Metabolomic Serum Profiling Detects Early-Stage High-Grade Serous Ovarian Cancer in a Mouse Model

Abstract

Ovarian cancer is a deadly disease killing more than any other gynecologic cancer. Nonspecific symptoms, combined with a lack of early detection methods, contribute to late diagnosis and low five-year survival rates. High-grade serous carcinoma (HGSC) is the most common and deadliest subtype that results in 90% of ovarian cancer deaths. To investigate metabolic patterns for early detection of this deadly ovarian cancer, Dicer- Pten double knockout (DKO) mice that phenocopy many of the features of metastatic HGSC observed in women were studied. Using ultraperformance liquid chromatography−mass spectrometry (UPLC−MS), serum samples from 14 early-stage tumor (ET) DKO mice and 11 controls were analyzed in depth to screen for metabolic signatures capable of differentiating early-stage HGSC from controls. Iterative multivariate classification selected 18 metabolites that, when considered as a panel, yielded 100% accuracy, sensitivity, and specificity for classification. Altered metabolic pathways reflected in that panel included those of fatty acids, bile acids, glycerophospholipids, peptides, and some dietary phytochemicals. These alterations revealed impacts to cellular energy storage and membrane stability, as well as changes in defenses against oxidative stress, shedding new light on the metabolic alterations associated with early ovarian cancer stages.