Abstract
Diabetic retinopathy remains the most frequent cause of new cases of blindness among adults aged 20–74 years. A number of large clinical trials have validated treatment methods now considered standard. However, the disease continues to progress in approximately 50% of the eyes treated by photocoagulation. Other forms of therapy targeted at the earliest stages of retinal disease are needed. The difficulties in defining and accepting surrogate outcomes appropriate to evaluate the earlier stages of retinopathy are discussed. Special attention is given to trial design and the most likely possibilities for surrogate outcomes: mean difference on the Early Treatment Diabetic Retinopathy Study retinopathy scale of at least 2 steps per eye, reduction in macular thickening and reduction in fluorescein leakage.