Reduced Ratio of Inner to Outer Surface Area of the Brain in Autism Spectrum Disorders (ASD)

Background:

The neuroanatomy of ASD is both complex and multifaceted, affecting multiple aspects of the neural architecture in vivo (Ecker, 2017). Among others, neuroanatomical abnormalities in ASD also include differences in cortical folding (e.g. Wallace et al., 2013; Ecker et al., 2016, Schaer et al., 2013). While several different mechanisms have been suggested to mediate cortical folding (for review see Lewitus et al., 2013), there is evidence to suggest that local gyrification is driven by a differential expansion of upper vs. lower neural layers (Richman et al., 1975; Kriegstein et al., 2006). This hypothesis is also of relevance to recent genetic investigations suggesting that ASD-related genes are not equally expressed across all cortical layers, but predominantly affect the superficial cortical layers (e.g. Parikshak et al., 2013).

Objectives:

The present study therefore aimed to establish whether the ratio of the inner to outer surface area (R_outer:inner) of the brain significantly differs in ASD individuals from neurotypical controls during childhood and adolescence.

Methods:

73 right-handed males with ASD (diagnosed using ADI-R (Lord et al., 1994)), and 75 neurotypical controls, aged 8-25 years were recruited and assessed at the Institute of Psychiatry, Psychology, and Neuroscience, King’s College, London. Both groups were matched for age, handedness, and full-scale IQ. For all 148 participants, high-resolution structural T1-weighted volumetric images were acquired.

Cortical surface models were derived using the FreeSurfer image analysis suite (http://surfer.nmr.mgh.harvard.edu/). The R_outer:inner was calculated as the ratio of inner (i.e. white matter) to outer (i.e. grey matter) surface area. Parameter estimates for R_outer:inner were estimated by regression of a general linear model (GLM) at each vertex with (1) group as categorical fixed-effects factor, (2) linear age, as well as an age-by-group interaction, and (3) FSIQ as continuous covariate. Corrections for multiple comparisons across the whole brain were performed using random-field theory (RFT)-based cluster-corrected analysis for non-isotropic images using a p=0.05 (two-tailed) cluster significance threshold.

Results:

There were no significant group differences in age [t(146) = -1.69, p = 0.09], or FSIQ [t(146) = 1.77, p = 0.08] between ASD individuals and controls. We found that R_outer:inner was significantly decreased in ASD individuals relative to controls in several clusters, which predominantly included prefrontal and temporal regions. Moreover, we observed significant age-by-group interactions in the (1) right middle temporal gyrus, (2) left anterior cingulate cortex, and in the (3) right superior cortex (RFT based cluster corrected, p<0.05). In addition, ASD individuals also had significantly reduced vertex-wise measures of absolute inner and outer surface area compared to healthy controls.

Conclusions:

Our findings suggest that a differential expansion of the inner and outer surface area in the brain of individuals with ASD may mediate the differences in local gyrification that have previously been reported in the condition. The R_outer:inner might also be used to guide future studies into the genetic and molecular underpinnings of ASD, and for the stratification of ASD individuals into biologically more homogeneous subgroups.