Neuroretinal and RPE changes and susceptibility to Age-Related Macular Degeneration: insights from the longitudinal Alienor Study.

Purpose: We assessed the associations of macular layer thicknesses, measured using spectral-domain OCT (SD-OCT), with incident age-related macular degeneration (AMD) and AMD polygenic risk scores (PRS).

Design: Population-based cohort study PARTICIPANTS: 653 participants of the Alienor study, with biennial eye imaging from 2009 to 2024.

Methods: Macular layer thicknesses of eight distinct layers and three compound layers were automatically segmented based on SD-OCT imaging of the macula. Total and pathway specific PRS were calculated from previous AMD genome-wide association studies summary statistics. Associations of macular layer thicknesses with incident intermediate and advanced AMD were analyzed using time-dependent Cox proportional hazards models. Associations of macular layer thicknesses with PRS were assessed using linear mixed models.

Main Outcome Measures: Incident intermediate and advanced AMD based on fundus colour photographs and SD-OCT.

Results: Mean age at first OCT examination of the 653 participants was 82.2 ± 4.2 years and 61.3 % were women. In multivariate adjusted models, incident intermediate AMD was associated with thicker retinal pigment epithelium (RPE) - Bruch's Membrane (BM) complex in the 1 mm central circle (Hazard ratio (HR)= 1.13 for 1 μm increase; P= 8.08 x 10). Incident advanced AMD was associated with thicker RPE-BM complex in both the central circle (HR= 1.09; P= 0.005) and the inner circle (1 mm - 3 mm) (HR= 1.28; P= 1.61 x 10). Over the study period, RPE-BM complex thickening in the inner circle was more pronounced in individuals with high total PRS (ß= 0.06 μm/year for 1 standard deviation increase, P= 1.61 x 10), high complement pathway PRS (ß= 0.04 μm/year, P=3.23 x 10), high lipid pathway PRS (ß= 0.03 μm/year, P= 3.74 x 10) and ARMS2 (ß= 0.03 μm/year, P= 0.002). Further, high total PRS and high complement-specific PRS were associated with thinner photoreceptor segment layer (PSL) at baseline and with thinning of the outer nuclear layer over the study period.

Conclusion: These results highlight the importance of RPE-BM complex thickening in the pathophysiological sequence of AMD. Further longitudinal studies are needed, in particular to determine the value of RPE-BM thickening and PSL thinning measured using SD-OCT for the clinical follow-up of AMD patients.