Single nucleotide polymorphism profiles associated with three common phenotypes of advanced age-related macular degeneration.

Abstract

Age-related macular degeneration (AMD) has been discovered to have strong genetic associations. Complement factor H and LOC387715/ARMS2 variants are shown to have strong association with development and progression of AMD (Seddon et al. 2007). Other loci in multiple chromosomes have also been associated with genetic predisposition to AMD, including five key gene variants to advanced AMD (Chen et al. 2010). Correlation between individual genetic profiles or cumulative measures of genetic load with specific AMD subtypes is still being elucidated. We performed a case–control study using RetnaGene AMD Test (Perlee et al. 2013) to elucidate single nucleotide polymorphism (SNP) profiles of three common phenotypes of advanced AMD [bilateral geographic atrophy (GA), Type-1 exudative AMD (wAMD) and Type-2 wAMD] to determine whether certain genetic variants are more frequently linked with specific phenotypes. Our secondary aim was to compare SNP profiles of good with poor responders to anti-vascular endothelial growth factor (VEGF) treatment. A total of 157 eyes met our inclusion criteria (37 eyes with GA, 63 eyes with Type-1 wAMD and 57 eyes with Type-2 wAMD). The demographic profiles among tested subjects associated with the 3 phenotypes of advanced AMD were comparable in terms of age, gender, race (all Caucasian) and vision. Complete eye examination, fundus photography, fluorescein angiography, indocyanine-green angiography and spectral domain ocular coherence tomography of the macula were performed at baseline. Buccal mucosal swabs were genotyped with a panel of 12 AMD-associated SNPs using matrix-assisted laser desorption ionization-time of flight mass spectrometry system, at Sequenom Center for Molecular Medicine, San Diego, CA. SNP frequencies and allelic odds ratios were analysed for significance using Fisher’s exact test. Good responders were defined as ≥10 letters improvement or ≥50% reduction of central subfield thickness in response to anti-vascular endothelial growth factor therapy. All study phenotypes of advanced AMD showed similar risk of neovascular development (p > 0.05, Pearson chi-square). Of the tested SNP’s, only C3 risk variant rs2230199 (Arg102Gly) was observed to be significantly more common in eyes with bilateral GA than Type-1 nAMD (p = 0.03, Fisher’s exact test). Furthermore, we found no differences in SNP profiles between good and poor responders when assessing entire cohort of Type-1 and 2 eyes, or stratifying according to Type-1 or Type-2 wAMD, or to type of anti-VEGF drug used (i.e. bevacizumab, ranibizumab, aflibercept). Despite distinct phenotypes, there appears to be a common genetic load shared among GA and Type-1 and Type-2 wAMD eyes. This may explain why our cohort showed similar or greater risk (categorical [chi-square] and Cox hazard ratio) for development of new choroidal neovascularization in GA eyes compared to Type-1 and Type-2 wAMD eyes. The inherent predisposition of GA eyes for exudative changes is consistent with frequent clinical observation of exudative development in certain GA eyes. The higher frequency of C3 risk variant rs2230199 observed in GA compared to eyes Type-1 wAMD suggests that inflammatory and immune mechanisms, which are frequently linked to the alternative complement pathway, may possibly contribute to GA formation in nonexudative AMD (Tan et al. 2016). The results of our study are intriguing, but limited by small sample size and lack of multiplicity testing. Additionally, comparison of results from readily available AMD genetic tests for the same subject appear to have considerable variation of estimated risks, raising questions regarding reliability of the data (Buitendijk et al. 2014). Therefore, our study conclusions are suggestive and appropriate for generating hypothesis. Further studies including much larger sample size and multiplicity testing are necessary to validate our findings.