Abstract C48: Defining mechanisms of adaptation to KRAS G12C inhibitors: Using quantitative proteomics to design combinatorial strategies in pancreatic cancer

Abstract

Abstract KRAS is mutated in 95% of pancreatic ductal adenocarcinoma (PDAC) tumors and is a critical driver of PDAC initiation, survival, and proliferation. Therapeutic efforts to target KRAS directly have been largely unsuccessful, leading to the thought that KRAS was “undruggable.” Recently, KRAS G12C inhibitors (KRASi) that covalently bind the mutant cysteine have been reported and are currently being tested in clinical trials. Unfortunately, as has been demonstrated with other targeted therapies, the efficacy of targeting an oncogenic driver can be limited by both tumor heterogeneity and development of resistance. Here, we have used a mass spectrometry-based quantitative proteomics workflow to identify pathways of adaptation to KRASi and to predict cytotoxic drug combinations both in 2D and 3D cell culture conditions. We treated multiple KRAS G12C mutant tumor lines (pancreatic, lung) with KRASi, which induced a cytostatic response with subsequent re-establishment of proliferation at longer time points. We profiled the proteomic adaptations acutely and at long term using a multiplexed quantitative proteomics workflow and identified and quantified over 8000 proteins. Pathway analysis by Gene Set Enrichment Analysis (GSEA) identified common mechanisms of acute drug response among KRAS G12C mutant tumor lines such as downregulation of cell cycle/transcription and increased lipid metabolism. Long-term adaptation was associated with increased DNA repair and oxidative metabolism, but further adaptations differed between tumor cell lines, suggesting selective programs are required to reactivate proliferation. Connectivity map (Cmap) analysis identified 30 perturbagen classes (connectivity score of Cmap class>90) that positively correlated with the MiaPaCa-2 KRASi profile at 24h. We identified correlations with elements in the KRAS pathway (MEK, RAF inhibitors) and with pathways previously described to synergize with KRAS inhibition (PI3K inhibitors), validating the utility of our approach. In addition, we identified combinations with HSP90, MET and EGFR inhibitors that, when co-targeted with KRASi, were able to suppress growth in long-term assays and induced cytotoxicity. Given that recent results suggest that 3D culture conditions may be most predictive of in vivo efficacy, we selected cell lines with greater KRASi sensitivity in 3D growth conditions and performed proteomic analysis in 3D versus 2D culture. Through GSEA analysis we determined differences in the 3D versus 2D proteome basally as well as after KRASi. Finally, we identified combinations by Cmap analysis that differentially correlated with the 3D signature, including CDK4/6 inhibitors, which reduced growth in 3D conditions. Overall, we employed a proteomics platform to characterize adaptation to KRASi and identified combinatorial regimens that induce cytotoxicity with potential therapeutic utility. Citation Format: Naiara Santana-Codina, Amrita Singh Chandhoke, Qijia Yu, Beata Malachowska, Miljan Kuljanin, Mark P. Jedrychowski, Ajami Gikandi, Marcin Stanczak, David A. Scott, Wojciech Fendler, Nathanael S. Gray, Joseph D. Mancias. Defining mechanisms of adaptation to KRAS G12C inhibitors: Using quantitative proteomics to design combinatorial strategies in pancreatic cancer [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Advances in Science and Clinical Care; 2019 Sept 6-9; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2019;79(24 Suppl):Abstract nr C48.