In vivo release and retinal toxicity of cyclosporine-loaded intravitreal device.

ABSTRACT

PURPOSE: To determine the in vivo release profile and retinal safety of cyclosporine A (CsA) delivered from a biodegradable poly-lactide-co-glycolide (PLGA) device in the vitreous cavity of rabbits' eyes. METHODS: A total of 60 animals (60 eyes) divided into two groups were used. For the in vivo release study, 32 eyes received PLGA implants containing 350 µg of CsA, and 16 eyes received the implants without drug (control). Four animals of CsA group and two of the control group were killed weekly until 8 weeks; the vitreous was removed, and CsA concentration was evaluated. Ophthalmological examination was performed in the animals prior to implant placement and weekly during the study period. Electroretinography (ERG) was performed in other six animals for each group, treated and control, at the beginning and at the end of the study (8 weeks) when they were killed and had their eyes processed for histology. RESULTS: No sign of inflammation was noticed on slit lamp examinations and the IOP maintained stable during the study period in CsA and control groups. CsA concentration in the vitreous (ng/ml) was 257.07 ± 117.23, 271.15 ± 98.96, 296.66 ± 86.25, 256.27 ± 99.22, 304.50 ± 88.18, 326.35 ± 105.24, 491.25 ± 119.90 and 589.93 ± 132.55 after 1, 2, 3, 4, 5, 6, 7 and 8 weeks of implantation, respectively. At the end of the study, 21.67 % of mass loss was found. The retina did not show any histological alteration in either group, but a significant reduction in dark-adapted b-wave amplitude was observed in the CsA group, with no changes in a-wave amplitude. CONCLUSIONS: These data show that the PLGA system is safe, but the selective reduction in ERG b-wave amplitude indicates that the PLGA with 350 µg CsA causes retinal function impairment, specifically on the rod postreceptor pathway, 8 weeks after implantation. These ERG changes were not associated with any histological damage as seen at the light microscopy level.