The causes of autism – including genetic and environmental factors – are not entirely understood. Many studies have already shown that serious maternal infection during pregnancy is associated with increased risk for offspring in both people and animals. New research however, shows that the lowest risk for autism is associated with mid-levels of an immune marker measured at birth – whereas too much or not enough were linked to increased risk.

The report from researchers at the Karolinska Institute in Sweden appears in Biological Psychiatry, published by Elsevier. The study hinges on the idea that the developing brain may be particularly vulnerable to disturbances in immune signaling and exposure to inflammation.

“We studied a set of molecules called acute phase proteins that are part of the innate immune system, which is our first line of defense against infections and is always monitoring the body for signs of invasion,” said lead author Renee Gardner, PhD. “These molecules are circulating in our bloodstream all the time, but they can rise rapidly after exposure to infection.”

The researchers examined the proteins from blood samples, taken at birth, of nearly 1,000 autistic children and over 1,000 healthy controls from the Stockholm Youth Cohort, a Swedish health registry. Babies born with high levels of a classical marker of inflammation, called C-reactive protein (CRP), were at highest risk for autism.

The thinking based on previous studies was simply that too much inflammation is bad for the developing brain. Surprisingly, however the lowest risk was associated with levels of CRP in the mid-range. “This means that too much inflammation may indeed be a bad thing for the developing brain, but so might too little, Dr. Gardner explained.

“Among newborn babies whose mothers had been hospitalized for an infection during pregnancy, those who were able to make a little more of these acute phase proteins tended to have a lower risk of autism. So it seems as if a greater ability to respond to the immediate environment might translate to a lower risk of autism,” she added.

In a second part of the study, the researchers compared the immune protein levels at birth between children with autism and their siblings without autistic. The unaffected siblings had higher levels of the immune markers than those with autistic . “This is interesting because siblings share about half their DNA, and the environment inside the womb and during the first few days of life are likely similar between siblings,” Dr. Gardner said.

Another interesting finding relates to the risk for autism posed by maternal anemia, or iron deficiency. Among babies whose mothers were anemic, those with the highest levels of the iron-binding protein ferritin in their blood – a proxy for iron levels – were protected from autism. That finding hints at the importance of iron status for the developing brain and may explain the anemia-linked risk of neurodevelopmental disorders.

“The association between markers of activation of the immune system at birth and the subsequent risk for autism could be important,” said Biological Psychiatry editor John Krystal, MD. “We have been seeking avenues for prevention for autsim such as mechanisms that could be targeted by medications before the onset of symptoms. However, we must be cautious as we do not yet know whether immune activation is a contributor or a marker of the risk for autism.”