Genetics and stem cells

Gene mutation linked to distinct type of autism

“Have scientists found the autism gene?" asks the Mail Online.

The news is based on a genetic study that found children with autism spectrum disorder (ASD) were more likely to have a mutation in a gene called CHD8 than children without the disorder.

ASD is an umbrella term for conditions affecting social interaction, communication, interests and behaviour.

However, talk of a single autism gene is premature. This is relatively early stage research. The genetic test will need to be further tested and validated in large and diverse groups to ensure it accurately identifies people with ASD.

The results linking the CHD8 mutation to a specific sub-type of ASD with a similar set of symptoms also made the news but was based on just 15 people. This means this sub-analysis cannot be viewed as reliable.

The hope is that if a test does reliably predict what set of symptoms a child with ASD is likely to develop, an individual treatment and care plan can be devised to meet their future needs.

Where did the story come from?

The study was carried out by a large international collaboration of researchers from medical and academic institutions and was funded by a variety of non-commercial institutions including Simons Foundation Autism Research Initiative, the National Institutes for Health (US) and the European Commission.

The study was published in the peer-reviewed science journal Cell.

As with many Mail Online health stories the headline was misleading. There is no such thing as a single autism gene, in the same way that there is no such thing a single type of autistic disorder.

The actual reporting of the study in the main body of the story was accurate and informative. (Mail Online journalists may want to have a quiet word with their headline writers).

What kind of research was this?

This was a genetic study looking for gene variants associated with autism spectrum disorder.

Autism spectrum disorder (ASD) is an umbrella term for a number of conditions that affect social interaction, communication, interests and behaviour. Symptoms can range from mild to severe.

In some cases ASD can lead to below average intelligence and learning difficulties. In other cases intelligence is unaffected or above average.

The researchers describe how different subtypes of ASD have been described based on behaviour but with limited success because behaviour varies so much within and between subtypes. As opposed to behaviour, the researchers thought genes may hold the key to the different subtypes of ASD and sought to investigate this hypothesis.

What did the research involve?

Researchers sequenced the DNA of 3,730 children with development delay or ASD looking for variations in a gene called CHD8 – a gene previously associated with ASD. They looked to see if any genetic variations were associated with being diagnosed with ASD overall, but also for any links to specific characteristics of subsets of people with ASD such as distinct faces, a larger than average head (marcoencephaly), and stomach or digestive complaints.

While the genetic study recruited a lot of children, the assessment of specific characteristics was based on just 15 people.

This sort of genetic study is often used to further scientific understanding of any genetic origins of a disease. Scientists know what many genes do, so if there is a genetic link it helps them understand the biology of the disease process, this can ultimately lead to ideas for new drugs and treatments.

What were the basic results?

The genetic analysis revealed 15 different and independent genetic variations (mutations) in the CHD8 gene in the children with development delay or ASD.

Thirteen of the 15 CHD8 mutations associated with ASD caused the resulting CHD8 protein to shorten (truncate). These truncating mutations were not found in 8,792 healthy controls – including 2,289 unaffected siblings – suggesting the mutations were relatively specific to the disease.

The presence of one of the genetic mutations identified was linked to an increased likelihood of an ASD diagnosis overall. 

Specific CHD8 mutations were also associated with distinct characteristics, but only using 15 people with ASD so should be taken with a pinch of salt. These were:

  • increased head size accompanied by rapid early postnatal growth
  • a facial phenotype marked by prominent forehead
  • wide-set eyes
  • pointed chin
  • increased rates of digestive complaints
  • sleep dysfunction

As a follow-up study, the researchers disrupted the CHD8 gene in zebra fish – a popular research animal for genetic manipulation as they breed very quickly and adapt well to being housed in aquariums. They found the effects of this were broadly similar to ASD type symptoms in humans including larger head size as results of expansion of the forebrain and midbrain, and impairment of gastrointestinal motility.

How did the researchers interpret the results?

The researchers concluded their findings “suggest that CHD8 disruptions represent a specific pathway in the development of ASD and define a distinct ASD subtype”.


This study used many thousands of children to identified mutations in the CHD8 gene that were associated with the development of ASD. Much less convincing was the finding that CHD8 mutations might be associated with distinct characteristics within the spectrum of disease, potentially representing a genetically defined subtype, as this was based on just 15 people.

Identifying subtypes of ASD is important to guide diagnosis, prognosis and disease management to maximise quality of life. This study is interesting because most categorisation of ASD has focused on behavioural dimensions. By contrast, this study flipped this on its head and looked at the genetic dimension of the disease, which revealed some statistically significant links.

Knowing this gene-disease link will help researchers understand the biological origins of the disease, which improves the chance of treatments being discovered or better ways to manage symptoms of the conditions to improve quality of life.

This is relatively early stage research; the genetic test will need to be further tested and validated in large and diverse groups to ensure it accurately identifies people with ASD and or any subtypes.

Today this test does not help people with ASD, but it does help pave the way for potential improvements in the future.

NHS Attribution