Some autism cases linked to genetic mutation

Up to 1% of autism cases may be blamed on too much or too little of a particular strand of an estimated 25 genes on chromosome 16, according to a pair of recent papers. Understanding how this creates the developmental disorder eventually may lead to improved treatment of autistic children, including those who don't carry these mutations.

"Each little piece of the puzzle gets us closer to the fundamental biological mechanisms, and if we understand the specific biological mechanism, we can consider how we might go about developing intelligent therapeutics for this disorder," said Mark J. Daly, PhD, assistant professor at Massachusetts General Hospital and Harvard Medical School.

One paper, published online Jan. 9 in the New England Journal of Medicine and co-authored by Dr. Daly, found that children with one or three copies of the 16p11.2 genetic segment, rather than the normal two, were at significant risk for autism. The other paper, published online Dec. 21, 2007, in Human Molecular Genetics, implicated having only one copy in this disorder.

These are two of the latest studies shedding some light on autism's genetic underpinnings. About 10% of cases can be explained by various genetic syndromes, such as fragile X or Prader-Willi syndromes, but the rest remain a mystery. Experts predict that no single cause will be found to account for a great number of the cases.

"Based on our experience with other human complex disorders, it's relatively infrequent that specific genes are much more than a percent or two of the whole story," Dr. Daly said.

The challenge now facing researchers is to replicate these findings and determine their clinical utility. Genetic testing for this variation is possible and performed at some institutions. Those who conduct the screening say it can help children receive a confirmed autism diagnosis and improve access to educational services. The results also may suggest increased risk in subsequent children and give comfort to parents.

"Mothers wonder, 'Is there something I did while I was pregnant? What could I have done differently?' This tells them that there's nothing that the parent could have done differently, and that's helpful information just in and of itself," said David Miller, MD, PhD, assistant director of the Genetics Diagnostic Laboratory at Children's Hospital Boston, one of the authors of the New England Journal of Medicine paper. He has been testing patients for this deletion or duplication for the past year.

But while there was praise for this study, there also was caution about genetic testing and how these data might be used.

"This study has meaning and significance, but it's still a tiny, little step forward. This is a very small percentage of kids," said Patricia Manning-Courtney, MD, director of the Kelly O'Leary Center for Pervasive Developmental Disorders at Cincinnati Children's Hospital Medical Center. "I'm not sure how we would apply this information clinically or what my next step would be. Everybody wants a biomarker for autism, but we're already learning that there is not going to be one."

Some experts question how specific these chromosomal abnormalities are to autism. For example, not all of the children with this genetic variation in the New England Journal of Medicine study had a diagnosis of autism. They were identified as having developmental delay or mental retardation. Some who had a diagnosis of autism also may have had some intellectual disability, although this is not clear from the paper.

"It would be interesting to know more details about the affected patients," said Scott Myers, MD, a neurodevelopmental pediatrician at the Janet Weis Children's Hospital in Danville, Pa. "I suspect that the more specific and common phenotypic aspect of 16p11.2 microdeletions or microduplications is intellectual disability. ... If the 16p11.2 abnormalities are more specific for the autism phenotype, there should be a substantial number of affected individuals who have autism spectrum disorders without mental retardation."

Doubt also was expressed because autism is considered an inherited genetic disease, and siblings of affected children tend to have a higher risk of developing it. The studies' authors said the deletion and duplication of this genetic material is something that, for the most part, is not passed down by either parent and may occur during the formation of the egg or sperm, at conception or very early in fetal development.

But "it's a pretty big leap to say [autism] is not inherited," Dr. Manning-Courtney said.

Candidate gene for autism identified

But while this genetic variation may not be inherited, another pair of papers, published online Jan. 10 in the American Journal of Human Genetics, reported a genetic factor that is. A variant of CNTNAP2, a gene that makes a protein allowing brain cells to communicate with each other, increases the risk of developing autism, researchers found. Not everyone with this gene variant develops the disorder, but the risk is higher if the gene is inherited from a child's mother.

"CNTNAP2 is an excellent candidate gene for autism," said Aravinda Chakravarti, PhD, senior author on one of the papers and professor of medicine, pediatrics, molecular biology and genetics at Johns Hopkins University School of Medicine in Baltimore. "It encodes a protein that's known to mediate interactions between brain cells and that appears to enable a crucial aspect of brain-cell development. A gene variant that altered either of these activities could have significant impact."

As the list of genetic causes of autism is growing, so are the data against the hypothesis that the vaccine preservative thimerosal could be involved. Researchers with the immunization branch of the California Dept. of Public Health analyzed data collected by the state's Dept. of Developmental Services. Despite the removal of nearly all thimerosal from childhood vaccines between 1999 and 2001, the rate of autism continued to increase, according to the study, published in the January Archives of General Psychiatry. The authors concluded that these data do not support the theory linking thimerosal with this disorder.