Association of Hyperautofluorescence Signals with Geographic Atrophy Progression in the METforMIN Trial

Abstract

Purpose : We aimed to investigate the association between rim area focal hyperautofluorescence (RAFH) signals and geographic atrophy (GA) growth rates. We also calculated associations between RAFH and patient-specific factors, as well as the impact of oral metformin on the growth of RAFH over time.

Methods : We included 71 eyes from 44 participants from the Metformin for the Minimization of GA Progression trial who had at least 6 months of follow-up. Two masked graders, each separately, identified and measured RAFH lesions with a proprietary semiautomatic segmentation software and ImageJ. We defined RAFH as the ratio of hyperautofluorescent areas within the 450 µm perimeter surrounding the GA and the overall area enclosed within the perimeter after subtracting GA area. We calculated the association between the GA growth rate and baseline RAFH using Spearman correlation and used linear mixed-effects models to investigate: (1) associations between baseline RAFH and patient-factors such as age, sex, body mass index, (2) the longitudinal change of RAFH, and (3) the impact of metformin on RAFH.

Results : The intraclass correlation between the two graders was 0.78. Baseline RAFH was positively associated with the baseline square root of GA area (P < 0.001) but not with age, sex, body mass index, number of baseline GA lesions, fundus autofluorescence pattern, as well as history of smoking or cardiovascular disease. The association between baseline RAFH and GA growth rate (in mm²/year) was statistically significant (Spearman’s r = 0.54; P < 0.001), but the association became non-significant after using the square root transformed GA growth rate (in mm/year) (Spearman’s r = 0.22; P = 0.071) as shown in Figure 1. RAFH increased significantly over time at 0.20/year (P < 0.01). Oral metformin did not significantly affect the change in RAFH over time (P = 0.89).

Conclusions : RAFH can be graded manually with moderate reproducibility. RAFH increased slightly but significantly over time. The growth rate of GA area was positively associated with RAFH, but the association was no longer significant after using square root transformed GA area. Metformin does not impact the overall growth rate of RAFH.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

Figure 1. Left: GA growth rate in (mm²/year). Right: Square root of GA growth rate in (mm/year). GA = geographic atrophy; RAFH = rim area focal hyperautofluorescence.