Extracellular vesicle–derived urinary metabolomes show distinctive changes with breast cancer

Abstract Introduction Breast Cancer (BC) remains a significant clinical challenge, and despite well-established screening strategies, new biomarkers could improve BC detection, treatment and management. Urine represents a headily accessible liquid biopsy for diagnosis and extracellular vesicle (EV) transfer of oncogenic proteins, RNAs, and metabolites that promote tumor growth, invasion, metastasis, and immune evasion. Aims To compare the whole urine and urinary EV metabolomes and identify BC specific metabolite changes. Methodology Urine samples were collected from four participant groups: breast cancer (BC) patients (n = 42), individuals with breast benign disease (BBD; n = 3), symptom controls (SC; n = 4), and healthy controls (HC; n = 6). EVs were isolated using differential centrifugation, ultrafiltration, and size-exclusion chromatography (SEC), and their morphology was confirmed by transmission electron microscopy (TEM). Metabolites from whole urine and from EVs derived from the same samples were extracted using methanol–water (70:30, v/v) and analyzed by direct-infusion mass spectrometry (DI-MS) in both positive and negative ESI modes. Metabolic features were processed with BinneR and annotated using the HMDB and KEGG databases. Integrated multi-omics analysis of whole-urine and EV-associated metabolomes was performed using the DIABLO framework within the MixOmics package in R platform. Results DI-MS profiling detected a broad spectrum of metabolites in both whole–urine and EV–derived fractions. Multivariate analyses revealed a clear separation of breast cancer (BC) patients from healthy controls and non–cancer groups in both matrices. Whole EV metabolites with area under the curves (AUC) of > 0.7 included glyceryl phosphoryl derivatives, N-eicosapentaenoyl species, sphinganine-1-phosphate and tetracosahexaenoic acid. EV–enriched metabolites included carnitine, histidine and adenosine monophosphate. DIABLO–based integrative analysis suggested that urinary and EV metabolomes were broadly similar with the discrete putative metabolite biomarkers representing minor, but specific changes with BC. Conclusions The whole urine and EV metabolomes suggested a small number of metabolite changes that were specific to BC. This could indicate that the urinary EVs describe distinctive aspects of the breast carcinogenic process.