Bisphosphonates do not affect healing of a critical-size defect in estrogen-deficient mice

Bisphosphonates (BP) are anti-resorptive drugs that are widely used to prevent bone loss in osteoporosis. Since inhibition of bone resorption will cause a decrease in bone formation through a process called coupling, it is hypothesized that extended treatment protocols may impair bone healing. In this study, β-tri‑calcium-phosphate (βTCP) ceramics were inserted into critical-size long bone defects in estrogen-deficient mice under BP therapy. The study assessed the benefits of coating the ceramics with Bone Morphogenetic Protein-2 (BMP2) and an engineered BMP2 analogue (L51P) that inactivates BMP antagonists on the healing process, implant resorption, and bone formation. Female NMRI mice (11-12 weeks of age) were ovariectomized (OVX) or sham operated. Eight weeks later, after the manifestation of ovariectomy-induced osteoporotic bone changes, BP therapy with Alendronate (ALN) was commenced. After another five weeks, a femoral critical-size defect was generated, rigidly fixed, and βTCP-cylinders loaded with 0.25 μg or 2.5 μg BMP2, 2.5 μg L51P, and 0.25 μg BMP2/2.5 μg L51P, respectively, were inserted. Unloaded βTCP-cylinders were used as controls. Femora were collected six and twelve weeks post-implantation. Histological and micro-computer tomography (MicroCT) evaluation revealed that insertion of cylinders coated with 2.5 μg BMP2 accelerated fracture repair and induced significant bone formation compared to controls (unloaded cylinders or coated with 2.5 μg L51P, 0.25 μg BMP2) already six weeks post-implantation, independent of estrogen-deficiency and BP therapy. The simultaneous administration of BMP2 and L51P (0.25 μg BMP2/2.5 μg L51P) did not promote fracture healing six and twelve weeks post-implantation. Moreover, new bone formation within the critical-size defect was directly linked to the removal of the βTCP-implant in all experimental groups. No evidence was found that long-term therapy with ALN impaired the resorption of the implanted graft. However, osteoclast transcriptome signature was elevated in sham and OVX animals upon treatment with BP, with transcript levels being higher at six weeks than at twelve weeks post-surgery. Furthermore, the transcriptome profile of the developing repair tissue confirmed an accelerated repair process in animals treated with 2.5 μg BMP2 implants. L51P did not increase the bioefficacy of BMP2 in the applied defect model. The present study provides evidence that continuous administration of BP does not inhibit implant resorption and does not alter the kinetics of the healing process of critical-size long bone defects. Furthermore, the BMP2 variant L51P did not enhance the bioefficacy of BMP2 when applied simultaneously to the femoral critical-size defect in sham and OVX mice.