Cumulative incidence and risk factors for medication-related osteonecrosis of the jaw during long-term prostate cancer management

Bone-modifying agents (BMA) are extensively used in treating patients with prostate cancer with bone metastases. However, this increases the risk of medication-related osteonecrosis of the jaw (MRONJ). The safety of long-term BMA administration in clinical practice remains unclear. We aimed to determine the cumulative incidence and risk factors of MRONJ. One hundred and seventy-nine patients with prostate cancer with bone metastases treated with BMA at our institution since 2008 were included in this study. Twenty-seven patients (15%) had MRONJ during the follow-up period (median, 19 months; interquartile range, 9–43 months). The 2-year, 5-year, and 10-year cumulative MRONJ incidence rates were 18%, 27%, and 61%, respectively. Multivariate analysis identified denosumab use as a risk factor for MRONJ, compared with zoledronic acid use (HR 4.64, 95% CI 1.93–11.1). Additionally, BMA use at longer than one-month intervals was associated with a lower risk of MRONJ (HR 0.08, 95% CI 0.01–0.64). Furthermore, six or more bone metastases (HR 3.65, 95% CI 1.13–11.7) and diabetes mellitus (HR 5.07, 95% CI 1.68–15.2) were risk factors for stage 2 or more severe MRONJ. MRONJ should be considered during long-term BMA administration in prostate cancer patients with bone metastases.

development of MRONJ 10,13 .A meta-analysis of 6 randomized controlled trials found a significantly higher incidence of MRONJ with denosumab compared to zoledronic acid over 3 years 7 .Additionally, a meta-analysis of 3 randomized trials indicated a lower incidence of adverse events, including MRONJ, when zoledronic acid was administered every 12 weeks rather than every 4 weeks 16 .However, the effects of BMA type and dosing interval on MRONJ remain unknown in cases treated with BMA for the long term.
The incidence of MRONJ in patients with prostate cancer (5-39%) has been reported to be higher than in patients with other cancers 1,11,17,18 .Patients with advanced prostate cancer now achieve prolonged survival owing to novel drugs and treatment modalities [19][20][21] .As a result, it is expected that the number of cases of long-term use of BMA would increase.However, the association between prostate cancer and a higher incidence of MRONJ remains unclear, and the risk factors for MRONJ have not been thoroughly discussed.Therefore, this study aimed to determine the cumulative incidence of MRONJ with the long-term use of BMAs in patients with prostate cancer with bone metastasis.We further made the null hypothesis that BMA type and dosing interval do not affect the occurrence of MRONJ.We also evaluate the cumulative incidence of stage 2 or more severe MRONJ and risk factors for MRONJ development.

Study design and data collection
This was a retrospective, observational study conducted at a single center to determine the cumulative incidence and risk factors for MRONJ in prostate cancer patients with bone metastases who were treated with BMAs.
The Department of Urology at our hospital has been using zoledronic acid since January 2008 and denosumab since July 2012 for patients with castration-sensitive or castration-resistant prostate cancer with bone metastases.In Japan, the utilization of zoledronic acid and denosumab for solid tumors with bone metastases is approved by the Japanese Drug Authority.Since July 2012, the attending physicians selected either zoledronic acid or denosumab, depending on the patient's preference and general condition.This study included 197 prostate cancer patients with bone metastases who were treated with BMAs between January 2008 and April 2023.Patients were enrolled at least 2 months after BMA administration.Patients who received BMAs for bone metastases from other carcinomas (n = 5) or for whom BMAs were used to treat hypercalcemia (n = 3) or for whom adequate clinical data were unavailable (n = 6) were excluded from the study.Ultimately, 179 patients met the study eligibility criteria (Fig. 1).Dental evaluation and care were discussed with the primary dentist before the initiation of BMAs.Zoledronic acid was administered intravenously at a dose of 4 mg every four weeks, and the dose was modified according to the manufacturer's instructions for patients with renal dysfunction.Denosumab was administered subcutaneously at a dose of 120 mg every four weeks.Physicians were allowed to extend the intervals of zoledronic acid and denosumab administration to 6-12 weeks at the start of BMA therapy.In principle, zoledronic acid and denosumab were continued, with discontinuation criteria being the occurrence of MRONJ or deterioration of the patient's Performance Status.www.nature.com/scientificreports/

Endpoints
The primary and secondary endpoints were the cumulative incidence of MRONJ and the occurrence of severe MRONJ, respectively.The risk factors for the development of MRONJ and severe MRONJ were also examined.When comparing the BMAs for analysis, the first-line BMA received by the patient was selected.MRONJ was diagnosed by dentists using the diagnostic criteria defined by the Japanese Allied Committee on Osteonecrosis of the jaw.The diagnosis was based on current or previous treatment with bisphosphonates or denosumab, the presence of exposed necrotic bone in the maxillofacial region lasting > 8 weeks, and no history of radiation therapy to the jawbone.The stage of MRONJ was indicated according to the 2017 Japanese Position Paper 22 , depending on the severity of symptoms.In this study, we defined severe MRONJ as MRONJ of stages 2 and 3.
We conducted analyses using data extracted from medical records to investigate the risk factors associated with MRONJ.

Statistical analysis
The cumulative incidence of MRONJ was calculated using Kaplan-Meier analysis.Risk factors for MRONJ were analyzed by Cox regression analysis using JMP software (JMP version 17.0.0;SAS Institute Inc., Cary, NC, USA).Statistical significance was set at p < 0.05.

Ethical approval
This study was approved by the Institutional Review Board of Osaka University (ethics review number: 13397-20).An opt-out consent procedure was employed for the use of patient data obtained in routine medical care.The informed consent was obtained from all participants prior to the initiation of BMA administration.The study was conducted in compliance with the Declaration of Helsinki.

Consent to participate
Informed consent was obtained from all participants included in the study.

Patient characteristics
The patient characteristics (n = 179) are summarized in Table 1.A total of 109 patients received zoledronic acid, and 70 patients received denosumab.Four patients who switched from zoledronic acid to denosumab during the study period were included in the zoledronic acid group: zoledronic acid duration (9, 17, 36, 36 months) and subsequent denosumab duration (37, 4, 4, 10 months).The median age of the patients was 72 years (interquartile range [IQR] 67-78), and the median observation period was 19 months (IQR 9-43 months).The treatment duration was longer, and the ALP, LDH, and BAP levels were higher at BMA induction in the zoledronic acid group than in the denosumab group.The Ca level at BMA induction was higher in the denosumab group than in the zoledronic acid group (Table 1).

Risk factors for the development of MRONJ of any stage
Table 3 presents the results of the analysis of factors associated with the development of MRONJ of any stage.In the univariate analysis, denosumab use rather than zoledronic acid use was a significant risk factor for the development of MRONJ (HR 3.26, 95% CI 1.46-7.24,p = 0.004).BMA administration at intervals longer than one month was associated with a lower risk of MRONJ than monthly BMA use (HR 0.13, 95% CI 0.02-0.95,p = 0.044).In the multivariate analysis, denosumab use was a significant predictor of MRONJ (HR 4.64, 95% CI 1.93-11.1,p < 0.001).In addition, BMA administration at longer than one-month intervals was associated with a lower risk of MRONJ (HR 0.08, 95% CI 0.01-0.64,p = 0.017).Kaplan-Meier curves are developed for the denosumab and zoledronic acid groups as well as for the monthly BMA and BMA administration at longer than one-month intervals groups (Figs.2C,D).www.nature.com/scientificreports/www.nature.com/scientificreports/

Cumulative incidence and risk factors for severe MRONJ
The cumulative incidence rates of severe MRONJ were 3%, 7%, 21%, and 51% at 1, 2, 5, and 10 years, respectively (Fig. 3A).Additionally, Table 4 outlines the results of analyzing the factors associated with severe MRONJ of stage 2 or higher.In the univariate analysis, an EOD grade of 2 or higher at BMA induction (six or more bone metastases) compared to an EOD grade of 0 or 1 (HR 3.73, 95% CI 1.16-11.9,p = 0.026) and the presence of diabetes mellitus (HR 4.38, 95% CI 1.53-12.5,p = 0.006) were the significant risk factors for the occurrence of severe MRONJ.In the multivariate analysis, an EOD grade of 2 or higher at BMA induction (HR 3.65, 95% CI 1.13-11.7,p = 0.030) and the presence of diabetes mellitus (HR 5.07, 95% CI 1.68-15.2,p = 0.004) were the significant predictors of severe MRONJ (Fig. 3B,C).

Discussion
This study focused on the cumulative incidence of MRONJ with long-term BMA use in prostate cancer patients with bone metastases.The cumulative incidence of MRONJ was approximately one-quarter at 5 years, but at 10 years, more than half of the patients had developed the disease.Denosumab use was the risk factor for MRONJ of any stage, whereas BMA administration at longer than one-month intervals was associated with a lower risk of MRONJ.Thus, the null hypothesis that BMA type and dosing interval do not affect the occurrence of MRONJ has been rejected.The study also found that the presence of diabetes and an EOD grade of 2 or higher at BMA induction were risk factors for severe MRONJ.www.nature.com/scientificreports/Recently, Nakai et al. 1 reported an MRONJ incidence rate of 27.5% in patients with prostate cancer and bone metastases at 5 years, which is similar to our findings.However, the incidence at 10 years remained unknown.In this study, the 10-year cumulative incidence of MRONJ was 61%.Although the majority of MRONJ cases occurred within the first 3 years, some MRONJ cases were observed even after long-term treatment.This highlights the substantial risk associated with long-term BMA use.The association between prolonged BMA use and an increased incidence of MRONJ has been increasingly reported 10,11,24,25 , making the optimal duration www.nature.com/scientificreports/ of BMA administration a crucial issue for the future, especially in patients with prostate cancer who undergo prolonged BMA therapy.Currently, 3-4-weekly zoledronic acid administration or 4-weekly denosumab administration is recommended to reduce the incidence of SREs in patients with CRPC with bone metastases.However, the optimal dosing intervals remain unclear.A systematic review of three randomized trials revealed a decreased occurrence of MRONJ, with zoledronic acid administered every 12 weeks compared to every 4 weeks 16 .In breast cancer, dosing intervals of zoledronic acid every 3-4 weeks and every 12 weeks are both recommended as first choices 26 .In prostate cancer, zoledronic acid and denosumab are recommended every 4 weeks according to the ASCO-CCO and ESMO guidelines 27,28 .However, given the increased incidence of MRONJ with prolonged administration of BMA, extending the dosing interval is an issue that should be considered.To date, there are no established views on extended BMA dosing intervals in prostate cancer.At our hospital, the dosing interval is currently extended based on the judgment of the clinician.This study found that patients with BMA administration intervals longer than one month had a lower risk of any stage of MRONJ than those that were administered BMAs every 4 weeks.Among patients who received BMA at intervals of longer than one month at the start of therapy, severe cases of MRONJ were rare (only one case).In patients who have been using BMAs for a long time, a decrease in the cumulative dose owing to longer dosing intervals may reduce the risk of developing MRONJ.
Although denosumab is more effective than zoledronic acid in reducing the incidence of SREs, an analysis of eight RCTs found a significantly higher incidence of MRONJ with denosumab than with zoledronic acid administration 7 .The incidence of MRONJ with denosumab use ranged from 0.5-2.1% at 1 year, 1.1-3.0%at 2 years, and 1.3-3.2% at 3 years 7 .Long-term denosumab use has been associated with an increased risk of MRONJ; however, observational studies on denosumab treatment beyond 3 years are limited.Our study identified denosumab use as an independent risk factor for MRONJ development compared with zoledronic acid use, even during long-term follow-up.The use of denosumab is expected to increase in the future because of its potential to mitigate nephrotoxicity and its non-requirement for intravenous infusion.The different degrees of osteoclast activity inhibition by zoledronic acid and denosumab may account for the differences in the incidence of MRONJ.In this study, there was a higher proportion of MRONJ stage 0 cases in the denosumab group.This could suggest that varying degrees of osteoclast inhibition led to an increase in non-exposed lesions, which do not exhibit bone exposure but show imaging evidence of bone resorption or osteonecrosis.However, the mechanisms underlying the increased risk of MRONJ associated with denosumab use remain unclear.When considering long-term BMA administration, it may be better to select zoledronic acid or increase the dosing interval for denosumab.A clinical trial investigating the incidence of MRONJ with 4-weekly denosumab administration versus 12-weekly administration is ongoing (NCT02051218), and its findings are eagerly anticipated.
This study revealed a 7% incidence of severe MRONJ at the 2-year mark, which escalated to a significant 51% cumulative incidence over 10 years.The American Association of Oral and Maxillofacial Surgeons (AAOMS) recently announced that the cumulative BMA dose was unrelated to severity, which differs from previous observations 29 .Reports of a higher incidence of severe MRONJ in patients receiving long-term zoledronic acid administration are consistent with the findings of the present study 5 .Depending on the disease stage, conservative or surgical therapy is recommended for MRONJ 29,30 .Many studies have reported better efficacy and outcomes in patients undergoing surgical therapy than in those who do not 31,32 , leading to a trend toward recommending surgical therapy.Palla et al. stated that early surgery has a higher cure rate, is less invasive, and is associated with a better quality of life than advanced-stage surgery 33 .Therefore, diagnosing MRONJ at an early stage is crucial.
Although the pathogenesis of MRONJ is not fully understood, it is believed to be related to the presence of numerous bacteria in the oral cavity and specific characteristics of the jawbone.Inflammation easily spreads to the jawbone because of tooth infection, and the jawbone is situated in an infection-prone area 34 .Furthermore, BMA administration, with its resultant bone remodeling inhibition, suppression of osteoclast activity, and angiogenesis inhibition combined with the abundance of oral bacteria and easily infected jawbones, is hypothesized to cause MRONJ 22,35 .Nashi et al. reported that age and low serum albumin levels were associated with severe MRONJ, reflecting delayed wound healing 36 .The carcinomatous state with a high number of bone metastases is a systemic inflammatory state in which inflammation delays wound healing and increases scar formation 37 .In this study, we found that the presence of numerous bone metastases with an EOD of grade 2 or higher at BMA induction was a risk factor for severe MRONJ.These results support the finding that MRONJ results from local inflammation that is associated with jawbone infection 38 and becomes more severe due to delayed wound healing.In this study, diabetes was identified as a risk factor for severe MRONJ.Diabetes has been reported to be associated with delayed wound healing 37 , and obesity has been associated with an increased number of oral microorganisms, contributing to periodontal disease 37,38 and poor oral hygiene 39,40 .These reports are consistent with our observation that diabetes is a risk factor for severe MRONJ.
This study had several limitations.Firstly, this was a single-center, retrospective, observational study with a relatively small number of patients who developed MRONJ, resulting in limited findings.Secondly, data on the outcomes of patients who developed MRONJ were lacking, making it unclear whether the disease improved, remained stable, or worsened.Thirdly, the MRONJ diagnoses were not made by a single dentist, introducing potential variability in the diagnostic process.Fourth, the assessment of MRONJ relied solely on AAOMS staging.Clinical symptoms such as pain and signs of infection, along with radiographical evaluation, were utilized to evaluate the disease, but some variability may exist.

Conclusion
In conclusion, this study revealed a significantly increased cumulative incidence of MRONJ in patients with prostate cancer and bone metastases after long-term BMA treatment.

Figure 2 .
Figure 2. Cumulative incidence rate of any stage of MRONJ.(A) While the majority of cases of MRONJ occurred within three years, there were also several instances of MRONJ occurring during long-term treatment.(B) The 2-year, 5-year, and 10-year cumulative incidence rates of MRONJ were 18%, 27%, and 61%, respectively.(C) In the patients receiving zoledronic acid or denosumab, denosumab use rather than zoledronic acid use was a significant risk factor for the development of MRONJ (p = 0.004).(D) In the patients with monthly BMA use or BMA use at longer than one-month intervals, BMA administration at longer than one-month intervals was associated with a lower risk of MRONJ than monthly BMA use (p = 0.044).MRONJ medication-related osteonecrosis of the jaw.

Figure 3 .
Figure 3.Cumulative incidence rate of severe MRONJ.(A) The 2-year, 5-year, and 10-year cumulative incidence rates of severe MRONJ were 7%, 21%, and 51%, respectively.(B) Among patients with or without diabetes, the incidence rate of severe MRONJ was significantly different between patients with diabetes and those without diabetes (p = 0.006).(C) Among patients with EOD grade at BMA induction ≧2 or EOD grade at BMA induction < 2, the incidence rate of MRONJ was significantly different between the patients with EOD grade ≧2 and those with EOD grade < 2 (p = 0.026).MRONJ medication-related osteonecrosis of the jaw; EOD extent of the disease.

Table 2 .
Cumulative incidence of MRONJ.MRONJ medication-related osteonecrosis of the jaw.

Table 3 .
Risk factors for any stage of MRONJ (univariate and multivariate Cox proportional hazard analyses).