Cone mosaic integrity in choroideremia following gene augmentation via subretinal injection of AAV2-hCHM

Abstract

Purpose : The subretinal injection used during delivery of gene therapy vectors forces the neural retina to detach from the retinal pigment epithelium posing a risk of photoreceptor damage. Here, we use adaptive optics scanning light ophthalmoscopy (AOSLO) to assess the integrity of the cone mosaic and evaluate cone densities before and one-month after subretinal injections of AAV2-hCHM in subjects with choroideremia (CHM).

Methods : Low dose (up to 5x10¹⁰ vector genome (vg) per eye, n=5) or high dose (up to 1x10¹¹ vg per eye, n=4) AAV2-hCHM was delivered via uniocular subfoveal injection in 9 CHM patients, ages 26-57. The macular regions of both eyes were imaged pre- and one-month post-injection using a custom-built, multimodal AOSLO. Images were montaged within each timepoint, then manually aligned across timepoints with cellular accuracy. Cone inner segments were manually identified in pre- and post-injection images at multiple regions of interest (ROIs) in both eyes in all subjects. Bound cone densities were compared between timepoints and between injected and uninjected eyes. AOSLO images were compared with spectral-domain optical coherence tomography (SD-OCT).

Results : AOSLO images showed preservation of the cone inner segment mosaic at one-month post injection in all 9 AAV2-hCHM injected eyes, with the exception of foveal disruption in one patient. Cone-by-cone longitudinal alignment was attained at multiple ROIs in every case. There was no significant difference in cone density across timepoints or between injected and uninjected eyes. Cone densities were the same for injected eyes pre- vs one-month post-injection (mean ± standard deviation = 22,400 ± 12,000 vs 24,000 ± 11,200 cones/mm²) and uninjected eyes (24,500 ± 11,700 vs 24,300 ± 12,100 cones/mm²). Post-injection SD-OCTs showed retention of the foveal inner segment ellipsoid zone (EZ) in all eyes, but loss of EZ definition and cone outer segment tip signal in the patient with foveal disruption.

Conclusions : Cone mosaic integrity is maintained following subretinal delivery of AAV2-hCHM, providing strong evidence in support of the safety of the surgical injections and short-term exposure to this viral vector. Individual vulnerability to the subfoveal injection may account for the foveal cone loss observed in one patient. Future studies will evaluate the integrity of the cone mosaic following long-term AAV2-hCHM gene augmentation.

This is a 2021 ARVO Annual Meeting abstract.