Autism is Associated with in vivo Changes in Gray Matter Neurite Architecture.

Postmortem investigations in autism have identified anomalies in neural cytoarchitecture across limbic, cerebellar, and neocortical networks. These anomalies include narrow cell mini-columns and variable neuron density. However, difficulty obtaining sufficient post-mortem samples has often prevented investigations from converging on reproducible measures. Recent advances in processing magnetic resonance diffusion weighted images (DWI) make in vivo characterization of neuronal cytoarchitecture a potential alternative to post-mortem studies. Using extensive DWI data from the Adolescent Brain Cognitive Developmentsm (ABCD(R)) study 142 individuals with an Autism diagnosis were compared with 8971 controls using a restriction spectrum imaging (RSI) framework that characterized total neurite density (TND), its component restricted normalized directional diffusion (RND), and restricted normalized isotropic diffusion (RNI). A significant decrease in TND was observed in Autism in the right cerebellar cortex ({beta}=-0.005, SE =0.0015, p=0.0267), with significant decreases in RNI and significant increases in RND found diffusely throughout posterior and anterior aspects of the brain, respectively. Furthermore, these regions remained significant in post-hoc analysis when the ASD sample was compared against a subset of 1404 individuals with other psychiatric conditions (pulled from the original 8971). These findings highlight the importance of characterizing neuron cytoarchitecture in Autism and the significance of their incorporation as physiological covariates in future studies. Lay abstractChildren with autism have differences in neuron structure unique from the general population and populations with attention, anxiety, and depression disorders. Brain imaging data on over 11,000 children was acquired at ages 9 and 11 years-of-age. Estimates of neuron density were derived from brain imaging data using recently validated techniques and comparative groups were composed using parent reported diagnosis of autism and other common psychiatric disorders. Consistent macro-structural changes in brain have been difficult to replicate and micro-structural changes have been historically difficult to acquire with other methodologies. We identified regional differences in the density of neuron cell bodies, neuron branching, and total neuron density in those with a reported diagnosis of ASD. Findings were consistent when compared against those with other psychiatric disorders in post-hoc analysis. These findings demonstrate the viability and importance of investigating in vivo changes to neurons in those with autism to advance our current understanding of related physiology.