Distinct immunopathology in the early stages between different antiresorptives-related osteonecrosis of the jaw-like lesions in mice.

There is limited information about denosumab-related osteonecrosis of the jaw (DRONJ), unlike bisphosphonate-related ONJ (BRONJ). The mode of action is clearly different between denosumab and bisphosphonates. DRONJ occurs mainly following tooth extraction in cancer patients treated with the combination of denosumab and other drugs including chemotherapy. However, DRONJ animal models similar to these clinical situations have not been developed. The aims of this study were to 1) create a new model of high-prevalence chemotherapy/anti-RANKL antibody-related ONJ-like lesions to mimic patients receiving a denosumab/chemotherapy combination; and 2) compare the histopathological and immunopathological findings in the early stages of BRONJ-like and anti-RANKL antibody-related ONJ-like lesions. Cyclophosphamide (CY) and anti-mouse RANKL monoclonal antibody (mAb) or zoledronate combination therapy (CY/mAb and CY/ZA, respectively) was performed to create ONJ-like lesions in female C57BL/6J mice. Both maxillary first molars were extracted at 3 weeks after drug administration. The animals were euthanized at either 2 or 4 weeks after tooth extraction. Increased necrotic bone and empty lacunae with decreased living bone and osteocyte numbers were common histopathological findings in CY/mAb- and CY/ZA-induced impaired wound healing at 4 weeks after tooth extraction, and they were diagnosed as ONJ-like lesions based on validation of BRONJ and DRONJ in humans. In areas of impaired healing at 2 weeks post-extraction, decreases in angiogenesis and F4/80LYVE-1 macrophages were noted as common immunopathological findings, although anti-angiogenesis was worse with CY/mAb than with CY/ZA. Interestingly, CY/mAb did not reduce F4/80LYVE-1 cells and normal lymphangiogenesis remained, whereas CY/ZA profoundly suppressed the larger size of F4/80LYVE-1 cells, similar to vessels with a concomitant decrease in lymphangiogenesis. Therefore, the distribution of the larger size of F4/80LYVE-1 cells differed in the early stages between different antiresorptive-induced ONJ-like lesions in conjunction with lymphangiogenesis, although the histopathological findings were similar. These findings suggest that the pathogenesis of BRONJ and DRONJ may differ due to the distributions of F4/80LYVE-1 tube-like-structured cells.