Metabolic Profiles and Blood Biomarkers to Discriminate between Benign Thyroid Nodules and Papillary Carcinoma, Based on UHPLC-QTOF-ESI+-MS Analysis

Abstract

In this study, serum metabolic profiling of patients diagnosed with papillary thyroid carcinoma (PTC) and benign thyroid pathologies (BT) aimed to identify specific biomarkers and altered pathways when compared with healthy controls (C). The blood was collected after a histological confirmation from PTC (n = 24) and BT patients (n = 31) in parallel with healthy controls (n = 81). The untargeted metabolomics protocol was applied by UHPLC-QTOF-ESI+-MS analysis and the statistical analysis was performed using the MetaboAnalyst 5.0 platform. The partial least squares-discrimination analysis, including VIP values, random forest graphs, and heatmaps (p < 0.05), was complemented with biomarker analysis (with AUROC ranking) and pathway analysis, suggesting a model for abnormal metabolic pathways in PTC and BT based on 166 identified metabolites. There were 11 classes of putative biomarkers selected that were involved in altered metabolic pathways, e.g., polar molecules (amino acids and glycolysis metabolites, purines and pyrimidines, and selenium complexes) and lipids including free fatty acids, bile acids, acylated carnitines, corticosteroids, prostaglandins, and phospholipids. Specific biomarkers of discrimination were identified in each class of metabolites and upregulated or downregulated comparative to controls, PTC group, and BT group. The lipidomic window was revealed to be more relevant for finding biomarkers related to thyroid carcinoma or benign thyroid nodules, since our study reflected a stronger involvement of lipids and selenium-related molecules in metabolic discrimination.