Abstract 2136: Diversity across the pancreatic ductal adenocarcinoma disease spectrum revealed by network-anchored functional genomics

Pancreatic ductal adenocarcinoma (PDAC) is an incurable disease characterized by poor survival, dense desmoplastic stroma and activating mutations in KRAS (<90%). These tumors are highly complex ecosystems composed of molecularly distinct sub-populations that exhibit a spectrum of genetic features and associated phenotypes. Despite recent advances in the transcriptomic characterization of PDAC into at least two tumor subtypes, this alone has been insufficient to define more specific patterns of oncogenic dependency. In order to substantially improve clinical outcomes, there is a need to comprehensively define inter- and intra-tumor phenotypic diversity and understand the genetic dependencies that underlie these discrete molecular sub-populations. To this end, we integrated CRISPR-based co-dependency annotations with a disease-specific co-expression network developed from patient-derived models to establish CoDEX, a framework to quantitatively associate gene-cluster patterns with genetic vulnerabilities. Utilizing CoDEX, we defined multiple prominent anti-correlating gene-cluster signatures and pathway-specific dependencies, both across genetically distinct PDAC models and intratumorally at the single-cell level. Of these network-defined cluster trends, one differential signature recapitulated the characteristics of classical and basal-like PDAC molecular subtypes on a continuous scale. Anchoring CRISPR-defined genetic dependencies within the gene-cluster signature highlighted fundamental vulnerabilities associated with transcriptomic signatures of PDAC subtypes. Subtype-associated dependencies were subsequently validated across multiple PDAC models intratumorally. For this, we utilized direct-capture scRNAseq to analyze the effect of CRISPR mediated knockout of basal-like associated dependencies (SMAD4, ILK, and ZEB1) on differential signature representation. Silencing these network-prioritized targets resulted in a significant and directional clonal shift toward the more indolent classical-like signature, with the knockout phenotype only observed in tumor models with a predominant basal-like population. These results demonstrate the utility of CoDEX as a novel and quantitative approach for characterizing specific genetic dependencies within defined molecular contexts, with the potential to guide future clinical positioning for targeted therapeutics. Citation Format: Sanjana Srinivasan, Johnathon Rose, Wantong Yao, Sahil Seth, Michael Peoples, Annette Machado, Chieh-Yuan Li, I Lin Ho, Jaewon J. Lee, Paola A. Guerrera, Eiru Kim, Mustafa Syed, Joseph Daniele, Angela K. Deem, Michael Kim, Christopher A. Bristow, Eugene Koay, Giannicola Genovese, Andrea Viale, Timothy P. Heffernan, Anirban Maitra, Traver Hart, Alessandro Carugo, Giulio Draetta. Diversity across the pancreatic ductal adenocarcinoma disease spectrum revealed by network-anchored functional genomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2136.