BC patients who have evidence of bone metastases should be treated with BMAs in order to reduce skeletal related events. ONJ is a serious complication associated the use of BMAs. In this large population-based study we evaluate the patterns of use of BMAs and ONJ rates in a cohort of older BC patients.
Patients diagnosed with de novo metastatic BC between 2007-2013 were identified in the SEER-Medicare database. We identified the presence of bone metastases using specific PEDSF variables and claims. All patients were required to receive systemic anticancer therapy within one year of diagnosis. HCPCS codes were used to identify the use of BMAs within 1 year of cancer diagnosis, ONJ was identified using an established ICD-9 codes. Descriptive statistics and regression models were used.
A total of 1,528 patients were included. The median age of the cohort was 74 yo. Within one year of diagnosis of bone metastases 71.1% of the patients received BMAs (68% bisphosphonates, 27% denosumab, 5% both). Older patients (OR = 0.64; 95%CI 0.46-0.89), those with more comorbidities (OR = 0.61; 95%CI 0.43-0.85) or with full/partial state buy-in (OR = 0.65; 95%CI 0.49-0.87) -surrogate for poverty- were less likely to receive BMAs. Nineteen cases of ONJ were identified, all of them occurred among BMAs-treated patients (1.7%), the 2-year and 4-year cumulative rates were 1.4% and 4.0%, respectively. Similar rates of ONJ were observed between patients treated with bisphosphonates and denosumab. Median duration of BMA therapy among patients who developed ONJ was 20 months (IQR 10-43). No clinical predictors of ONJ were identified.
In this large, nationally-representative cohort, the majority of the BC patients with evidence of bone metastases received treatment with BMA according to current guidelines. Similar rates of ONJ were observed among bisphosphonates and denosumab users. ONJ occurred in approximately 2% of this SEER-Medicare population of patients with BC and bone metastases treated with BMA. Our findings are of significant value since they reflect the patterns of care and complications associated with these commonly used agents in the general population.
Clinical trial identification
Legal entity responsible for the study
Mariana Cgavez Mac Gregor.
CERCIT, Sudan G. Komen.
All authors have declared no conflicts of interest.