Jan. 26, 2015 -- Scientists searching for the genetic roots of autism have found something surprising: In families where two children have been diagnosed with the disorder, siblings don’t often share the same gene changes.
The new study fits into a larger body of research suggesting the genetics of autism are still poorly understood. It also raises the possibility that the disorder isn’t usually inherited, even when it runs in families.
Instead, it could be that some other factor, like a shared environment or common conditions during pregnancy, might explain why children with an older brother or sister with autism have a 7 times' higher risk of getting the condition themselves, compared to kids in the general population.
Exactly what the shared risk might be is anyone’s guess.
“This is still the million dollar question,” says Stephen Scherer, PhD. He’s the director of the Centre for Applied Genomics at the Hospital for Sick Children and the senior author of the study, which is published in the journal Nature Medicine.
The researchers also found some intriguing new clues about biological causes of the disorder, which affects 1 in 68 children and appears to be on the rise in the U.S.
For the study, they sequenced all the genes of parents and children in 85 families where two children had been diagnosed with autism.
They found gene changes known to be linked to autism in 36 -- or slightly less than half -- of the affected families.
Of the 36 families where the cause of autism could be tied to specific gene changes, only 11 sets of siblings shared the same changes, and only 10 sets -- or less than a third -- got those changes from their parents.
In two-thirds of those families, there's at least one different gene change between siblings. “So they have their own form of autism,” Scherer says.
And siblings who didn’t share the same "risk genes" tended to have autism that looked very different, with different problems and challenges. One brother or sister might not be able to speak or communicate, for example, while the other might have close-to-normal language development.
“I think when physicians who see the kids hear that, they’re not surprised so much because they really see the variability, even between siblings,” Scherer says.
Scherer says the whole genome approach also uncovered some new points of risk. One change in a gene called THRA affects the thyroid. Another group of mutations, called Lo-F, are involved in hearing and deafness.
“When we saw this, I got excited right away, because it makes sense that some of the kids might not be responding or interacting, because they’re deaf,” Scherer says.
Previous studies have shown that only about 10% of autism cases can be explained by penetrant genes -- genes in which a specific change nearly always leads to the same symptom or disease.
But most cases of autism seem to be rooted in 150 or so genes changes that are too weak to cause symptoms on their own. They only become a problem when a person has a combination of changes that tip them into a disease or a condition like autism, says Charis Eng, MD, PhD. She's the chair of the Genomic Medicine Institute at Cleveland Clinic in Ohio. She wasn't involved in the research.
Eng says another possibility is that the genes between brothers and sisters may be different, but they affect the same biological processes -- for example, the way the brain processes sound -- ultimately causing autism.
And she says it could be that some common environmental factor, like exposure to a chemical, tips a child with several weak "risk genes" into autism. Because so many different families were studied, though, it seems unlikely that it would be the same factor in every case.
Another expert says the real value of the new research is that it shows just how much we still don’t know about the condition or how children get it.
“This study confirms there are many, many different ways to get to the behaviors we call autism,” says Chris Gunter, PhD. She's the associate director of research at the Marcus Autism Center in Atlanta. She wasn't involved in the study.
“Maybe what we’re calling autism is too big a label,” she says.
The data from the study will be added to a major program called the MSSNG project, which is a collaboration between Google and the nonprofit advocacy group Autism Speaks. The goal of the project is to sequence the genomes of 10,000 families with autism and make the data freely available to researchers around the world.