UCSF

Purpose

To investigate the relationship between geographic atrophy (GA) progression and change in best-corrected visual acuity (BCVA) over 4 years and to identify factors associated with faster BCVA loss.

Design

Secondary analysis of 2 randomized controlled clinical trials.

Participants

Age-Related Eye Diseases Study (AREDS) and AREDS2 participants with GA secondary to nonexudative age-related macular degeneration.

Methods

Baseline and annual color fundus photographs were assessed for GA area and proximity to the foveal center. BCVA was measured using Early Treatment Diabetic Retinopathy Study logMAR charts. Analyses included BCVA change over 4 years, with the relationship of BCVA decline with GA progression and other baseline factors examined using multivariable linear mixed-effects models.

Main outcome measures

The primary outcome was BCVA change over 4 years. Secondary outcomes included BCVA change from baseline to years 1, 2, and 3.

Results

We included 1351 eyes from 994 participants, including 594 eyes from 464 participants with 4-year BCVA follow-up. Higher baseline BCVA, smaller baseline GA proximity to the foveal center, and greater GA growth rate were each independently associated with larger BCVA loss over 4 years (each P < 0.001). Among the 594 eyes with 4-year BCVA data, 69 eyes with a baseline BCVA < 40 letters (Snellen equivalent of 20/160 or worse) and 42 eyes with baseline GA located more than 1 mm from the foveal center did not experience significant BCVA loss over 4 years. In contrast, 483 eyes that met both criteria of baseline BCVA ≥ 40 letters and GA lesions involving or within 1 mm of the foveal center showed significant BCVA loss over 4 years (mean change = -11.33 letters [95% confidence interval = -12.80 to -9.84]), with faster GA progression associated with larger BCVA loss (P < 0.001).

Conclusions

In this cohort, eyes with GA involving or within 1 mm of the foveal center and a baseline BCVA of ≥ 40 letters appeared more likely to experience significant BCVA loss, suggesting these eyes may benefit more from therapies that slow GA progression. Our findings support a personalized approach to managing GA patients, potentially guiding the design of future GA trials.