OBJECTIVES: To estimate the cost-effectiveness of once-daily abiraterone acetate (AA) plus prednisolone for the treatment of advanced metastatic castration-resistant prostate cancer (mCRPC) after failure of taxane-based chemotherapy such as docetaxel, under the Brazilian Private Health System perspective. METHODS: A cost-effectiveness analysis was developed based in a Markov model to simulate the disease progression and patient mortality. A systematic revision of the literature was developed over the efficacy and safety of the use of AA and cabazitaxel (C), both combined with prednisolone (P), in patients diagnosed with advanced mCRPC. Efficacy data is informed by the Phase III trials (C + P versus mitoxantrone (M) + P and AA + P versus P).  Data is combined and adjusted via a mixed treatment comparison network meta-analysis to determine the relative efficacy of each comparator front a controlled therapy used as efficacy reference for the clinical tests (HR for overall survival (OS): C+P vs M = 0.703 (IC95%: 0.59-0.83); AA + P vsP = 0.649 (IC95%: 0.543-0.768)). It was assumed that M + P is equal to P alone. The costs and consequences of the disease treatment were computed for each treatment alternative. Only direct medical costs were considered. Costs and outcomes were discounted at 5% yearly. The outcomes considered were life years (LY) and quality adjusted life years (QALY). RESULTS: The incremental cost-effectiveness analysis demonstrated that AA is the most economically attractive medication. When the incremental cost-effectiveness ratio (ICER) for LY and QALY gained was evaluated, AA was dominant with regards to C, being more effective (LY: 1.3559 vs 1.2895; QALY: 0.7977 vs 0.7329) with lower costs (R$79,974 vs R$90,025).  CONCLUSIONS: AA is the best therapeutic option, with the best cost-effectiveness ratio, versus C for the treatment of patients diagnosed with advanced mCRPC under Brazilian private perspective.