Advertisement

Test Identifies Newborns Likely to Develop Autism

Share via
TIMES MEDICAL WRITER

California researchers have developed the first test that can identify newborns who are likely to develop autism or mental retardation, providing powerful new clues to the causes of the devastating developmental disorders.

The discovery confirms that a predisposition for autism--once thought to be the result of poor parenting--is present at birth. It also suggests that physicians may eventually be able to screen newborns for the disorder.

A team from the California Birth Defects Monitoring Program examined stored blood samples collected from 249 infants during the 1980s. The researchers found unusually high levels of four proteins associated with brain development in nearly all the samples from children who later were diagnosed with autism or mental retardation, but in none from infants who developed normally.

Advertisement

The findings, reported Wednesday at a San Diego meeting of the American Academy of Neurology, suggest that autism probably arises from a combination of genetic defects and exposure to toxic chemicals, viruses or other environmental substances. It also indicates that the crucial period for such factors is during the early weeks of pregnancy, when the central nervous system is forming.

“The bottom line is that they [the infants] had this at birth, and that is very significant,” said Dr. Isabelle Rapin of Albert Einstein Medical Center in New York City. “It pulls you in a direction that autism has not been pulled before.”

Autism is a severe developmental disorder in which children seem isolated from the world around them. It is marked by poor language skills and an inability to handle social relations. No cure exists, but many problems can be alleviated with intensive behavioral therapy.

Advertisement

The discoveries do not rule out the possibility that the onset of symptoms can be triggered by vaccinations--an idea that has recently gained great currency among parents of autistic children. But they do indicate that other factors play a much greater role.

“I don’t use the word ‘breakthrough’ lightly, but that is what this looks like to me,” said Dr. John Harris, chief of the Birth Defects Monitoring Program. “This could potentially have a huge impact on society.”

Experts cautioned, however, that the results have not yet been formally written up and reviewed by other scientists. And even if the research is deemed acceptable, the findings must be replicated before they will be widely accepted.

Advertisement

The monitoring program is already gearing up to study as many as 5,000 autistic children--and several thousand healthy children--to validate the finding, said Dr. Judith K. Grether, the principal investigator. But such a study may take five years, she said.

“This really is an exciting finding,” but it doesn’t mean scientists have a specific test for autism, since mentally retarded children had the same elevated protein levels, added Dr. David Amaral, director of the MIND Institute at UC Davis, which provided part of the funding for the study. “But this may be an early warning signal that there will eventually be a developmental disorder” and physicians can use other kinds of testing to determine which one.

“It may even be possible to intervene and prevent it, just like we do now for phenylketonuria,” he added. Phenylketonuria, a genetic defect that can be detected at birth, produces mental impairment if the infants are given certain foods. Identifying it allows parents to avoid those foods.

Similarly, the hope is that detection of a predisposition toward autism at birth might allow therapy to forestall its development.

The conventional wisdom is that about one in every 1,000 children suffers from autism, but several recent studies have suggested that the incidence is much higher.

A report by the Centers for Disease Control and Prevention released last month found an incidence of 67 cases per 1,000 in Brick, N.J., where parents had reported an autism “cluster.” Studies in other communities suggest rates nearly as high, however, and autism activists argue that the nation is in the midst of an epidemic.

Advertisement

Genetics clearly play a role, although no one has identified a specific gene link. Environmental toxins may also be important, but the CDC could find none in Brick.

Whether infants are born autistic or develop the disorder later has been hotly debated. Studies have pointed to a genetic component, but symptoms don’t normally develop until age 1 or 2.

Many parents argue that vaccinations play a major role because many children develop the first signs of autism shortly after immunization at 18 months, particularly with the MMR (mumps, measles and rubella) vaccine.

But no one has developed a convincing biochemical explanation of how the disease originates. Experts hope the new findings may provide crucial guideposts pointing toward new areas of biological research into autism’s origins.

“This opens the window to learning the role of some of these triggering events,” said Dr. Bernard Rimland of the Autism Research Institute in San Diego. In particular, Grether noted, the genes that serve as the blueprints for the four proteins would be a good starting point in searching for a genetic origin.

The new study could probably have been done only in California, Grether said. Blood samples are taken from every newborn in the state--which accounts for one in every seven infants born in the United States--to test for phenylketonuria and two other genetic defects, thalassemia and thyroid disorders. Other states discard the samples, but California retains them and now has a library of millions of blood samples on file.

Advertisement

Grether and her colleagues, working with Dr. Karin Nelson of the National Institute of Neurological Diseases and Stroke, collected the samples for 64 infants who developed autism, 66 who developed mental retardation, 65 who developed cerebral palsy (which is also characterized by mental impairment), and 54 healthy babies.

After consulting various experts on neural development, they identified eight target proteins that play a role in development of the brain. Researchers at George Washington University used highly sensitive microchemical techniques to screen each sample for the presence of the eight proteins.

Nelson reported Wednesday at the meeting that four of the proteins--vasoactive intestinal peptide, calcitonin-related gene peptide, brain-derived neurotrophic factor and neurotrophin 4--were present at elevated levels in 96.9% of the autistic children and 92.4% of those with mental retardation. In contrast, they were elevated in only 9.2% of the children with cerebral palsy and in none of the healthy control subjects.

“For both mental retardation and autism, this is the first time anyone has identified a clear biological marker at such an early age,” said Walter Herlihy, president of Repligen Corp., which is testing a hormone called secretin as a potential treatment for autism. “I’m going to go back and incorporate this in our clinical trials to see if the markers are present in older children as well.”

Such clear-cut results “startled” the researchers, Harris said, but their meaning is not yet clear.

“What we don’t know is whether what we have measured is part of the causal pathway or a byproduct of something going wrong that we haven’t measured directly,” Grether said. Either way, “if a lab procedure could be developed that would make testing easy, cheap and fast, it has potential to be used for routine newborn screening.”

Advertisement

Virtually all researchers agree that early detection and intervention offer the greatest hope.

Early detection “would also give the parents more time to adjust,” added Herlihy, who has two autistic daughters.

Advertisement