29 | The extracellular vesicle potential for the early detection of resistance to immunotherapy in melanoma patients

Background: Although immunotherapy has revolutionized melanoma management, there is an urgent need to find useful biomarkers to follow disease evolution and response to therapy. Extracellular Vesicles (EVs) have been emerging as a promising biomarker giving their established role in tumor progression, with specific regard to the critical role in tumor escaping from immune surveillance.Methods: Twenty patients with unresectable stage III or stage IV melanoma, treated with anti-PD-1 therapy as first-line treatment were enrolled and followed for up to 24 months. A longitudinal screening at different timepoints was applied to explore the usefulness of PD-L1-positive circulating melanoma-derived EVs in monitoring patient response. Melanoma EVs were counted and characterized directly from a few microliters of diluted patient plasma, taking advantage of the ExoView platform, an innovative technology that immunocaptures tetraspanins positive EVs and, then, evaluates the presence of EV surface-specific antigens through a well-defined cocktail of antibodies (CD146, Mart1, PD-L1).Results: EVs revealed a size of 55 to 116 nm in diameter for both melanoma patients and a cohort of healthy donors (HD). Nevertheless, a higher number of total EVs was found at the baseline in non-responding (NR) patients when compared to those found in HD and responding (R) patients. The pre-treatment levels of total PD-L1- or melanoma antigen positive EVs were found to be higher in NR than in R patients. Moreover, the pre-treatment levels of tumor-derived (CD146+, Mart1+) PD-L1-positive EVs were higher in NR than in R patients.Conclusions: This pilot study provides proof-of-principle of the possibility to directly analyze EVs in the whole plasma with no need of previous purification, thus avoiding issues related to EV isolation and contamination. Moreover, a subpopulation of circulating melanoma EVs was identified as PD-L1-positive, which could be used as a biomarker to distinguish between non-responders and responders. Once a sufficient number of patients is available, it will be possible to stratify them by disease stage to explore potential differences in the levels of tumor-derived PD-L1-positive EVs.