Cost to Medicare from bone modifying agent overuse in castration-sensitive prostate cancer.

12 Background: Bone modifying agents (BMAs) prevent skeletal related events among patients with castration-resistant prostate cancer (mCRPC) and bone metastasis, but not among patients with castration sensitive prostate cancer (CSPC). However, approximately 1/3 of Medicare patients with CSPC receive BMAs unnecessarily, and the costs to Medicare resulting from this overuse have not been assessed. Methods: We used linked SEER registry and Medicare claims data for persons diagnosed with stage IV prostate adenocarcinoma during 2011-2015, identified those who subsequently received a cancer therapy used exclusively for mCRPC, and measured 1) the number of doses of zoledronic acid and denosumab received between diagnosis and initiation of mCRPC therapy and 2) presence of patient comorbidities which may warrant BMA therapy for fragility fracture prevention (eg., osteoporosis, osteopenia). We estimated excess BMA costs to Medicare by assessing the proportion of patients who received BMA therapy during their CSPC interval and did NOT have an indication for fragility fracture prevention, using the Medicare fee schedule for drug prices. We estimated adverse event costs using event frequency and costs estimates drawn from the peer-reviewed literature. The modeled outcome was the cost to Medicare from excess BMA use for the cohort of patients diagnosed with stage IV prostate cancer each calendar year. We conducted one-way sensitivity analysis for each model input by using the highest and lowest estimates available from peer-reviewed sources, monthly vs. every-three-month dosing frequencies for both drugs, and by +/-20% for all other inputs. Results: The median time from diagnosis to initiation of CRPC therapy was 387 days (IQR 253,573), during which time 42% received >=1 dose of denosumab (mean doses = 7) and 18% received >=1 dose of zoledronic acid (mean doses = 7). The estimated, annual excess BMA cost to Medicare was $45,392,377, comprised of $43,303,078 and $45,512 in drug costs for denosumab and zoledronic acid, respectively, and $1,625,869 and $419,917 in adverse event costs, respectively. The estimate was most sensitive to denosumab administration schedule (monthly vs. every-three-month) [estimate range: $29,487,144 - $101,060,692] and the duration of CSPC [estimate range: $37,853,180 - $101,754,921]. Conclusions: BMA overuse in CSPC incurs substantial cost to Medicare, largely due to denosumab drug costs. Our sensitivity analyses suggest we may overestimate costs if denosumab dosing frequency is now lower than during our observation period (eg., if more providers now use 3-month dosing), and we may underestimate costs if patients now experience CSPC for a longer duration (a plausible consequence from the introduction of ARSi therapy for CSPC in the post-LATITUDE era). Avoidable costs and toxicity may be reduced by greater adherence to guideline-concordant use of BMAs.