tailoring mutant specific drug regimens in the clinic. Citation Format: Kanika Sharma, Katie Beam, Nicole Fer, Matthew Holderfield. Correlating RAS oncogenic allele dependence with drug sensitivity. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2147. doi:10.1158/1538-7445.AM2015-2147