Abstract Multiple Myeloma (MM) is an often incurable plasma cell proliferative disorder, representing about 10% of all hematologic malignancies. Osteolysis is one of the hallmarks of this malignancy, meaning that almost every patient will manifest with an osteolytic lesion during their disease course, resulting in increased morbidity and pain with ultimately a severe impact on the quality of life. The development and progression of MM is largely dependent on the bone marrow (BM) microenvironment wherein communication through different factors including extracellular vesicles (EVs) takes place. This crosstalk not only leads to drug resistance but also to the development of osteolysis. Targeting vesicle secretion could therefore simultaneously ameliorate drug response and bone disease. Here we examined the effects of MM exosomes on different aspects of osteolysis using the 5TGM1 murine model. This syngeneic murine model highly mimics human MM disease and presents with typical MM characteristics such as osteolysis, angiogenesis and a serum M-spike. We first investigated the effects of 5TGM1 small EVs or exosomes, sized 50-120nm in diameter, on osteoclasts and osteoblasts in vitro. 5TGM1 exosomes were able to enhance the resorptive active of osteoclasts. In contrast, these exosomes induced apoptosis in pre-osteoblasts while also blocking their differentiation to mature osteoblasts. RT-PCR showed a downregulation of Runx2, Osterix and Collagen 1A1 expression, while DKK-1 expression was upregulated. Mechanistically, we confirmed the presence of DKK-1 on the 5TGM1 exosomes, which led to a downregulation of the Wnt pathway in osteoblasts. In vivo, we uncovered that 5TGM1 exosomes could induce osteolysis in a similar pattern as the MM cells themselves. We injected C57BL/KalwRij intravenously during three weeks with 5TGM1 exosomes, and analyzed the femurs by micro-computed tomography. These mice had a significantly lower trabecular bone volume, as the result of a lower trabecular number combined with a higher trabecular separation. Next, we inhibited exosome secretion in 5TGM1-inoculated mice to prevent bone loss and increase sensitivity to bortezomib. For blocking exosome secretion we used the neutral sphingomyelinase inhibitor GW4869. This increased cortical bone volume and decreased bone resorption markers (C-terminal telopeptide of collagen type I) without significantly affecting tumor load. This indicates that inhibiting exosome secretion in the MM microenvironment has a protective effect on myeloma bone disease, not associated with effects on tumor load. Furthermore, blocking exosome secretion by GW4869 also sensitized the myeloma cells to bortezomib, leading to a strong anti-tumor response in vivo when GW4869 and bortezomib were combined. Altogether, our results indicate a key role for exosomes in the myeloma BM microenvironment and suggest a novel therapeutic target for anti-myeloma therapy. Disclosures No relevant conflicts of interest to declare.