The neuroanatomical substrates of autism and ADHD and their link to putative genomic underpinnings

Abstract

Background: Autism spectrum disorders (ASD) are neurodevelopmental conditions accompanied by differences in brain development. Neuroanatomical differences in autism are variable across individuals and likely underpin distinct clinical phenotypes. To parse heterogeneity, it is essential to establish how the neurobiology of ASD is modulated by differences associated with co-occurring conditions, such as attention-deficit/hyperactivity disorder (ADHD). This study aimed to (1) investigate between-group differences in autistic individuals with and without co-occurring ADHD, and to (2) link these variances to putative genomic underpinnings.

Methods: We examined differences in cortical thickness (CT) and surface area (SA) and their genomic associations in a sample of 533 individuals from the Longitudinal European Autism Project. Using a general linear model including main effects of autism and ADHD, and an ASD-by-ADHD interaction, we examined to which degree ADHD modulates the autism-related neuroanatomy. Further, leveraging the spatial gene expression data of the Allen Human Brain Atlas, we identified genes whose spatial expression patterns resemble our neuroimaging findings.

Results: In addition to significant main effects for ASD and ADHD in fronto-temporal, limbic, and occipital regions, we observed a significant ASD-by-ADHD interaction in the left precentral gyrus and the right frontal gyrus for measures of CT and SA, respectively. Moreover, individuals with ASD + ADHD differed in CT to those without. Both main effects and the interaction were enriched for ASD-but not for ADHD-related genes.

Limitations: Although we employed a multicenter design to overcome single-site recruitment limitations, our sample size of N = 25 individuals in the ADHD only group is relatively small compared to the other subgroups, which limits the generalizability of the results. Also, we assigned subjects into ADHD positive groupings according to the DSM-5 rating scale. While this is sufficient for obtaining a research diagnosis of ADHD, our approach did not take into account for how long the symptoms have been present, which is typically considered when assessing ADHD in the clinical setting.

Conclusion: Thus, our findings suggest that the neuroanatomy of ASD is significantly modulated by ADHD, and that autistic individuals with co-occurring ADHD may have specific neuroanatomical underpinnings potentially mediated by atypical gene expression.

Keywords: ADHD; ASD; Comorbidity; Imaging-genetics; Neurodevelopmental disorders; Structural MRI.

Fig. 1

Differences of cortical thickness for main effects of ASD, ADHD, and ASD-by-ADHD interaction. A–C Random field theory (RFT)-based cluster corrected t-maps (p < 0.05, 2-tailed) for CT. A Significant decreases in CT in ASD compared to non-ASD are displayed in blue and significant increases are displayed in orange. B Significant decrease in CT in ADHD compared to non-ADHD is displayed in blue and significant increase is displayed in orange, C the ASD × ADHD interaction effect for CT, D mean CT at right cingulate cortex for the interaction effect, E mean CT at right temporal lobe for the interaction effect and F mean CT at left parietal cortex for the interaction effect. ASD autism spectrum disorder, ADHD attention-deficit/hyperactivity disorder, TD typically developing, t t-statistic, CT cortical thickness, L left, R right

Fig. 2

Differences of surface area for main effects of ASD, ADHD, and ASD-by-ADHD interaction. A–C Random field theory (RFT)-based cluster corrected t-maps (p < 0.05, 2-tailed) for SA. A Significant decrease in SA in ASD compared to non-ASD is displayed in blue and significant increase is displayed in orange. B Significant decrease in SA in ADHD compared to non-ADHD is displayed in blue and significant increase is displayed in orange, C the ASD × ADHD interaction effect for SA. D Mean SA at right frontal gyrus for the interaction effect. ASD autism spectrum disorder, ADHD attention-deficit/hyperactivity disorder, TD typically developing, t t-statistic, SA surface area, L left, R right

Fig. 3

Differences of cortical thickness for ASD + ADHD versus ASD only. Random-field-theory-based cluster-corrected t-maps (p < 0.05, 2-tailed). A Significant decrease in CT in ASD + ADHD compared to ASD only is displayed in blue and significant increase is displayed in orange, B mean CT at left parietal cortex for the ASD + ADHD subgroup (blue) and the ASD only subgroup (yellow). ASD autism spectrum disorder, ADHD attention-deficit/hyperactivity disorder, t t-statistic, CT cortical thickness, L left, R right. **p < 0.01

Fig. 4

Genomic underpinnings of neurodevelopmental deviations in cortical thickness and surface area in ASD, ADHD, and ASD + ADHD. Panel A and C show the t-maps of the main effect of ASD, the main effect of ADHD and the ASD-by-ADHD interaction term for measures of CT (A) and SA (C). Panel B (CT) and D (SA) show significant odds ratios at a false discovery rate (FDR) corrected p threshold of 0.05 resulting from the gene set enrichment analyses for genes expressed in the different output maps. Gene sets were subdivided into sets with differential gene expression in ASD, sets representing ASD risk genes, and a set representing ADHD risk genes. Gene sets are annotated and labeled based on their original publication. ASD autism spectrum disorder, ADHD attention-deficit/hyperactivity-disorder, CTX cortex, DEG differentially expressed gene, down downregulated expression in ASD, up upregulated expression in ASD, NGenes number of genes in each gene set, CT cortical thickness, SA surface area. *p < 0.05 (FDR-corrected), **p < 0.01 (FDR-corrected); red squares indicate consistently significant enrichments across both approaches (see Additional file11: Figure S6)