SA5 - GAENETICS OF WORKING MEMORY AND VIGILANCE EDOPHENOTYPES IN ADHD

Abstract

Background Joint genetic analysis of correlated endophenotypes markedly enhances insight into the relationships among them and the shared and distinct genetic influences on phenotypically related complex traits. In addition, identifying pleiotropic effects among endophenotypes can help improve our biological understanding of the genetic architecture of a complex disease, namely childhood Attention Deficit Hyperactivity Disorder (ADHD). Cognitive functions, in particular aspects of executive functioning, have long been proposed as endophenotypes for ADHD, in some cases with distinct components of ADHD emphasized (hyperactivity and inattention-disorganization). This study identifies and characterizes the individual and shared genetic variants associated with two major cognitive endophenotypes—Working Memory (WM) and arousal — in relation to both ADHD sub-dimensions. Methods Composite scores for working memory and arousal were created on the basis of a latent variable model using multiple laboratory-obtained measures of each construct. Inattention and hyperactivity were defined on the basis of a composite of parental and teacher ratings. A confirmatory factor model fit well, supporting the internal validity of the composite scores. Subjects included 656 unrelated community recruited volunteer children ranging 7 to 13 years old (n=435 ADHD, 221 non-ADHD). SNPs included in our analysis (n=193,657) were from a polygenic risk score (PGS) computed using the reported discovery sample of 20,183 ADHD cases and 35,191 controls from the Psychiatric Genetics Consortium. A linear regression analysis accounting for age, sex and population stratification was used to test for association to each endophenotype. Pleiotropic relationships were identified using linear regression analysis accounting for the correlation among all four phenotype measures. A likelihood ratio test framework was used for inferring the number of traits associated with a genetic variant and for partitioning traits associated with the different SNPs. Results The PGS was associated with ADHD diagnosis (Nagelkerke R2=.04, p=0.0000006) and multi-indicator dimensional ADHD latent variables by parent report (beta=.196, SE=.041) and teacher report (beta=.176, SE=.040). Univariate SNP associations (p Discussion Our study provides motivation for pleiotropic effects in ADHD endophenotypes and helps begin to account for the mechanism by which cognitive endophenotypes participate in ADHD development.