Workshop of European task force on medication‐related osteonecrosis of the jaw—Current challenges

This paper reports on the conclusions of two workshops held in Copenhagen in September 2017 and November 2018 focused on medication-related osteonecrosis of the jaws (MRONJ). The workshops were organized and attended by a European task force on MRONJ, i.e. a multidisciplinary group of European clinical investigators with a special interest in the diagnosis and management of MRONJ and a track record of relevant research and publications. The aim of the workshops were to (i) highlight some of the most controversial aspects of current knowledge on MRONJ, including definition and classification, risk factors and management, and (ii) provide an expert opinion-based consensus with a view to inform clinicians and advise researchers, as a first step of reaching solutions. It should be pointed out that all results and comments presented are the authors (the workshop group members) personal views and the present form of this publication is based on genuine consensus of all authors. This article is protected by copyright. All rights reserved.

consensus on these topics with a view to help clinicians making informed decisions on patient's care and inspire future investigators to design better clinical studies. The Group agreed to focus upon three highly controversial aspects of MRONJ: (a) definition and classification, (b) risk factors, and (c) management/treatment of MRONJ.

| CONTROVER S IE S ON DEFINITI ON AND CL A SS IFIC ATI ON
The consensus papers by Ruggiero et al. representing (2014) includes (a) the MRONJ case definition as the presence of exposed jaw bone or bone that can be probed through an intraoral or extraoral fistula(e) for at least 8 weeks in a patient receiving anti-resorptive and/or anti-angiogenic therapy who had not received radiotherapy to the head and neck, and (b) a disease classification into 4 clinical stages (stage 0-3). The most notable change introduced in the 2014 AAOMS consensus was the modified MRONJ definition so to include patients presenting with an intraoral or extraoral fistula(e). This important amendment was inspired by a number of reports highlighting that a sub-group of patients can in fact present with MRONJ disease characterized by the absence of exposed bone on visual inspection (so-called non-exposed MRONJ, including the presence of an intraoral fistula, mandibular fracture, dentally unexplained pain, and swelling, among other manifestations), and therefore, they would not fulfill the case definition of MRONJ as suggested by the initial version of the AAOMS consensus (2007) (Ascani, Campisi, & Junquera Gutierrez, 2014;Bedogni, Fusco, Agrillo, & Campisi, 2012;Fedele et al., 2015;S. Patel et al., 2012;Schiodt, Reibel, Oturai, & Kofod, 2014;Yarom, Fedele, Lazarovici, & Elad, 2010).
The background was the obviously different interpretations of the term "bone exposure" by different author groups and adjudicators in clinical and epidemiological studies. Some authors regarded bone that can be probed through a fistula as exposed and diagnosed MRONJ in the respective cases, while other authors did not include those patients.
The 2009 update of the AAOMS consensus papers (Ruggiero et al., 2009) partially addressed this issue as they added the new classification stage (stage 0) to include patients presenting with the nonexposed variant of MRONJ. However, the MRONJ case definition remained paradoxically unchanged, therefore preventing non-exposed MRONJ cases to be formally diagnosed, especially in clinical trials and epidemiological studies (Fedele et al., 2015).
Although the 2014 update of the AAOMS consensus represents a notable improvement, patients presenting with non-exposed MRONJ without fistulas (e.g., dentally unexplained pain, mobile teeth not due to periodontitis, numbness of the lip, mandibular fracture) continue to remain excluded from MRONJ case definition (Fedele et al., 2015) ( Table 1 and 2). There is therefore an urgent need for expanding the case definition of MRONJ so to encompass the other manifestations of non-exposed MRONJ and ensure that these patients can (a) be formally diagnosed and treated, and (b) be included in clinical and epidemiological studies. The Group appreciated that this might be a difficult task as an accurate case definition should ensure the exclusion of etiopathologically different disorders presenting with similar clinical manifestations, which include plaque-related gingivitis/periodontitis, dental and periapical disease, benign fibro-osseous lesion of the jawbones, chronic sclerosing osteomyelitis, infectious osteomyelitis, primary jawbone malignancy, metastatic disease, and TMJ disorders (Fedele et al., 2015;Patel et al., 2012;Ruggiero et al., 2014;Schiodt et al., 2014). Excluding these conditions as well as describing the MRONJ lesions requires imaging. The value of imaging is described later under controversies on management/treatment. Some authors have suggested that that up to one-quarter of MRONJ patients can present with the non-exposed variant (Fedele et al., 2015). Although this proportion is expected to be somewhat reduced after the inclusion of fistula in the definition (2014 AAOMS paper), efforts should be made to improve and expand case definition so to capture diagnosis in these patients including those with non-exposed MRONJ without fistulas.
The Group also suggested that the requirement of 8-week observation of potential MRONJ manifestation to fit the case definition may no longer be necessary. About one third to half of the affected individuals currently develop MRONJ without a history of dental extraction or other trauma (Otto, Pautke, Van den Wyngaert, Niepel, & Schiødt, 2018), and differential diagnosis with other dental and jawbone disease can be achieved without having to wait for 8 weeks (Bedogni et al., 2012).

MRONJ a Staging
At-risk category No apparent necrotic bone in patients who have been treated with either oral or IV bisphosphonates Stage 0 No clinical evidence of necrotic bone, but non-specific clinical findings, radiographic changes, and symptoms Stage 1 Exposed and necrotic bone, or fistulae that probe to bone, in patients who are asymptomatic and have no evidence of infection Stage 2 Exposed and necrotic bone, or fistulae that probe to bone, associated with infection as evidenced by pain and erythema in the region of the exposed bone with or without purulent drainage Stage 3 Exposed and necrotic bone or a fistula that probes to bone in patients with pain, infection, and one or more of the following: exposed and necrotic bone extending beyond the region of alveolar bone (i.e., inferior border and ramus in the mandible, maxillary sinus, and zygoma in the maxilla), resulting in pathologic fracture, extraoral fistula, oral antral/oral-nasal communication, or osteolysis extending to the inferior border of the mandible of sinus floor a Exposed or probable bone in the maxillofacial region without resolution for >8 weeks in patients treated with an anti-resorptive and/or an anti-angiogenic agent who have not received radiation therapy to the jaws.

| CONTROVER S IE S ON RIS K FAC TOR S
According to the literature, tooth extraction, infection, type and dosage of AR, and duration of treatment are considered to be risk factors.
Approximately half to two-thirds of MRONJ cases are reported to develop following a tooth extraction (Otto et al., 2018). Dental extraction was reported as a main risk factor in 73% of the cases of ONJ (Nicolatou-Galitis et al., 2011), and historically, these cases have been identified as a non-healing extraction socket (Bedogni et al., 2012). Accordingly, the vast majority of recommendations on dental treatment of patients on anti-resorptive or anti-angiogenic therapy have included advice against dental extractions as a mean to resolve dental infection (Khan et al., 2008;Khosla et al., 2007;Matsuo et al., 2014;Ruggiero et al., 2014;Yoneda et al., 2010).
However, a growing body of evidence suggests that dental infection, rather than dental extraction per se, might represent the main local risk factor for MRONJ (Otto et al., 2015;Panya et al., 2017;Saia et al., 2010). For example, a 2011 case-control study with three dental practice-based research networks in the United States found that the likelihood of developing osteonecrosis was higher (almost double) in patients with a history of suppuration compared to those with a history of dental extractions (OR 11.9 vs. 6.6) (Barasch et al., 2011). It is also increasingly reported that dental extractions in patients exposed to anti-resorptive therapy usually do not translate into MRONJ development, when tooth extraction is performed using alveolectomy and primary surgical mucosal closure (Heufelder et al., 2014;Otto et al., 2015).
Thus, surgical intervention per se should not be overemphasized as the main risk factor for MRONJ development. Similarly, it has been suggested that infection around the implants (peri-implantitis) represents a notable risk factor for MRONJ development (Giovannacci et al., 2016;Troeltzsch et al., 2016). This is also in line with the high success rate after surgery on the jawbone to cure MRONJ lesions (see later).
The Group suggested that dental infection might currently be a more common and relevant risk factor for MRONJ compared to extraction and that a notable proportion of MRONJ cases believed to have been triggered by dental extraction in fact represent cases of non-exposed MRONJ that had already developed because of dental/periodontal infection before the actual extraction took place.
Recent studies have reported the presence of histologically proven alveolar necrotic bone associated with dental/periodontal infection    AAOMS, 2007;Ruggiero et al., 2009Ruggiero et al., , 2014. The Group highlighted that AAOMS treatment recommendations, which are based on a clinically driven staging system, may fail to reflect the actual bone extension of MRONJ disease, with the risk of assigning patients to "inappropriate" treatments .

| CONTROVER S IE S ON MANAG EMENT/ TRE ATMENT
Accordingly, some authors advocated the adoption of treatments based also on the radiological aspects of MRONJ disease  in order to pick up early signs of disease or base therapeutic decisions on accurate assessment of disease extent. However, there remains no consensus on the efficacy of different radiological imaging modalities (e.g., CT, MRI, or nuclear imaging) in assessing, with high accuracy, the "true" MRONJ disease extent. (Bisdas et al., 2008;Devlin et al., 2018). A number of studies have compared specific imaging modalities and found inconsistent results in terms of overestimation/underestimation of the extension of MRONJ (Guggenberger et al., 2013;Stockmann et al., 2010).
The group advised that clinicians should be careful in adopting treat-

| RECOMMENDATIONS
Based on the discussion points summarized above, the group has produced a number of consensus key statements and recommendations so to inform clinicians and advice researchers, as a first step of reaching solutions.

| Key statements and recommendations relevant to definition and classifications of MRONJ
• Current widely adopted definition does not identify all patients affected by MRONJ • The current description of stage 0 is controversial, does not fulfill the definition of the disease, and may be misleading and difficult to interpret.
• Stage 0 of the AAOMS classification is a diagnostic challenge, as there are overlaps with dental and non-dental diseases.
Stage 0 may ultimately need confirmation by imaging and/or histopathology.
• Cases of non-exposed MRONJ without fistula should be included in the definition, possibly in terms of suspected or probable MRONJ after ruling out other dental and non-dental disease. The only ultimate proof of non-exposed MRONJ might be the histopathologic confirmation of necrotic bone. Decision on biopsy should be taken on an individual basis.
• The definition criterion of 8-week bone exposure/probing of bone does not apply to all cases and may delay diagnosis and consequently treatment.
• The role of imaging in the definition and classification of MRONJ needs further refinement. Imaging may aid in diagnosis (especially for non-exposed cases) and help determining disease extension and planning treatment.
• Present classification/staging does not adequately capture the extension and severity of MRONJ lesions. This may potentially affect treatment and prognosis.

| Key statements and recommendations relevant to risk factors for MRONJ
• Tooth extraction does not automatically translate into an increased risk of developing MRONJ, as certain surgical procedures notably reduce the risk.
• The reported high risk of developing MRONJ after tooth extraction might be related to an underlying pre-existing dental/ periodontal infection rather than to the surgery per se.
• The risk of developing MRONJ is not related to the way of administration as single factor; an accurate risk assessment should include an evaluation of the cumulative dosage and duration of anti-resorptive treatment. Typically, high-dose anti-resorptive therapy given to cancer patients with metastases is associated with higher risk of MRONJ development as compared to low-dose therapy given to osteoporosis patients.

| Key statements and recommendations relevant to management/treatment of MRONJ
• Because there is no accurate staging system reflecting the extension of MRONJ bone disease, it is problematic, and possibly misleading, to inform treatment recommendations on the basis of currently available staging systems.
• The term "conservative treatment" is used inconsistently in the literature and might include a number of different interventions ranging from topical antimicrobial mouthwashes to removal of superficial loose sequestra.
• The group recommends using the terms non-surgical versus surgical treatment.
• Recent literature suggests that non-surgical treatment may lead to disease progression.
• Surgical treatment is superior to non-surgical management in promoting long-term mucosal healing as well as absence of symptoms or radiological signs indicative of bone necrosis.
• If the aim of treatment is reduction of symptoms (pain) and control of infection, non-surgical treatment may be a valid management option. This seems particularly appropriate in frail elderly patients and in end-of-life oncology palliative setting.
• Early surgical intervention on localized disease may prevent progression and the need for subsequent extensive surgery (consider to treat surgically and early).

| G ENER AL SUMMARY
The Group has highlighted a number of controversial aspects of current knowledge and practice relevant to MRONJ, which have the potential to affect clinical management of patients as well as research.
The Groups suggest that key statement and recommendations presented in this paper might represent a useful tool so to stimulate a proactive discussion and inspire new and better-designed research, as first step to reach a consensus and improve the management of patients with MRONJ.

ACK N OWLED G EM ENTS
The workshops were supported by a donation from Amgen. All authors have received a fee for their contributions. The assistance from research nurse Line Eriksen in preparing the manuscript is highly acknowledged.

CO N FLI C T O F I NTE R E S T
No conflict of interest.

AUTH O R CO NTR I B UTI O N S
Each author has participated in the workshops and has actively contributed to the creative process of establishing the paper. The first author has also been responsible for heading the editing and the revision process.