Project Summary/Abstract Colorectal cancer (CRC) kills more than 50,000 Americans each year. Fluorouracil-based therapy remains the standard of care and there have been no targeted therapies approved for use in CRC in the past half decade. Mutational activation of the KRAS oncogene – which occurs more than in 40% of cases – is a major source of intrinsic and acquired resistance to both conventional and targeted therapies in CRC. Since there are no effective therapies that directly or indirectly target K-Ras or its downstream effector pathways, KRAS mutation is the single greatest barrier to medical treatment for CRC. Large scale sequencing of cancer genomes has revealed that, among those 40% of CRCs that express mutant K-Ras, the diversity of KRAS alleles is greater than in any other type of cancer. Epidemiological studies demonstrate that survival and response to therapy varies depending on the KRAS genotype of the patient's cancer, suggesting that different mutant forms of the K-Ras oncoprotein could exhibit distinct oncogenic properties. Experimental validation of allele-specific behaviors has never been achieved, however. We will use primary human and mouse organoids and genetically engineered mouse models to address three key questions relating to K-Ras oncogenicity: (1) Are different mutant forms of K-Ras equivalent in their ability to promote colorectal cancer initiation and progression? (2) Are genetic interactions between KRAS and other genes allele-specific? (3) How do mutant forms of K-Ras influence the tumor microenvironment in a non-cell- autonomous manner to promote cancer progression? The ultimate goal of this work is to decipher the “KRAS Allele Code” in order to identify therapeutic strategies that are effective for cancers expressing specific K-Ras mutants. Precision medicine, where a physician tailors a patient's therapy to the genes that are mutated in his/her cancer, requires this level of understanding, especially for mutant oncoprotein that, like K-Ras, cannot be targeted with direct inhibitors.