DESCRIPTION (provided by applicant): Mutations in the KRAS locus are amongst the most pervasive genetic alterations in human cancer. In colorectal cancer (CRC), KRAS mutations are not only prevalent (~42%) but crucially impact therapeutic response to EGFR inhibitors. Despite the attractiveness of KRAS as a therapeutic target, direct pharmacologic inhibition of constitutively active mutant KRAS or its downstream effector pathways has proven challenging, considerable effort and recent promising progress notwithstanding. Since KRAS mutant cancer cells may possess additional dependencies on either downstream pathways or acquire secondary mutations, such "synthetic lethal" vulnerabilities represent valuable therapeutic opportunities. Unfortunately, prior KRAS synthetic lethal screens have suffered from complications of reproducibility and off-target effects, compounded by potential context-dependent effects from the complex and heterogeneous 2D cancer cell lines employed. As explored in the current application, primary 3D colon organoid cultures represent a novel approach to identification of KRAS synthetic lethal interactions in CRC. Such organoids are particularly well characterized for colon and combine native differentiation and tissue architecture with experimental tractability. Importantly, they provide an unusual opportunity for mutant KRAS to be studied in the setting of primary human colon tissue with a limited subset of deliberately engineered oncogenic mutations opposed to transformed cell lines with potentially large numbers of confounding genetic alterations. Our goal is to explore KRAS synthetic lethality in colorectal cancer, via a synergistic effort from Multi- PIs Calvin Kuo and Kevan Shokat, the Co-Investigators Haian Fu, Michael McManus and Olivier Gevaert and collaborators Stanley Qi and Michael Bassik. In Aim 1, we generate "engineered" primary human colon organoids with CRISPR APC mutation or more complex genotypes ± KRAS mutation (G12D, G12C, G12V, etc.), as well as colorectal cancer patient-derived organoids from surgical specimens (CRC-PDO) with wild- type or mutant KRAS status. Aim 2 pursues activity-based proteomic characterization of kinases that are altered by mutant KRAS in engineered colon or CRC-PDO organoids ± mutant KRAS and functionally evaluates these kinases as synthetic lethal targets. In parallel, Aim 3 robustly multiplexes organoids into multiwell format to identify small molecule synthetic lethal compounds by high-throughput screening. Lastly, Aim 4 utilizes a complementary genetic approach to CRC KRAS synthetic lethality based on shRNA and CRISPRi-mediated gene silencing, both for orthogonal functional validation of synthetic lethal candidates arising in Aims 2 and 3, but also for unbiased druggable genome screens to identify novel synthetic lethal interactions. Overall, we present a highly integrated approach to KRAS synthetic lethality in CRC where novel organoid methods are directly coupled to proteomic, small molecule and genomic screening technologies and bound by multi-center cross-validation of candidate loci and small molecules.

PUBLIC HEALTH RELEVANCE: Mutations in the KRAS gene are a leading cause of colon cancer and can critically affect treatment efficacy. This application uses 3-dimensional "organoid" cultures to discover weaknesses that are specifically present in KRAS-mutant colon cancers that can be used to develop new therapies.