There is clearly an increasing awareness within the myeloma community of the effects of myeloma on the skeleton. At the 8th International Myeloma Workshop, there was a full session devoted to this topic, and one of the preworkshop symposia dealt with bone-active drugs in the bone microenvironment. Attendance at these sessions was excellent, and there is obvious interest in this area.
A major impetus for the increased interest in bone-active drugs has been the advent of bisphosphonates, which were approved for use in myeloma bone disease by the Food and Drug Administration (FDA) 6 years ago. This session focused on the current status of bisphosphonates, their potential as direct antitumor agents, their effects on tumor burden, the molecular targets for bisphosphonates in osteoclasts and myeloma cells, and consideration of newly described mediators of osteoclastic bone resorption whose production may be excessive in myeloma, namely RANK ligand and macrophage-inflammatory protein. There was also discussion of the potential utility of new therapeutic agents that interfere with the cell-cell interactions responsible for bone-resorbing activity in myeloma and particularly of the specific inhibitors of RANK ligand, namely RANK.Fc and osteoprotegerin, and the use of antagonists to macrophage-inflammatory protein. The potential of markers of bone resorption for staging disease and assessing prognostic risk was addressed as well.
Dr. Jonathan Green of Novartis Oncology reviewed the structure and properties of the bisphosphonates in current clinical use and discussed in some detail their potential as antitumor agents. The bisphosphonates have now achieved widespread use, with a total market of more than $1.5 billion, more than one third of which is for cancer-related indications. This rather simple group of compounds falls into two classes, one of which contains nitrogen. The nitrogen-containing bisphosphonates, which by and large comprise the newer bisphosphonates, have a different molecular mechanism of action from that of the older compounds etidronate and clodronate. This was discussed in detail by Dr. Roger Pearse of Cornell University Medical College. These bisphosphonates inhibit an enzyme in the mevalonate pathway of cholesterol biosynthesis. The major enzyme targeted appears to be farnesyl diphosphate synthase, and the capacity of these compounds to inhibit this enzyme apparently correlates very closely with their potency and efficacy as antiresorbtion agents. The most potent of the antiresorptive bisphosphonates are zoledronic acid and risedronate. Dr. Pearse’s work suggests that these compounds then inhibit a chemical modification of small membrane-bound proteins (GTPase proteins); this is known as prenylation. This in turn has a number of serious consequences for the cell that ultimately lead to its death by apoptosis and loss of its capacity to stimulate bone resorption. Dr. Pearse and his colleagues have found that similar changes may also occur in myeloma cells and suggest that part of the antitumor effect of the bisphosphonates may be related to this capacity to cause tumor cell apoptosis. Dr. Green reviewed other in vitro effects of the bisphosphonates that could be linked to their capacity to cause tumor cell apoptosis, including decreased release of interleukin-6 by osteoclasts, myeloma cells, and osteoblasts; effects on cell proliferation; effects on tumor cell apoptosis; and effects on angiogenesis. These effects may be enhanced by other agents such as dexamethasone and chemotherapeutic agents such as Taxol.
The major issue is whether these anti-tumor effects have relevance to the use of bisphosphonates in vivo. Although most of them occur at quite large concentrations of the bisphosphonates in in vitro experiments, it is possible that these concentrations may be achieved in vivo, since bisphosphonates are concentrated in the bone microenvironment by their capacity to bind to calcified matrices. However, in most of the preclinical models of myeloma in which bisphosphonates have been tested so far, the bisphosphonates do not have major effects on myeloma cell apoptosis or myeloma cell death, whereas they are very effective inhibitors of bone resorption. This is an important area that clearly needs further investigation. One of the issues may be that there are a number of different animal models of myeloma including the 5T models as well as the human severe combined immunodeficiency disease (SCID) models.
Dr. Babatunde O. Oyajobi of the University of Texas Health Science Center at San Antonio discussed some of the newly described mediators responsible for bone destruction in myeloma, with particular emphasis on RANK ligand. He has found that myeloma cells when cocultured with bone marrow stromal cells increase expression of RANK ligand, which has the capacity to stimulate osteoclast formation and activity.
To determine the significance of RANK ligand and osteoclastic bone resorption, he used an antagonist to RANK ligand known as RANK.Fc, which comprises a genetically engineered hybrid molecule consisting of the extracellular domain of RANK linked to a fragment of immunoglobulin (Ig). This soluble chimeric protein then acts as a functional antagonist to RANK ligand, binding to its cognate-receptor rank. He used this in the 5T murine model of myeloma bone disease and found very effective inhibition of bone resorption, similar to what is seen in this model with the powerful bisphosphonates. However, he also observed a marked reduction in tumor burden reflected by a decrease in the tumor volume in bone and in the circulating concentrations of IgG2b, the monoclonal protein produced by myeloma cells that serves as a marker of tumor burden in this model.
Dr. Pearse found very similar results. He observed enhanced expression of RANK ligand (also called TRANCE) by myeloma cells and stromal cells when they were cocultured. He also found that RANK.Fc caused marked inhibition of bone destruction induced by myeloma in two murine models. He used the model in which SCID immunocompromised mice are irradiated and inoculated with human fetal bone and human myeloma cells. In these circumstances he saw a marked reduction in bone lesions and tumor burden. He also noted that RANK.Fc prevented the development of paralysis in the hind limbs of the tumor-bearing mice and blocked the progression of multiple myeloma; he concluded that RANK ligand deregulation may be responsible for tumor progression. In addition, he noted that in myeloma marrow there was decreased expression of the soluble decoy receptor for RANK ligand, osteoprotegrin.
A number of posters also addressed this issue. They included work by Penser, Anderson, Barsley, Guiliani, Abe, and Croucher. There are some differences in the findings from the different groups, possibly related to differences in the antibodies used to detect RANK ligand, the cell lines used, and the animal models. However, all investigators agree that there is increased and deregulated expression of RANK ligand in myeloma, this expression is enhanced by cell-cell contact and co-culture of myeloma cells with stromal cells, and RANK.Fc is a very effective inhibitor of osteoclast stimulation and possibly of bone resorption. Open issues include the importance of down regulation of OPG, the relative importance of the myeloma cell as a source of RANK ligand, and whether OPG has similar effects to those of RANK.Fc in decreasing tumor burden.
Dr. Gregory Mundy presented data from Dr. David Roodman focusing on the role of macrophage-inflammatory protein in the osteolysis associated with myeloma. Investigators found that a biologically active mediator of osteoclast formation in a cDNA expression library from a patient with myeloma was detected and characterized to be a macrophage-inflammatory protein. Dr. Roodman found that bone marrow plasma supernatants from patients with myeloma had powerful osteoclast-stimulating activity that increased osteoclast formation from human marrow cultures and whose production correlated with severity of disease. When this soluble factor was incubated with antibodies to macrophage-inflammatory protein, the bone-destructive lesions were totally inhibited. The human ARH-77 cell line was stably transfected with antisense constructs to macrophage-inflammatory protein and then inoculated into irradiated SCID mice. There was a marked reduction in associated osteolysis and development of bone lesions. Tumor burden was also reduced. This work raises the possibility that macrophage-inflammatory protein may be an important mediator of the osteoclast stimulation that occurs in myeloma.
Finally, Dr. Facon presented an evaluation of markers for bone resorption in patients with monoclonal gammopathy of undetermined significance and multiple myeloma. These markers are thought to reflect the rate and extent of bone resorption and the activity of bone-resorbing and bone-forming cells. The report indicated that three markers of bone resorption, namely deoxypyridinylene, NTX and CTX gave useful information. These urinary markers correlated very well, and the investigators recommended that they play an important role in the staging of the disease and may provide extremely useful information when a patient is undergoing prognostic assessment.