TOPLINE:

A novel outcome measure for geographic atrophy, the Geographic Atrophy Weighted-by-Acuity Index, aligned with best-corrected visual acuity, showing better correlation and predictive accuracy than the area of atrophy. This new measure, validated internally and externally, might offer improved tracking of vision changes, researchers reported. 

METHODOLOGY:

• Researchers conducted a prognostic study to assess a new geographic atrophy variable for potential use in clinical trials and to aid clinical decision-making; eyes with geographic atrophy were included.
• They analyzed data of 1528 eyes of 1120 participants (mean age, 77.1 years; 57.6% women) from the AREDS2 trial for training and internal validation to develop the Geographic Atrophy Weighted-by-Acuity Index. External validation was performed on 35 eyes of 35 participants (mean age, 74.3 years; 60% women) from the Geographic Atrophy Minocycline Trial.
• Best-corrected visual acuity was measured using Early Treatment Diabetic Retinopathy Study charts, and both color fundus photographs and fundus autofluorescence images were graded at specialized reading centers.
• Analysis included the division of the macula into 60 annuli, calculation of geographic atrophy occupancy for each annulus, and derivation of the novel outcome measure using ridge regression with geographic atrophy occupancy and age as predictors. The outcome for model development was deficit in the best-corrected visual acuity letter score (letter score, 100).
• The correlation between the novel geographic atrophy outcome measure and best-corrected visual acuity letter score (range of interest, 35-85) was assessed, with similar assessment performed for the geographic atrophy area. The area under the receiver operating characteristic curve was used to evaluate the ability to predict clinical visual acuity thresholds.

TAKEAWAY:

• The novel geographic atrophy outcome measure was more accurate at predicting best-corrected visual acuity letter scores cross-sectionally than the geographic atrophy area in both internal and external validation.
• The novel geographic atrophy outcome measure was more strongly correlated with best-corrected visual acuity letter score deficit at baseline than the geographic atrophy area on internal validation (Pearson correlation coefficient [r], 0.58 vs 0.32; difference, 0.27; P < .001).
• External validation confirmed a better correlation between the novel outcome measure and best-corrected visual acuity letter score deficit for both color fundus photograph-defined geographic atrophy (Pearson r, 0.69 vs 0.58; difference, 0.12; P = .005) and fundus autofluorescence-defined geographic atrophy (Pearson r, 0.70 vs 0.56; difference, 0.13; P = .002).
• In the AREDS2, longitudinal analysis revealed a stronger correlation between an increase in the novel outcome measure and a decline in best-corrected visual acuity letter scores compared with geographic atrophy area changes (Pearson r, 0.37 vs 0.28; difference, 0.09; P < .001).

IN PRACTICE:

“The novel GA [geographic atrophy] outcome measure could play a complementary role, as a secondary measure, alongside existing measures. Then, meaningful structural measures would be available for therapies that slow GA expansion macula-wide or preferentially near the center,” the researchers reported. “This novel GA outcome measure is designed to combine the advantages of a structural outcome measure with those of” best-corrected visual acuity, they added.

SOURCE:

The study was led by Jintong Hou, PhD, of the Division of Epidemiology and Clinical Applications at the National Eye Institute in Bethesda, Maryland. It was published online on January 29, 2026, in JAMA Ophthalmology.

LIMITATIONS:

Color fundus photographs were used as grading may show less variation in area measurements on fundus autofluorescence. Optical coherence tomography was not available for foveal localization in the AREDS2. The annulus approach assumed equal importance of all pixels within each annulus.

DISCLOSURES:

This study was supported by the Intramural Research Program of the National Institutes of Health, Department of Health and Human Services, including contributions from various NIH institutes. One author reported receiving an unrestricted grant from Research to Prevent Blindness, Inc., to the Department of Ophthalmology and Visual Sciences, University of Wisconsin-Madison. No relevant conflicts of interest were disclosed by the authors.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.