Determination of exosomal cancer stem cell biomarkers with cancer diagnostic utility

Resumen

Malignant melanoma (MM) is the most aggressive and life-threatening form of skin cancer and its incidence rapidly continues increasing worldwide. This type of cancer is often curable, but only if detected early. When diagnosed in early stages of disease (I and II), it has an excellent prognosis, but the mortality rate drastically rises in advanced stages (III and IV) after distant spread. Despite the progress made in recent years, diagnosis of MM remains challenging. To date, there are no suitable clinical diagnostic, prognostic or predictive biomarkers for this neoplasia. Several candidate biomarkers have been proposed but few have reached clinical application. Thus, there is still an urgent need for discovering specific useful biomarkers and for developing methods that can sensitively detect this neoplasia at the earlier pre-metastatic stages. It is known that the extraordinary metastasis capacity and chemotherapy resistance of this type of cancer are mainly due to intratumoral heterogeneity and, specifically, to MM cancer stem cells (CSCs). These cells represent a subpopulation of cancer cells which possess stem-like functional properties such as self-renewal ability and multipotency, and they are responsible for tumor initiation, progression, therapy resistance, relapse and metastasis. In addition, accumulating evidence supports the idea of the multiple roles that exosomes play in cancer’s biology. These small extracellular vesicles are involved in cell-to-cell communication and, by transferring their cargo, cancer cell-derived exosomes are able to induce in target cells diverse signalling pathways involved in tumor initiation, sustenance, progression and metastasis. These tiny messengers are actively released by all cells, including CSCs, to the extracellular matrix and to the bloodstream, making of them an interesting source of circulating biomarkers that can be easily isolated from the different body fluids by liquid biopsies. In this context, several omics technologies and analytical platforms, such as metabolomics and LC-HRMS, represent powerful tools for detecting and analyzing the expression levels of circulating metabolites in liquid biopsies that might serve as potential biomarkers not only for cancer, but also for other diseases. Taking all these facts into account, the main goal of this study was to identify potential CSC-derived exosomal and serological biomarkers using metabolomics techniques (LC-HRMS) with diagnostic, prognostic or predictive value for malignant melanoma. First, a patient-derived MM cell population enriched in CSCs was appropriately characterized. Next, exosomes derived from those MM-CSCs and serums from patients with MM were also characterized and their metabolomic profile was analyzed by LC-HRMS following an untargeted approach and applying univariate and multivariate statistical analyses. These analyses revealed significant metabolomic differences between exosomes derived from MM-CSCs compared to those from MM differentiated cells, as well as between serum-derived exosomes from patients with MM at several stages of disease and those derived from healthy individuals. Various metabolites showed a clear differential pattern of expression across the different sample groups of comparison and some of them were tentatively identified. Interestingly, we identified similarities in some structural lipids from both CSC-derived exosomes and those derived from patients with MM, such as the glycerophosphocoline PC 16:0/0:0. According to biomarker evaluation models, all those differential metabolites showed an excellent discrimination capacity between the two groups of samples (patients with MM and healthy controls), suggesting that they could be considered as potential exosomal biomarkers for this disease. Taking together, these results proved that metabolomic characterization of CSC-derived exosomes sets an open door to the discovery of clinically useful biomarkers for MM diagnosis. Along the same line and aiming at the same objective, another similar metabolomic study was also carried out but using a larger sample size. A total of 105 serum samples from both healthy individuals and patients with MM were analysed by LC-HRMS, following a similar nontargeted approach. Here again, significant metabolomic differences were found between patients in stage I of disease and healthy controls, and 10 candidates were consequently selected and identified as different classes of glycerophospolipids such as glycerophosphoethanolamines (PEs), glycerophosphocholines (PCs), cytidine diphosphate diacylglycerol (CDP-DG), or oxidized glycerophospholipids (POB-PS and PKOHA-PG). They all were suggested as potential biomarkers for early diagnosis of MM, since they yielded remarkable diagnostic capacities based on ROC curves analyses and biomarker evaluation models. All together, the results presented herein provide evidence that metabolomics and LCHRMS analytical platforms represent powerful tools for the identification and quantification of exosomal and serological metabolites in liquid biopsies, which could serve as potential biomarkers not only for MM, but also for other types of cancer. Given we are living in the era of effective molecular targeted treatments and immunotherapies, the discovery, validation and implementation into clinic of suitable biomarkers is more necessary than ever. Consequently, more funding and research are needed in order to find suitable biomarkers that could aid or improve early and accurate MM diagnosis. In this regard, untargeted LC-MS-based metabolomics applied to liquid biopsies and circulating compounds could pave the way for those achievements, since studies like this demonstrate its value for biomarker research, as well as its general applicability for translational research and precision medicine in oncology.