66 - GENOME-WIDE ASSOCIATION STUDY OF ANXIETY AND STRESS-RELATED DISORDERS

Abstract

Background In Europe, over 70 million people are annually suffering from anxiety and stress-related disorders and over 90% of patients with these disorders have another comorbid mental or somatic disorder. Therefore, it is of strong interest to identify the risk factors of these disorders. Heritability estimates vary between 30% and 50%, however first Genome-Wide Association Studies (GWAS) have been limited by their small sample sizes. As anxiety and stress-related disorders are likely to configure various expressions of abnormalities in the basic stress-response system, we aimed to conduct a GWAS aggregating a large number of cases with varying diagnoses of anxiety and stress-related disorders. Methods The study is making use of a Danish nation-wide population-based sample collected by the Lundbeck Foundation Initiative for Integrative Psychiatric Research including 14,711 cases with anxiety and stress-related diagnoses and 17,857 controls. Data were processed using the Ricopili pipeline; imputation is based on the 1000 genomes phase 3 as reference panel. GWAS was conducted using the imputed marker dosages and an additive logistic regression model. Analyses were supplemented by comparing severe mental disorder with comorbid anxiety and stress-related disorder to those without, using a propensity-score-matched design. SNP heritability and genetic correlation with other traits were computed using LD-score regression. Results In case-control design several SNPs located in the PDE4B gene achieved genome-wide significance. The analyses comparing mental disorders with and without comorbid anxiety and stress-related disorders reveled no significant signal. The SNP heritability estimate for anxiety was 14%, 33% for stress-related disorders and 26% for the combined phenotype. We found a positive significant genetic correlation of anxiety and stress-related disorders with neuroticism, depression, bipolar disorder, schizophrenia and smoking; and a negative genetic correlation with college completion, years of schooling, age at menarche and age of first birth. Discussion This is the first GWAS meta-analysis on anxiety and stress-related disorders demonstrating influences of common genetic variation in the etiology of these disorders and indicating PDE4B as a susceptibility locus. Consistent with phenotypic observations, anxiety and stress-related disorders share genetic variants with psychiatric traits in the expected direction.