Background: Therapies previously reserved for castration resistant prostate cancer (CRPC) have demonstrated improved progression-free (PFS) and overall survival (OS) in mHNPC. Four randomized trials have provided level 1 evidence for using either D or AA in addition to androgen deprivation therapy (ADT) for mHNPC, but the cost-effectiveness of these options has not been compared. Methods: A Markov cohort model was developed to project cost-effectiveness of each treatment until disease progression. Survival curves for progression/death were abstracted and digitized from the CHAARTED and LATITUDE studies. Clinically or financially significant adverse events (AEs) were modeled (neutropenia, neutropenic fever, and severe fatigue); utility values were obtained from the literature. Drug costs were obtained from a range of sources (Average Wholesale Price; VA costs). Effectiveness was measured in PFS quality adjusted life years (PFS QALYs) and cost-effectiveness was calculated using incremental cost-effectiveness ratios (ICER). Results: Adding D or AA to ADT improved PFS QALYs by 0.26 and 0.54, and increased cost by $12,185 and $208,684, respectively. Resulting ICERs were $46,519/QALY (D vs ADT) and $705,323/QALY (AA vs D). Results were highly sensitive to AA price, although even under lowest prices, the ICER was $404,451/QALY (AA vs DC) (Table 1). AA cost must be reduced by 76% for it to fall below a willingness-to-pay threshold of $150,000/QALY. Conclusions: Addition of AA modestly increases PFS QALYs compared with D, but substantially increases costs. While therapy subsequent to progression will impact the overall cost-effectiveness of the respective frontline options, the relative durations of treatment for CRPC are shorter. Thus, cost-effectiveness of mHNPC therapy is an important consideration given that OS is similar between studies for D and AA.

*Base case; LB = lower bound, UB = upper bound