Adhd beyond diagnosis: insights from genetics of adhd domains in a well-characterized youth clinical sample

There is a growing recognition that dichotomous psychiatric diagnoses do not accurately reflect the underlying biology of psychopathology and may even hinder our understanding of pathophysiology by collapsing phenotypes with potentially different etiologic underpinnings. This is particularly relevant in ADHD, one of the most prevalent neurodevelopmental disorders. The diagnosis of ADHD is based on two separate symptom domains: hyperactivity/impulsivity (HI) and inattention (IN) that are continuously distributed across individuals in the population. In clinics, patients often present with symptoms that vary on these distinct domains, including some who do not meet the full diagnostic criteria, despite experiencing functional impairment. Thus, polygenic scores (PGSes) derived from the specific ADHD domains may be informative in referred youth if they can serve as objective indicators linking symptoms to biological mechanisms. If so, PGSes of ADHD domains may extend the identification of children with emerging psychopathology and/or at high risk for developing adverse outcomes. In this study, we aimed to gain insight into the relevance of ADHD domain-specific PGSes to ADHD-related phenotypes in an outpatient child psychiatry sample. The Longitudinal Study of Genetic Influences on Cognition (LOGIC), a cohort of youth consecutively referred for neuropsychiatric evaluation, was used. The sample of 1,137 youth (NPsychchip=505; NGSA=632) was genotyped and imputed in two batches following standard procedures. PGSes of ADHD domains (PGS-IN and PGS-HI) were calculated using PRScs software based on summary statistics from a genome-wide association study on the ADHD domains in children and adolescents in the general population (N up to 53,258). Only unrelated probands aged 6 – 17 without intellectual disability were included in the analyses (N=976; meanage = 11.0 ± 3.10; 35.5% girls). ADHD diagnoses (N = 596 (61.1%)) were established based on DSM criteria and parent rated symptoms of IN and HI were measured via the Child Symptom Inventory or Child and Adolescent Symptom Inventory. The associations between these PGSes and ADHD domains was examined using linear regression, adjusting for sex, age, first five principal components and medication. The results from the two batches were meta-analyzed using a random effects inverse variance method. Preliminary results show that PGS-IN was significantly associated with IN and nominally associated with HI (IN: beta=0.33,95%CI=0.15-0.51, p=0.0003; HI: beta=0.22,95%CI=0.03-0.40,p=0.0232), while PGS-HI did not show any significant associations (IN: beta=0.16, 95%CI=-0.25-0.57, p=0.4383; HI: beta=0.18, 95%CI=-0.15-0.52,p=0.2806). This pattern of associations was also observed after controlling for ADHD diagnosis. Further, PGS-IN also revealed the same pattern of associations when the analyses were restricted to individuals with ADHD (IN: beta=0.38, 95%CI=0.18-0.59, p=0.0003; HI: beta=0.25,95%CI=0.05-0.44, p=0.0147). The PGS-HI did not show any associations (all p > 0.40). These results add to the notion that the genetic underpinnings of specific ADHD symptoms can be dissected and related to phenotypes in the clinical setting, even after accounting for diagnosed disorder. Studies are needed to further clarify how the genetics of ADHD domains may have utility as objective indices of risk for manifesting or emerging ADHD psychopathology in help seeking youth.