Abstract
Purpose :
To investigate the association between perilesional fundus autofluorescence (FAF) patterns and the ratio of photoreceptor (PR) and retinal pigment epithelium (RPE) integrity loss on optical coherence tomography (OCT) in patients with geographic atrophy (GA) secondary to age-related macular degeneration (AMD) in the phase 3 OAKS and DERBY trials.
Methods :
Spectralis OCT and FAF images from the OAKS (NCT03525613) and DERBY (NCT03525600) trials, two phase 3, randomized controlled trials of GA secondary to AMD with intravitreal pegcetacoplan were analyzed. RPE and PR integrity loss was measured on OCT in an automated way by previously validated deep learning-based algorithms. Perilesional FAF patterns were graded by a centralized reading center from the trials according to the Classification of Atrophy Meetings criteria. Pooled study eyes and fellow eyes that met the inclusion/exclusion criteria from both trials were divided into four quartiles (Q1-Q4) determined using baseline PR/RPE loss ratios on OCT. The distribution of FAF patterns between the quartiles was analyzed. The association between baseline FAF pattern and growth of automatically measured RPE and PR integrity loss on OCT will be presented at the meeting for the first time.
Results :
FAF and OCT gradings were available for 1317 pooled study and fellow eyes from both trials together. All study eyes had to present mandatory FAF patterns for inclusion. There was a significantly higher number of eyes with the focal pattern in small quartiles compared to high quartiles (p<0.001). The diffuse-trickling pattern was significantly more present in high quartiles compared to low quartiles (p=0.005) reflecting the high density of pseudodrusen as a known marker of GA progression. The banded and patchy pattern did not show a significant difference between the quartiles. The distribution of FAF patterns between PR/RPE loss ratios per quartile is summarized in Figure 1.
Conclusions :
Perilesional FAF patterns are associated with baseline PR/RPE loss ratios and disease activity on OCT in a very variable manner. This finding supports the use of automated measurements of PR/RPE loss ratios on OCT as a reliable biomarker for risk stratification in GA patients rather than specific FAF patterns. This will help in personalized patient selection as well as treatment management of GA patients in clinical routine.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.