A systems-level omics and hallmark–based analysis of pancreatic cancer for the identification of potential multi-target drugs.

e16441 Background: Pancreatic cancer is a highly aggressive malignancy with limited therapeutic options. Advances in omics technologies enable comprehensive characterization of molecular alterations, while the cancer hallmarks framework provides a biologically informed basis for identifying potential therapeutic targets. Methods: We collected the following data for pancreatic cancer: transcriptomic data from The Cancer Genome Atlas (TCGA), proteomic data from the Human Protein Atlas (HPA) and Clinical Proteomic Tumor Analysis Consortium, survival data from HPA and TCGA, and genetic alteration from the Genomic Data Commons Data. We also collected the frequencies of Single Nucleotide Polymorphism (SNP) and copy number variation (CNV). Next, curated contributions of each omics to ten cancer hallmarks (e.g. self-sufficiency in growth) were scored, combining omics layer and their contribution to each hallmark, yielding cumulative scores that were combined multiplicatively for gene prioritization. Prioritized genes were filtered using curated lists of druggable genes from the Human Protein Atlas and the Drug–Gene Interaction Database. Potential therapeutic compounds targeting these genes were then identified using Enrichr with the MAGMA drug database. Results: Survival and transcriptomics data, proteomic data, and CNV/SNP data were collected from 176, 235, and 420 pancreatic cancer patients, respectively. TP53, COL17A1, S100P, PLA2G1B, and CLPS showed the highest omics scores, while MAP2K1, KRAS, AKT1, TP53, and HRAS exhibited the highest genetic alteration scores. Following multiplicative integration of omics and hallmark scores and evaluation of therapeutic target candidacy, TP53, KRAS, MET, RAC1, PIK3CB, ERBB2, CCND1, and FAS emerged as the top-ranked genes. Interestingly, a phase three trial for a new drug targeting RAS in patients with pancreatic cancer is in progress. The results of Enrichment analysis with the top 25 genes are listed in Table 1. Conclusions: This multi-omics and hallmark-directed analysis identifies key potential therapeutic targets in pancreatic cancer, highlighting potential utility of biologically-informed multi-target strategies in new drug discovery. Top-scoring therapeutic targets identified in pancreatic cancer and their potential interacting compounds. Name Combined Score Affected Genes CDK Inhibitor 27481 CCND1;CDKN2A;ERBB2;KRAS;ESR1;TP53 Abemaciclib 27481 CCND1;CDKN2A;ERBB2;KRAS;ESR1;TP53 Osimertinib 25687 ERBB2;KRAS;MET Crizotinib 23266 NPM1;SMAD4;CDKN2A;ERBB2;KRAS;TP53;MET ALK Inhibitors 22887 NPM1;SMAD4;CDKN2A;MET Alectinib 22887 NPM1;SMAD4;CDKN2A;MET Vesnarinone 11950 FAS;TP53;THBS1 Olaparib 7691 MYC;ERBB2;KRAS;ESR1;TP53;MET Venetoclax 7243 NPM1;BCL2;KRAS;TP53 BCL Inhibitor 7243 NPM1;BCL2;KRAS;TP53