OBJECTIVES: An economic model was developed to evaluate the cost-effectiveness of voretigene neparvovec (VN) compared to best supportive care (BSC) in individuals with RPE65-mediated inherited retinal dystrophy (IRD), from a French healthcare system perspective. Methods were based on HAS guidelines and international good research practices for modelling.

METHODS: A Markov model was used and included six health states (HS) based on AMA visual deficiencies classification: Moderate Visual Impairment (VI), Severe VI, Profound VI, vision limited to “counting fingers”, vision limited to “hand motion” to “no light perception” and death. A lifetime horizon was used. Transition probabilities were calculated based on the results from the VN phase III trial and a natural history study. The economic endpoints used in the model were blindness-free years (BFY) and quality-adjusted life years (QALYs). Utility data from a vignette study were used. Resource utilization and costs included: disease-related costs (diagnosis, mutation testing, paramedical care and follow-up), drug costs (acquisition, administration and adverse events), costs of medical transport and accommodation and death.

RESULTS: Patients treated with VN stayed longer in better HS while patients in the BSC arm progressed more quickly to worse HS. The treatment with VN results in a gain of 11.7 BFY and 4.5 QALY versus BSC. The ICER for VN versus BSC was €51, 552 per BFY and €132, 607 per QALY. Deterministic and probabilistic sensitivity analyses generally showed consistency with base case findings. When additional scenarios were explored, ICERs were most sensitive to variations in the multi-state model parameters and duration of treatment effect.

CONCLUSIONS: VN is the first gene therapy approved in RPE65-mediated IRD and represents a clinically significant advancement in the management of this disease. VN was associated with an important gain of QALYs versus BSC and can be considered a cost-effective therapy in this orphan disease.