Geographic atrophy growth rates in the Amish population: findings from the Amish eye study

Purpose To characterise the natural history and 2 year enlargement rate of geographic atrophy (GA) in the Amish population using fundus autofluorescence (FAF) imaging. Methods This retrospective analysis included participants from the Amish Eye Study. Of the 2668 eyes imaged at baseline and 1370 eyes with 2 year follow-up imaging, 80 eyes demonstrated GA at one or both visits. GA area and square-root–transformed (SQRT) GA area were measured at baseline and at 2 years. Focality and foveal involvement were also documented. Univariate and multivariate linear regression analyses were performed to evaluate the impact of FAF-derived and optical coherence tomography (OCT)-derived biomarkers on GA growth. Results GA lesions (n=32) demonstrated significant enlargement over 2 years on FAF (baseline 6.86 mm²; 2 years 9.56 mm²; mean change 2.70 mm²; p<0.001). Mean annual GA growth was 1.35 mm²/year on the raw scale and 0.25 mm/year after SQRT. In univariable mixed-effects analysis of SQRT growth, foveal involvement was the only significant predictor (β=0.115, p=0.031). In multivariable analysis, foveal involvement remained independently associated with faster growth (β=0.139, p=0.042), whereas baseline lesion size was not significant. Findings were consistent in the sensitivity model using untransformed baseline GA area. Conclusions In this Amish cohort, GA demonstrated consistent enlargement over the 2 year period, with mean annual growth rates of 1.35 mm2/year. Baseline foveal involvement was the strongest predictor of subsequent GA enlargement rate; specifically, GA lesions involving the fovea at baseline enlarged faster than lesions sparing the fovea. These findings provide important natural-history data for atrophy progression within a genetically and environmentally homogeneous population.