Novel Approaches for Investigating the Neurology of Autism: Detailed Morphometric Analysis and Correlation with Motor Impairment

In order to understand how to best diagnose and treat individuals with autism, it is important to identify the brain abnormalities associated with the disorder. Studies of postmortem brain tissue from individuals with autism who died unexpectedly have been informative; however, while postmortem examination allows for detailed analysis of brain structure, it is limited by the small number of available brains and by the inability to directly examine the in vivo (“living”) brain. The latter may be crucial; understanding the brain basis of a complex and variable disorder such as autism may require directly examining how differences in brain structure are associated with characteristic behavioral features.  

Magnetic resonance imaging (MRI) offers a complementary approach that addresses these limitations. Anatomic MRI investigations of autism have thus far focused on measuring the size (total volume) of particular brain regions. This “volumetric” approach has led to the discovery that autism is associated with an increase in total brain volume during early childhood; however, it has proven less useful in more precisely identifying the brain abnormalities associated with autism. More advanced approaches for MRI analysis of brain structure in autism are clearly needed.  

We propose to use a combination of innovative approaches to identify brain abnormalities in autism. First, we propose to use a powerful tool, “LDDMM,” that allows for detailed examination of the shape of brain structures. Doing so brings much greater precision, effectively “turning up the microscope” - by looking at differences in shape, one can identify the location of specific abnormalities in brain structures that would be overlooked in analysis of total volume. Analyses with LDDMM, which was developed by Dr. Michael Miller who is working on this project, have lead to important advances in early diagnosis and evaluation of treatment effects in Alzheimer’s and schizophrenia; however, this approach has only very recently been applied to studying autism. We will use LDDMM to analyze autism-associated differences in the shape of both grey matter structures (Specific Aim 1) and white matter connections between these structures (Specific Aim 2).  

Second, we propose to apply an innovative approach for examining how these differences in brain structure predict the behavioral features of autism; moving beyond measures of social, communicative, and behavioral dysfunction traditionally used in studies of autism, we will also examine associations with motor skill deficits (Specific Aim 3). Motor impairment is a highly consistent finding across studies of autism; people with autism tend to be clumsy and awkward, in parallel to the awkward nature of their social interaction. Motor signs can be measured with a high degree of precision and examining how differences in brain structure predict motor deficits offers an important window into understanding the brain basis of autism.  

The combination of these innovative approaches provides a promising new direction for identifying the brain basis of autism. Doing so can lead to earlier and more accurate diagnosis as well as help to focus efforts to identify the genetic, metabolic and/or environmental causes of autism, improving chances for prevention and treatment.