Quotes of the Day

An international group of researchers reports the discovery of a set of rare genetic variations that may increase susceptibility to autism spectrum disorders. The new study, which came out of the Autism Genome Project Consortium, the largest database of genetic information from families with autism, was published Wednesday in the journal Nature.

Autism researchers welcomed the new paper as valuable progress in understanding some of the causes of a largely inscrutable disorder. "I think this paper is important because it's one in series of ongoing large studies that continue to really chip away very nicely at the mystery of autism," says Dr. Daniel Geschwind, director of the Center for Autism Research and Treatment at the University of California, Los Angeles.

Since the launch of the international consortium, a collaboration of more than 120 scientists and 50 institutions, at least two dozen or so genes have been identified and associated with autism spectrum disorders (ASDs). Past studies of twins and families have shown that susceptibility to ASDs is primarily genetic, and researchers estimate that 5% to 15% of autism cases can be traced back to specific, known genes. But the complex genetics of the other 85% cases of ASD have been difficult to untangle. Genes may play a role in these children's vulnerability to autism, but many other factors, such as parental age and other unknown environmental influences, may impact whether they actually develop autism's hallmark deficits in social, cognitive and communication skills.

In the new study, researchers have taken a step toward quantifying the influence of certain genetic patterns. After conducting detailed genetic analyses and comparing the genomes of 996 people affected with autism and 1,287 matched controls, they found that those with ASDs were 20% more likely to have so-called copy number variations — abnormalities in the number of copies — of specific genes. These changes affect about 1% of people in the general population.

Copy number variations (CNVs) may include either missing copies or additional copies of genes; most genes have exactly two copies, one inherited from the mother and the other from the father. CNVs may also be inherited, but they may pop up spontaneously — in fact, some of the autism-related changes discovered in the current study were found in children but not their parents.

If such variations occur in important genes, such as those involved in development, they could have deleterious effects. These are precisely the CNVs that the consortium scientists have associated with ASDs: most of the genes showing CNVs in autism patients were linked with intellectual ability — and had been previously associated with autism. In the current study, these genes carried a 70% greater likelihood of harboring such copy variations than other autism-related genes.

By conducting a deeper investigation into the genes that showed the CNVs, the consortium also stumbled upon four new genes associated with autism. The new genes, in addition to the nearly two dozen genetic defects already linked to ASDs, will guide researchers toward new targets for treatment, the authors hope. "Prior to 2007, we knew nothing about autism. We knew there was some genetic involvement but we had no details," says senior author Stephen Scherer, a senior scientist at Hospital for Sick Children at University of Toronto. "This study and others that are coming are really starting to crack open the black box and give us a better understanding of what is going on in autism."

What has particularly energized the research community is that all of the genes linked to autism so far, including those discovered in the Nature paper, appear to affect similar biochemical pathways in the brain — pathways involved in intellectual development or cell signaling and communication. In the current paper, the consortium scientists overlayed the map of the various genes affected by CNVs on that of previously identified autism genes involved in intellectual development, and saw a gratifying convergence between the two sets.

Such common ground, says Scherer, may offer shared targets for new drugs, so that even people whose autism has slightly different genetic or environmental causes may benefit from the same therapies. If the biology of the disorder converges on similar pathways in the body, a single treatment or therapy targeted at these pathways could hopefully treat a good proportion of cases.

"We have no good therapeutics for autism and the reason is because we didn't know the pathways involved before, so we couldn't rationally design therapeutics," says Scherer. "Now we have...more targets identified. Some were known before, and some are entirely new, but they all link to known genes in the pathway."

That knowledge comes at a critical time, especially since experts are discovering that autism is a progressive disease, which in many cases can worsen if left untreated. Recent studies suggest that early intervention in infants — even before age 1 — who are suspected of having ASDs could reverse or dampen some of the disorder's more severe symptoms. "Families want to understand what caused their child's autism," says Geraldine Dawson, chief science officer for Autism Speaks and a psychiatrist at University North Carolina Chapel Hill. "They want to have good information about what that means for family planning, for their next child if they are planning one. They also want to be able to detect autism as early as possible because we know that if we provide early intervention in the first years of life, this can have a huge impact on the outcome of the disorder."

Ultimately, Dawson says, genetic testing for the CNVs identified by the consortium may become an important part of diagnostic and treatment decisions. But she notes that as with all genetic testing, the results must be presented in the proper context and with adequate counseling, so parents understand the true risks of autism-related genes. Even genes that are highly associated with autism don't always lead to the disorder; in most cases, these genes confer a vulnerability to autism — but not a certain diagnosis.

Dawson expects that the next step toward standardizing a CNV-based test is to screen for the CNVs in autism patients, and gather data on how effective the genetic markers are in diagnosing ASDs — and, in the future, how useful they may be in predicting new cases. "It's very exciting, but also very important as we move into a new era that we do so thoughtfully, and always with the families in mind," she says.

• Alice Park