The Waterfall in Disease Progression of Early Onset Stargardt Macular Dystrophy: Implications for Clinical Trial Design

Abstract

Purpose : Autosomal recessive Stargardt disease has a wide phenotypic range, but clinical features can be used to group patients into several distinct disease classes. To design clinical trials with the best chance of measuring meaningful treatment effects for each subject, we will need to characterize the shape, variability, and ordering of phenotype progression curves for each disease class. Here, we present functional and structural progression curves for two sibships that manifest with a severe, early childhood onset, "thickened ELM" phenotype with parafoveal loss of outer retina.

Methods : Two sibships were studied, with 4 affected members in Sibship A and 2 affected members in Sibship B. Disease status was classified through clinical evaluation, and mutations in ABCA4 were identified using a tiered-genetic testing strategy. Visual acuity, infrared scanning laser ophthalmoscopy images, and optical coherence tomography scans (Heidelberg Spectralis) were collected as part of routine clinical care. Total retinal thickness was extracted from segmentation data embedded in Heidelberg OCT VOL files.

Results : Across both families, five out of six individuals manifested loss of visual acuity in three phases (slow, rapid, and slow/stable), with the beginning and end of the rapid phase variable between family members (see Figure). For total retinal thickness, progression curves vary with eccentricity from the fovea, with gradual loss of the parafovea preceding the rapid loss of the fovea. In Patient A4, the rapid loss of total retinal thickness in the fovea precedes the rapid loss of visual acuity, highlighting a discrepancy between structural and functional metrics to consider in clinical trial design.

Conclusions : Progression of visual acuity and total retinal thickness both follow an approximately S-shaped curve, reminiscent of a waterfall. This type of curve is consistent with cellular toxicity effects from accumulating bisretinoids. The tight correlation between eyes and variability between family members suggests the effect of one or more modifying factors. Successful treatment trials will need to account for both the shape and ordering of the progression curves and the effects of modifiers if meaningful changes in visual function and structure are to be measured over the duration of the trial.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.