Comparative metabolomics of serum and serum extracellular vesicles of melanoma and triple-negative breast cancer utilizing biobank samples

Abstract Extracellular vesicles (EVs), abundant in circulation and obtainable through liquid biopsy, carry metabolites that reflect both the activity of EVs and state of the cells of origin. Thereby, metabolomics offers a sensitive means to detect dynamic changes associated with cancer. We compared serum and matched serum-derived EV metabolite profiles from patients with melanoma or triple-negative breast cancer (TNBC) and age- and sex-matched cancer-free controls to evaluate the suitability of biobank serum samples and the corresponding EVs to identify cancer-type -associated metabolic alterations. For untargeted metabolomics and metabolite/pathway identification, EVs were isolated from biobank-derived serum samples by size-exclusion chromatography and ultrafiltration, and preanalytical variables of samples were monitored according to Minimal Information for Blood EV research guidelines. More metabolite differences were found between patients and controls in melanoma than in TNBC. Melanoma serum contained reduced levels of several metabolite groups (amino acids, lipids, and organic acids), whereas melanoma EVs were enriched in polyunsaturated phospholipids. In TNBC, serum contained increased levels of long-chain fatty acid amides, while EVs showed no significant metabolite alterations. These results demonstrate that serum and EVs offer complementary metabolite information, and biobank samples are suitable for EV metabolomics. Monitoring and recording preanalytical factors will enable improved cross-study comparability and biomarker validation.