Abstract
Purpose :
Adeno-associated viral (AAV) gene therapy has shown promise in treating a wide range of ocular disease. Intravitreal injection would ideally be the route of administration however it has proven not to be as efficacious as the technically challenging subretinal injection. We hypothesize that the reduced efficacy is due to the physical and dilutive barriers of the vitreous and inner limiting membrane (ILM) preventing the diffusion of virus particles to the underlying retina. This study tests the hypothesis that mechanical removal of the ILM in non human primates (NHP) will result in safe and effective transfection of AAV using green fluorescent protein (GFP) as a reporter.
Methods :
6 NHP eyes underwent vitrectomy, posterior vitreous detachment, ILM peel with pooling of 1.7x1013 GC/ml of AAV-GFP under air. 4 additional eyes served as controls, undergoing only vitrectomy and pooling under air. The eyes were examined at regular intervals post-surgery and fundus autofluorescence (FAF) imaging was used to quantify the intensity and area of transfection. the NHP were euthanized 16 weeks post-surgery and immunohistochemical analysis was performed to ascertain the GFP expression at the cellular level.
Results :
Green fluorescence was detected as early as 2 weeks post-surgery in all eyes and was sustained throughout the study. At 16 weeks, eyes that underwent ILM peel showed a larger transfected area compared to non peeled eyes (50.7 (33.1 – 58.4) pixel2 versus 5.1 (0.6-7.6) pixel2 (P < 0.01) and a higher fluorescence intensity ratio (10.3 (2.2-18.5)) versus (1.9 (0.6 – 4.4)), p=0.05). There was better colocalisation in ILM peeled eyes compared to the non peeled eyes in the mueller cells (0.77(0.52 - 0.82) versus 0.39 (0.25 - 0.53), p = <0.05) and photoreceptor cell layer (0.579 (0.24 - 0.92) versus 0.16 (-0.19 - 0.53), p=<0.05). There was no significant difference in the colocalisation of cells in the retinal pigment epithelial layer (0.73 (0.32- 0.96) versus 0.51 (-0.30 - 1.69), p = 0.08) and the ganglion cell layer (0.62 (0.41 - 0.83) versus 0.61 (-0.50 - 1.72), p = 0.96).
Conclusions :
The results are consistent with our hypothesis that the vitreous and ILM serve as physical barriers to AAV transfection. We describe a practical and safe method for intravitreal AAV gene delivery which would also facilitate the potential need for repeated treatments.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.