Autism's unexpected link

Gina Kolata / New York Times News Service /

Published Aug 14, 2013 at 05:00AM

Researchers studying two seemingly unrelated conditions — autism and cancer — have unexpectedly converged on a surprising discovery. Some people with autism have mutated cancer or tumor genes that apparently caused their brain disorder.

Ten percent of children with mutations in a gene called PTEN, which causes cancers of the breast, colon, thyroid and other organs, have autism. So do about half of children with gene mutations that can lead to some kinds of brain and kidney cancer and large tumors in several organs, including the brain. That is many times the rate of autism in the general population.

“It’s eerie,” Evan Eichler, a professor of genome science at the University of Washington, said about the convergence.

He and others caution that the findings apply to only a small proportion of people with autism; in most cases, the cause remains a mystery. And as with nearly all genetic disorders, not everyone with the mutations develops autism or cancer, or other disorders associated with the genes, like epilepsy, enlarged brains and benign brain tumors.

But researchers say the findings are intriguing, given that there are no animals that naturally get autism, no way of analyzing what might cause autism in developing brains and no cure. The newly discovered link has enabled scientists to genetically engineer mice with many symptoms of the human disorder.

And it has led to the first clinical trial of a treatment for children with autism, using the drug that treats tumors that share the same genetic basis.

‘Honest-to-God’ diagnosis

Richard Ewing, of Nashville, Tenn., a 10-year-old who has a form of autism caused by a tumor-causing gene, is among those in the new study. His parents, Alexandra and Rick Ewing, know he is at risk for tumors in the brain, heart, kidney, skin and eyes. But that bad news was tempered by his eligibility for the clinical trial, which has only just started.

“There is a big difference between us and the rest of the autism community,” Rick Ewing said. “We have an honest-to-God genetic diagnosis.”

Not everyone agrees that the discovery is so promising. Steven McCarroll, a geneticist at Harvard, notes that autistic children with the cancer gene mutation have “a brain that is failing in many ways.” Autism in these children could be a manifestation of a general brain malfunction, he said, adding, “The fact that autism is one of the many neurological problems that arise in these patients doesn’t necessarily tell us anything penetrating about the social and language deficits that are specific to autism.”

But other scientists who are not involved in the research that produced these findings say the work is changing their understanding of autism and why it develops. Like cancer, autism can involve unregulated growth of cells, in this case neurons in the brain.

Jonathan Sebat, chief of the Center for Molecular Genomics of Neuropsychiatric Diseases at the University of California, San Diego, describes the parallels between cancer and autism as “quite uncanny.”

“We haven’t solved it all; we have only solved a tiny bit,” he added. “But the small bit we solved has been very illuminating.”

One discovery

It was Dr. Charis Eng, a cancer geneticist at the Cleveland Clinic, who first noticed a surprising incidence of autism in children whose parents had the PTEN mutation (pronounced p-10). Eventually, investigators discovered that the rate of autism was 10 percent, about 10 times what would normally be expected.

At the same time, researchers found that another genetic disorder was even more likely to result in autism. That disorder, tuberous sclerosis, increases the risk for kidney cancer and a type of brain cancer; half of tuberous sclerosis patients had autism.

Although PTEN and tuberous sclerosis genes are not the same, they are part of the same network of genes that put a brake on cell growth. Disabling PTEN or one of the tuberous sclerosis genes releases that brake. One result can be cancer or tumors. Another can be abnormal wiring of nerve fibers in the brain and autism.

Dr. Mustafa Sahin of Boston Children’s Hospital decided to test whether drugs used to treat tumors caused by tuberous sclerosis gene mutations might also treat autism in people with the same mutated genes.

He started with mice, deleting tuberous sclerosis genes in their cerebellums. Nerve fibers in the animals’ brains grew wildly, and the mice had unusual behaviors, reminiscent of autism. They had repetitive movements and groomed themselves constantly, so much that they sometimes rubbed their skin raw. And unlike normal mice, which prefer other mice to an inanimate object, these mice liked a plastic cup just as much.

But rapamycin, which targets the tuberous sclerosis gene and blocks a protein involved in cell division, changed the animals. They no longer compulsively groomed themselves, and they no longer liked the plastic cup as much as a live mouse. The animals did better on tests of learning and memory, and the growth of nerve fibers in their brains was controlled.

Now Sahin is giving a similar drug, everolimus, to autistic children with a tuberous sclerosis gene mutation, asking if it can improve their mental abilities. Richard is among the children. Each child takes the drug or a placebo for six months. The study is scheduled to be completed by December 2014.

Another discovery

While Eng started with cancer gene mutations and discovered a link to autism, Eichler, of the University of Washington, started with autism and found a connection to cancer genes.

He focused on what he calls “out-of-the-blue autism,” which occurs with no family history, recruiting 209 families with autistic children.

He saw a striking genetic difference. Compared with their parents and normal siblings, the autistic children had two to three times as many mutations that disabled a gene. The mutated genes were often part of a pathway that controls cell growth. At first, the researchers thought the pathway was ubiquitous, and its link to autism was murky.

“We were a bit bummed,” Eichler said. “Then I said: ‘Wait, some of those genes are cancer genes.’”

But he does not yet know whether these children with autism are also at risk for cancer.

“It’s obviously a significant issue,” Eichler said. “But we need to let the science nail it first.”

The Ewings, whose son is in the autism clinical trial, have learned to live with the tumor threat. For now, their biggest problems are dealing with Richard’s autism.

When Richard’s parents heard about Sahin’s study, they immediately signed him up, though it meant traveling to Boston from Nashville nine times in six months.

They had not dared to take their son on planes before, worried that he could not handle crowded airports.

But the study was too important to pass up, Rick Ewing said.

“Traveling with a kid who can’t talk, who has food issues, who is not patient: We hadn’t really done these things,” Ewing said.

They hope the drug will make a difference.

“We always thought Richard has a lot going on in his brain,” Alexandra Ewing said. “We feel there is a lot of untapped potential.”

For Andrew and Lucy Dabinett’s 9-year-old son, Tommy, whose autism is caused by a PTEN gene mutation, there are no clinical trials as of yet.

Tommy, who lives with his family in Rye, N.Y., has a limited vocabulary, flaps his arms, rocks back and forth, and needs diapers.

When he was 3, a doctor told his parents that he had a PTEN mutation and that in addition to autism, he had a high risk of cancer.

“Of course it is terrifying,” Lucy Dabinett said. “But I already knew there was something terribly wrong with my child. I just needed an answer.”

“Honestly,” she said, “it was a relief to have an answer.”

As surprised researchers have discovered, some people with autism have mutated cancer or tumor genes that apparently caused their brain disorder.