Table 2.

Drummond et al. checklist for assessing gene therapies completed for four economic evaluations of VN

Item	US-ICER [7, 8]	Johnson et al. [12] (US)	UK [9]	The Netherlands [11]
Clinical effectiveness
Surrogate endpoint used	VA and VF (secondary endpoints in phase III clinical trial) instead of multi-luminance mobility test (primary endpoint in RCT)
Rare disease (number of eligible patients)	1000–3000 [8]	1000–2000 [12]	86 [9]	45 [11]
Serious condition	Severe visual impairment throughout childhood with deteriorating vision over time resulting in complete blindness by 30–40 years of age
Single-arm trial	RCT with intervention and control arm
Paediatric population	Age in RCT: mean 15, median 11, interquartile range 6–20 years
Reporting of adverse consequences and risks	Only short-term adverse events included	No adverse events included	Only short-term adverse events included	Only short-term adverse events included
Size of clinical trial	Intervention arm n = 29; control arm n = 9
Length of clinical trial	1-year data used in model; follow-up data until 3–4 years available
Extrapolation to long-term outcomes, number of years extrapolation of treatment effect	10 years, followed by 10-year waning period from full treatment effect to natural history (i.e. no treatment effect)	Lifetime, i.e. > 70 years	40 years, followed by natural history (i.e. no treatment effect)	20 years, followed by natural history (i.e. no treatment effect)
Elements of value
Severe disease	Not considered			Considered in cost-per-QALY threshold
Value to caregivers	Not considered			Considered in a scenario analysis
Insurance value	Not considered
Scientific spillovers	Not considered
Lack of alternatives	Before the introduction of VN, there were no interventions that alter the natural history of RPE65-mediated IRD. Patients were treated with BSC
Substantial improvement in life expectancy	RPE65-mediated IRD does not affect mortality risk and therefore treatment with VN does not improve life expectancy directly
Other considerations
Discounting
Different discount rates explored	Not varied in scenario analysis	Not varied in scenario analysis	1.5% and 0% in scenario analyses	0%, 1.5% and 5% for costs and outcomes in scenario analyses
Uncertainty
Alternative payment models explored	Threshold analysis to estimate the maximum price of VN in order to be cost effective	Not considered	Confidential simple discount PAS	Pay for performance agreement (‘no cure, no pay’) considered in a scenario analysis

BSC best supportive care, IRD inherited retinal dystrophy, PAS patient access scheme, QALY quality-adjusted life-year, RCT randomized controlled trial, UK United Kingdom, US United States, US ICER United States Institute for Cost-Effectiveness Research, VA visual acuity, VF visual field, VN voretigene neparvovec