Abstract 3169: Dependency-agnostic drug discovery: A dual-screening strategy to uncover novel cancer vulnerabilities

The current landscape of small-molecule drug discovery prioritizes genetic dependencies considered as viable protein targets. While this approach has driven significant therapeutic advancements, it has also constrained the exploration of non-dependency targets harboring tractable vulnerabilities. Moreover, many genetic dependencies remain undruggable, precluding their therapeutic exploitation and leaving critical gaps in addressing cancers with unmet clinical needs. To overcome these limitations, we developed a tandem screening approach that integrates large-scale chemical and genetic screens to identify novel vulnerabilities independent of traditional dependency paradigms. Using PRISM, a multiplexed drug-screening platform, we identify compounds with selective toxicity against specific cancer lineages, revealing novel biology readily amenable to small-molecule drug discovery. Subsequent genome-wide CRISPR screens elucidate the mechanistic underpinnings of these compounds, offering actionable insights into new therapeutic opportunities. We applied this approach to Ewing sarcoma, an aggressive bone tumor characterized by the fusion oncoprotein EWS-FLI1, a known dependency that drives epigenetic reprogramming and disease progression. Despite its critical role in disease initiation and maintenance, therapeutic targeting of EWS-FLI1 has remained elusive, underscoring the urgent need for alternative strategies. From our PRISM screen of 4,518 compounds across 935 cancer cell lines, including 17 Ewing sarcoma lines, we identified several compounds with selective toxicity against Ewing sarcoma. Chemical clustering and mechanism-of-action analysis revealed diverse chemical motifs and biological targets, none of which were directly linked to Ewing sarcoma dependencies. A CRISPR rescue screen with three top compounds identified a shared vulnerability: UGT3A2, a UDP-glucose transferase specifically expressed in Ewing sarcoma under the transcriptional control of EWS-FLI1.Our mechanistic studies revealed that UGT3A2-mediated glycosylation of these compounds depletes intracellular UDP-glucose, disrupting glycolysis and triggering ER stress via the PERK-EIF2A pathway, culminating in cell death. This unique vulnerability of UGT3A2, while not a dependency itself, represents a tractable target for therapeutic exploitation in Ewing sarcoma. Building on these findings, we have developed a library of lead compounds exploiting the vulnerability imposed by UGT3A2 in Ewing sarcoma and are currently advancing them in later-stage preclinical optimization. This study underscores the potential of our tandem screening approach to uncover vulnerabilities in cancers with undruggable targets and unmet clinical needs. Beyond Ewing sarcoma, this platform offers a promising strategy to expand the therapeutic landscape for difficult-to-treat malignancies. Citation Format: Olasunkanmi Olaoye, Colleen Harrington, Kenneth N. Ross, Allen T. Basanthakumar, Jamie O. Brett, Shakchhi Joshi, Jonathan H. Shrimp, Min Shen, Lucy Merickel, Audrey Taillon, Silvi Salhotra, Bo Kyung A. Seong, Ji Seul Han, Yara R. Zabala, Ashleigh Meyer, Qi Liu, Luc de Waal, Jun Qi, Matthew D. Hall, Marcia C. Haigis, Jennifer Roth, Kimberly Stegmaier. Dependency-agnostic drug discovery: A dual-screening strategy to uncover novel cancer vulnerabilities [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 3169.