Brain Scans Show Early Signs of Autism Spectrum Disorder
Posted on by Dr. Francis Collins
For children with autism spectrum disorder (ASD), early diagnosis is critical to allow for possible interventions at a time when the brain is most amenable to change. But that’s been tough to implement for a simple reason: the symptoms of ASD, such as communication difficulties, social deficits, and repetitive behaviors, often do not show up until a child turns 2 or even 3 years old.
Now, an NIH-funded research team has news that may pave the way for earlier detection of ASD. The key is to shift the diagnostic focus from how kids act to how their brains grow. In their brain imaging study, the researchers found that, compared to other children, youngsters with ASD showed unusually rapid brain growth from infancy to age 2. In fact, the growth differences were already evident by their first birthdays, well before autistic behaviors typically emerge.
Autism spectrum disorder includes a range of developmental conditions, such as autism and Asperger syndrome, that are characterized by challenges in social skills and communication. Scientists have long known that teens and adults with ASD have unusually large brain volumes. Researchers, including Heather Hazlett and Joseph Piven of the University of North Carolina, Chapel Hill, found more than a decade ago that those differences in brain size emerge by about age 2 . However, no one had ever visually tracked those developmental differences.
In the new study reported in Nature , Hazlett, Piven, and their colleagues set out to collect that visual evidence. They examined 106 infants at high risk of ASD, based on an older sibling with that diagnosis. Fifteen of the study’s high-risk infants went on to be diagnosed with ASD at age 2. The study also included another 42 infants with no family history of ASD and a low risk for the disorder. In all groups, the infants were mostly white, had similar birth weights, and comparable family backgrounds.
Each infant underwent detailed behavioral assessments for early signs of ASD, such as trouble babbling or making eye contact, at 6, 12, and 24 months of age. At each visit, the researchers also used a magnetic resonance imaging (MRI) scanner to capture detailed images of each child’s brain while napping.
Between the first and second scans, or just 6 to 12 months into the study, the MRIs showed something remarkable. There was a significant increase in the surface area of the brains of kids who would later develop ASD compared to other children. By age 2, the brains of these kids were obviously larger. The researchers found that kids whose brains grew the fastest also had the most severe social deficits.
But could these observations be translated into early diagnosis? To begin looking for an answer, the researchers turned to machine learning. They wanted to find out whether a computer could use features captured in those MRI scans, including the surface area and volume of the brain, to predict accurately which kids would develop ASD and which ones wouldn’t. They found that 8 times out of 10, the computer got it right. Importantly, the computer-derived algorithm relied primarily on the changes in brain surface area between the ages of 6 months and 1 year to make those calls .
If the new findings can be confirmed in more children, it may lead to a much-needed new approach to early diagnosis for kids at high risk of ASD. Piven says the findings also highlight that ASD doesn’t occur suddenly and spontaneously, as sometimes incorrectly thought. Rather, the disorder develops over time—beginning in the first year of life or likely earlier—from genetic and environmental influences that the North Carolina team and others are working hard to understand.
Why should the development of a larger brain lead to ASD? We don’t really know yet. Perhaps in the process of molding the brain for optimum function, it’s not just the expansion of neurons and synaptic junctions that matter, but also the “pruning” that allows this complex network—the most complicated structure in the known universe—to achieve maximum efficacy for human abilities and social interactions.The findings also add to evidence from studies in mice that ASD may be related to abnormalities in the progenitor cells that allow the brain to grow . The hope is that, as more is learned about ASD’s underlying biology, researchers will make even greater strides toward improving its diagnosis and discovering entirely new kinds of treatments to help these kids and their parents.
 Magnetic resonance imaging and head circumference study of brain size in autism: birth through age 2 years. Hazlett HC, Poe M, Gerig G, Smith RG, Provenzale J, Ross A, Gilmore J, Piven J. Arch Gen Psychiatry. 2005 Dec;62(12):1366-76.
 Early brain development in infants at high risk for autism spectrum disorder. Hazlett HC, Gu H, Munsell BC, Kim SH, Styner M, Wolff JJ, Elison JT, Swanson MR, Zhu H, Botteron KN, Collins DL, Constantino JN, Dager SR, Estes AM, Evans AC, Fonov VS, Gerig G, Kostopoulos P, McKinstry RC, Pandey J, Paterson S, Pruett JR, Schultz RT, Shaw DW, Zwaigenbaum L, Piven J; IBIS Network; Clinical Sites; Data Coordinating Center; Image Processing Core; Statistical Analysis.Nature. 2017 Feb 15;542(7641):348-351.
 Regulation of cerebral cortical size by control of cell cycle exit in neural precursors. Chenn A, Walsh CA. Science. 2002 Jul 19;297(5580):365-9.
Autism Spectrum Disorder (National Institute of Mental Health/NIH)
Heather Hazlett (University of North Carolina, Chapel Hill)
Joseph Piven (University of North Carolina, Chapel Hill)
NIH Support: National Institute of Mental Health; Eunice Kennedy Shriver National Institute of Child Health and Human Development; National Institute of Biomedical Imaging and Bioengineering
Tags: ASD, Asperger syndrome, autism, Autism Spectrum Disorder, brain, brain imaging, child health, communication disorder, diagnostics, infant, Infant Brain Imaging Study, machine learning, MRI, pediatrics