Abstract
OBJECTIVES: To assess the cost-effectiveness of aducanumab at its updated price for treating patients with mild cognitive impairment (MCI) and mild Alzheimer's disease (AD). DESIGN: Cost-effectiveness analysis. SETTINGS: A five-state Markov model was constructed using 10 000 virtual patients to assess the cost-effectiveness of aducanumab from the perspective of the US healthcare system. The model employed a one-year cycle time and a lifetime time horizon. Transition probabilities and mortality rates were derived from a literature review. To address uncertainty and generalise the base case results, both one-way and probabilistic sensitivity analyses were conducted. PARTICIPANTS: 10 000 virtual patients with MCI and mild AD. INTERVENTIONS: The study group consisted of patients using aducanumab, while the control group consisted of those using a placebo. PRIMARY AND SECONDARY OUTCOME MEASURES: Primary outcomes included costs and quality-adjusted life years (QALYs). In line with the healthcare system perspective, only direct medical costs were included. Drug costs were obtained from official records, while other medical costs were derived from literature reviews. Utilities used to calculate QALYs were also obtained from the literature. Incremental analysis was conducted to assess cost-effectiveness in the base case analysis by comparing the incremental cost-effectiveness ratio (ICER) against the willingness-to-pay (WTP) threshold. A discount rate of 3% was applied to both costs and effectiveness. RESULTS: From the perspective of the US healthcare system, compared with the control group, the study group had an incremental cost of US$143 821.1 and an incremental QALY of 0.10. The ICER of patients using aducanumab compared with those using placebo was US$1 012 219.0 per QALY gained, which was much greater than the WTP threshold of US$50 000 to US$150 000, indicating that using aducanumab was not cost-effective. One-way sensitivity analysis showed the five most sensitive parameters were relative risk of progressing from MCI to mild AD, the utility of MCI, initial age, discount rate and the price of aducanumab. In the probabilistic sensitivity analysis, when the WTP was the WTP threshold of US$150 000, the probability of aducanumab being cost-effective was 0%. In addition, when the probability of aducanumab being cost-effective was 50%, the WTP was US$1 180 000, and when the probability of aducanumab being cost-effective was 95%, the WTP was US$1 906 000. CONCLUSIONS: Even with the updated price being half of the original, aducanumab is still not cost-effective, underscoring the need for affordable, evidence-based AD treatments.