Abstract
Purpose: :
Risks of Age-related Macular Degeneration (AMD) for genetic variants vary widely among studies. This study aims to improve the risk estimates for CFH and ARMS2 by calculation of incidence risk ratio’s for the various genotypes in the prospective population-based Rotterdam Study.
Methods: :
Participants aged 55+ years underwent 5 identical examinations during 1990-2010 with response rate 70-73%. Fundus photographs were graded for features of AMD using the International Classification and Grading System. CFH Y402H (rs1061170) and ARMS2 A69S (rs10490924) SNPs were genotyped using Taqman assays. Genotypes were stratified into groups of 0-4 risk alleles. Risk ratio’s of incident late AMD were calculated using Cox proportional hazard ratio (HR) with adjustment for age and sex.
Results: :
5545 participants were successfully genotyped. The frequency of CFH Y402H and ARMS2 A69S alleles was 36.2% and 20.2%; the frequencies of the 0-4 genotypes were 26.4, 42.3, 24.3, 6.5 and 0.6%(s), respectively. The total number of incident late AMD was 123. Hazard ratio of incident late AMD during mean follow-up 8.9 yrs (SD 5.07 years) was highest for participants with 4 risk alleles: HR 38.90 (95% CI 10.90-138.83). Participants with 3, 2, 1, or 0 risk allele(s) had HRs of 6.73, 3.89 and 2.61, respectively (P trend < 0.0001). Risk ratio’s of late AMD for homozygous and heterozygous genotypes of CFH Y402H polymorphism were HR 6.36 (95% CI 5.59-13.15) and 2.27 (95% CI. 1.54-3.39). For any carriership of ARMS2, HR was 1.24 (95%CI 0.87-1.76).
Conclusions: :
Use of population-based incidence data and disentanglement of the risk genotypes give better insight in the risk of AMD for CFH Y402H and ARMS2 A69S.
Keywords: age-related macular degeneration • genetics • clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology