评估卵巢癌中的旁系同源基因依赖性以推动靶点发现

Ovarian cancer is the deadliest gynecologic cancer, and the fifth leading cause of cancer-related mortality in individuals assigned female at birth. Though some targeted therapies are available for the treatment of ovarian cancer, there is an unmet need especially for non-RAS or BRCA1/2 mutant cases. Recent developments in CRISPR technology have allowed for high-throughput identification of genetic dependencies, accelerating the identification of new drug targets for cancer treatment. However, we and others have shown that many cellular dependencies are missed due to genetic redundancy within the human genome. Indeed, paralog genes can compensate for each other's functions and obscure gene-dependent phenotypes when using only single gene knockout tools. To assess paralog dependency, multiple genes must be knocked out at the same time. In this study, we developed the focused dual gene-targeting library pgMI, which is comprised of 3,800 paired-guide RNAs (pgRNAs) targeting 116 paralog pairs that have been identified as synthetic lethal in at least two published double knockout CRISPR screens. To aid in genetic interaction mapping, pgMI contains both single and double-targeting pgRNAs. In an initial analysis of six non-small cell lung cancer cell lines, we uncovered several cell line specific and pan-essential paralog dependencies. To assess paralog dependencies specific to ovarian cancer, we then performed pgMI CRISPR screens in three common ovarian cancer cell lines - SKOV3, JHOS4, and OVCAR8. These cell lines represent a diverse set of cancer driver genes, which include activating mutations in both the Wnt and RAS signaling pathways. pgMI screening uncovered paralog dependencies both unique and shared across ovarian cancer cell lines, providing a narrowed list of potential therapeutic targets for further study.