MM is a hematologic malignancy characterized by clonal proliferation of plasma cells in the BM [29]. However, different trafficking events of MM cells allow them to reach distinct niches from the BM, re-circulate to the BM, and finally egress from the BM during the extramedullary stage of the disease [28]. When MM cells re-enter the BM, they use the BM sinusoids, where the interaction CXCR4/CXCL12 is critical to promote both MM cell homing and retention in the BM [30]. In the BM, MM cells will interact first with proteins in the ECM, a complex layer of proteins that serves as a scaffold for many cells. Interactions between MM cells and the ECM are required for cell proliferation, migration, and survival [31]. Specifically, CD138 and VLA-4 on MM cells directly interact with the ECM proteins, such as collagen type 1 and fibronectin. The binding of VLA-4 to fibronectin induces activation of nuclear factor-kB (NFkB), inducing tumor cell survival and cell adhesion-mediated drug resistance [32]. These interactions generate a welcome and growth-supporting environment that stimulates the dissemination of the malignant plasma cells and results in the upregulation of anti-apoptotic proteins and cell cycle dysregulation [33]. Strategies used in the clinic to disrupt these MM–ECM interactions and reduce cell adhesion-mediated drug resistance include the CXCR4 inhibitor AMD3100 or the proteasome inhibitor bortezomib, which downregulates VLA-4 on MM cells [34], leading to the de-adhesion of MM cells from the BM and turning them more sensitive to therapeutic agents [35]. However, although these agents can enhance the efficacy of treatments by disrupting these interactions, they also contribute to the mobilization of MM cells from the BM into the circulation, promoting extramedullary disease [36].

In physiological conditions, the primary cell population in the BM stroma, known as bone marrow mesenchymal stromal cells (BM-MSCs), support the maintenance and differentiation of hematopoietic lineages, regulate bone homeostasis and contribute to the spatial delimitation of the endosteal and vascular niches [37]. However, in MM, BM-MSCs, as part of the BM microenvironment, play a crucial role in the pathology of the disease. Despite being at low proportions in the BM (0.01 to 0.001% of mononuclear cells) [38], BM-MSCs are the main population among BM stromal cells that interact with MM cells by direct cell–cell contact or through paracrine secretion of different pro-survival cytokines. Thus, for instance, binding VLA-4 on MM cells to VCAM-1 on BM-MSCs promotes activation of NFkB increasing MM cell survival and proliferation [39]. Moreover, the integrin lymphocyte function-associated antigen 1 (LFA-1) on MM cells and its transmembrane binding partner Mucin 1 (MUC1) bind to ICAM-1 in adjacent BM-MSCs, resulting in the activation of different pathways associated with poor prognosis and disease progression in patients [40]. The strong impact of the interactions with BM-MSCs in the physiology of MM cells and their acquisition of multidrug resistance phenotype justifies their consideration as targets for MM therapy. Indeed, some drugs have been developed to disrupt these interactions and tested in MM patients, such as Natalizumab, a recombinant humanized IgG4 monoclonal antibody (MoAb) which binds α4 integrin impairing the interaction VLA-4/VCAM-1 (NCT00675428). Other promising approaches have been preclinically evaluated, such as the LFA-1 inhibitor LFA878 [41].

Soluble mediators are also required for MM plasma cell survival and proliferation in the BM. Thus, MM cells induce BM-MSCs to secrete cytokines that will be used for their benefit. Specifically, the main secreted cytokine is interleukin-6 (IL6), which is involved in MM growth, survival, migration, and drug resistance [42]. In turn, MM cells use IL6 to enhance the secretion of vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF). Then, both VEGF and bFGF bind to their receptors on BM-MSCs, re-stimulating IL6 production [43]. Whereas inhibition of IL6 has not shown clinical benefit in MM [44], blocking of IL6 receptor with tocilizumab has shown efficacy in MM patients [45]. Furthermore, MM cell interactions with BM-MSCs cells are mediated through Notch pathways and Dickkopf-1 (DKK1), which induce the secretion of IL6, VEGF, and insulin-like growth factor (IGF-1) in BM-MSCs [46,47]. Moreover, MSC-derived exosomes contain the long intergenic noncoding RNA LINC00461, which promotes MM cell proliferation and suppresses the beneficial effect of dexamethasone treatment. Indeed, the knockdown of LINC00461 enhances the beneficial impact of dexamethasone in preclinical studies [48].

B-cell activating factor (BAFF) and a proliferation inducing ligand (APRIL) are additional mediators with a protective effect on MM cells [49]. BAFF is a member of the tumor necrosis factor (TNF) family expressed on the surface of BM-MSCs and as a soluble form. BAFF stimulates B cell growth, and ligation of BAFF leads to increased proliferation and survival of MM cells [50]. APRIL is a secreted protein by BM-MSC that binds to B-cell maturation antigen (BCMA) and to transmembrane activator and calcium-modulator and cyclophilin ligand (TACI) on MM cells [51]. Therefore, APRIL-based CARs target MM cells expressing either BCMA or TACI with high efficacy at pre-clinical levels [52]. Moreover, BM-MSCs also protect MM cells against the lytic machinery of CAR-T cells [53].

Another member of the TNF family involved in this stromal training is TNFα, which induces the expression of adhesion molecules, such as LFA-1, ICAM-1, VCAM-1, and VLA-4 on MM cells, as well as ICAM-1 on BM-MSCs, resulting in increased binding of MM cells to BM-MSCs and further enhancing IL6 secretion [54]. These paracrine loops are critical for maintaining the constant growth of MM cells through the activation of different signaling pathways. In addition, MM cell interactions with BM-MSCs, added to the senescent status of cells in the BM in MM, further enhance the secretion of cytokines, chemokines, and soluble factors secreted by BM-MSCs to the BM milieu, which induce further MM proliferation and survival. TNFα, crucial in inflammation, is related to bone resorption and enhanced in MM patients. Thus, targeting TNFα could improve MM responses to treatments [55]. However, reports in inflammatory diseases suggest that anti-TNF-α inhibitors enhance the risk of having future hematological malignancies [56].

In summary, these interleukins and growth factors secreted by BM-MSCs cause tumor growth and drug resistance, limiting current MM treatments’ impact. Indeed, they are promising targets for developing anti-MM therapies that avoid the negative effect of BM-MSCs on dexamethasone treatment [48], on CAR-T cell therapies [53], or the negative impact of IL6 secretion. Thus, tocilizumab, an anti-IL6R [57], BHQ880, a monoclonal antibody against DKK1 [58], or tabalumab, a potent and selective fully human IgG4 MoAb with neutralizing activity against membrane-bound and soluble BAFF [59] are strategies that could be added to MM treatment.

As previously mentioned, MM cells not only interact with the stromal compartment they also alter the endosteal and vascular niches in the BM. In the endosteal niche, healthy bone remodeling in the BM is maintained by a balance between bone formation (osteoblastogenesis) versus bone degradation (osteoclastogenesis). However, MM cells alter this dynamic balance, enhancing bone resorption to enable space for MM proliferation, causing the osteolytic lesions characteristic of myeloma bone disease (MBD) in around 80–90% of MM patients [60]. The negative impact of MBD on patient survival, quality of life, and public health costs has led to the development of different approaches to block MM-endosteal niche interactions. Strategies for patients to treat and avoid MBD have recently been reviewed [61]. Here, we describe which interactions of MM cells with cells in the endosteal niche, including BM-MSCs and other bone populations, such as osteoclasts and osteoblasts, accelerate MBD.

MM cells, through different mechanisms, upregulate osteoclast activity and differentiation resulting in imbalanced bone resorption, causing the osteolytic lesions of the MBD [62]. Specifically, MM cells secrete macrophage inflammatory protein-1α (MIP1α) and MIP1β that directly activate osteoclast formation and activity [63,64]. In turn, osteoclasts secrete IL6 to stimulate their self-proliferation and the proliferation of MM cells [65]. This interaction upregulates Chondroitin synthase 1 (CHSY1), which induces Notch signaling promoting MM cell survival and stimulating the recruitment of osteoclast precursors to increase bone resorption [66]. Macrophage-colony stimulating factor (M-CSF) and receptor activator of NFkB (RANK) ligand (RANKL) are additional factors required for osteoclast differentiation. Osteocytes produce RANKL, which promotes osteoclast activity through binding to RANK on osteoclastic lineage cells [67]. Nevertheless, MM cells’ interaction with BM-MSCs leads to the secretion of RANKL by BM-MSCs, further stimulating osteoclast activation and differentiation and enhancing bone lysis. This interaction also leads to the production of cytokines by BM-MSCs, such as IL6, which further promotes osteoclast growth [68]. In this way, amino-bisphosphonates have been administered in MM patients as first-line therapy for MBD due to their capacity to inhibit osteoclast activity [69]. Moreover, Denosumab, a fully human monoclonal antibody against RANKL, has also shown clinical benefit in MM patients [70]. Denosumab inhibits the development and activity of osteoclasts, decreases bone resorption, and increases bone density [71].

On the other hand, MM cells prevent osteoblast progenitor cell maturation and inhibit osteoblast activation, to continue impairing bone formation. Direct cell–cell interactions of MM cells through binding to VCAM-1 on osteoblast progenitors downregulate RUNX2 activity, essential for osteoblast differentiation [72]. Moreover, osteoblasts and BM-MSCs produce osteoprotegerin (OPG), which prevents the development of bone alterations caused by osteoclast/osteoblast imbalance. However, the binding of VLA-4 on MM cells to VCAM-1 on BM-MSCs decreases OPG secretion, forcing the balance in favor of osteoclasts and bone degradation [73]. Disrupting this VLA-4/VCAM-1 interaction with monoclonal antibodies, such as Natalizumab, could prevent bone lysis in MM patients, as described in preclinical models [74]. On the other hand, BHQ880, the DKK1-neutralizing antibody, can increase osteoblast differentiation, blocking the negative effect of MM cells on osteoblastogenesis and reducing IL6 secretion in MM patients [75].

During the development of MM, an alteration in the neovascularization process occurs that affects the vascular niche. Neovascularization is the formation of new vessels from existing ones through endothelial cells (angiogenesis) or from endothelial precursors (vasculogenesis). Interactions between plasma cells and the BM microenvironment can modify this biological process [76,77,78].

Angiogenesis in cancer involves an early balanced avascular phase that gives rise to an uncontrolled and unlimited in-time vascular phase [79]. In the context of MM, Rajkumar et al. demonstrated that the BM microvascular density is increased in MM patients [80]. In this environment, the accumulation of MM cells in the BM generates hypoxic tumors highly expressing hypoxia-inducible factor-1 alpha (HIF-1α). HIF-1α will upregulate angiogenesis to deliver oxygen and nutrients and remove catabolites [81]. Different cytokines control angiogenesis, such as VEGF, fibroblast growth factor-2 (FGF-2), and hepatocyte growth factor (HGF). In MM, MM plasma cells become CD45-negative and produce VEGF [82]. Moreover, endothelial cells in the BM of MM modify their phenotype, expressing surface receptors related to angiogenesis, such as VEGFR-2 and Tie2/Tek, and increased expression of the β3-integrin and endoglin [83]. This differentiated phenotype in endothelial cells of the BM enhances MM cell interaction with the new-formed blood vessels and favors the entry and dissemination of MM cells into the circulation. This angiogenic phenotype in MM cells can also be induced by oncogenes, such as C-MYC, C-FOS, C-JUN, and ETS-1, which become active as a consequence of the genetic instability and immunoglobulin translocations in MM [84].

On the other hand, the differentiation of endothelial progenitors termed angioblasts during embryogenesis causes the development of the vascular system, known as vasculogenesis [85]. Studies suggest that vasculogenesis is responsible for the neovascularization in the BM in MM [86,87]. Indeed, endothelial markers such as VIII-related antigen (FVIII-RA), vascular endothelial-cadherin (VE-cadherin), VEGFR-2, TIE/Tek, and CD133 are expressed in endothelial cells of the neovessel wall [88]. Moreover, interactions of MM cells with BM-MSCs in the BM also impact vasculogenesis. Thus, MM cells stimulate BM-MSCs in the vascular niche to secrete HGF, VEGF, and IL8, further inducing neovascularization [89]. In turn, endothelial cells in MM will produce IGF1 and IL6 to promote MM cell growth, causing an autocrine loop in endothelial cells, which will enhance their production of VEGF, platelet-derived growth factor (PDGF), Ang-1, HGF, and IL1 to promote angiogenesis constantly [90].

The relevance of angiogenesis in the development of MM has led to the development of different treatments targeting this process. For instance, amino-bisphosphonates that inhibit osteoclasts also present anti-angiogenic activities and are administered in MM patients as supportive therapy for bone disease [69]. Ria et al. reviewed different strategies in MM mainly based on VEGF inhibition, such as monoclonal antibodies anti-VEGF (Bevacizumab) [91]. However, the addition of bevacizumab to anti-MM therapies did not result in a significant improvement in the outcome of patients [92,93]. Derivatives of quinolone and quinazoline, which inhibit a variety of tyrosine kinases, including VEGFRs, EGFR, and PDGFR have also been tested in MM patients. Despite their in vitro activity and reduced plasma levels of VEGF in treated MM patients, no responses or clinical benefits were achieved [94,95]. These disappointing results inhibiting a single proangiogenic cytokine could be related to the role played by hypoxia and other active pro-angiogenic pathways in the BM microenvironment, and greater efficacy could be feasible with drugs that simultaneously block multiple cytokines. Moreover, immunomodulators (IMIDs), such as thalidomide or lenalidomide, have revealed anti-angiogenic activity and inhibition of the secretion of angiogenic cytokines in MM patients [96,97].