Abstract A08: in vivo CRISPR screening enables precision medicine by identifying novel drug combinations and modeling anticancer drug sensitivity

Identifying novel targets to synergistically improve the efficacy of existing chemotherapy is a major goal in cancer research. Here, we report in vivo CRISPR knockout (KO) screening in a patient derived xenograft (PDX) model of pancreatic cancer to identify genes that mediate the resistance to trametinib, an inhibitor of the MEK signaling pathway, which is aberrantly active in pancreatic cancers due to oncogenic KRAS mutations. The screening and subsequent genetic as well as in vitro and in vivo pharmacological validations identified a set of kinetochore function genes whose depletion is synthetic lethal with MEK inhibition. by triggering mitotic slippage and polyploidy. Notably, using the unbiased knowledge of gene-specific survival scores from the in vivo KO screening under the pressure of a chemotherapeutic agent, we devised a novel Drug Response Evaluation approach Based on in vivo CRISPR screening (DREBIC). The DREBIC is a major detour from conventional machine learning and statistical prediction models as it integrates functional screening based gene viability scores and basal expression levels to project cancer type- and mutation-specific drug sensitivity profiles. Using large-scale experimentally identified drug-response data from the Cancer Genome Project (CGP) and Cancer Cell Line Encyclopedia (CCLE) as a gold standard, DREBIC models trametinib response with high sensitivity and specificity across thousands of cancer cell lines. DREBIC identifies intrinsic fitness p...
Source: Molecular Cancer Therapeutics - Category: Cancer & Oncology Authors: Tags: New Technology and Bioinformatics: Poster Presentations - Proffered Abstracts Source Type: research