Background: Ewing’s sarcoma (ES) is the second most frequent primitive malignant bone tumor in adolescents with a very poor prognosis for high risk patients, mainly when lung metastases are detected (overall survival <15% at 5 years). Zoledronic acid (ZA) is a potent inhibitor of bone resorption which induces osteoclast apoptosis. Our previous studies showed a strong therapeutic potential of ZA as it inhibits ES cell growth in vitro and ES primary tumor growth in vivo in a mouse model developed in bone site. However, no data are available on lung metastasis. Therefore, the aim of this study was to determine the effect of ZA on ES cell invasion and metastatic properties. Methods: Invasion assays were performed in vitro in Boyden’s chambers covered with Matrigel. Matrix Metalloproteinase (MMP) activity was analyzed by zymography in ES cell culture supernatant. In vivo, a relevant model of spontaneous lung metastases which disseminate from primary ES tumor was induced by the orthotopic injection of 10⁶ human ES cells in the tibia medullar cavity of nude mice. The effect of ZA (50 μg/kg, 3x/week) was studied over a 4-week period. Lung metastases were observed macroscopically at autopsy and analysed by histology. Results: ZA induced a strong inhibition of ES cell invasion, probably due to down regulation of MMP-2 and −9 activities as analyzed by zymography. In vivo, ZA inhibits the dissemination of spontaneous lung metastases from a primary ES tumor but had no effect on the growth of established lung metastases. Conclusion: These results suggest that ZA could be used early in the treatment of ES to inhibit bone tumor growth but also to prevent the early metastatic events to the lungs.

The effect of amino-bisphosphonates on osteoblastic lineage and its potential contribution to the pathogenesis of bisphosphonate-associated osteonecrosis of the jaw (BONJ) remain controversial. We assessed the effects of zoledronic acid (ZOL) on bone and vascular cells of the alveolar socket using a mouse model of BONJ. Thirty-two mice were treated twice a week with either 100 μg/kg of ZOL or saline for 12 weeks. The first left maxillary molar was extracted at the third week. Alveolar sockets were assessed at both 3 weeks (intermediate) and 9 weeks (long-term) after molar extraction by semi-quantitative histomorphometry for empty lacunae, preosteoblasts (Osterix), osteoclasts (TRAP), and pericyte-like cells (CD146). Also, the bone microarchitecture was assessed by micro-CT. Osteonecrotic-like lesions were observed in 21% of mice. Moreover, a decreased number of preosteoblasts contrasted with the increased number of osteoclasts at both time points. In addition, osteoclasts display multinucleation and detachment from the endosteal surface. Furthermore, the number of pericyte-like cells increased at the intermediate time point. The alveolar bone mass increased exclusively with long-term ZOL treatment. The severe imbalance between bone-forming cells and bone-resorbing cells shown in this study could contribute to the pathogenesis of BONJ.

Osteosarcoma (OS) is the most common cancer of bone. Jaw osteosarcoma (JOS) is rare and it differs from other OS in terms of the time of occurrence (two decades later) and better survival. The aim of our work was to develop and characterize specific mouse models of JOS. Syngenic and xenogenic models of JOS were developed in mice using mouse (MOS-J) and human (HOS1544) osteosarcoma cell lines, respectively. An orthotopic patient-derived xenograft model (PDX) was also developed from a mandibular biopsy. These models were characterized at the histological and micro-CT imaging levels, as well as in terms of tumor growth and metastatic spread. Homogeneous tumor growth was observed in both the HOS1544 and the MOS-J JOS models by injection of 0.25 × 106 and 0.50 × 106 tumor cells, respectively, at perimandibular sites. Histological characterization of the tumors revealed features consistent with high grade conventional osteosarcoma, and the micro-CT analysis revealed both osteogenic and osteolytic lesions. Early metastasis was encountered at day 14 in the xenogenic model, while there were no metastatic lesions in the syngenic model and in the PDX models. We describe the first animal model of JOS and its potential use for therapeutic applications. This model needs to be compared with the usual long-bone osteosarcoma models to investigate potential differences in the bone microenvironment.