Characterization of Medication Related Osteonecrosis of the Jaws Pathophysiology and Local RANKL Delivery as an Experimental Intervention

Author(s): Soundia, Akrivoula | Advisor(s): Tetradis, Sotirios | Abstract: Medication related osteonecrosis of the jaws (MRONJ) is a severe adverse effect of antiresorptive and angiogenic medication prescribed to patients with osteoporosis or bone malignancies. Our laboratory has been working on the elucidation of the mechanisms of this disease for the past decade. In this dissertation, we further characterized pathogenesis in rodents and humans and explored the delivery of RANKL as a therapeutic approach. Tooth extraction was performed on diseased or healthy teeth in mice treated with saline, OPG-Fc (the equivalent of denosumab in rodents) or bisphosphonates. Extraction of healthy teeth did not suffice to induce MRONJ in mice treated with antiresorptives. In contrast, extraction of diseased teeth induced mucosal defects, absence of radiographic healing, periosteal reaction, significant osteonecrosis and bone exposure in mice treated with antiresorptives. Extraction of periodontally diseased teeth in zoledronate (ZA) treated rats resulted in altered socket healing with a significant amount of osteonecrosis. An increased signal of collagen type III, MMP-9, MMP-13 and α-SMA was seen in the sockets of these rats compared to saline treated rats or zoledronate treated rats with extraction of healthy teeth. We also took advantage of our patient database to investigate the radiographic appearance of stage 0 MRONJ patients. We concluded that sclerosis, cortical erosion, periosteal reaction, sequestration and crater-like defect are more commonly present in stage 0 MRONJ patients as opposed to dental disease patients. Additionally, if a stage 0 MRONJ patient presents with radiographic sequestration at initial visit, they are at a higher risk to progress to frank bone exposure in less than a year.Next, we explored local RANKL delivery as a potential therapy for MRONJ. RANKL was delivered in the sockets of ZA or saline treated rats in a collagen tape. In the extraction sockets of ZA treated rats with RANKL treatment, there was an improvement in mucosal and radiographic healing, a decrease in osteonecrosis and bone exposure and a simultaneous increase in osteoclast numbers compared to the ZA treated rats that had not received local RANKL treatment. We also investigated the role of macrophages in MRONJ development. In a model of periodontitis, zoledronate treated mice demonstrated a predominance of M1 macrophages (macs) and an absence of M2 macs compared to the saline treated mice. The above M1 macrophages also expressed MMP-13. Rosiglitazone treatment reversed the M1/M2 ratio in ZA treated mice 2 weeks after periodontal disease was induced and reduced epithelium to crest distance in ZA mice. However, a difference was not detected in the bone exposure incidence, the epithelium to crest distance or the osteonecrotic areas between ZA mice with or without rosiglitazone treatment. Our data provide insight to the pathogenesis of MRONJ, describe the radiographic profile of stage 0 MRONJ patients and suggest that local RANKL delivery may be efficient in minimizing MRONJ incidence and improving socket healing.