Autism is generally considered to be a neuropsychiatric disorder. Yet it
also seems to involve some immune system abnormalities, suggest new findings
reported at the Fourth International Meeting for Autism Research, held in
Boston in May.
The abnormalities concern both B cells (which make antibodies) and
cytokines (proteins that respond to viruses, bacteria, and other antigens;
engage in cross-talk between immune cells and neurons; and affect mood and
The results come from David Amaral, Ph.D., Judy Van de Water, Ph.D., and
co-workers. Amaral, a neuroscientist who specializes in the study of the
autistic brain, is research director of the University of California at Davis
M.I.N.D. Institute. Van de Water is an immunologist at the M.I.N.D.
Amaral and a colleague recruited 70 children with autism and 35 normally
developing children aged 4 to 6 for their study.
They drew a sample of blood from each subject, examined each sample for the
number of B cells present, and then compared the number of B cells in the
blood of the autism subjects with the number in the blood of the healthy
They found about 20 percent more B cells in the blood from the autism
subjects than in the blood from the healthy ones.
Van de Water and her co-workers drew blood samples from autism subjects and
from normally developing subjects aged 2 to 5, incubated the blood samples
with viral and bacterial antigens, and then determined how many cytokines were
produced in the samples in response to the antigens. They found that fewer
cytokines were produced in the blood from the autism subjects than in the
blood from the healthy ones.
What such immune anomalies mean remains to be determined. For example,
might an abnormally large number of B cells in the blood and a decreased
response of cytokines to antigen provocation play a role in autism's
causation? And if such immune quirks do play a role in the development of
autism, is that so in all cases or only in some?
"I think we are getting more and more convinced that autism is a
multiple causal disorder... whether the immune system is involved or
not," Amaral said at a press conference in conjunction with the autism
meeting. "Thirty percent of kids with autism have seizures, but 70
percent don't.... A certain percentage of children have severe
gastrointestinal problems, but some don't.... There is a regressive form
versus a more innate form; our best guess now is that the regressive form is
20 percent to 30 percent."
He added that in one of the M.I.N.D. Institute's new projects, researchers"
will evaluate 1,800 children [with autism] at all levels—genes,
the immune system, the brain—so that we can become more knowledgeable
about how many subtypes there are."
Eventually those immune anomalies flagged in autism subjects might turn out
to be early diagnostic markers for the disorder—something that does not
currently exist. And the reason why a biological test for autism at birth
could be so valuable, Amaral said, is that it would then put scientists in a
position of intervening at this very early date, when the brain is still
In fact, Amaral envisions that a biological test for autism at birth might
be useful for preventing autism in children who are susceptible to the
regressive form of the illness—the kind that develops between ages 2 and
In other words, "Such children may be born with a vulnerability to
autism, but not doomed to get it," he explained. "Something in the
environment may be a second hit. So if one could detect at birth those
children who are vulnerable and at the same time understand more about what it
is in the environment that triggers autism, it might be possible to keep those
children from ever experiencing the trigger and thus prevent their