Abstract 3643: AI driven discovery of novel synthetic lethal gene pairs for targeted cancer therapy

Abstract Synthetic lethality (SL) offers a promising avenue for cancer therapy by exploiting vulnerabilities unique to tumor cells and minimizing off-target effects on normal tissue. However, the application of SL-based therapies remains constrained by a limited repertoire of validated targets and biomarkers. Using data from the Cancer Dependency Map (DepMap) and advanced in silico modeling, Evariste’s Frobenius Target platform identified 50 potential SL gene pairs with druggable nodes and rational mechanisms linking targets and biomarkers. From this set, 24 gene pairs were prioritized for validation based on translational relevance and impact on the patient population. Validation was achieved through a collaborative high-throughput workflow with EditCo Bio, leveraging their CRISPR-based cell line engineering platform to generate 72 knockout pools in under three weeks. Validation studies in multiple cell lines demonstrated >10 high-confidence SL pairs, including a novel SL relationship involving focal adhesion kinase (FAK). To further substantiate this relationship, novel inhibitors were developed with best-in-class selectivity over the SL partner gene. Functional assays confirmed these inhibitors’ effectiveness in biomarker-low cell lines, with promising implications for clinical translation. This study highlights the potential of AI-driven approaches to uncover the next generation of SL targets, expanding the therapeutic landscape in cancer research. By focusing on biomarker-low gene expression as a predictor of SL, our work establishes a robust framework for the discovery and validation of clinically actionable SL relationships. Citation Format: Amalia Soenens, Daniel Miller, Oliver Vipond, Florencia Skuras, Alfie Brennan, Anna Hercot. AI driven discovery of novel synthetic lethal gene pairs for targeted cancer therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3643.