Safety and pharmacokinetics of intravitreal 2-methoxyestradiol implants in normal rabbit and pharmacodynamics in a rat model of choroidal neovascularization.

Choroidal neovascularization (CNV) is the leading cause of severe vision loss associated with age-related macular degeneration. As the pathogenesis of CNV formation is better understood, mechanism-based therapies, including the use of antiangiogenesis inhibitors, have been investigated. 2-methoxyestradiol (2ME2), an endogenous metabolite of estradiol, has been shown in the chick allantoic membrane model and the corneal micropocket assay to have antiangiogenic properties. The authors sought to determine the safety and pharmacokinetics of sustained-release intravitreal 2ME2 implants in normal rabbit and their efficacy in a rat model of CNV. 2ME2 implants were constructed using two designs: implant A, a silicone-based reservoir implant for the rabnbit eye, and implant B, a microimplant matrix design for the rat eye. In vitro release rates of both implants were determined. New Zealand white (NZW) rabbits had implant A placed in the vitreous cavity of one eye and the ocular toxicity was evaluated by clinical examination, serial electroretinography (ERG), and histopathology over a 28 week period. The steady state clearance of 2ME2 in the rabbit eye was calculated from in vivo release rates divided by steady state vitreous concentrations. A CNV model in the Brown-Norway rat was performed by injecting an adenoviral vector encoding human vascular endothelial growth factor in the subretinal space. Following the injection, a 2ME2 or sham (no drug) microimplant was placed in the vitreous cavity. Animals were killed over a 3 week period and the eyes examined for CNV by histopathology. Results showed that following a short burst, the release rate of implant A followed zero-order kinetics, typical of reservoir devices, and the cumulative release of implant B was proportional to the square root of time, as expected for a matrix delivery device. The safety studies in normal rabbit showed no ocular toxicities by clinical examination, ERG, and histopathology. Pharmacokinetic evaluation in the rabbit showed mean 2ME2 vitreous levels within the therapeutic range for the inhibition of endothelial cell proliferation. The experimental rat model showed a significant reduction in CNV in eyes treated with the 2ME2 implant. In conclusion, sustained-release 2ME2 intravitreal implants, which can be designed to deliver potentially therapeutic vitreous levels of 2ME2 for an extended period of time, appeared to be safe in normal rabbit and effective in a rat model of CNV. Sustained-release 2ME2 intravitreal implants may hold promise in the treatment of recurrent CNV refractory to standard therapy. Copyright 2002 Elsevier Science Ltd.