Asymmetry of Fusiform Structure in ASD and Its Association with Symptom Severity

Background:  

Deviations from typical brain asymmetry have been implicated in neurodevelopmental disorders (Renteria 2012). Of the brain regions implicated in autism spectrum disorder (ASD), the fusiform gyrus, is associated with abnormalities in face processing (Schultz 2003). Although the fusiform may underlie ASD symptoms, only a few studies have examined its structural asymmetry (Herbert 2002; 2005). The asymmetry of the fusiform in terms of volume, surface area (SA), cortical thickness (CT), or developmental trajectory are not well characterized. Furthermore, the relationship between fusiform asymmetry and ASD symptom severity is unclear.

Objectives:  

We examine the following: 1) Left-Right (L-R) asymmetry of volume, SA, and CT of the fusiform gyrus in ASD compared to typically developing controls (TDC); 2) developmental differences in fusiform asymmetry between ASD and TDC; 3) correlation between fusiform asymmetry and symptom severity as measured by Gotham Autism Severity Scores (Gotham 2009).

Methods:  

Structural MRIs and demographics of 128 ASDs (average age ± std: 15.5 ± 4.1 years) and 127 TDCs (15.6 ± 3.8 years) from the ABIDE dataset were used in this study. All subjects were right handed males between the ages of 6 to 25 years. We processed the MRIs with FreeSurfer and derived the volumes, SAs and CTs of the fusiform gyri. Symmetry index (SI) (Galaburda 1987) of the above measures was calculated as the difference between the left and right hemispheric measures as a percentage of the average of the left and right measures. 

A generalized linear model was fitted for SI with the following regressors: age, ASD/TDC status, verbal IQ (VIQ), performance IQ (PIQ), and age status interaction. For correlations between SI and symptom severity, SIs from 69 ASDs were fitted with the following regressors: age, VIQ, PIQ, and ADOS. The aforementioned model was repeated with autism symptom severity scores from 28 ASDs. ADOS and severity score analyses were limited to a subset of subjects for whom these data were available.

Results:  

Significant group differences in volume (p = 0.031) and SA (p = 0.031) asymmetries emerged between ASD and TDC. ASD subjects exhibited greater leftward asymmetry for both measures (Figure 1A). Although significant overall group differences are evident, asymmetries for ASDs were spread both leftward and rightward (see distribution in Figure 1A). A significant interaction with age (p = 0.049) emerged in SA for ASD and TDC with decreasing leftward asymmetry in ASD with increasing age (Figure 1B). ADOS scores and fusiform volume asymmetry were negatively correlated (p = 0.047). A stronger correlation (p = 0.0097) was observed between autism symptom severity and fusiform volume asymmetry. See Figure 2A, B.

Conclusions:  

We provide evidence for abnormal patterns of asymmetry in fusiform gyrus volume and SA in ASD compared to TDC. The wide distribution of asymmetry in ASD may indicate the presence of ASD subtypes, and closer evaluation is needed. Furthermore, the development of asymmetry in fusiform SA may differ in ASD. Finally, we provide evidence that asymmetry in fusiform volume may be related to autism symptom expression using both ADOS and autism severity scores.