1768MO Incidence of fracture related hospitalisations in men with de novo high risk localised and metastatic hormone sensitive prostate cancer: Analysis of routinely collected healthcare data from the STAMPEDE docetaxel and zoledronic acid comparisons

Androgen deprivation therapy (ADT) is the mainstay medical treatment for men with locally advanced (M0) and metastatic hormone sensitive prostate cancer (M1 mHSPC): bone loss and increased fracture risk are recognised complications. The STAMPEDE trial compared patients (pts) treated with ADT ± docetaxel (Doc) ± zoledronic acid (ZA). No survival benefit was demonstrated with the addition of ZA however, long term effects on bone health and fracture risk were not formally collected within the trial. Health systems data through Hospital Episode Statistics (HES) for pts in England provides data with demonstrated integrity and provenance beyond standard trial follow up permitting evaluation of fracture risk. HES data were obtained (up to Mar 2021) for pts randomised to ADT (Arm A), ADT+ZA (Arm B), ADT+Doc (Arm C) and ADT+Doc+ZA (Arm E). ZA (4mg) was given as six 3 weekly cycles, then 4 weekly for 2 years. Fracture related hospitalisations (FRH) were identified using a prespecified coding framework of ICD10 diagnosis and OPCS procedure codes. Flexible parametric competing risks models were used to estimate 5- and 10yr cumulative incidence of FRH and subdistribution hazard ratios (SDHR). Linked data were available for 2,042 (734 M0, 1,308 M1) out of 2,140 eligible pts (95%). The 5-year cumulative incidence of FRH for M1 and M0 pts treated with ADT were 23% (95% CI 19-28%) and 11% (95% CI 8-15%) respectively. The 10-year cumulative incidence with ADT in M0 pts was 26% (95% CI 20-33%). Addition of ZA significantly reduced the incidence of FRH in M1 pts (SDHR 0.73, 95% CI 0.55-0.97; p=0.015). Data were inconclusive in M0 pts (SDHR 0.88, 95% CI 0.59-1.32, p=0.549). Doc had no significant effect on FRH in both M1 (p=0.264) and M0 (p=0.570), with no evidence of interaction between ZA and Doc in both M1 (p=0.526) and M0 (p=0.805). High incidence of 10yr fracture related hospitalisations was observed in STAMPEDE trial participants with either M0 or M1 prostate cancer. Addition of ZA reduced risk of FRH in people with M1 disease but not with M0 disease. These data support the use of bone protection agents to reduce fracture risk in men with mHSPC.