Purchase this article with an account.
G. Jun, S. M. Adams, I. K. Kim, J. W. Miller, G. S. Hageman, M. G. Kotoula, E. E. Tsironi, L. A. Farrer, M. M. DeAngelis; Extended Analysis in the 10q26 Region for Neovascular Age-Related Macular Degeneration (AMD). Invest. Ophthalmol. Vis. Sci. 2010;51(13):1261.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To examine the PLEKHA1/ARMS2/HTRA1 10q26 region in geographically and ethnically diverse populations in an effort to pinpoint variation contributing to AMD risk.
We ascertained 1,011 individuals including 500 sibpairs from New England and 344 unrelated subjects from central Greece that included those with normal maculas, the diagnosis of AREDS category 2, 3 and neovascular AMD. Eighteen SNPs, spanning 6.8 Mb including previously reported AMD risk variants, were genotyped by direct sequencing, Sequenom iPLEX and/or TaqMan assays. Association of these SNPs with neovascular AMD after adjusting for age was evaluated using the FBAT in the family sample (226 discordant, 87 concordantly affected, and 48 concordantly unaffected sibpairs) and PLINK for the case-control cohort (139 affected and 121 unaffected). Haplotype analysis was performed in two datasets using sliding windows with 2 to 5 SNPs to identify critical regions that might harbor significant signals.
Association signals in the family sample encompassed the entire 10q26 region including 7 SNPs with a nominal P < 10-7 under the additive genetic model for neovascular AMD. The most significant findings in the families were with SNPs rs10490924, rs11200638, and rs1019331 (P < 10-13 for each). In the Greek case-control cohort, rs10490924, rs3750848, and rs2672587 were most significantly associated with neovascular AMD (P < 10-4 for each). Meta-analysis revealed that seven SNPs were significant at P < 10-7, and the most significant result was with rs10490924 (P = 2 x 10-14) showing a consistent pattern of association in the two datasets. The most significant haplotype in the unrelated subjects was found with two SNPs in HTRA1, rs1019331 and rs2672587 (P = 8 x 10-6). In the families, however, haplotype analysis did not strengthen the results from the SNPs analyzed individually.
These results confirm that SNPs in the 10q26 region are important risk factors for neovascular AMD in US Caucasians and Greeks, although the rank order of significant SNPs varied between the cohorts. Therefore, functional experiments along with comprehensive analysis in the region are necessary to identify true causal variants.
This PDF is available to Subscribers Only