Identification of proteomic and metabolic signatures associated with chemoresistance of human epithelial ovarian cancer

Emerging drug resistance in epithelial ovarian cancer (EOC) thwarted progress in platinum-based chemotherapy, resulting in increased mortality, morbidity and healthcare costs. The aim of this study was to detect the responses induced by chemotherapy at protein and metabolite levels, and to search for new plasma markers that can predict resistance to platinum-based chemotherapy in EOC patients, leading to improved clinical response rates. Serum samples were collected and subjected to proteomic relative quantitation analysis and metabolomic analysis. Differentially expressed proteins and metabolites were subjected to bioinformatics and statistical analysis. Proteins that played a key role in platinum resistance were validated by western blotting and enzyme-linked immunosorbent assay (ELISA). Metabolites that were the main contributors to the groups and closely with clinical characteristics were identified based on the database using nuclear magnetic resonance (NMR). In total, 248 proteins from two independent experiments were identified using isobaric tags for relative and absolute quantitation (iTRAQ)-based quantitative proteomic approach. Among them, FN1, SERPINA1, GPX3 and ORM1 were chosen for western blotting and ELISA validation. Platinum resistance likely associated with differentially expressed proteins and FN1, SERPINA1 and ORM1 may play a positive role in chemotherapy. HPLC-MS analysis of four groups revealed a total of 25,800 metabolic features, of which six compounds were chosen for candidate biomarkers and identified based on the database using NMR. The metabolic signatures of normal control (NC), platinum-sensitive (PTS) and platinum-resistant (PTR) groups were clearly separated from each other. Those findings may provide theoretical clues for the prediction of chemotherapeutic response and reverse of drug resistance, even lead to novel targets for therapeutic intervention.