Safety and Vision Outcomes of Subretinal Gene Therapy Targeting Cone Photoreceptors in Achromatopsia: A Nonrandomized Controlled Trial

Abstract

Importance: Achromatopsia linked to variations in the CNGA3 gene is associated with day blindness, poor visual acuity, photophobia, and involuntary eye movements owing to lack of cone photoreceptor function. No treatment is currently available.

Objective: To assess safety and vision outcomes of supplemental gene therapy with adeno-associated virus (AAV) encoding CNGA3 (AAV8.CNGA3) in patients with CNGA3-linked achromatopsia.

Design, setting, and participants: This open-label, exploratory nonrandomized controlled trial tested safety and vision outcomes of gene therapy vector AAV8.CNGA3 administered by subretinal injection at a single center. Nine patients (3 per dose group) with a clinical diagnosis of achromatopsia and confirmed biallelic disease-linked variants in CNGA3 were enrolled between November 5, 2015, and September 22, 2016. Data analysis was performed from June 6, 2017, to March 12, 2018.

Intervention: Patients received a single unilateral injection of 1.0 × 1010, 5.0 × 1010, or 1.0 × 1011 total vector genomes of AAV8.CNGA3 and were followed up for a period of 12 months (November 11, 2015, to October 10, 2017).

Main outcomes and measures: Safety as the primary end point was assessed by clinical examination of ocular inflammation. Systemic safety was assessed by vital signs, routine clinical chemistry testing, and full and differential blood cell counts. Secondary outcomes were change in visual function from baseline in terms of spatial and temporal resolution and chromatic, luminance, and contrast sensitivity throughout a period of 12 months after treatment.

Results: Nine patients (mean [SD] age, 39.6 [11.9] years; age range, 24-59 years; 8 [89%] male) were included in the study. Baseline visual acuity letter score (approximate Snellen equivalent) ranged from 34 (20/200) to 49 (20/100), whereas baseline contrast sensitivity log scores ranged from 0.1 to 0.9. All 9 patients underwent surgery and subretinal injection of AAV8.CNGA3 without complications. No substantial safety problems were observed during the 12-month follow-up period. Despite the congenital deprivation of cone photoreceptor-mediated vision in achromatopsia, all 9 treated eyes demonstrated some level of improvement in secondary end points regarding cone function, including mean change in visual acuity of 2.9 letters (95% CI, 1.65-4.13; P = .006, 2-sided t test paired samples). Contrast sensitivity improved by a mean of 0.33 log (95% CI, 0.14-0.51 log; P = .003, 2-sided t test paired samples).

Conclusions and relevance: Subretinal gene therapy with AAV8.CNGA3 was not associated with substantial safety problems and was associated with cone photoreceptor activation in adult patients, as reflected by visual acuity and contrast sensitivity gains.

Trial registration: ClinicalTrials.gov Identifier: NCT02610582.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Fischer reported receiving grants from the Kerstan Foundation during the conduct of the study; receiving personal fees from Novartis, Adelphia Values, Sanofi, and Retina Implant; receiving grants from Biogen and Casebia Therapeutics; receiving grants and personal fees from Bayer outside the submitted work; and having a patent to treat Retinitis Pigmentosa licensed. Dr Michalakis reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study and having a patent to PCT/EP2017/054229 pending. Dr Wilhelm reported receiving grants from the Tistou and Charlotte Kerstan Stiftung during the conduct of the study. Dr Kohl reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. Dr Weisschuh reported receiving grants from the Kerstan Foundation during the conduct of the study. Dr Schoen reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. Dr Kuehlewein reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. Dr Kahle reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. Dr Werner reported receiving grants from Kerstan Stiftung during the conduct of the study. Dr Peters reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. Dr Seeliger reported receiving grants from the Kerstan Foundation during the conduct of the study and having a patent to PCT/EP2017/054229 pending. Dr Zrenner reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study and having a patent to CNGA3 pending. Dr Biel reported receiving grants from the Kerstan Foundation during the conduct of the study and having a patent to PCT/EP2017/054229 pending. Dr Wissinger reported receiving grants from the Tistou and Charlotte Kerstan Foundation during the conduct of the study. No other disclosures were reported.

Figures

Figure 1.. CONSORT Flow Diagram

vg indicates vector genomes.

Figure 2.. Treatment Area and Association With Foveal Anatomic Features

Representative indocyanine green angiographic images from 2 patients at 1 year after treatment with low (1 × 1010 vector genomes) (A) and high (1 × 1011 vector genomes) (B) doses of adeno-associated virus encoding CNGA3 gene therapy showing no change in retinal and choroidal perfusion. The target area included the cone photoreceptor–rich macula in all cases (small dashed circles indicate prebleb area; larger circles, extent of full bleb; white X, retinotomy; blue X, preferred retinal locus at baseline; blue X with apostrophe, preferred retinal locus at 1 year after treatment). Foveal thickness as the mean (±2 SDs [error bars]) of all patients (C) and of individual patients (D) over time. Foveal thickness measurement at day 0 (immediately after surgery) was only available for 1 patient. BL indicates baseline.

Figure 3.. Visual Acuity and Contrast Sensitivity

A and B, Improvement of best-corrected visual acuity (BCVA) as the sum score of identified letters on a standardized chart over time. C and D, Increase in contrast sensitivity in patients from baseline over time. All data are presented as mean (±2 SDs [error bars]) of all 9 patients (C) and mean of 3 by dose group (D). Dose 1 was 1 × 1010 vector genomes, dose 2 was 5 × 1010 vector genomes, and dose 3 was 1 × 1011 vector genomes. BL indicates baseline.