Bortezomib Reduces Serum Dickkopf-1 and RANKL Concentrations and Normalizes Indices of Bone Remodeling in Patients with Relapsed Multiple Myeloma.

Abstract Bortezomib is a proteasome inhibitor, which is currently indicated for the treatment of relapsed/refractory myeloma (MM). Although the anti-myeloma effect of bortezomib has been clearly demonstrated, its effect on bone metabolism is still unclear. There are recent reports that bortezomib increases serum alkaline phosphatase (ALP) activity, which is consistent with enhanced osteoblast function. The aim of this study was to evaluate the effect of bortezomib on bone turnover in 34 patients with relapsed MM. Bortezomib was given alone at a dose of 1.3 mg/m2 on days 1, 4, 8, and 11 of a 3-week cycle for 4 cycles. Responders could continue for 4 more cycles, while non-responders could continue therapy with the addition of dexamethasone. The following serum indices were measured on day 1 of cycle 1, and then on day 21 of cycles 4 and 8: osteoblast inhibitor dickkopf-1 (DKK-1); osteoclast regulators: soluble RANKL (sRANKL) and osteoprotegerin (OPG); bone resorption markers: C-telopeptide of collagen type-I (CTX) and tartrate-resistant acid phosphatase type-5b (TRACP-5b); and bone formation markers: bone-specific ALP (bALP) and osteocalcin (OC). We also studied 33 healthy controls of similar gender and age. The objective response rate after 4 cycles of therapy was 66%: CR 8% and PR 58%. Sixteen responders and 3 non-responders continued on therapy for 4 more cycles. Myeloma patients at baseline had increased values of DKK-1 (p=0.007), sRANKL, sRANKL/OPG ratio, and both markers of bone resorption (p<0.0001) when compared to controls. In contrast, bone formation as assessed by serum bALP and OC was significantly reduced (p<0.001). There was a strong correlation between bone lytic disease and serum CTX (r=0.59, p<0.01), and sRANKL (r=0.4, p=0.03). Patients with severe bone disease (>9 lytic lesions, n=7) had elevated values of DKK-1 compared with all others (mean±SD: 223.4±264.4 ng/mL vs. 84±62.4 ng/mL; p=0.01). Moreover, serum levels of DKK-1 correlated with CTX levels (r=0.39, p=0.04), and weakly with bALP concentrations (r=−0.32, p=0.09). The administration of bortezomib produced a significant reduction of DKK-1 (p=0.035), sRANKL (p=0.01), CTX and TRACP-5b (p<0.001) after 4 cycles, which was still seen after 8 cycles of treatment (p<0.01). Bortezomib also produced a dramatic increase in both markers of bone formation, bALP and OC, after 4 and 8 cycles of therapy (p<0.01). Responders tended to have lower initial levels of DKK-1 compared with non-responders. Patients who achieved a CR or vgPR after 4 cycles of bortezomib had greater elevation of bALP than all others: mean±SD of increase: 306.3%±556.9% vs. 45.8%±56.5%; p=0.02. It is of interest that 3/4 non responders also had an increase in bALP (mean: 39.6%) after 4 cycles of bortezomib. There was no other correlation between response to therapy and alteration of bone markers. No healing of the lytic lesions was observed even in CR patients. This study suggests that bortezomib reduces serum levels of DKK-1 and RANKL, irrespective of response to therapy, in patients with relapsed myeloma and thus leads to normalization of abnormal bone remodeling through the increase of bone formation and reduction of bone resorption.