Lipidomic profiling of extracellular vesicles from breast and metastatic triple-negative breast cancer cell lines for identification of potential biomarkers.

Triple-negative breast cancer (TNBC) remains a highly aggressive disease with limited therapeutic options, and metastasis is the leading cause of patient mortality which highlights a major unmet medical need. Nevertheless, the mechanisms regulating metastatic progression remain poorly understood. Extracellular vesicles (EVs) have recently emerged as key mediators of the intercellular communication. Ubiquitously secreted, these nanoparticles provide a convenient and informative glimpse into cellular physiology given that their molecular composition shall reflect the state of their parent cell. Therefore, in this study, we aim to show that lipid profiles of EVs may differentiate malignant from non-malignant cells and to reveal biomarkers associated with TNBC. For this reason, we investigated the lipid composition of EVs released by cancer cells (MDA-MB-231, MDA-MB-453 and MDA-MB-468 representing TNBC) and non-malignant (MCF10A) human breast epithelial cells, with the goal of highlighting specific lipid profiles that might serve as potential biomarkers for TNBC. EV isolation and characterization were performed in accordance with MISEV2023 guidelines, while lipidomic profiling was carried out using an untargeted liquid chromatography and high-resolution mass spectrometry approach. We annotated over 500 complex lipid species. Multivariate analysis was used to explore major trends in the dataset, separating EVs derived from cancerous and non-cancerous cells and capturing differences among TNBC-derived EVs, reflecting their distinct genetic background. Further supervised modelling revealed distinct TNBC lipidomic signatures including phosphatidylcholines with alkyl chains of C18:0_C19:1, C20:1_C22:5, C18:1_C22:6 or C15:0_C22:5, C17:0_C20:3 and phosphatidylserine C32:0 which consistently emerged as shared lipid species across TNBC-derived EVs, highlighting their potential as biomarkers for the detection of TNBC. Furthermore, we highlighted lipid subsets specific to each of the investigated TNBC's source of EVs for in-depth molecular subtyping of TNBC.