31576
New-Born Infants at Risk for Autism Spectrum Disorders Exhibit Altered Grey Matter Maturation
Autism spectrum disorders (ASD) alter the trajectory of brain maturation. This is based, not only on evidence that there are structural differences in brain regional grey matter (GM) in children and adults with ASD, but also differences in inter-regional GM correlations, suggesting that how brain regions connect and grow together is also distinct. However, it is not known to what extent the maturational profile observed even in children with ASD is primary to the condition, and/or a secondary or compensatory consequence of living with ASD. To understand what makes a brain vulnerable to dysmaturation we need to look as early as possible in development.
Objectives:
Therefore, in this exploratory study we tested the hypothesis that infants vulnerable to developing ASD traits have differences in inter-regional GM correlations already around birth.
Methods:
Brain MRI scans of term-born neonates were acquired during sleep, on a 3T Phillips scanner. GM volumes for 44 cortical and subcortical regions were extracted using a neonatal dedicated atlas (22 regions per hemisphere). An exploratory analysis compared inter-regional GM volume correlations (using Pearson's r) in 15 ASD high-risk (HR) neonates (i.e., with a parent or sibling diagnosed with ASD) and 15 low-risk (LR) neonates matched for sex and gestational age at birth (±0.3), with no significant differences in age at scan between the two groups (postmenstrual age mean=42.4±1.6). To examine if group differences were significantly different from the null hypothesis using non-parametric methods, permutation testing of 1000 random group assignments were performed. Significance level for group differences was set at p<0.01.
Results:
The HR group had an altered pattern of GM inter-regional correlations concentrated in temporal and limbic regions, especially involving the anterior cingulate. Specifically, 5 inter-regional GM correlations between the anterior cingulate, hippocampus, anterior temporal and superior temporal regions were significantly higher in the HR group. However, 7 inter-correlations between a wider network of regions were significantly lower in the HR group, including between temporal-occipital-parietal cortices, the insula and basal ganglia regions.
Conclusions:
Our preliminary analysis suggests that the inter-relationship between brain regions within networks which contribute to primary sensory and multimodal processes is already altered in neonates at HR of ASD. Moreover, there appears to be an anterior/posterior gradient in that anterior cingulate correlations with temporal lobe targets are higher in the HR group whereas correlations between temporal regions, the subcortex and posterior cortical targets are lower. Whether this pattern has behavioural or developmental correlates remains to be determined. In addition, how the inter-relationship between brain regions changes as the infants grow and whether there are even earlier (fetal) maturation differences in GM-based structural connectivity will be examined in a larger cohort.