Safety and durability of effect of contralateral-eye administration of AAV2 gene therapy in patients with childhood-onset blindness caused by RPE65 mutations: a follow-on phase 1 trial

Abstract

Background: Safety and efficacy have been shown in a phase 1 dose-escalation study involving a unilateral subretinal injection of a recombinant adeno-associated virus (AAV) vector containing the RPE65 gene (AAV2-hRPE65v2) in individuals with inherited retinal dystrophy caused by RPE65 mutations. This finding, along with the bilateral nature of the disease and intended use in treatment, prompted us to determine the safety of administration of AAV2-hRPE65v2 to the contralateral eye in patients enrolled in the phase 1 study.

Methods: In this follow-on phase 1 trial, one dose of AAV2-hRPE65v2 (1.5 × 10(11) vector genomes) in a total volume of 300 μL was subretinally injected into the contralateral, previously uninjected, eyes of 11 children and adults (aged 11-46 years at second administration) with inherited retinal dystrophy caused by RPE65 mutations, 1.71-4.58 years after the initial subretinal injection. We assessed safety, immune response, retinal and visual function, functional vision, and activation of the visual cortex from baseline until 3 year follow-up, with observations ongoing. This study is registered with ClinicalTrials.gov, number NCT01208389.

Findings: No adverse events related to the AAV were reported, and those related to the procedure were mostly mild (dellen formation in three patients and cataracts in two). One patient developed bacterial endophthalmitis and was excluded from analyses. We noted improvements in efficacy outcomes in most patients without significant immunogenicity. Compared with baseline, pooled analysis of ten participants showed improvements in mean mobility and full-field light sensitivity in the injected eye by day 30 that persisted to year 3 (mobility p=0.0003, white light full-field sensitivity p<0.0001), but no significant change was seen in the previously injected eyes over the same time period (mobility p=0.7398, white light full-field sensitivity p=0.6709). Changes in visual acuity from baseline to year 3 were not significant in pooled analysis in the second eyes or the previously injected eyes (p>0.49 for all time-points compared with baseline).

Interpretation: To our knowledge, AAV2-hRPE65v2 is the first successful gene therapy administered to the contralateral eye. The results highlight the use of several outcome measures and help to delineate the variables that contribute to maximal benefit from gene augmentation therapy in this disease.

Funding: Center for Cellular and Molecular Therapeutics at The Children's Hospital of Philadelphia, Spark Therapeutics, US National Institutes of Health, Foundation Fighting Blindness, Institute for Translational Medicine and Therapeutics, Research to Prevent Blindness, Center for Advanced Retinal and Ocular Therapeutics, Mackall Foundation Trust, F M Kirby Foundation, and The Research Foundation-Flanders.

Figure 1. Subretinal areas of second-eye injection and Goldmann visual fields

Scotomas and the natural blind spot are shown in black.

Figure 2. FST with (A) white, (B) blue, and (C) red stimuli, (D) visual acuity measurements, and (E) mobility change score

(A–C) Note that the scales of the y axes are different. The more negative the threshold, the higher the sensitivity. (D) Visual acuity measurements are shown as mean logMAR scores for each visit. (E) The mobility change score reflects the change in the ability of the patients to pass the mobility test under lower illuminance than at baseline. Error bars represent 1 SE. Only significant p values (ie, p<0·05) are shown. See appendix p 6 for detailed statistical analyses. FST=full-field light sensitivity thresholds.

Figure 3

Full-field light sensitivity threshold results immediately before and after administration of AAV2-hRPE65v2 to the second, contralateral eye in patients given low-dose treatment in their first eye

Figure 4

Full-field light sensitivity threshold results immediately before and after administration of AAV2-hRPE65v2 to the second, contralateral eye in patients given medium-dose treatment in their first eye

Figure 5

Full-field light sensitivity threshold results immediately before and after administration of AAV2-hRPE65v2 to the second, contralateral eye in patients given high-dose treatment in their first eye

Figure 6. Change in mobility score

Participants are listed according to their original dose assignment when the first eye was injected: (A) low dose, (B) medium dose, and (C) high dose. A positive change score reflects the ability to navigate more accurately and quickly at dimmer light levels than at baseline. See appendix p 6 for detailed statistical analyses. *Mobility testing for NP01, CH11, and CH12 was not standardised until their follow-on visits in year 1. All other participants began the follow-on study after mobility testing standardisation.

Figure 7. fMRI results at baseline of the follow-on study and at 1 year after administration to the second eye

Participants are listed according to their original dose assignment when the first eye was injected: (A) low dose, (B) medium dose, and (C) high dose. Significant areas of cortical activations are shown as orange clusters overlaid onto the medial and lateral representations of the inflated cortex. Results are presented for the left eye for all but one participant (CH09), who received administration to his right eye in the follow-on study. For follow-up fMRI results, a stringent statistical threshold of false discovery rate <5%, corrected p<0·004, and continuous cluster area ≥100 mm² was used for all patients (except NP01).^, For baseline fMRI results, a relaxed threshold of uncorrected p<0·05 and continuous cluster area ≥50 was used for all patients (except NP15 and NP02) to reveal cortical activations.