P062 Fragility fractures, atypical femoral fractures, and osteonecrosis of jaw after initial bisphosphonate treatment for three and five years: nested case-control and cohort studies using linked English primary-care, hospitalisation and mortality data

Background/Aims Oral bisphosphonates reduce fracture risk, but long-term use is associated with the risk of rare, atypical fragility fracture (AFF) and osteonecrosis of the jaw (ONJ), prompting guidance to consider “treatment pause”. Evidence underpinning treatment pause is limited to small extension trials and heterogeneous observational studies. Signals for increased fracture risk with longer interruptions are inconsistent. We evaluated the association between fragility fracture and bisphosphonate treatment interruption after three and five years of prescription and explored the incidence of AFF and ONJ during continued bisphosphonate prescription. Methods Design: Nested case control and retrospective cohort studies. Data source: Clinical Practice Research Datalink (Aurum) data that were collected from primary care and linked to hospitalization and mortality records in England. Participants: Adults newly prescribed alendronate or risedronate (1997-2022). Outcomes: Fragility fracture, AFF, ONJ. Cohort establishment: Participants with a calculated medication possession ratio (MPR) ≥67% in each of the first three and five years were followed up from the three-/five-year timepoints (start of follow-up) to the earliest of fragility fracture, death, study end, two-years, or last data availability. Cases experienced fragility fracture and were matched to up to four controls on age, sex, and year of start of follow-up using risk set matching. Interrupted, intermittent, and continued prescription was defined as MPR ≤33%, 34-66%, and ≥67% between three-/five-year timepoints and fragility fracture or matched date. Multivariable conditional logistic regression was used to assess the association between fragility fracture and bisphosphonate prescription. Potential confounders controlled for included, demographics, lifestyle factors, comorbidities, including the Charlson’s comorbidity index, medication such as vitamin D +/-calcium prescription and healthcare-seeking behaviour. Ascertainment of AFF or ONJ: Patients included in the above-mentioned cohort, who had not experienced an AFF or ONJ before start of follow-up, were followed up to the earliest of outcome (AFF or ONJ as appropriate, date of last bisphosphonate prescription, study-end, death, two years or last data availability. The incidence of AFF and ONJ was estimated. Results Data for 26,809 and 13,408 participants prescribed bisphosphonate for three and five years, respectively, were included. Fragility fracture was not associated with treatment interruption or intermittent treatment compared to continuous treatment in the three-years (adjusted odds ratio (aOR) (95% CI) 1.02 (0·85-1.24) and 1.13 (0.93-1.38)) or five-years (aOR (95% CI) 1.02 (0·85-1.24) and 1.13 (0.93-1.38)) bisphosphonate prescribed cohorts. After three- and five-years of bisphosphonate prescription, the incidence (95% CI) of AFFs was 1.93 (1.55-2.41) and 2.50 (1.92-3.26)/1,000 person-years, and the incidence of ONJ was 0.07 (0.02-0.21) and 0.00/1,000 person-years, respectively. Conclusion Bisphosphonate interruption for up to two years, after either three or five years of prescription, was not associated with an increased risk of fragility fractures. AFF and ONJ were uncommon. These findings support decision-making regarding treatment pause after adherent bisphosphonate therapy. Disclosure G. Nakafero: None. T. McKeever: None. N.S. Kumar: None. V. Welsh: None. S. Muller: Grants/research support; SM is partly funded by the NIHR Applied Research Consortium West Midlands (NIHR200165).. A. Abhishek: None.