Epithelial ovarian cancer (EOC) is the leading cause of gynecologic cancer-related death due to its nonspecific characteristics compared to benign cases and poor prognosis after conventional therapies. Small peptides (SPs) demonstrated to have potential for diagnosis and prognosis were focused on in our study for the discovery of biomarkers that address these issues. Metabolic profiles of 15 SPs, including nine dipeptides and six tripeptides were acquired from plasma samples of 140 EOC and 158 benign ovarian tumor (BOT) patients. Partial least square discriminant analysis showed separations between EOC and BOT subjects of different age brackets. Hyp-Leu, Glu-Trp and Phe-Phe were selected as promising predictive SP-biomarkers for better EOC diagnosis compared to conventional biomarkers. Combined Hyp-Leu, Glu-Trp and CA125 presented an area under the curve (AUC) of 0.904, with a sensitivity and specificity of 0.804 and 0.944, respectively. This finding suggested that the combination of these biomarkers performed much better than CA125 alone. Hyp-Leu and Gly-Phe-Trp showed significantly improved performances in the log-rank tests and Kaplan-Meier curves demonstrating their prognostic potential. All SP-biomarkers proved to have excellent stabilities at room temperature. Correlation network analysis implied latent conversions among amino acids, dipeptides and tripeptides during EOC. In conclusion, the selected SPs in combination with CA125 show profound promise for discriminating EOCs from BOTs and for predicting the progression after surgery, which provides invaluable information for clinicians in the precision diagnosis and treatment of EOC.
Keywords: biomarker; epithelial ovarian cancer; metabolic profiling analysis; small peptide.
© 2018 UICC.