Alternation of Gray Matter in Autism Spectrum Disorder: Subcortical-Cortical of Structural Covariance and Age Effect Analysis

Objective: Autism spectrum disorder (ASD) is a neurodevelopmental disorder with variable phenotypes including atypical brain structure and function. Magnetic Resonance Imaging (MRI) studies on the atypical brain structure over developmental age in individuals with ASD have yielded inconsistent results. In this study, we aimed to better characterize the developmental effects on the neuroanatomy of males with ASD by comprehensively considering various brain morphological parameters as well as their age-diagnosis effect and cortical - subcortical associations over childhood to young adulthood (Alexander-Bloch, et al., 2013c). We postulated that ASD is associated with altered trajectories of neurodevelopmental processes involving pruning and establishing new connections. Based on this, we anticipate that individuals with ASD would differ from typically developing individuals (TD) in the patterns of cortical and subcortical morphology with age- related effect including gray matter volume that reflect these neurodevelopmental differences. Moreover, if present, these differences should be more apparent in childhood and even out with greater age. Methods: Structural MRI (3T) images of 117 males with ASD (mean age ± SD), 14.6±4.4 years) and 108 TD males (mean age ± SD, 15.0 ± 5.9 years), with age ranging from 7 to 30 years were acquired. We separated the participants for structural covariance approach into childhood (ASD: n = 48, mean age ± SD, 10.8 ± 1.2 years; TD: n = 51, mean age ± SD, 10.5 ± 1.5 years), and to evaluate the brain morphological differences over these different stages of the developmental lifespan. Volume based morphometry analysis was implemented using FreeSurfer ver. 5.2.0, which percolated gray matter into various brain regions with volume. The statistical analysis was using multiple regressions. Results: Participants with ASD in childhood had significantly diagnosis- by-age interaction effects that smaller relative regional volumes than TDC in the left lateral orbital frontal cortex, right rostral middle frontal gyrus, insular cortex, left superior temporal gyrus, bilateral inferior parietal cortex, isthmus–cingulate cortex and right pericalarine cortex which that ASD had more slowly decrease through the maturation, and the cross point consistently shown in adolescents (from 12-15 years old). As expected, for cortical subcortical associations, during childhood, the thalamus, putamen and nucleus accumbens volumes were positively correlated with regional cortical volumes more extensively in ASD than TD during childhood and only specifically in right hemisphere. Conclusions: Our findings support a systemic alteration in neurodevelopmental morphology in ASD. The age modulated the global volume difference in ASD and TDC, however the pattern of age- related effect was opposite from child to young adult in two groups. Importantly, we report a novel finding of distinct patterns of cortical-subcortical associations between ASD and TD. Moreover, these effects of ASD were somewhat masked during adolescence, consistent with a more heterogeneous phase of neurodevelopment as some ASD individuals begin to ""catch up"" with TD while some others do not.