Development of a Subretinal Drug Delivery System

Abstract

Abstract: : Purpose: To optimize a subretinal (SR) drug delivery method designed to overcome systemic toxicity and diffusion limits of larger molecules. Methods: Flexible poly(methyl methacrylate) or poly(caprolactone) or rigid Nitinol (polymer coated) filaments were implanted in the SR space of 15 rabbits. These materials can either be matrix–loaded or coated with therapeutic drugs. Rabbits underwent fundus examination, FA and OCT 1, 2 and 4 weeks postoperatively. Several surgical methods were investigated. A superior conjunctival peritomy was created, and sclerotomies were made with a 20 g MVR blade 1 mm posterior to the limbus in the superotemporal quadrant (and superonasally in eyes that underwent vitrectomy). The filament was inserted into the SR space with intraocular microforceps. A vitreous cutter and infusion line were used in rabbits that had a vitrectomy before implantation. A 25–39 g needle was used to administer SR BSS solution in rabbits that received a local retinal fluid detachment (bleb) before implantation. In some eyes a 30g needle was used to create a retinotomy before implantation; in others, the filament tip was used to puncture the retina. Results:There was no SR fluid accumulation, exudate, haemorrhage or fibrosis surrounding the device at any follow–up points. FA showed no vascular leakage. Fundus photos and OCT showed the implants in position without inflammation or migration (13/13 eyes). The flexible polymer–only filaments allowed conformal positioning in the SR space. The rigid Nitinol filaments were easiest to implant – though this was at the expense of potentially penetrating deeper tissue layers (2/6 eyes). SubR blebs minimized this problem. SubR filament implantations were initially carried out without any vitrectomy, separate retinotomy or bleb. The rabbit vitreous presented itself as an unforgiving mucous–adhesive barrier that made precise positioning of the SubR implant very difficult, but nonetheless possible (1/3 eyes). A vitrectomy before SR implantation improved surgical outcome (10/10 eyes), but a retinotomy did not. Tapered filament tips were quintessential for self–starting retinotomy inserts. Implant lengths >3 mm were hard to place SR, and part of the filament generally protruded epiretinally after the procedure. Conclusions: Best results were obtained with SubR implants that were flexible, in the range of 2–2.5 mm in length, with a tapered tip, when inserted following vitrectomy and the creation of a SubR fluid detachment.

Keywords: retina • retinal neovascularization