Doctors in a national study were able to predict with 80 percent accuracy which children would develop autism by age 2 through analysis of MRI brain scans at ages 6 months and 1 year — the earliest yet indicator of possible development of the condition.
The first-of-its-kind research paves the way to predicting which children will develop the neurodevelopmental condition significantly before common behavioral symptoms emerge. Two is the standard age of diagnosis using conventional measures, which include a parental questionnaire.
About 150 children were included in the research, and the subjects were limited to those considered at high risk of autism because of an older sibling’s diagnosis. The results were reported in the journal Nature.
The findings hold importance for thousands of families across the country. Experts at the national Centers for Disease Control and Prevention estimate that 1 in 68 children in the United States has an autism spectrum disorder.
“These investigators have shown that an MRI of the brain can better identify the likelihood that the younger sibling will also have autism, even before that infant or young toddler shows any clinical signs,” said Dr. Andrew Adesman, chief of developmental and behavioral pediatrics at Cohen Children’s Medical Center in New Hyde Park.
“It is hoped that early identification of a child will . . . allow for even earlier interventions than are currently provided,” said Adesman, who underscored that the findings need to be replicated by other researchers.
The new investigation hinged on older observations that have gone unexplained for years.
Brain enlargement was known to occur in children with autism, but that phenomenon and its timing remained confounding, as has the enlargement’s relationship to autism’s behavioral symptoms.
The brain scans examined in the new analysis revealed enlargement in specific regions of the brain. The scans also revealed changes in brain volume, surface area and thickness of the organ at both 6 and 12 months of age. These features helped scientists predict the condition with unprecedented accuracy.
Currently, doctors monitor the infant siblings of children who have autism but do not intervene until there are clinical signs of the condition.
Typical therapeutic interventions focus on language and behavior-skills training, which have become the core of care for children on the autism spectrum, Adesman and other experts said.
Evelyn Ain of Oyster Bay, the mother of a teenage son with an autism spectrum disorder, applauded the new research but voiced some qualifications.
“First of all, it’s wonderful that this research is being done because, really, there is not enough research,” said Ain, a national advocate for children with autism. “Even with something as positive as this, I can’t say that it gives parents much hope, but it does give them a jump-start on the services and therapy.
“Early intervention is one of the best tools to help children succeed,” she said, “and the earlier you start, the better it is for the child.”
One issue, she added, is that parents can’t secure interventional services based on MRI scan data alone. She expressed doubt that those services would be effective in a child of only 6 months, even with early evidence of autism.
Previous observational studies have suggested that children first show evidence of autism around their first year, when they do not meet many of the milestones of others their age, such as not responding to their name when called or not making eye contact when spoken to. Symptoms vary from one child to the next.
An inability to use language, or to speak at all, remains one of the hallmarks of the condition, as are obsessive interests, such as an affinity for certain toys or subjects. Some children show repetitive behaviors, such as rocking back and forth, head banging or “walking only on tippy toes,” Ain said.
Team members in the study, led by medical investigators at the University of North Carolina at Chapel Hill and including 10 other institutions, credited their predictions’ accuracy to a customized algorithm that was applied to the data in classifying children most likely to meet criteria for autism by age 2.
Dr. Joseph Piven, the study’s senior author, said the findings open a new window into the diagnosis of autism.
“Typically, the earliest an autism diagnosis can be made is between ages 2 and 3,” he said. “But for babies with older autistic siblings, our imaging approach may help predict during the first year of life which babies are most likely to receive an autism diagnosis at 24 months.”