NEURAL CORRELATES OF COMORBID AUTISM SPECTRUM DISORDER AND ATTENTION DEFICIT HYPERACTIVITY DISORDER

Autism Spectrum Disorder (ASD) and Attention-Deficit/Hyperactivity Disorder (ADHD) are two of the most commonly diagnosed developmental disorders and co-occur in greater than 37% of children with either disorder. However, until the DSM-5 allowed for a comorbid diagnosis, study of the etiology of comorbid ASD and ADHD has been a challenge. Neuroimaging research on both disorders has consistently found the cerebellum to be a site of structural and functional abnormality. In the present study, we conducted three experiments to investigate the structural and functional abnormalities of comorbid ASD and ADHD with a particular focus on the cerebellum and cerebro-cerebellar circuits. In experiment 1 we used whole brain voxel-based morphometry (VBM) and cerebellar lobular analysis of grey matter (GM), comparing 20 children with ADHD-only (12M; Age: 10.7±1.3), ASD-only (17M; Age: 10.6±1.8) and comorbid ASD+ADHD (13M; Age: 10.8±1.3) with 20 typically developing children (10M; Age: 10.6±1.1) aged 7-13 years old. Individual t-tests comparing clinical groups to typically developing children were performed, and post-hoc correlations with ASD and ADHD symptoms were performed in statistically significant GM clusters and lobules. VBM results indicate increased GM in the precuneus in both ASD-only and ADHD-only children compared to TD children. Increased precuneus GM correlated with a greater number of ADHD symptoms, but no relationship was found with ASD symptoms. The precuneus was also found to be enlarged in ASD and ADHD when compared to the comorbid group. The comorbid group showed increased GM bilaterally in the caudate, in a region that that overlapped with increased GM in the ASD group, and correlated with all three symptom domains of ASD. The ASD+ADHD group also had increased GM in left cerebellar lobules IV, V and VI, a finding which was unique to this group. Cerebellar lobular analysis revealed increases in GM volumes in right VIIb, vermis VIIIa, left/right/vermis VIIIb, and left/right/vermis IX in the ADHD group. These results suggest that increased precuneus GM present in both ASD and ADHD groups may be related to ADHD-related symptoms, whereas increased caudate GM present in ASD and ASD+ADHD groups was related to worse ASD symptoms. Further, all three clinical groups showed diverging abnormalities in the cerebellum. These findings suggest that there are both shared and unique underlying neural correlates in the comorbid group. In experiment 2 we performed whole-brain resting state functional connectivity (rsFC) analyses examining the Default Mode, Cognitive Control and Ventral and Dorsal Attention networks based on seeds located in the cerebellum. Graph theory analysis was also performed on the resting state data. Participants included 16 boys with ADHD (Age: 9.7±1.3), 15 boys with ASD (Age: 9.9±1.3), 14 boys with comorbid ASD+ADHD (Age: 10.1±1.1), and 14 typically developing boys (Age: 9.6±1.0). Primary findings include significant decreases in cerebro-cerebellar Default Mode Network connectivity in medial superior frontal regions in all three clinical groups, as well as increased out of network connectivity in the ADHD and comorbid groups. Additionally, all three clinical groups showed increased connectivity in the occipital lobe with the Cognitive Control Network seed. For the Ventral Attention Network seed, the ASD group showed a decrease in connectivity with the right lingual gyrus and the ADHD group showed decreased connectivity with the medial supplementary motor cortex and cingulate. The ADHD group also showed increased connectivity between the Ventral Attention Network seed and cerebellar right Crus I and vermis VI/VII. Correlations revealed a significant negative correlation between rsFC increases in the postcentral gyrus of the comorbid group for the DMN and ADI – Social scores. No significant results emerged from the Dorsal Attention Network seed, direct comparisons between the comorbid and ASD and ADHD groups, or the graph theory analysis. These findings suggest all three clinical groups showed patterns of long-range intranetwork under-connectivity with unique patterns of abnormal out-of-network connectivity. A common factor in all clinical groups was decreased functional connectivity in the Default Mode Network, which is consistent with deficits of inhibition and attention common to both disorders. Lastly, in experiment 3 we investigated structural connectivity using region of interest (ROI) and Tract-Based Spatial Statistics (TBSS) analyses of Diffusion Tensor Imaging (DTI) data in 22 children with ADHD (16M; Age: 10.1±1.1), 14 children with ASD (11M; Age: 10.2±1.4), 22 children with comorbid ASD+ADHD (18M; Age: 10.1±1.1), and 22 typically developing peers (17M; Age: 10.3±1.4). TBSS analyses yielded no significant whole-brain differences at corrected thresholds. Using the Johns Hopkins University white matter atlas for ROI analyses, differences in Fractional Anisotropy (FA) were found between the comorbid group and ADHD groups in the bilateral anterior thalamic radiation (ATR), forceps minor, corona radiata, and internal capsules. In addition, increased MD was found in the internal and external capsules in the comorbid group compared to ADHD. The ASD group showed increased MD in the corpus callosum and decreased MD in the uncinate fasiculus. These results suggest that increased WM structural connectivity abnormalities in the comorbid group may be better accounted for by ASD and not ADHD. Taken together, these results suggest unique structural brain abnormalities may arise in comorbid ASD+ADHD, while shared abnormalities in GM structure between the individual groups may be due to symptom overlap. In addition, functional network analyses revealed weakened long-range intranetwork and heightened out-of-network connectivity in all three clinical groups across a number of well-established resting state networks. Our data indicate that comorbidity may exhibit characteristics of shared liability and suggest careful assessment of both ASD and ADHD symptoms in future ASD and ADHD research. These results also add to the increasing literature that the cerebellum plays an important role in these developmental disorders. Future research must be careful to include the cerebellum when performing “whole-brain” analyses, especially when making claims of global functional and structural connectivity.