Mom's ovaries could hold clues to some autism cases, new research suggests—and this time it's not because of genetic vulnerabilities carried in her eggs. A new, large-scale study out of Sweden suggests that women with polycystic ovarian syndrome (PCOS)—an endocrine disorder that affects 5 to 10 percent of women of childbearing age—have an increased risk of giving birth to children with autism spectrum disorder (ASD).
The Karolinska Institute's Renee Gardner, along with colleagues from Sweden and the U.S., tapped into a Swedish national population health database to look at potential ties between PCOS and ASD. As they reported online December 8 in Molecular Psychiatry, the team looked at 23,748 individuals with ASD and nearly 209,000 unaffected individuals, all born in Sweden between 1984 and 2007.
Although identifying information about the individuals was removed, the researchers had access to information about their relationships to others in the database as well as documented diagnoses and use of health care services. The group found that ASD was 59 percent more prevalent in children born to women with PCOS—a relationship that was independent of PCOS complications such as increased neonatal distress or C-section delivery.
This risk level is roughly comparable with that of having a father over age 50 (estimated to be 66 percent) but lower than it is in those with certain rare genetic syndromes or mutations. The authors of the analysis believe PCOS increases ASD risk in offspring to a greater extent than maternal infection, one of many factors previously implicated in autism.
The apparent PCOS link may seem odd at first glance. But it fits with a theory for autism development that centers on androgens—male sex hormones. Many researchers suspect conditions that boost androgen levels or otherwise upset hormone balance during pregnancy may alter fetal brain formation in ways that contribute to social deficits, delayed language development and other autism-associated symptoms later in life. The theory is supported by several lines of evidence including epidemiological, genetic and mouse model data as well as direct measurements of male hormones and hormonal precursors in amniotic fluid from children with autism and Asperger syndrome.
All women naturally produce some male sex hormones, typically at low levels. Along with ovarian cysts, women with PCOS are prone to increased androgen levels, which can contribute to symptoms such as acne, excess hair growth, weight gain and, in some cases, problems becoming pregnant. For women who do conceive, both PCOS and obesity have been implicated in a bump in androgen levels during pregnancy.
In that framework, results from the Swedish study suggest that "maternal testosterone, which can cross the placenta, is one source of the elevated prenatal steroids to which children who later develop autism are exposed," Simon Baron-Cohen, director of the University of Cambridge Autism Research Center, noted via e-mail.
A longtime proponent of the fetal androgen theory, Baron-Cohen has been involved with numerous studies related to the hypothesis, including yet-to-be-published work by a former graduate student in his lab, who saw similar ties between PCOS and ASD using population data from the U.K.'s National Health Service. In addition, a 2007 study in Hormones and Behavior, co-authored by Baron-Cohen, also found an overrepresentation of PCOS among mothers with autistic children, Although the finding did not reach statistic significance.
The new study from Sweden, however, offers robust evidence that maternal PCOS increases autism risk. Hakon Hakonarson of The Children's Hospital of Philadelphia, who studies ASD genomics and was not involved in the work, agrees that the reported association is convincing, given the size of the study and high quality of the data from the Swedish population resource. He adds that the findings fit with a role for altered fetal hormone exposure in a subset of ASD cases.
But Hakonarson is quick to point out the full gamut of ASD cases likely involves many different genetic and environmental contributors, because the absolute ASD risk in children of PCOS sufferers remains small. With ASD rates hovering around 1 to 2 percent in the general population, he explains, even an almost 60 percent increase in ASD rates in children born to PCOS-affected moms translates into just a fraction of a percent higher chance of the disease overall. Experts say additional contributors are almost certainly present, because the vast majority of women with PCOS give birth to children who don't develop autism. "For any one woman with PCOS, this doesn't mean that her child is definitely going to get autism," Karolinska’s Gardner agrees. "It just means that the risk of this relatively rare disorder is somewhat increased." For her part, she hopes the study will raise awareness amongst women with PCOS without causing alarm or stress.
Investigators are hesitant to tout any clinical interventions at the moment, although Gardner says her group is interested in using population data to compare ASD prevalence in children born to women who have or haven't received pharmaceutical treatment for PCOS. And because the association described in the Swedish study appeared more pronounced in PCOS sufferers who carried excess weight during pregnancy, she suggests women who have the condition may benefit from making an extra effort to maintain a healthy weight before and during pregnancy.
Finally, both Gardner and Baron-Cohen note that the PCOS and ASD link may help to identify ASD-prone children who might benefit from early interventions.