Abstract
Anxiety disorders are one of the most common, debilitating and costly classes of psychiatric disorders worldwide. Twin studies estimate heritability of anxiety disorders to be between 30%-60%, depending on specific disorder, age, and level of impairment. Although individual anxiety disorders are considered clinically distinct, they share much of their phenotypic and genetic variance, potentially reflecting an underlying liability distribution. The UK Biobank has collected symptom and disorder level anxiety data on 157,366 individuals across the UK who have contributed their genetic data. We used this dataset to investigate genome-wide associations, SNP based heritability, and genetic correlations in four anxiety phenotypes. These reflect population level current anxiety symptoms as a quantitative phenotype, and three case control phenotypes; severe current anxiety symptoms, probable lifetime generalised anxiety disorder and self-reported lifetime diagnosis of any anxiety disorder. Probable lifetime generalised anxiety disorder and selfreported lifetime diagnosis of any anxiety disorder were meta-analysed with a comparable genome-wide association study of anxiety. Genetic analyses included unrelated Caucasian individuals of Western European ancestry.
Estimates of SNP heritability from common variants ranged between 4% (for population level anxiety symptoms) and 32% (for probable generalised anxiety disorder), and all four UK Biobank anxiety phenotypes are highly genetically correlated. Three genome-wide significant loci were found to be associated with anxiety. Both rs3807866 located in the TMEM106B protein coding region on chromosome 7, and rs2996471 located in the NTRK2 protein coding region on chromosome 9, were associated with self-report of any lifetime anxiety diagnosis. An additional non characterised region on chromosome 9 was associated with both self report of any lifetime anxiety diagnosis (rs10809485), and severe anxiety symptoms (rs17189482). Meta-analysis with a comparable genome-wide association study of anxiety did not result in additional findings. This represents the largest genetic study of anxiety to date-however larger sample sizes will be required to further examine the common genetic architecture underlying anxiety.