Using Goldmann Visual Field Volume to Track Disease Progression in Choroideremia

PURPOSE: Choroideremia is an X-linked choroidopathy caused by pathogenic variants in the CHM gene. It is characterized by the early appearance of multiple scotomas in the peripheral visual field that spread and coalesce, usually sparing central vision until late in the disease. These features make quantitative monitoring of visual decline particularly challenging. Here, we describe a novel computational approach to convert Goldmann visual field (GVF) data into quantitative volumetric measurements. With this approach, we analyzed visual field loss in a longitudinal, retrospective cohort of patients with choroideremia. DESIGN: Single-center, retrospective, cohort study. PARTICIPANTS: We analyzed data from 238 clinic visits of 56 molecularly-confirmed male patients with choroideremia from 41 families (range, 1–27 visits per patient). Patients had a median follow up of 4 years (range, 0–56 years) with an age range of 5 to 76 years at the time of their visits. METHODS: Clinical data from molecularly-confirmed patients with choroideremia, including GVF data, were included for analysis. Goldmann visual field records were traced using a tablet-based application, and the 3-dimensional hill of vision was interpolated for each trace. This procedure allowed quantification of visual field loss from data collected over decades with differing protocols, including different or incomplete isopters. Visual acuity (VA) data were collected and converted to logarithm of the minimum angle of resolution values. A delayed exponential mixed-effects model was used to evaluate the loss of visual field volume over time. MAIN OUTCOME MEASURES: Visual acuity and GVF volume. RESULTS: The estimated mean age at disease onset was 12.6 years (standard deviation, 9.1 years; 95% quantile interval, 6.5–36.4 years). The mean field volume loss was 6.8% per year (standard deviation, 4.5%; 95% quantile interval, 1.9%–18.8%) based on exponential modeling. Field volume was more strongly correlated between eyes (r(2) = 0.935) than best-corrected VA (r(2) = 0.285). CONCLUSIONS: Volumetric analysis of GVF data enabled quantification of peripheral visual function in patients with choroideremia and evaluation of disease progression. The methods presented here may facilitate the analysis of historical GVF data from patients with inherited retinal disease and other diseases associated with visual field loss. This work informs the creation of appropriate outcome measures in choroideremia therapeutic trials, particularly in trial designs. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.