Non-targeted metabolomic analysis predicts the therapeutic effects of exenatide on endothelial injury in patients with type 2 diabetes.

AIMS: We aimed to investigate whether treatment with exenatide could ameliorate endothelial injury in patients with type 2 diabetes mellitus (T2DM), and to identify biomarkers for predicting amelioration of the endothelial injury induced by the treatment. METHODS: Ninety-three patients with T2DM were recruited and treated with exenatide for 16 weeks. Enzyme-linked immunosorbent assays were performed at baseline and after the treatment to measure serum levels of endothelial injury markers, including soluble thrombomodulin (sTM). Patients were categorized as responders (n = 47) or non-responders (n = 46) based on median changes in their sTM levels. Serum levels of metabolites at baseline were measured with non-targeted liquid chromatography-mass spectrometry. The results obtained were evaluated with multivariate analysis. RESULTS: Treatment with exenatide for 16 weeks resulted in reduced body weight and improved levels of fasting plasma glucose, 2-hour postprandial plasma glucose, and HbA1c in patients with T2DM (all P < 0.05). Compared with baseline, serum levels of endothelial injury markers including sTM were significantly lowered after the treatment. Metabolites presented at significantly different levels in responders versus non-responders were considered as biomarkers for a therapeutic response of sTM to the exenatide treatment. Among those identified, 4-hydroxyproline and 12-oxo-9(Z)-dodecenoic acid were found to correlate most closely with the exenatide-induced endothelial protection response. The specificity and sensitivity of the multi-metabolite signature model contained higher 4-hydroxyproline and lower 12-oxo-9(Z)-dodecenoic acid were 53.3% and 92.3%, respectively, and the area under receiver operating characteristic curve was 69.2% (P < 0.001). CONCLUSIONS: Treatment with exenatide for 16 weeks ameliorates endothelial injury in patients with T2DM. Endothelial protection benefit from exenatide treatment was effectively predicted by the specific metabolomic combination of higher 4-hydroxyproline and lower 12-oxo-9(Z)-dodecenoic acid.