STANFORD, Calif. - The brains of autistic children have a distinctive topography that a team of Stanford University scientists was able to capture using new imaging techniques, with the hope of someday creating a template for the autistic brain that could be used to diagnose kids at an early age.
Detailed, computerized analyses of MRI scans showed a pattern of organization, especially in regions of the brain dedicated to communication and self-awareness, that was particular to children with autism, according to the new research, which was published online in the journal Biological Psychiatry.
The results are still years away from being applied to children in the real world. But they could be used not just for diagnoses, but to refine treatments and offer a better understanding of how autism affects the brain.
"Older kids, you just talk to them and you know they have autism. But the 2-year-old where you don't have a good idea what's going on with him, whether he's autistic or not, this tool could help," said Dr. Antonio Hardan, a child psychiatrist and an author of the Stanford study. "And having this kind of tool might also help you determine what kind of treatment the individual will be getting."
Autism affects roughly 1 in 100 children in the United States, and rates have increased significantly in the past 20 years. As parents and doctors become increasingly aware of the disorder and its lifelong complications, research into diagnosing and treating autism has exploded.
Diagnosis has improved significantly, and on average, children are diagnosed at about age 3. But the diagnostic process is difficult and dependent on the skill of the doctor or therapist.
Autism experts would prefer that most children be diagnosed well before age 3, when the disorder has already started to affect their verbal and social skills. If doctors can start treating autism before brain damage has occurred, they may be able to avoid some of the worst effects and improve the long-term outlook for children.
"Lots of people are looking for biologic fingerprints that we can use reliably to identify kids very early on, ideally before any manifestation of the disorder," said Lisa Croen, director of the Autism Research Program for Kaiser Permanente Northern California.
Croen and other Kaiser researchers are looking at other types of biomarkers, such as molecular or genetic signatures that might be found with a simple blood or saliva test. Such tests could be done on newborns or even pregnant mothers to help identify children who are at risk of developing autism.
The Stanford study looked at the brain scans of 24 autistic children between ages 8 and 18, and compared them with scans of 24 children without autism.
Previous studies comparing scans analyzed only the overall volume of the brain, or a large section of the brain. Those studies have shown that autistic children often have larger than average brains, but the information isn't refined enough to be useful in diagnosing or treating autism.
The new study sectioned the scans into tiny cubes and then used computer analysis to compare the size and structure of individual cubes in autistic and non-autistic brains. That allowed scientists to get a much more detailed picture of the specific areas that differ between the brains.
The resulting brain maps applied to 80 percent to 90 percent of the children with autism. If the maps can be replicated in a larger group of children, and at a younger age, they could be used to help diagnose autism.
"This is a potential biomarker, although how much of a predictive value it has remains to be investigated," said Vinod Menon, a professor of neurology and psychiatry who led the Stanford study. "In principle, it will tell you whether the individual is more likely to be on the typically developing side or the autism side."