Proteogenomic Characterization of Human Pancreatic Cancer

Abstract

Introduction: Recent genomic and transcriptomic analyses of human pancreatic ductal adenocarcinoma (PDAC) identified substantial genomic alterations and genes for which expression was altered with a cancer-related profile. However, these data have yet to provide molecular targets for diagnosis or clinical therapy. Here, we report proteogenomic analysis of PDAC. Methods: We collected PDAC tumor tissues and matched blood samples from consecutive 196 patients with PDAC who underwent surgery at Seoul National University Hospital from March 2010 to December 2016. DNA, RNA, and proteins were extracted from these samples, and proteogenomic analyses were performed. Subtypes of PDAC were identified using integrated clustering and pathway analysis. Functional experiments for polymorphonuclear myeloid-derived suppressor cells (PMN-MDSC) mediated inhibition of T cell proliferation were done using orthotopic pancreatic cancer models. Results: Mutation-phosphorylation correlations identified signaling pathways associated with somatic mutations in significantly mutated genes. mRNA-protein abundance correlations revealed oncogene and tumor suppressor candidates correlating with patient survival. Integrated clustering of mRNA, protein, and phosphorylation data identified six PDAC subtypes (Sub1-6), which were indistinguishable using mRNA data alone. mRNA and protein signatures defining Sub1-6 revealed that Sub1, Sub2-4, and Sub5-6 were precursor, invasive, and immunogenic tumors, respectively. In the Sub2-4 group, proliferation was highest for Sub4; Sub6 in the Sub5-6 group had an increased pancreatic secretion capacity. Orthotopic mouse PDAC models revealed higher numbers of pro-tumorigenic immune cells in Sub4 tumors, inhibiting T cell proliferation. Conclusions: Our proteogenomic analysis provides therapeutic targets and improves understanding of cancer biology and patient stratification in PDAC.