Genes in Diabetic Retinopathy Project (DRCRnet-Genes)

The "Genetics in Retinal Diseases Project" (Protocol-GEN) is an ongoing effort to create a repository of genetic material and clinical phenotype information as a resource for the research community, both public and private institutions. Blood samples are collected from consented participants from Diabetic Retinopathy Clinical Research (DRCR) Retina Network trials and sent to a qualified laboratory to harvest DNA and store the materials. This bioresource is available to the research community (e.g., government, academic centers, and industry) who may request samples, together with detailed clinical data, for genotypic analyses. When complete, the project will create a resource that combines high-quality genetic material and uniquely detailed phenotype information.

The Network collaborated with Regeneron's Genetics Center for the sequencing of 2,084 samples, with both whole exome and genotype array data, as well as relevant phenotype data for the 8* completed studies (n = 2,074) are available for further analysis efforts. Of these, 850 samples of DME-only patients treated with anti-VEGF agents for 6 or 12 months were included in a GWAS study by the Network, in collaboration with the Mass Eye & Ear Infirmary (MEEI), for statistical analysis of the pharmacogenetics of treatment responses. Quality controlled genotypes from Illumina's Infinium Global Screening Array were imputed using the TOPMed reference panel. Both change in central subfield thickness (CST) and visual acuity (VA) were examined as co-primary phenotype endpoints, with several additional secondary outcomes and time to response. Linear and logistic regression were applied (PLINK2.0) to quantitative and dichotomous outcomes, respectively, on all samples and subpopulations (EUR, AFR, AMR) separately, adjusting for 10 genotype PCs, age, sex, HbA1c, diabetes type and duration, hypertension prior treatment, DR severity, baseline ETDRS and CST, and type/number of anti-VEGF agents received in 6 and 12 months. A cross-ancestry meta-analysis was performed on the 6-month time point using METAL. Replication of significant associations were pursued in a second independent sample set. To prioritize causal genes and pathways, gene and gene-set level association testing was performed using MAGMA. Colocalization analysis between expression and slicing QTLs (e/sQTLs) and significant GWAS loci was performed to further prioritize causal genes for genome-wide significant loci.

Multiple loci were significant at a genome-wide significance (P<5x10-8) across the primary and secondary CST and VA outcomes at 6 and 12 months. There was no overlap between CST and ETDRS loci and the loci associated with 6 and 12 months. Gene level analyses found using MAGMA found significant associations among some of the genes that were found to be significant in the GWAS including PPIL3 and BICC1. Colocalization analyses with expression and splicing QTLs from GTEx tissues and retina further proposed putative causal genes for several loci. In replication analyses with a second, independent study, one locus on chromosome 5 passed genome-wide significance with a consistent direction of effect between the discovery cohort and replication cohort.

Data available through dbGaP includes the entire genotype/sequencing dataset (2,084 samples), supporting documentation including SNP array and CNV calls derived from the sequencing, a linking table for accessing associated phenotype from the completed clinical trials (2,074 samples; URLs are listed in table below), and the GWAS/linkage results (851 samples).

*Protocol AD (PROMINENT-Eye Ancillary Study, NCT03345901) ended prematurely due to low recruitment. Samples were collected but insufficient clinical data is available to create a phenotype dataset.

• Type: Clinical Trial
• Archive: The database of Genotypes and Phenotypes (dbGaP)