Abstract
PURPOSE: Classify the appearance and quantify the growth rate of chorioretinal atrophy in patients who received voretigene neparvovec-rzyl (VN) for RPE65-mediated retinal degeneration. DESIGN: Multicenter retrospective analysis. SUBJECTS: Patients who underwent subretinal VN injection at 5 institutions and demonstrated posterior-pole chorioretinal atrophy. METHODS: Ultrawidefield scanning laser ophthalmoscopy or color fundus photos were assessed before and after subretinal VN. Atrophy was defined as regions with ≥ 2 of the following: (1) partial or complete retinal pigment epithelial depigmentation; (2) round shape; (3) sharp margins; and (4) increased visibility of choroidal vessels. Atrophy was qualitatively classified into different subtypes. All atrophy was manually segmented. Linear mixed-effects models with random slopes and intercepts were fit using atrophy area and square root of atrophy area. MAIN OUTCOME MEASURES: Number of eyes with each atrophy pattern, and slopes of linear mixed-effects models. RESULTS: Twenty-seven eyes from 14 patients across 5 centers developed chorioretinal atrophy after subretinal VN. A mean of 5.8 ± 2.7 images per eye obtained over 2.2 ± 0.8 years were reviewed, and atrophy was categorized into touchdown (14 eyes), nummular (15 eyes), and perifoveal (12 eyes) subtypes. Fifteen eyes demonstrated > 1 type of atrophy. Thirteen of 14 patients demonstrated bilateral atrophy. The slopes of the mixed-effects models of atrophy area and square root of atrophy area (estimate ± standard error) were 1.7 ± 1.3 mm(2)/year and 0.6 ± 0.2 mm/year for touchdown atrophy, 5.5 ± 1.3 mm(2)/year and 1.2 ± 0.2 mm/year for nummular atrophy, and 16.7 ± 1.8 mm(2)/year and 2.3 ± 0.2 mm/year for perifoveal atrophy. The slopes for each type of atrophy were significantly different in the square root of atrophy model, which best fit the data (P < 0.05). CONCLUSIONS: Chorioretinal atrophy after subretinal VN for RPE65-mediated retinal degeneration developed according to a touchdown, nummular, and/or perifoveal pattern. Perifoveal atrophy grew the most rapidly, while touchdown atrophy grew the least rapidly. Understanding the causes of these findings, which are present in a minority of patients, merits further investigation. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.