DNA testing for siblings of children with autism may predict diagnosis before symptoms appear

Ten-year study is first to measure predictive value of genetic alterations related to autism spectrum disorder.


A new study shows that DNA testing could help predict the likelihood of autism spectrum disorder in younger siblings of children who have already been diagnosed. (Photo: Getty Images)

New research led by the Hospital for Sick Children and the University of Alberta shows that testing the DNA of siblings of children with autism spectrum disorder (ASD) may predict a future diagnosis even if symptoms aren't yet apparent.

"Genetic factors are the most likely reason we see a clustering of ASD-related traits in families," explained Stephen Scherer, senior scientist and director of The Centre for Applied Genomics (TCAG) at SickKids.

"We wanted to investigate the possible benefits of genetic testing for infants whose older sibling had already been diagnosed with ASD. If we can identify those children early, we may be able to enrol them earlier in therapies," said Scherer, who is also the director of the McLaughlin Centre at the University of Toronto and principal investigator of the study

ASD refers to a group of neurodevelopmental conditions that cause challenges with communication, social understanding and behaviour. Studies show families who have a child with ASD have a 6.9 to 19.5 per cent chance of another child having ASD and a 30 to 40 per cent chance of another child having atypical development.

In the new study, published in Nature Communications, researchers tested nearly 300 infant siblings from 253 families for the presence of genetic alterations called copy number variations (CNVs) that have been linked to ASD.

They found that the presence of an ASD-relevant CNV in a sibling had a high likelihood of predicting a future diagnosis of ASD or atypical development. Indeed, by age three, 157 siblings were either diagnosed with ASD or developing atypically. DNA testing revealed CNVs in genes relevant to ASD in 11 of the 157 siblings who were eventually diagnosed.

This marks the first time scientists have been able to quantify the predictive value of CNVs in determining these diagnoses.

Early identification could lead to earlier intervention

One of the priorities of interventions for ASD is starting early; evidence shows infants as young as seven months old could benefit. Yet, most children in North America aren't diagnosed with ASD until they're over four years old.

"These findings add to a growing body of evidence that biomarkers might be helpful in identifying pre-symptomatic infants who are likely to develop ASD or other developmental challenges," said Lonnie Zwaigenbaum, a professor of pediatrics at the U of A, the Stollery Children's Hospital Foundation Chair in Autism and a Stollery Science Lab Distinguished Researcher.

"At this point, we can't fully determine the anticipated severity of a child's future symptoms. What we can say is that it's important to closely monitor their development and start therapeutic interventions early to support their skill development and address emerging functional impairments related to ASD."

The research team has confirmed similar findings in a separate group of 2,110 families having one child with ASD and a second child without it.

Their next step will be to look beyond CNVs and determine how newer technologies like whole genome sequencing might increase the early genetic detection rate.

The research was supported by Autism Speaks, Autism Speaks Canada, the Simons Foundation Autism Research Initiative, National Institutes of Health, National Institute of Mental Health, Canadian Institutes of Health Research, Canada Foundation for Innovation, Genome Canada, Ontario Genomics, Kids Brain Health Network, Canadian Institutes for Advanced Research, Stollery Children's Hospital Foundation through the Women and Children's Health Research Institute, Ontario Brain Institute, the Government of Ontario, the McLaughlin Centre at the University of Toronto and SickKids Foundation.