Abstract Leber congenital amaurosis is a group of inherited retinal dystrophies that cause severe sight impairment in childhood; RPE65-deficiency causes impaired rod photoreceptor function from birth and progressive impairment of cone photoreceptor function associated with retinal degeneration. In animal models of RPE65 deficiency, subretinal injection of recombinant adeno-associated virus (AAV) 2/2 vectors carrying RPE65 cDNA improves rod photoreceptor function, and intervention at an early stage of disease provides sustained benefit by protecting cone photoreceptors against retinal degeneration. In affected humans, administration of these vectors has resulted to date in relatively modest improvements in photoreceptor function, even when retinal degeneration is comparatively mild, and the duration of benefit is limited by progressive retinal degeneration. We conclude that the demand for RPE65 in humans is not fully met by current vectors, and predict that a more powerful vector will provide more durable benefit. With this aim we have modified the original AAV2/2 vector to generate AAV2/5-OPTIRPE65. The new configuration consists of an AAV vector serotype 5 carrying an optimized hRPE65 promoter and a codon-optimized hRPE65 gene. In mice, AAV2/5-OPTIRPE65 is at least 300-fold more potent than our original AAV2/2 vector.