Abstract 3911: Proteomic and transcriptomic subtypes of KRASG12C mutant lung cancer

Abstract

Abstract Sotorasib (AMG510) is the first KRASG12C covalent inhibitor (KRASG12Ci) approved by the FDA in lung cancers harboring oncogenic KRASG12C mutation. However, it remains unclear why some patients show upfront resistance to the KRASG12Ci, and the mechanisms of adaptive resistance. To enhance the efficacy of the KRASG12Ci, several combination regimens are in clinical trials which includes Sotorasib combination with MEK, PD1, SHP2, pan-HER, PD-L1, and EGFR inhibitors. However, it remains unclear how to precisely identify an appropriate cohort for analysis and predict tumor-centric combination strategies that would be effective in the clinic. We hypothesized that subsets of KRASG12C lung cancer could be identified using transcriptomics and proteomics. To test our hypothesis, we performed hierarchical clustering of the microarray dataset on 87 NSCLC KRASG12C tumors and identified three novel transcriptomic subtypes - Subtype 1 had higher expression of extracellular matrix (ECM) genes, which are related to cellular adhesion, and lower expression of EMT genes. Subtype 2 had low ECM expression, high expression in EMT and cell cycle pathway genes suggesting that this is a more invasive subtype. Subtype 2 had lowest expression of small molecule metabolism, suggesting it may be more responsive to KRASG12Ci. Subtype 3 had the lowest expression in interferon gamma/cytotoxic T-cell, T-cell, and B-cell pathways. Subtype 3 had the highest expression of small molecule metabolism, suggesting that it could be more resistant to KRASG12Ci. Having observed these unique clusters, we next hypothesized that mass spectrometry-based expression and phosphoproteomic analysis (pY and pS/T/Y) will further refine these clusters and help identifying new subsets. Our previous work by Stewart et al., demonstrated the importance of proteomics integrated with genomic and transcriptomic analyses to define molecular subtypes in a cohort of 108 squamous cell lung cancer patients. We believe that a similar integrated proteogenomic study is important in context of KRASG12C tumors to understand the signaling diversity and phenotypically categorized them to optimize patient enrichment schemes for effective combination therapy. For that purpose, we performed an integrative proteogenomic analysis of 75 KRASG12C tumors (55 lung, 12 large bowel adenocarcinomas and 8 PDXs) with full clinical follow up and pathology data. We are presently classifying the 75 KRASG12C tumors into phosphoproteomic subtypes. We also carried out targeted exome sequencing for cancer-specific genes. Next, we plan to leverage existing gene expression microarray data and add exome sequencing proteomics, and phosphoproteomics for a comprehensive, protein-centric characterization on these tumors. Importantly, the inclusion of 8 PDXs being co-processed and co-analyzed along with the tumors will enable rapid follow-up experiments to validate the findings into preclinical therapy trials. Citation Format: Hitendra Singh Solanki, Paul A. Stewart, Eric A. Welsh, Victoria Izumi, Bin Fang, Sean J. Yoder, John M. Koomen, Eric B. Haura. Proteomic and transcriptomic subtypes of KRASG12C mutant lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3911.