Abstract B106: Personalized functional screens for gene drivers of pancreatic cancer

Abstract

Abstract Pancreatic ductal adenocarcinoma (PDAC) is a devastatingly lethal disease that remains one of the most challenging malignancies to treat successfully. The identification of oncogenic “driver” genes and their activated pathways has been the moving force behind the development of therapies for other cancers. Recognizing this, The Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC) are cataloging genomic aberrations in PDAC with the goal of identifying new therapeutic targets and detection biomarkers. These efforts have confirmed that the majority of PDAC tumors harbor early activating mutations in the KRAS oncogene, which is not druggable through current targeting modalities, along with other moderate to high frequency aberrations in TP53 and SMAD4 among others. We are now faced with the formidable challenge of identifying rare pathogenic aberrations from among the numerous biologically neutral “passengers” in these PDAC sequencing datasets, as identifying new drug-actionable events and understanding their mechanisms-of-action offers great promise for improving patient outcomes. However, comprehensive assessment of low frequency aberrations is difficult given their large number and the fact that they may either directly or indirectly influence tumor behavior through modifying activities of other drivers like KRAS. While RNAi-based screening platforms have successfully validated new tumor suppressors and other genetic liabilities in cancer, less progress has been made toward developing high-throughput, gain-of-function (GOF) screening systems for validating hyper-activated oncogenes that are especially attractive given the efficacy of antibody and small molecule inhibitor therapies tailored toward such factors. To address these challenges, we established a High-Throughput Mutagenesis and Molecular Barcoding (HiTMMoB) platform enabling GOF annotation of PDAC gene aberrations through (1) accurate modeling of numerous aberrations (amplifications, mutations, insertions, deletions) using our robotics-driven platform of >32,000 sequenced-verified open reading frame (ORF or “gene” clones) and (2) a molecular barcoding and sequencing-based detection strategy that permits rapid DNA tagging of wild-type and mutant ORFs for virus-based, pooled functional screens and ORF barcode quantitation. We employed HiTMMoB to build patient-specific aberration ORF sets based on ICGC sequencing data from four PDAC patients (i.e., all cloneable, uncharacterized somatic mutations reported in four separate tumors). These aberration ORFs were screened for their ability to promote in vivo tumorigenesis using a human pancreatic ductal epithelial cell line engineered with a doxycycline (DOX)-inducible KRASG12D allele, which after injection into mice on DOX diet leads to rapid tumor growth and tumor regression upon DOX withdrawal. Using this model we simultaneously screened for PDAC drivers that (1) cooperate with KRASG12D in mice on DOX, (2) drive tumor escape from DOX withdrawal (KRASG12D extinction) and (3) promote tumorigenesis in a KRAS-independent manner in the absence of DOX. This “personalized functionalization” approach coupled with our barcode detection strategy revealed two potent driver aberrations initially reported in two separate patients by the ICGC. Our screening approach, driver validation and mechanistic data support the notion that discovery of low frequency, functional aberrations may intersect or otherwise lead to important pathways representing known or novel therapeutic liabilities. When applied more broadly to aberration gene sets informed by biological importance and computational analyses, our functional screening technologies are revealing high priority PDAC targets to enroll in deep mechanistic biology studies and drug development programs with the ultimate goal of developing personalized treatment strategies critically needed for PDAC patients. Citation Format: Yiu Huen Tsang, Turgut Dogruluk, Hengyu Lu, Rosalba Minelli, Nikitha Nair, Marie-Claude Gingras, Agda Karina Eterovic, Gordon Mills, Kenneth Scott. Personalized functional screens for gene drivers of pancreatic cancer. [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Innovations in Research and Treatment; May 18-21, 2014; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2015;75(13 Suppl):Abstract nr B106.