LAUSR

An early genotype-to-phenotype relationship in cancer personalized medicine was that KRAS mutations confer resistance to EGFR inhibitors in colorectal cancer. Confusingly, phase 3 clinical trials found that patients with the KRAS G13D mutation were an exception to this rule. That constitutively active KRAS mutants would cause resistance to EGFR inhibitors seemed intuitive, and it appeared inconsistent with known mechanisms of RAS biology for any constitutively active KRAS mutant to behave differently. We have developed a computational model that describes the biochemical mechanisms of RAS signal regulation. Simulations of this mass-action model reproduce known patterns of RAS pathway activation. The incorporation of the specific biochemical parameters for different KRAS mutants has allowed us to uncover novel, non-obvious insights into the response of KRAS G13D and KRAS G12C colorectal cancers to EGFR inhibition. For KRAS G13D, the model revealed EGFR inhibition leads to reduced wild-type HRAS and NRAS signaling in KRAS G13D cancers, but not KRAS G12D or KRAS G12V cancers. The model also revealed a critical dependence upon tumor suppressor NF1, and that the impaired binding of KRAS G13D to NF1 frees NF1 to act on wild-type RAS. Importantly, our computational predictions have been prospectively validated experimentally. Our computational model has also revealed that cotreatment of KRAS G12C colorectal cancer with EGFR and KRAS G12C inhibitors results in a depletion of the active forms of both wild-type and mutant RAS proteins, which we have also confirmed experimentally. For both KRAS G13D and KRAS G12C colorectal cancers, our work demonstrates how computational models can identify behaviors that logically follow from what is known, but which the complexity of the RAS signaling network has prevented from previously being uncovered despite extensive, empirical, experimental efforts. This abstract is also being presented as Poster B51. Citation Format: Edward C. Stites. Mechanism-based computational models to study interplay between EGFR inhibitors and KRAS mutants [abstract]. In: Proceedings of the AACR Special Conference on the Evolving Landscape of Cancer Modeling; 2020 Mar 2-5; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2020;80(11 Suppl):Abstract nr PHB05.