Abstract
Purpose::
A recent study by Maller et al. estimated the risk of advanced Age-related Macular Degeneration (AMD) based on five variants in the Complement Factor H (CFH), LOC387751 and C2/FB genes in a setting comparing advanced AMD cases to non-AMD controls. In this model, subjects who were homozygous for all risk alleles had a 285-fold greater risk of advanced AMD than subjects who carried only low-risk genotypes (i.e., 3% of the general population). Our aim was to determine which genotypic profiles have an increased or decreased risk of advanced AMD compared to the average risk in the population, and to evaluate the value of genetic screening of these variants for the prediction of advanced AMD.
Methods::
Recalculation of risks was performed with logistic regression analysis using the observed genotype frequencies provided by Maller et al. The predictive value of the risk alleles, or discriminative accuracy, was evaluated by the area under the receiver-operating characteristic curve.
Results::
Compared to the population-risk, subjects homozygous for all risk alleles had a 14-fold increased risk of advanced AMD, and subjects carrying only non-risk alleles had a 20-fold decreased risk. Absolute AMD risks were 35% and 0.17%, respectively. The discriminative accuracy of testing all five alleles was 79.6%. When considering only one SNP, the discriminative accuracy of Y402H of CFH was highest (68.9%). When considering two SNPs, the discriminative accuracy was highest for Y402H and A69S (78.8%).
Conclusions::
In reference to the population-risk, subjects carrying these five high-risk alleles have a much less augmented risk of advanced AMD than as presented by Maller et al. Nevertheless, testing these variants will help discriminate those who will develop advanced AMD from those who will stay disease-free. To assess further usefulness for clinical practice, these calculations should be repeated in patients with early features of AMD.
Keywords: age-related macular degeneration • genetics • clinical (human) or epidemiologic studies: risk factor assessment