Economic Evaluation of Three Frequently Used Gonadotrophins in Assisted Reproduction Techniques in the Management of Infertility in the Netherlands

Background and Objective Subfertility represents a multidimensional problem associated with significant distress and impaired social well-being. In the Netherlands, an estimated 50,000 couples visit their general practitioner and 30,000 couples seek medical specialist care for sub-fertility. We conducted an economic evaluation comparing recombinant human follicle-stimulating hormone (follitropin alfa, r-hFSH, Gonal-F (R)) with two classes of urinary gonadotrophins-highly purified human menopausal gonadotrophin (hp-HMG, Menopur (R)) and urinary follicle-stimulating hormone (uFSH, Fostimon (R))-for ovarian stimulation in women undergoing in vitro fertilization (IVF) treatment in the Netherlands. Methods A pharmacoeconomic model was developed, simulating each step in the IVF protocol from the start of therapy until either a live birth, a new IVF treatment cycle or cessation of IVF, following a long down-regulation protocol. A decision tree combined with a Markov model details progress through each health state, including ovum pickup, fresh embryo transfer, up to two subsequent cryo-preserved embryo transfers, and (ongoing) pregnancy or miscarriage. A health insurer perspective was chosen, and the time horizon was set at a maximum of three consecutive treatment cycles, in accordance with Dutch reimbursement policy. Transition probabilities and costing data were derived from a real-world observational outcomes database (from Germany) and official tariff lists (from the Netherlands). Adverse events were considered equal among the comparators and were therefore excluded from the economic analysis. A Monte Carlo simulation of 5000 iterations was undertaken for each strategy to explore uncertainty and to construct uncertainty intervals (UIs). All cost data were valued in 2013 Euros. The model's structure, parameters and assumptions were assessed and confirmed by an external clinician with experience in health economics modelling, to inform on the appropriateness of the outcomes and the applicability of the model in the chosen setting. Results The mean total treatment costs were estimated as (sic)5664 for follitropin alfa (95 % UI (sic)5167-6151), (sic)5990 for hp-HMG (95 % UI (sic)5498-6488) and (sic)5760 for uFSH (95 % UI (sic)5256-6246). The probability of a live birth was estimated at 36.1 % (95 % UI 27.4-44.3 %), 33.9 % (95 % UI 26.2-41.5 %) and 34.1 % (95 % UI 25.9-41.8 %) for follitropin alfa, hp-HMG and uFSH, respectively. The costs per live birth estimates were (sic)15,674 for follitropin alfa, (sic)17,636 for hp-HMG and (sic)16,878 for uFSH. Probabilistic sensitivity analysis indicated a probability of 72.5 % that follitropin alfa is cost effective at a willingness to pay of (sic)20,000 per live birth. The probabilistic results remained constant under several analyses. Conclusion The present analysis shows that follitropin alfa may represent a cost-effective option in comparison with uFSH and hp-HMG for IVF treatment in the Netherlands healthcare system.