PROJECT SUMMARY
Autism Spectrum Disorder (ASD) is a highly heritable and heterogeneous neurodevelopmental condition that is
defined by difficulties in social interaction and communication as well as patterns of repetitive/stereotyped
behaviors and restricted interests. Multiple lines of evidence have pointed to the cerebellum as playing a
crucial role in the pathophysiology of autism. In particular, there has been long-standing interest in the
cerebellar dentate nucleus, which is one of the most commonly reported brain regions affected in post-mortem
studies of autism. In vivo investigation of the dentate in autistic individuals has been limited by difficulty
visualizing its structure using standard 3T magnetic resonance imaging (MRI) scanning due to its signal
properties (high iron concentration with resulting susceptibility artifacts) and convoluted structure with poor
boundary resolution at 3T. This has left us with a major gap in understanding how the cerebellum might
contribute to autism pathophysiology, as the dentate nucleus comprises nearly the entirety of cerebellar
projection to the cerebral cortex and is thereby crucial to development of a wide range of sensorimotor, social-
emotional and cognitive skills often impaired in autism. Recognizing the importance of investigating the
dentate, we propose to apply, in autistic children (and typically developing (TD) control children), a highly
innovative high-field 7T MRI approach that includes novel susceptibility-weighted imaging (SWI) and
quantitative susceptibility mapping (QSM) methods developed by our research team, which are designed to
optimize visualization of iron-rich brain regions such as the dentate nucleus. Leveraging these innovations, we
propose (Aim 1) to then employ complementary univariate (region of interest; ROI-based) and multivariate
(voxel-based) approaches to examine for autism-associated differences in total and regional dentate volume.
We further propose (Aim 2) to leverage this vastly improved dentate definition, combined with resting state
(BOLD) at 7T, to assess for autism-associated differences in intrinsic functional connectivity of the dentate,
including ventral vs. dorsal components. Finally, we propose (Aim 3) to investigate, among ASD children,
associations between measures of dentate structure and functional connectivity (derived in Aims 1 and 2,
respectively) with core autism features. We have successfully acquired 7T scans using our innovative imaging
methods in both adults and in children (both ASD and TD) and have established high levels of intra- and inter-
rater reliability for delineating dentate nucleus volume, providing strong feasibility for the proposed
investigation. Our innovative proposal thereby has strong potential for “high reward” impactful discovery: It will
lead to crucial advances in understanding how differences in cerebellar dentate structure/function might
contribute to the pathophysiology of autism and, in doing so, help identify novel brain-based biomarkers
important for improving diagnosis, prognosis and targeted intervention, including offering a potential
therapeutic target for neuromodulation and other therapies.
