Absence of Age-Related Cortical Thinning In Autism Spanning Into Adulthood

Background: While there is substantial data suggesting that brain development of children with autism proceeds abnormally, there is less data charting its development through adulthood.  For example, the most replicable structural MRI finding in children with autism is megencephaly which is presumably driven by rapid overgrowth of brain tissue in early childhood.  This is followed an abrupt cessation while growth continues in typically developing individuals.  What is unclear is how this abnormal developmental trajectory continues through adulthood.  Two recent studies have addressed this issue by examining the relationship between age and cortical thickness in children, adolescents, and adults with autism.  The first study (N=127, age range=10-60 years) reported cortical thickness to be negatively correlated with age in controls but not in autism.  The second study (N=81, age range=12-24 years) reported accelerated age-related cortical thinning in autism.  Thus, further research is needed to reconcile these findings.

Objectives: The purpose of this preliminary investigation was to chart the relationship between age and cortical thickness in a sample of individuals with autism with ages spanning into adulthood.  It is hypothesized that age will correlate negatively with cortical thickness in typically developing individuals.  In contrast, it is hypothesized that age-related cortical thinning would be absent in individuals with autism.

Methods: Participants included 18 adolescent and adult males with autism (mean age=22.8 ±6.9 years, range=16.1-37.8 years) and 14 gender- and age-matched controls (mean age=25.9 ±7.6 years, age range=17.9-42.4 years).  High-resolution, T1-weighted images were acquired using a 3-Tesla MRI scanner.  All images were assessed for excessive motion artifact by an experienced rater.  Structural MRI data was processed and analyzed using the FreeSurfer image analysis suite (http://surfer.nmr.mgh.harvard.edu).  FreeSurfer consists of automated tools for reconstruction of the brain from structural MRI data, facilitating the quantification of cortical thickness using a spatially-unbiased analysis.  Group-wise comparison of cortical thickness data was conducted using the QDEC (Query, Design, Estimate, Contrast) application included in the FreeSurfer image analysis suite.  QDEC fits general linear model at each surface vertex where the discrete variable consisted of group membership and age as the continuous variable.  Images were generated displaying areas where the thickness-age correlation significantly differed between groups (p<0.01, two-tailed).

Results: There were numerous areas in both left and right hemispheres where the age-thickness correlation significantly differed between groups.  For the vast majority of these areas, this correlation was negative in the control group, but zero or positive in the autism group.  For the left hemisphere, these cortical areas included: prefrontal, cingulate, superior temporal, superior parietal, and paracentral.  For the right hemisphere, these cortical areas included: prefrontal, cingulate, superior temporal, middle temporal, inferior parietal, and lateral occipital.  Many of these areas involve cortical structures known to underlie social cognition.

Conclusions: Contrary to typically developing individuals where there is age-related cortical thinning, individuals with autism display an increase or no change in cortical thickness with age.  These preliminary results support the presence of an abnormal neurodevelopmental trajectory that extends into adulthood, affecting the neural structures involved in social cognition.