The correlation of genetic risk scores with OCT biomarkers in AMD

Abstract

Purpose : Fifty-two genetic variants associated with different biological pathways were associated with the risk for developing age related macular degeneration (AMD). Yet, there is limited data on their association with specific features of the disease. We aimed to evaluate for genotype-phenotype correlations among patients with AMD to gain further insights to the pathogenic pathways involved in the disease.

Methods : To assess genetic risk, we utilized a weighted genetic risk score method that quantifies the cumulative genetic predisposition of individuals by summing the number of risk alleles multiplied by their effect size derived from the International AMD Genomic Consortium's analysis. The dataset consisted of 917 AMD patients (mean age±SD: 77±9 years) and 432 controls (mean age±SD: 71±8 years). A weighted score was calculated for 52 AMD risk variants (Global score), 19 complement-related variants (Complement score), 7 lipid-metabolism related variants (Lipid score), and 26 genetic variants of other pathways (other pathways score). For 578 AMD patients with available optical coherence tomography (OCT) images, OCT-based phenotypic features were annotated: typical drusen, complete retinal pigmented epithelium and outer retinal atrophy (cRORA), sub-retinal drusenoid deposits (SDD) and hyperreflective foci (HRF). Correlation tests and Logistic regression were used to explore the relationship between disease features and genetic scores.

Results : Higher mean±SD complement score was present in AMD patients (0.67±0.05) compared with controls (0.27±0.06; P<0.0001), as well as a higher global score (AMD: 0.65±0.07, Controls: 1.45±0.11; P<0.0001). Drusen presencecorrelated with lipid score (r=0.09, P=0.02). cRORA and HRF displayed a positive correlation with ARMS2/HTRA1 (r=0.11, P=0.004; r=0.08, P=0.03 respectively). Logistic regression analysis indicated associations between cRORA and complement score (OR=1.25, 95%CI 1.05-1.50, p=0.01) and between HRF and ARMS2/HTRA1 (OR=1.53, 95%CI 1.03-2.27, p=0.04).

Conclusions : The study provides insights into genotype-OCT biomarkers relationships in AMD. Genetic risk variants for the disease are associated with specific OCT-based phenotypic features, as illustrated by cRORA's association with a high complement score or HRF's link withARMS2/HTRA1. Such insight on the interplay between genetics and specific AMD characteristics, may have potential future implications for disease prediction and management.

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