Association of Single-Nucleotide Polymorphisms in Age-Related Macular Degeneration With Pseudodrusen: Secondary Analysis of Data From the Comparison of AMD Treatments Trials.

Importance: Previous studies investigating the association of single-nucleotide polymorphisms (SNPs) that confer increased risk of age-related macular degeneration (AMD) with pseudodrusen have yielded conflicting results and have not evaluated other AMD SNPs or pseudodrusen subtypes. Objective: To determine the association of SNPs in the complement factor H (CFH), age-related maculopathy susceptibility 2 (ARMS2), HtrA serine peptidase 1 (HTRA1), complement C2 (C2), complement C3 (C3), lipase C (LIPC), and complement factor B (CFB) genes with the presence of pseudodrusen and pseudodrusen subtypes (ie, dot, reticular, and confluent). Design, Setting, and Participants: In this post hoc analysis of cross-sectional data from US participants in the Comparison of AMD Treatments Trials, genotyping was performed in 835 participants with TaqMan assays for the SNPs rs1061170 (Y402H variant in CFH), rs800292 (I62V variant in CFH), rs10490924 (A69S variant in ARMS2), rs11200638 (HTRA1), rs547154 (C2), rs2230199 (R102G variant in C3), rs10468017 (LIPC), and rs4151667 (L9H variant in CFB). Main Outcomes and Measures: Presence and subtype of baseline pseudodrusen in either eye determined using color fundus photography, red-free images, and fluorescein angiograms. Results: Among 835 participants enrolled for genotyping, 755 (90.4%) were evaluated for pseudodrusen. Of these, 471 (62.4%) were female and 750 (99.3%) were white, and the mean (SD) age was 78.3 (7.5) years. A total of 213 of 755 participants (28.2%) had pseudodrusen (107 [14.2%] had dot pseudodrusen, 180 [23.8%] had reticular pseudodrusen, and 102 [13.5%] had confluent pseudodrusen). After adjusting for age, sex, and smoking status, the ARMS2 risk allele T was associated with higher risk of pseudodrusen (odds ratio [OR], 1.93; 95% CI, 1.19-3.12) for TT vs GG (P = .04). A similar association was found for HTRA1 (OR, 2.04; 95% CI, 1.26-3.31) for AA vs GG (P = .03). The CFH Y402H risk allele C was associated with lower risk of pseudodrusen (OR, 0.61; 95% CI, 0.38-0.97) for CC vs TT but was not statistically significant after correcting for multiple comparison (P = .20). CFH Y402H, ARMS2, HTRA1, and C3 were significantly associated with reticular pseudodrusen. Conclusions and Relevance: Among patients with neovascular AMD, the AMD risk alleles ARMS2 and HTRA1 were associated with an increased risk of pseudodrusen and the risk allele CFH Y402H was associated with lower risk of pseudodrusen, supporting findings from previous studies. Understanding the role of these SNPs in the development of pseudodrusen might improve our understanding of the pathogenesis of AMD and help develop future therapies.