Inhibition of DKK1 Activity Is Associated with Increased Osteoblast Numbers and Bone Formation, Reduced Osteoclast Activity and Inhibition of Tumor Growth in the SCID-rab Model for Primary Myeloma.

We have previously demonstrated that myeloma (MM) cells produce the Wnt signaling antagonist, Dickkopf-1 (DKK1), and that DKK1 inhibits differentiation of osteoblasts (Tian et al. NEJM 2003). This is thought to result in an uncoupling process that may lead to induction of lytic bone disease in MM and possibly promote tumor growth. The aim of this study was to investigate the role of DKK1 in our established SCID-rab model for primary MM. Growth of primary MM in this system is restricted to the implanted bone and associated with typical disease manifestations including increased osteoclast activity, reduced osteoblast numbers and induction of osteolytic bone disease (Yata & Yaccoby, Leukemia 2004). In this study, SCID-rab mice were engrafted with MM cells expressing DKK1 (assessed by global gene expression profiling) from 8 patients. The level of DKK1 expression was correlated with numbers of MRI and x-ray focal lesions in these patients. Following establishment of MM growth, as monitored by weekly measurement of human monoclonal immunoglobulins (hIg) in mouse sera and radiographically, mice were injected subcutaneously into the surrounding area of the implanted bone with neutralizing antibody (AB, R&D) against DKK1 (polyclonal AB: n=4, 50 μg/injection/2 days; monoclonal AB: n=4, 100 μg/injection/day) or control IgG AB, for 4–6 weeks. Whereas bone mineral density (BMD) in control mice was reduced by 7.1%±4.6% from pre-treatment levels, BMD in mice treated with anti-DKK1 was increased by 5.6%±6.7% from pre-treatment levels (p<0.03). The bone anabolic effect of DKK1 AB was also visualized on x-rays and was detected in mice transplanted with cells from patients with low and high degree of bone disease. Histological examination revealed that myelomatous bones of DKK1 AB-treated mice had increased numbers of osteocalcin-expressing osteoblasts (45±5 vs. 16±2 per mm bone in control mice, p<0.02) and reduced number of multinucleated TRAP-expressing multinucleated osteoclasts (5±3 vs. 13±2 per mm bone in control mice, p<0.004). These results support recent reports demonstrating the critical involvement of Wnt signaling in osteoblasts in regulation of osteoclastogenesis as well (Glass et al., Dev. Cell 2005). Whereas in control mice myeloma burden increased in all experiments, anti-DKK1 treatment was associated with reduced tumor growth from pre-treatment levels in 4 of 8 experiments. Overall, myeloma tumor burden increased by 325%±122% and 165%±53% from pre-treatment levels in control and DKK1 AB-treated mice, respectively (p<0.03). Furthermore, bone marrow areas with locally high numbers of differentiated osteoblasts were depleted of myeloma cells in DKK1 AB-treated hosts. We conclude that DKK1 is a key player in myeloma bone disease and that blocking DKK1 activity in myelomatous bones reduces osteolytic bone resorption, increases bone formation and helps control myeloma progression.