Abstract
Purpose:
Extramacular phenotype variations are not part of most AMD classification algorithms. We obtained wide angle imaging and conducted genetic studies in a Croatian population in Zagreb and in age matched controls to investigate possible genetic and phenotype relationships.
Methods:
After obtaining IRB approval and informed consent, 150 subjects with posterior AMD (macular drusen or pigment) and 150 age matched controls from the same region were studied at a single center as part of the Croatian Opera Study (Optos PEripheral RetinA). All patients were imaged with the Optos P200 MA system (Optos Plc, Waltham, MA) to obtain color images of the entire retina and AMD subjects underwent fluorescein angiography as well. Images were assessed by masked readers for the presence and extent of peripheral retina findings, including drusen, peripheral reticular pigmentation (PREP), hyper and hypopigmentation, and paving stone degeneration (cobblestone). Several single nucleotide polymorphisms (SNPs) associated with AMD were genotyped, including rs10490924 (ARMS2), 11200638 (HTRA1), and Y402H (rs1061170) and 1410996 (CFH). We compared the anatomic features of the AMD and control cohorts and assessed their genetic differences. We developed odds ratios for each anatomic feature where appropriate.
Results:
We found that subjects with AMD were significantly more likely to have the above SNPs compared to controls by Chi squared statistics. Patients homozygous for Y402H had a 4.5X greater likelihood of having drusen in the periphery (OR 1.55-13.02, P=0.006). Those homozygous for Y402H had a 4.49X higher likelihood of having peripheral reticular pigmentation (OR 1.66-12.13, P=0.003) while heterozygotes had a 3.51X chance of having PREP (OR 1.35-9.14 P=0.01). We had only a few patients with pavingstone (PS) degeneration in each group, so we pooled the two groups (n=48) and found that homozygotes for rs1040924 had a 2.5X higher chance of having PS (OR 1.04-5.76, P=0.041).
Conclusions:
Peripheral retina phenotypes reflect the genetic risk factors for AMD, and greater attention to the periphery appears warranted when studying AMD. Paving stone degeneration, peripheral drusen, and PREP appear to be associated with CFH, ARMS2, and HTRA1.