Decreased Gray Matter in Orbitofrontal and Superior Temporal Areas in Autism: A Voxel-Based Morphometry Study

Background: Despite the abundance of studies examining brain anatomy in autism spectrum disorders (ASD), a biomarker for the disorder has been rather elusive. Of late, some studies have proposed structures, such as the superior temporal sulcus, as an endophenotype that can predict the severity of autism symptoms (Kaiser et al., 2010). Similarly, some other studies have found gray and white matter abnormalities in orbitofrontal cortex (OFC) to have a negative correlation with social behavior in ASD (Girgis et al., 2006; Rojas et al., 2006). The current study is an attempt to examine structural integrity of these two social brain structures, and possibly relate our findings to underlying connections, to developmental trajectory, and to symptom severity in ASD.

Objectives: The objective of our study is to determine if gray and white matter composition of the social brain, specifically the superior temporal cortex (STC) and OFC, can provide clues to potential neural signatures of ASD. 

Methods: Structural images were acquired from ten high-functioning adults with ASD and ten typically developing control participants (data collection in progress). To obtain region-based measures of gray and white matter volumes, participants’ T1-weighted structural scans were segmented using FreeSurfer 5.1 (Fischl& Dale, 2000).  All measurements were normalized to cortical, total, and hemispheric gray and white matter to adjust for individual differences in brain anatomy. We calculated the gray and white matter volumes of different areas in the social brain (amygdala, STC, OFC, insula, inferior frontal gyrus, medial prefrontal cortex, and anterior cingulate cortex). Simple regression analyses were used to assess the predictive relationship between ASD symptoms (measured by the Autism Quotient) and social brain area volumes. 

Results: The main results are as follows: 1) Participants with ASD had significantly reduced gray matter, relative to controls, in two areas of the social brain: STC and lateral OFC (STC: t(18)=2.07, p=0.05; OFC: t(18)=2.39, p=0.02). However, volumes of the other previously mentioned social brain regions were not significantly different between the two groups. 2) Our ASD group had significantly reduced total white matter (t(18)=2.27, p=0.03), as well as reduced left and right hemisphere white matter relative to control participants (Left: t(18)=2.31, p=0.03; Right: t(18)=2.21, p=0.04); 3) A simple regression analysis revealed that greater AQ scores marginally predicted reduced OFC volume (β=-0.46, t(18)=-1.91, p=0.07). 

Conclusions: These results support previous studies of anatomical alterations in the ASD brain, specifically the OFC (Salmond, et al., 2003).  Although the STC was consistently smaller in volume in the ASD group, we did not find any significant relationship between this and symptom severity in ASD. While the evidence for an endophenotype may be mixed at this point, our findings may suggest the OFC and STC as potential targets for future research in ASD. We also plan to examine the developmental trajectory of brain organization in ASD by comparing data of children and adults with ASD. Finally, our finding of reduced total white matter in the ASD group may impact functional and anatomical cortical connectivity (Kana, Libero, & Moore, 2011).