Abstract
Purpose :
To identify genetic variants associated with anti-VEGF drug response as measured by visual acuity, anatomic outcomes and treatment frequency in the VIEW 1 study.
Methods :
A genome wide association study (GWAS) was conducted on 362 VIEW 1 patients. DNA samples were genotyped using the Illumina Omni Express Exome Chip. Logistic regression with baseline values was performed to establish the association between genetic variants and efficacy variables. For each SNP, genotypes were coded according to an additive mode of inheritance. Variants associated with gaining ≥15 ETDRS letters at week 52, presence of intraretinal cystoid edema (fluid as measured by time domain optical coherence tomography (TD-OCT)) at week 52 and frequency of treatment at week 96 were evaluated.
Results :
An X-chromosome SNP (rs2056688) revealed the highest association with anatomical outcome, demonstrating an odds ratio (OR) of 0.2578 and a point-wise association (p-value 7.27 x 10-7) with presence of intraretinal fluid at week 52. Four neighboring SNPs (rs5962084, rs5962087, rs5915722, rs5962095) revealed similar ORs (0.3151-0.3461) and point-wise associations (5.48 x 10 -6 – 8.59 x 10 -6). The rs2056688 SNP was located in a non-coding region, with the closest relevant functional gene (Protein Kinase X-Linked (PRK-X)) mapping ~400kb upstream of the putative variant. Additional SNPs with lower significance were found in association with proportion of patients with ≥15 ETDRS letters gains in vision at week 52 and frequency of treatment at week 96.
Conclusions :
A GWAS in neovascular AMD patients undergoing anti-VEGF treatment in the VIEW 1 trial identified an association between a genetic variant and the presence of intraretinal fluid at week 52 as measured by TD-OCT. The variant was located at a position on the X chromosome near the gene for PRK-X, a serine/threonine protein kinase involved in angiogenesis. Given the relatively small number of samples evaluated in this study, the findings require validation in an independent dataset.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.