Abstract SY44-02: Proteogenomic and phosphoproteomic analysis of breast cancer

Abstract

Abstract The genetic landscape of human breast cancer has been well defined in The Cancer Genome Atlas (TCGA) project. Mass spectrometry (MS)-based global proteome and phosphoproteome analyses may provide an orthogonal approach to genomic studies to further improve the molecular taxonomy and our understanding of breast cancer. Central questions in breast cancer biology that will be addressed in this study are: (1) Which genomic characteristics are executed at the protein level? (2) How is the molecular taxonomy of breast cancer reinforced and revised by protein and phosphorylation data? and (3) What phosphorylation-driven signaling networks emerge from genetic alterations? We analyzed human breast cancer samples that have been previously genetically characterized by the TCGA project. Tumor samples were analyzed by global shotgun-proteomics and phosphoproteomics using iTRAQ4-plex peptide labeling. All mass spectrometry data was acquired using nearly 5,000 h of measurement time on a Q Exactive instrument and the data was analyzed in Spectrum Mill using patient-specific RNA-sequencing derived protein databases. In total we quantified >16,000 proteins and >70,000 phosphorylation sites, with an average of >12,000 quantified proteins and >20,000 phosphorylation-sites for each tumor. Only 1-2% of all 9,600 genetically encoded somatic single amino acid variants and 1-2% of 36,000 alternative splice junctions were detected at the protein level despite the very comprehensive proteome coverage obtained. While the global mRNA protein abundance correlation was rather low (Spearman's correlation of 0.4), we found very good correlation for most protein kinase gene amplifications for mRNA, protein and phosphoprotein abundance. Hierarchical clustering analysis of both the proteome and the phosphoproteome data yielded an overlapping set of three major clusters: a basal-enriched, a luminal-enriched and a normal-inclusive group. The two most recurrently mutated genes in human breast cancer are PIK3CA and TP53 at frequencies of 30-40%. Comparison of PIK3CA or TP53 mutated vs non-mutated tumors highlights specific phosphorylation signaling events downstream of mutated PI3-kinase and increased phosphorylation of cell cycle check point kinases in p53-mutated tumors. Network and pathway analysis is being performed to comprehensively integrate genetic and phospho-/proteomic alterations in one model. The effects observed in human tumors will be compared to a set of 24 patient-derived xenograft tumors that will allow drug efficacy studies in fully genetically characterized tumors in the future. Citation Format: Philipp Mertins, DR Mani, Karl Clauser, Michael Gillette, Pei Wang, Jana Qiao, David Fenyo, Kelly Ruggles, Sherri Davies, Bing Zhang, Michael Gatza, Sean Wang, Ping Yan, Chenwei Lin, Michael McLellan, Reid Townsend, Li Ding, Song Cao, Henry Rodriguez, Amanda Paulovich, Matthew Ellis, Steven A. Carr, Clinical Proteomics Tumor Analysis Consortium: CPTAC. Proteogenomic and phosphoproteomic analysis of breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr SY44-02. doi:10.1158/1538-7445.AM2015-SY44-02