NEW YORK – A large international team has analyzed the genomes of multi-ancestry cohorts to track down dozens of new and known genetic risk loci for age-related macular degeneration (AMD), an eye condition affecting the central retina that tends to be diagnosed more frequently in non-Hispanic White individuals than in individuals from Black, Hispanic/Latino, or Asian populations.
"AMD is a progressive disease, although progression rates vary by age, sex, and ancestry," co-senior and co-corresponding author Panos Roussos, a researcher at the Icahn School of Medicine at Mount Sinai and the James J. Peters Veterans Affairs Medical Center, and his colleagues wrote in a paper published in Nature Genetics on Monday, adding that the condition "shows differential risk across diverse populations."
For the study, investigators with the Department of Veterans Affairs Million Veteran Program, the International AMD Genomics Consortium, and elsewhere brought together array-based genotyping and electronic health record data from six research cohorts comprising 61,248 individuals with AMD and 364,472 unaffected controls. Using those data, they conducted GWAS and GWAS meta-analyses that led to 30 new AMD risk loci and 33 loci linked to AMD in the past.
"We performed the largest GWAS meta-analysis to date and the first GWAS of AMD in [African ancestry] and [Hispanic/Latino ancestry] populations, nearly quadrupling the number of cases and doubling associated loci," the authors explained, noting that their search "uncovered marked differences in risk between major AMD loci across populations even at loci with large effect sizes … which may partially explain lower rates of AMD in these groups."
For example, the researchers found AMD-linked alleles in class II major histocompatibility complex genes in participants of African ancestry. They also saw ancestry-associated differences in effect size at new and known AMD risk loci.
In particular, they highlighted ancestry-related heterogeneity for risk variants involving the complement factor H-coding gene CFH and the ARMS2/HTRA1 locus, which appeared to have significantly smaller or insignificant effect sizes in participants of African American ancestry compared to their European American counterparts.
The team went on to explore these differences with haplotype-based local ancestry analyses aimed at untangling individual chromosome ancestry, bringing in African, European, and Native American reference sequence data from the 1000 Genomes project and other studies.
To explore pleiotropic contributors to AMD, meanwhile, the researchers performed phenome-wide and laboratory-wide association studies informed by AMD polygenic risk scores, uncovering dozens of other phenotypes with ties to AMD-associated variants.
Likewise, with transcriptome-wide association studies spanning retinal tissue and dozens of other tissue types, gene set enrichment analyses, rare variant analyses, and fine-mapping approaches, the investigators focused in on the genes and pathways with ties to AMD — from genes involved in complement cascade, humoral immunity, and other immune and inflammatory pathways to apoptosis and the regulation of high-density lipoprotein cholesterol.
More generally, the authors explained, the new work "illustrates the importance of considering diverse admixed genomes and using local ancestry-aware analyses, which can leverage differences in effect size, frequency, and [linkage disequilibrium] to tease apart mechanisms of genetic risk and improve risk prediction."