DENOSUMAB-RELATED OSTEONECROSIS OF JAW (ONJ) VERSUS ZOLEDRONIC ACID-RELATED ONJ: DIFFERENT CHARACTERISTICS AT COMPUTED TOMOGRAPHY SCAN EVALUATION? A PROPOSAL FOR A MULTICENTRE STUDY

Background: Osteonecrosis of the Jaw (ONJ) - related to Bisphosphonates (BPs), denosumab, and other drugs - is a bone disease, that should be evaluated not only on base of clinical features (bone exposure and other signs or symptoms) but also by imaging tools, mostly Computed Tomography (CT) or Cone Beam Computed Tomography (CBCT). An Italian proposal of definition and staging is based on extension of bone alterations at CT scan study. Some scarce literature suggests that ONJ cases observed after denosumab might have different characteristics at CT scan.   Methods: We reviewed CT relevant images at ONJ sites of patients developing ONJ after denosumab treatment only, looking for presence of radiological alteration signs. “Early signs” included: cortical disruption; markedly thickened and sclerotic lamina dura; trabecular thickening; focal bone marrow sclerosis; persisting alveolar socket; sequestra formation; widening of periodontal ligament space. “Late signs” included: oro-antral fistula; pathologic fracture; prominence of the inferior alveolar nerve canal; osteolysis extending to the sinus floor; diffuse osteosclerosis; osteosclerosis of adjacent bones (zygoma; hard palate); periosteal reaction; sinusitis. Cases of patients receiving BPs and shifted to denosumab were excluded. The results were compared with similar data collected by CT scans of ONJ patients treated with zoledronic acid only. The ONJ stage according to Italian staging system (stage 1, focal ONJ; stage 2, diffuse ONJ; stage 3, complicated ONJ) was adopted in both the groups. Results:  We evaluated CT scans of patients treated with denosumab, developing ONJ.  After exclusion of patients shifted to denosumab after biphosphonates, 8 patients were found as receiving denosumab only and their CT scan findings were compared with those of 10 ONJ cases observed after zoledronic acid treatment, randomly selected in a database of 103 ONJ cases. Tha stage resulted more advanced in cases after zoledronic acid treatment. Preliminary analysis indicate that frequency of the single characteristics might be slightly different in the two patient populations; particularly diffuse osteosclerosis was less frequent in this little sample of patients receiving denosumab (2 out of 8 cases) in comparison with those ones receiving zoledronic acid (7 out of 10 cases). Conclusions: A large multicentre retrospective and prospective study is warranted and hopeful, aimed to clarify the differences on CT alterations in denosumab-related and BP-related ONJ cases.

osteosclerosis of adjacent bones (zygoma; hard palate); periosteal reaction; sinusitis. Cases of patients receiving BPs and shifted to denosumab were excluded. The results were compared with similar data collected by CT scans of ONJ patients treated with zoledronic acid only. The ONJ stage according to Italian staging system (stage 1, focal ONJ; stage 2, diffuse ONJ; stage 3, complicated ONJ) was adopted in both the groups. Results: We evaluated CT scans of patients treated with denosumab, developing ONJ. After exclusion of patients shifted to denosumab after biphosphonates, 8 patients were found as receiving denosumab only and their CT scan findings were compared with those of 10 ONJ cases observed after zoledronic acid treatment, randomly selected in a database of 103 ONJ cases.
Tha stage resulted more advanced in cases after zoledronic acid treatment. Preliminary analysis indicate that frequency of the single characteristics might be slightly different in the two patient populations; particularly diffuse osteosclerosis was less frequent in this little sample of patients receiving denosumab (2 out of 8 cases) in comparison with those ones receiving zoledronic acid (7 out of 10 cases). Conclusions: A large multicentre retrospective and prospective study is warranted and hopeful, aimed to clarify the differences on CT alterations in denosumab-related and BP-related ONJ cases.
Background: ONJ after treatment with Bispshosphonates (BPs) was firstly recognized on 2003, and it was largely described as BRONJ (BP-related ONJ). Definition and staging of BRONJ are controversial; the AAOMS definition [1][2] [3] is much restricted, based only on clinical features (bone exposure, lasting more than six weeks), not assuming the importance of bone study with imaging tools. An alternative definition has been published by a committee of members of Italian Societies of Oral Medicine (SIPMO) and Maxillofacial Surgery (SICMF) [4][5] [6] , based on imaging evaluation, essen tially Computed Tomography (CT) scan [7] [8] . After 2009, ONJ cases were also observed after treatment with denosumab (an antiresorptive drug different from BPs) and after antiangiogenic drugs (even without association of BPs or denosumab). A recent issue is: do denosumab-related ONJ cases have the same characteristics of BRONJ? Are different both as clinical behavior and at imaging exams? Some scarce literature data are suggesting of somewhat radiological differences, as recently reported [9] . However, the reports are unconsistent, varying from lower bone density values [10] to higher frequency of sequestrum and periosteal reaction [11] , to vicecersa less common sequestra and lysis of cortical border [12] for ONJ cases observed after denosumab treatment, in comparison with BP-related cases. We have experience of studies of CT scan evaluation in ONJ cases [13][14] [15] [16][17] [18] and we decided to analyse the CT scan findings of recently observed ONJ cases related to denosumab treatment in comparison with some cases related to zoledronic acid, already evaluated and present in our database [19] .

Methods:
We reviewed relevant images at ONJ sites of patients developing ONJ after zoledronic acid or denosumab treatment, looking for presence of radiological alteration signs, described by Bedogni et al [4] and included in ONJ management recommendations by Italian Societies of Oral Medicine (SIPMO) and Maxillofacial Surgery (SICMF) [5] [6] .
"Early signs" included: cortical disruption; markedly thickened and sclerotic lamina dura; trabecular thickening; focal bone marrow sclerosis; persisting alveolar socket; sequestrum formation; widening of periodontal ligament space. "Late signs" included: oro-antral fistula; pathologic fracture; prominence of the inferior alveolar nerve canal; osteolysis extending to the sinus floor; diffuse osteosclerosis; osteosclerosis of adjacent bones (zygoma; hard palate); periosteal reaction; sinusitis. complicated ONJ) was adopted in both the groups [5] [6] .  Table 1).  bone marrow sclerosis in none; persisting alveolar socket in none; sequestrum formation in none; widening of periodontal ligament space in 7.The "late signs" were as follows : oro-antral fistula in none; pathologic fracture in none; prominence of the inferior alveolar nerve canal in 1; osteolysis extending to the sinus floor in none; diffuse osteosclerosis in 7;

Conclusions:
Preliminary analysis indicate that frequency of the single characteristics might be slightly different in the two patient populations: particularly diffuse osteosclerosis was less frequent in cases observed after denosumab in comparison with those ones receiving zoledronic acid, as well as for other signs (i.e., markedly thickened and sclerotic lamina dura, and trabecular thickening). However, our case series is limited and not well balanced for stage, disease (breast cancer versus other cancers), sex, and length of drug exposure. A large multicentre retrospective and prospective study is warranted and hopeful, aimed to clarify the differences on CT alterations in denosumab-related and BP-related ONJ cases.