Diabetes mellitus invariably induces retinopathy which causes a loss of vision that is the major cause of blindness in people of working age across most ethnic groups. Although there have been major advances in gene therapy technologies, there is still no effective cure-all gene therapy for diabetes mellitus. This may be due to (i) involvement of multiple genes that may have different influences on diabetes across different ethnic groups, (ii) immune response to viral vectors, (iii) local, specific transfection only and not into systemic circulation, (iv) lack of stable long-term expression, and (v) lack of control of gene expression. Hence, a separate approach to gene therapy of diabetic retinopathy is necessary due to the difficulties in treating the underlying diabetes. Diabetic retinopathy is the inevitable microvascular complication in the retina from diabetes mellitus. There are possible genetic bases in several pathophysiological pathways for diabetic retinopathy, including oxidation of retinal cells, polyol accumulation pathways, increased non-enzymatic glycation in retinal cells and the release of growth factors by endothelial cells. We review the candidate genes in these putative pathways for diabetic retinopathy and discuss the challenges for gene therapy. The eye is an isolated system with a strong blood-retinal barrier and therefore provides a challenge for delivery of drugs and vectors from the systemic circulation using traditional approaches. Newer delivery approaches include the use of nanoparticles, liposomes, and iontophoresis. We also consider the social and health economic dimension of diabetic retinopathy gene therapy. Diabetic retinopathy is the most common cause of blindness for people of working age. The loss of visual acuity caused by diabetic retinopathy creates a detrimental impact on the patients quality of life. This results in quality-of-life costs to the individual, the health care system and to society. Significant progress has been made in gene therapy approaches for diabetic retinopathy, and it appears that this is an important area for continued research in order to improve visual outcomes and reduce the healthcare costs of diabetic retinopathy in our communities.