Effect of Autism Risk CNVs on Autistic and Neuropsychiatric Traits in a Population Based Adolescent Sample

Background:   Like many neurodevelopmental and neuropsychiatric disorders, Autism Spectrum Disorder (ASD) is viewed as lying on a continuum that shades into sub-clinical traits in the ‘normal’ population. Current estimates suggest that 10-20% of cases are caused by one or more copy number variants (CNVs) in the genome (see Simons Foundation Autism Research Initiative, SFARI). However, it is not known whether these same CNVs also affect autistic and related social-affective traits in the ‘general population’.  Furthermore, possible sex differences are indicated by recent findings suggesting that the increased male:female ratio in ASD may be due to a ‘protective’ effect in females, who may require a greater burden of CNVs.  Finally, our understanding of potentially mediating neural and neurocognitive mechanisms is still incomplete.

Objectives: To investigate in an adolescent sample from the general population (1) the effect of ASD-risk CNVs (as compiled by SFARI) on ASD symptoms and other neuropsychiatric traits (overall, and stratified by sex); and (2) to examine whether differences in cognitive profile, brain structure and/ or function in CNV-carriers mediate the link between genetic risk factors and clinical symptoms. 

Methods: Participants are 1,713 adolescents (841 males, 872 females) from the IMAGEN cohort. CNV detection was performed based on the Illumnia 610/660 arrays, using two different algorithms (QuantiSNP and PennCNV) on samples that passed Quality Control to minimize the number of potential false discoveries. For each participant we computed the number of intronic and exonic deletions/ duplications in SFARI risk-genes. In a first step, SFARI risk-gene carriers were compared to no-risk gene carriers on ASD symptoms (Social Responsiveness Scale); emotional, conduct, peer, hyperkinetic and overall problems impacting daily adaptive functioning (Strength and Difficulties Questionnaire), and IQ (WASI). In a second step, we plan to compare the groups on a number of a) neurocognitive domains: empathy/ perspective-taking, impulsivity, reward processing,  attention, and working memory; (b) cortical thickness/ surface area, and (c) and strength of functional connectivity. 

Results: 1,369 adolescents carried no deletion/duplication in SFARI risk genes, 297 one, 32 two, 6 three, and 9 four deletions/ duplications. We found that boys carrying one or more ASD-linked CNVs had nominally significantly higher ASD symptoms than no-risk carriers (p=.036). This effect was not observed in girls.  At age 13-14 years, both male and female SFARI risk gene-carriers also had significantly more peer problems (p=.001), emotional problems (p=.028), and overall problems that impacted daily functioning (p=.00001). However, SFARI-risk gene carriers did not differ from non-carriers in terms of IQ. 

Conclusions:

In a population-based sample, boys carrying one or more ASD-linked risk genes exhibited higher ASD traits than non-carriers. However, these CNVs also increased broader emotional and social (peer) problems, and overall problem impact in boys and girls at a similar rate.  Consistent with findings from clinical groups, this indicates that these CNVs may not map onto a  ‘specific’ vulnerability for ASD but a broader range of (sub-clinical) social-emotional dysfunctions. We will also report findings from ongoing analyses on the potentially mediating role of brain structure and function, and neurocognitive profile in these associations.