Intra-Individual Response Variability Assessed by Ex-Gaussian Analysis may be a New Endophenotype for Attention-Deficit/Hyperactivity Disorder.

Marcela Patricia Henrã­Quez-Henrã­Quez,Hugo Henrã­Quez,Francisco Aboitiz,Francisco Javier Zamorano,Francisco Rothhammer,Pablo Billeke

doi:10.3389/fpsyt.2014.00197

Abstract

Intra-individual variability of response times (RTisv) is considered as potential endophenotype for attentional deficit/hyperactivity disorder (ADHD). Traditional methods for estimating RTisv lose information regarding response times (RTs) distribution along the task, with eventual effects on statistical power. Ex-Gaussian analysis captures the dynamic nature of RTisv, estimating normal and exponential components for RT distribution, with specific phenomenological correlates. Here, we applied ex-Gaussian analysis to explore whether intra-individual variability of RTs agrees with criteria proposed by Gottesman and Gould for endophenotypes. Specifically, we evaluated if normal and/or exponential components of RTs may (a) present the stair-like distribution expected for endophenotypes (ADHD > siblings > typically developing children (TD) without familiar history of ADHD) and (b) represent a phenotypic correlate for previously described genetic risk variants. This is a pilot study including 55 subjects (20 ADHD-discordant sibling-pairs and 15 TD children), all aged between 8 and 13 years. Participants resolved a visual Go/Nogo with 10% Nogo probability. Ex-Gaussian distributions were fitted to individual RT data and compared among the three samples. In order to test whether intra-individual variability may represent a correlate for previously described genetic risk variants, VNTRs at DRD4 and SLC6A3 were identified in all sibling-pairs following standard protocols. Groups were compared adjusting independent general linear models for the exponential and normal components from the ex-Gaussian analysis. Identified trends were confirmed by the non-parametric Jonckheere–Terpstra test. Stair-like distributions were observed for μ (p = 0.036) and σ (p = 0.009). An additional “DRD4-genotype” × “clinical status” interaction was present for τ (p = 0.014) reflecting a possible severity factor. Thus, normal and exponential RTisv components are suitable as ADHD endophenotypes.

Highlights

Symptom-based diagnostic systems such as the current versions of the Diagnostic and Statistical Manual of Mental Disorders (DSM) [1] and the International Classification of Diseases (ICD) [2] have proved to be useful in the clinical approach to the neuropsychiatric patient and in epidemiological settings
We explored whether RTisv may represent a phenotypic correlate for previously described genetic risk variants for attentional deficit/hyperactivity disorder (ADHD) in the genes encoding for the dopaminergic receptor D4 (DRD4) and for the dopaminergic transporter 1 (SLC6A3/DAT1)
PARTICIPANTS In order to evaluate eventual differences in regards to the response times (RTs) distribution observed among ADHD patients, their non-affected siblings, and typically developing children (TD) without family history of ADHD, we evaluated a total of 55 subjects, corresponding to 20 discordant sibling-pairs and 15 unaffected children from the general population

Summary

Introduction

Symptom-based diagnostic systems such as the current versions of the Diagnostic and Statistical Manual of Mental Disorders (DSM) [1] and the International Classification of Diseases (ICD) [2] have proved to be useful in the clinical approach to the neuropsychiatric patient and in epidemiological settings. These markers, generically called “endophenotypes” constitute a more “direct” expression of the gene effect, since they conceptually lie between the gene and the disorder and – in consequence – are influenced by fewer genetic and environmental variables than the disorder itself In addition to their potential role improving the statistical power of molecular-genetic studies, endophenotypes should be a valuable tool when studying how the already-known genetic risk variants are related to the neurobiological and neuro-physiological phenotypes that underlie psychiatric disorders, which could be the first step to elucidate the specific domains of brain function influenced by these variants. This kind of approach has been successfully applied, for example, in the functional charwww.frontiersin.org

Methods

Results

Discussion

Conclusion