MAPK inhibitors dynamically affect melanoma release of immune NKG2D-ligands, as soluble protein and extracellular vesicle-associated

Abstract

Metastatic melanoma presents, in many cases, oncogenic mutations in BRAF, a MAPK involved in proliferation of tumour cells. BRAF inhibitors, used as therapy in patients with these mutations, often lead to tumour resistance and, thus, the use of MEK inhibitors was introduced in clinics. BRAFi/MEKi, a combination that has modestly increased overall survival in patients, has been proven to differentially affect immune ligands, such as NKG2D-ligands, in drug-sensitive vs. drug-resistant cells. However, the fact that NKG2D-ligands can be released as soluble molecules or in extracellular vesicles represents an additional level of complexity that has not been explored. Here we demonstrate that inhibition of MAPK using MEKi, and the combination of BRAFi with MEKi in vitro, modulates NKG2D-ligands in BRAF-mutant and WT melanoma cells, together with other NK activating ligands. These observations reinforce a role of the immune system in the generation of resistance to directed therapies and support the potential benefit of MAPK inhibition in combination with immunotherapies. Both soluble and EV-associated NKG2D-ligands, generally decreased in BRAF-mutant melanoma cell supernatants after MAPKi in vitro, replicating cell surface expression. Because potential NKG2D-ligand fluctuation during MAPKi treatment could have different consequences for the immune response, a pilot study to measure NKG2D-ligand variation in plasma or serum from metastatic melanoma patients, at different time points during MAPKi treatment, was performed. Not all NKG2D-ligands were equally detected. Further, EV detection did not parallel soluble protein. Altogether, our data confirm the heterogeneity between melanoma lesions, and suggest testing several NKG2D-ligands and other melanoma antigens in serum, both as soluble or vesicle-released proteins, to help classifying immune competence of patients.