Intravitreal sustained release of VEGF causes retinal neovascularization in rabbits and breakdown of the blood-retinal barrier in rabbits and primates

Vascular endothelial growth factor (VEGF) has been identified as a possible mediator of retinal neovascularization (NV), but it is not certain if VEGF alone is sufficient to cause retinal NV. We sought to investigate this issue by implanting ethylene-vinyl acetate copolymer pellets that slowly release VEGF into the vitreous cavity of rabbits and primates. Eyes were examined by indirect ophthalmoscopy, fundus photography, and fluorescein angiography and then animals were killed at various time points and immunocytochemical and ultrastructural evaluations were carried out. Seven days after implantation of a pellet containing 30 μg of human recombinant VEGF into the vitreous cavity of rabbits, retinal blood vessels became dilated and tortuous, and between days 14 and 21, retinal NV was noted in all eyes. Fluorescein angiography showed profuse leakage of dye from the anomalous vessels. Immunohistochemical staining for proliferating cell nuclear antigen (PCNA) showed positively staining nuclei in many of the endothelial cells of new blood vessels on the surface of the retina. Six eyes implanted with control pellets containing vehicle and two eyes implanted with pellets containing 30 μg of human serum albumin alone showed no retinal vascular abnormalities. Implantation of pellets containing 100 μg of VEGF into the vitreous cavity of primates resulted in iris NV and retinal vascular dilation and tortuosity very much like that seen in humans with ischemic retinopathies. Immunohistochemical staining for serum albumin showed widespread severe breakdown of the blood-retinal barrier (BRB). Histology showed dilated thin-walled retinal vessels. but unequivocal retinal NV could not be identified and staining for PCNA was negative. These findings indicate that sustained intravitreal release of VEGF causes widespread retinal vascular dilation and breakdown of the BRB. Retinal NV seems to require persistent high levels of VEGF at the retinal surface and can be achieved in rabbits providing a potentially useful model of retinal NV, but is difficult to achieve in primates. The extensive VEGF-induced disruption of the blood- retinal barrier suggests that VEGF antagonists may provide a new therapy for patients with ischemic retinopathies and macular edema.