The research is already leading to what could be the first biological test for autism; the antibodies are found almost exclusively in mothers of autistic children, and not in children with other types of disorders or in mothers of non-autistic children. Only 1% of mothers whose children were not affected by autism had the antibodies in their blood, compared to 23% of mothers of autistic children. Judith Van de Water, an immunologist and professor of internal medicine at the University of California Davis MIND Institute and the study's lead author, has consulted for a company, Pediatric Bioscience, that is developing a commercial version of the test, but the research was not funded by that organization and was supported primarily by the National Institute on Environmental Health Sciences.
"We haven't found any [mothers] who have these antibodies and don't have children with some sort of developmental disability issue," says Van de Water. "We feel this really identifies a subtype of autism."
The antibodies belong to a class of compounds called autoantibodies, which are immune cells that the body makes to target — often mistakenly — its own cells. Scientists do not know why or when the mothers produce these antibodies, which appear to monkey with normal nerve development in the fetal brain by interfering with their growth, migration and genetic replication. It is possible that infections during pregnancy — a known risk factor for autism —can prompt the immune system to produce them. Exposure to toxic chemicals can also cause immune defenders to mistake healthy cells for invaders, Van de Water notes.
The study involved 246 autistic children and their mothers, as well as 149 typically developing children. Of the mothers tested, all but one with the antibodies had an autistic child— and the child of the remaining mother had ADHD, a condition that often occurs along with autism. That suggests that a positive test almost certainly indicates a developmental disability. However, since 77% of the mothers of autistic children did not have these antibodies, Van de Water says, a negative test would not rule out all risk of autism.
And so far, the presence of the antibodies do not seem to be associated with any particular form of autism. "Certain behaviors seem to be associated with this, including stereotyped repetitive behavior like hand-flapping and lower levels of expressive language," says Van de Water, but no unique behavioral signature has been found so far. The children also did not seem to score differently on cognitive tests than other youngsters with autism.
Image: Pregnant woman having blood drawn, via Shutterstock