PRO24 COST-EFFECTIVENESS OF LANADELUMAB AND C1-INHIBITORS FOR PROPHYLAXIS AGAINST ACUTE ATTACKS OF HEREDITARY ANGIOEDEMA

Abstract

Hereditary angioedema (HAE) is a rare disorder characterized by recurrent episodes of body tissue swelling. We evaluated the cost-effectiveness of lanadelumab and the C1-inhibitors, Cinryze and Haegarda, for long-term prophylaxis against acute HAE attacks versus no prophylaxis in adult patients with HAE type 1/2 from a US health system perspective. We developed a Markov model with two health states: alive with HAE and dead with monthly cycles over a life-time horizon. Baseline characteristics were: mean age of 39.6 years, 68.4% female, mean weights of 88.8kg (male) and 76.4 kg (female), and a baseline attack rate of 3.39 per month. We tracked number of attacks, time attack free, survival, quality-of-life, and costs. Parameters were obtained from publicly available databases, published literature, and expert opinion. Outcomes included discounted (3% per year) quality-adjusted life-years (QALYs) and costs (2018 US$), and incremental cost per QALY gained (ICER). We conducted univariate and probabilistic sensitivity analyses, and scenario analyses. Expected QALYs were 17.47 for no prophylaxis, and ranged from 18.21 (Cinryze) to 18.66 (lanadelumab) for patients receiving prophylaxis. Expected costs were $9,953,000 for no prophylaxis and ranged from $10,343,000 (Haegarda) to $14,396,000 (Cinryze) yielding ICERs ranging from $328,000 (Haegarda) to $5,954,000 (Cinryze). ICERs were extremely sensitive to drug acquisition costs, the baseline attack rate and treatment effects. The probability that Haegarda and lanadelumab were cost-effective at a $150,000/QALY threshold was 36% and 5%, respectively. A scenario analysis showed that 75% of patients would need to be switched to every four-week dosing for lanadelumab to be cost-effective at a $150,000/QALY threshold. In the base case, drugs for prophylaxis against acute attacks of HAE were likely to exceed the cost-effectiveness threshold of $150,000/QALY. However, the model outcomes were highly sensitive to changes in model inputs and assumptions and should be interpreted with caution.