Cross-Disorder Comparison of Idiopathic Autism, Fragile X, Angelman Syndrome, and Typical Development: An MRI and DTI Study of Brain Structure

Background: Brain abnormalities are common in autism and related neurodevelopmental disorders, but no study has directly compared the brain structure of children with idiopathic autism, Angelman syndrome, Fragile X, and typical development.

Objectives: We conducted an MRI and diffusion tensor imaging (DTI) study of school-age children with related neurodevelopmental disorders to determine the extent and specificity of neural alterations and their associations with clinical symptoms.

Methods: Participants were n=11 children with Angelman syndrome (AS; 5 female), n=19 with Fragile X syndrome (FXS; all male), n=31 with autism spectrum disorder (ASD; 1 female), and n=31 age-matched typically developing controls (TD; 3 female). MRIs were conducted at 4-12 years of age, including T1- and T2-weighted (1mm³ voxels) and diffusion-weighted scans (2mm³ voxels; 45 directions). Tissue segmentation yielded cerebral white matter (WM), grey matter (GM), and total cerebral volume (TCV=WM+GM). Diffusion tensor tractography of specific fiber tracts yielded fractional anisotropy (FA), axial diffusivity (AD), and radial diffusivity (RD). Analysis of covariance tested for group differences in dependent variables (WM, GM, TCV, DTI measures) while accounting for differences in age and sex; hierarchical multiple regression tested associations between fiber tracts of interest and behavioral data, while testing for effects of age and sex.

Results: There was a significant main effect of group on WM volume (F_3,87=21.74, p<.0001), with the AS group having significantly smaller WM volume compared to all other groups (p<.0001), and the FXS group having significantly larger WM volume compared to all other groups (p<.05) (Fig. 1). There was a significant main effect of group on GM volume (F_3,87=17.75, p<.0001), with the AS group having significantly smaller GM (p<.0001), but there were no significant differences between the other groups. The AS group had significantly smaller TCV (F_3,87=22.36, p<.0001). Figure 1 illustrates the least-squares means, adjusted for age and sex. The AS group showed a disproportionate reduction in WM volume (32% smaller vs. TD group) compared to GM volume (18% smaller vs. TD group).

Following this finding of WM reduction in children with Angelman syndrome, tractography was performed in several WM tracts including the corpus callosum, corticospinal tract, and corticothalamic tracts. In each of these tracts, the AS group showed significantly lower FA (p<.0001) and higher RD (p<.0001) compared to TD controls. In the AS group, higher RD in these motor tracts was significantly associated with poorer motor ability (r=-.58, p<.05) (Fig. 2).

Conclusions: Differences in white matter are evident in monogenic syndromes such as Angelman syndrome and Fragile X, even using coarse measures such as overall WM volume. However, more detailed attempts must be made to tease apart the heterogeneity of idiopathic autism. DTI, for example, has been previously shown to identify connectivity alterations associated with specific clinical symptoms in ASD and FXS, and here we show associations between corticospinal connectivity and motor ability in Angelman syndrome, a disorder marked by pervasive and disabling motor deficits. DTI analyses of the ASD and FXS group are underway and will be presented along with associations with repetitive behaviors, anxiety measures, and social and communicative ability.