Subscribe to A Grey Matter or catch up on QBI talks & lectures


What is autism exactly, and what causes it? We talk to international autism expert Professor Elliott Sherr, from Neurology and Pediatrics at the Institute of Human Genetics, University of California, San Francisco to find out more.



Donna Lu:  Hi! You're listening to a Grey Matter, the Queensland Brain Institute's neuroscience podcast. This week we're talking autism. What it is, and what do we know causes it? We sat down with Professor Elliott Sherr, an international autism expert and a professor of neurology, paediatrics and genetics at the University of California, San Francisco. 

Prof. Sherr:  Well, I mean I'll start out by saying that it's not a simple issue and that there are a lot of people much smarter than I am, who still have disagreements about exactly what the diagnosis says, but I will say that autism is a behavioral diagnosis. Though the word by itself does not link it directly to a specific cause or a specific group of causes but rather it's behavior that you see in children and it's you know usually young children where you can see this most readily.  And it's behavior that addresses their social cognition or their social adaptability that being sort of one main area and we see deficits in that and that also includes language; and then there is another domain of self-stimulatory behavior or difficulties with sensory integration and these sort of two general categories make up autism.

Donna Lu:  Can you give some examples of what children with autism may do? 

Prof. Sherr:  Sure. So, to go back to the issue of language, a lot of kids with autism won't have any language production. No words at all or some of the higher functioning kids with autism might have language but might not be able to understand the subtleties that we call pragmatics.  Like the inflection in your voice or the double meaning in words and they have a hard time understanding that.  Another example would be kids to need to engage in self- stimulatory activity, so kids that for instance hand flap? That's a common sign of children with autism or they'll make vocalisations that are probably satisfying to them but can be quite disruptive you know, depending on where they are.

Donna Lu:  In terms of recognising the signs and symptoms, is there an average age of onset or when these symptoms become apparent?

Prof. Sherr:  Yeah, that's a great question. You know if indeed we think that autism or at least some of autism is caused by genetic perturbations then those genes are there or you know different from the very get go, so arguably somebody might have a susceptibility, a genetic susceptibility for autism from before birth right so from the time of conception.  Often times people aren't able to see the signs of it until children are a little bit older. One possibility is that's because there's actually changes in brain function that show the autism. The other possibility that I would favor is that our ability as clinicians and scientists to observe the differences is much harder in the little kids who aren't doing quite as much.

Donna Lu:  In recent decades, there's been quite a sharp uptake in the incidence of autism diagnosis. Is that because we're diagnosing it more accurately or would you argue that actually, it might be over diagnosed?

Prof. Sherr:  Well, I actually think that both are probably at play. There's a colleague of mine who is an epidemiologist in autism and she showed that over a period of about ten to fifteen years, there was a steady increase in the diagnosis of autism in the state of California, but at the same time there was an equal decrease in the rate of intellectual disability suggesting that some of it was just a matter of definition.  But I also think that we live in a society where there are limited resources for kids with developmental disabilities.  And kids with autism often have an easier shot at getting those resources and so lots of clinicians will think about you know, identifying their kid, their patients as autistic particularly if they're right on the borderline. Because they know they really need those services and can benefit from them. 

Donna Lu:  You mentioned borderline in terms of symptoms. Where does something like Asperger’s syndrome fit into this?

Prof. Sherr:  You know, that's a that's a great question. So, just so you know there's something called the DSM, the Diagnostic and Statistical Manual, so this is kind of like the you know, the "Bible" for the psychiatry community to come up with the behavioral criteria for a diagnosis. In DSM IV, which was the version one back of where we are now, there was an Asperger's diagnosis and that referred to individuals who had good language skills and had good intelligence, normal or above intelligence, but who still had the social or the self-stimulatory behaviors that you find typical in individuals with autism as opposed to other individuals with autism diagnosis who wouldn't have that requirement for high IQ.  In the current version, the team of clinicians and investigators who came up with the consensus rolled those two in together and so, there is technically no longer an Asperger's diagnosis which a lot of people find confusin,  because how could a diagnosis just disappear? But the team who make these designations decided to roll them up into one. So, it's not like these people are gone but they're just they're part of a larger group.

Donna Lu:  So, does that mean that under the new definition or diagnosis, if you like, of autism, there's a spectrum from mild to more severe?

Prof. Sherr:  Yes absolutely and that's actually designated in the criteria that there is mild, moderate and severe. 

Donna Lu:  I want to jump back to diagnosis. While the rise in incidence might be a matter of misdiagnosis or over diagnosis what we do know is that boys are diagnosed at a far higher rate than girls. Do we know why that is?

Prof. Sherr:  I think there are a lot of theories I don't think that there's strong enough evidence yet to really pin it down.  I think that as people get better at noticing the traits of individuals with autism that the differential between boys and girls in terms of the frequency, has actually lessened; and so the frequency is getting closer and closer to one to one; it's not there but it's…

Donna Lu:  I believe at the moment it's something around four to one or something?

Prof. Sherr:  Yeah. Depending on who you talk to, it’s three to one or slightly less than that. 

Donna Lu:  Ok.

Prof. Sherr:  And it used to be higher, right? So, I think some of it is that girls with autism don't necessarily display the same overt behaviors. So, they may not get picked up. I think that that's one possibility. Another one is that, girls generally are more socially adept than boys; and so, if you start with a higher skill level and your autism brings it down a bit, the girls are more likely to escape a diagnosis.

Donna Lu:  Autism is such a complex condition but can you talk generally about the risk factors that are associated with it?

Prof. Sherr:  I mean it's actually really complicated. I think that lots of different things can lead to an autism diagnosis so there is evidence that disruption in the way that nerve cells communicate with each other. So, there are structures in the brain called synapses which are basically chemical communications between two nerve cells that the proteins that do that communication as well as, regulate that…that they're disrupted. There are other kinds of mutations that can result in autism, where their early patterning in brain development is disrupted. So, for example, kids with autism can have larger heads than kids without autism. There's a subset where they have much larger heads and those individuals probably have a disruption in patterning in the brain and that influences kids to have autism.

Donna Lu:  In terms of environmental factors or genetics in say the parents or the mother, what do we know about increasing the risk of getting autism?

Prof. Sherr:  Right. So, that's a great question. I think the most striking evidence so far has been the genetic contribution and I think genetics comes in two general ways of thinking about it: One is that it's inherited from mom or dad or a combination thereof.  And so, there's definitely evidence of that and then the other possibility is that it's under the genetics category that children will have new genetic mutations that are just in them that aren't inherited from mom and dad and those kids are at risk for developing autism. There is evidence unfortunately, that the older the parents are – both moms and dads – that there is an increased risk for the child to develop autism to have one of these genetic reasons. 

Donna Lu:  So, other than having children younger, is there anything you can do to prevent or decrease the risk?

Prof. Sherr:  Yeah, that's a great question. I don't think we know enough yet to say for sure. Certainly, staying healthy and you know taking a daily multi-vitamin.  Particularly you know women who need extra follate during pregnancy should continue to do that. 

Donna Lu:  Another vitamin that QBI research is studying in relation to autism is vitamin D.  Professor John McGrath explains a recent study of pregnant women.

Prof. McGrath:  That was a very productive collaboration with my colleagues in Rotterdam. So, they have a very large birth cohorts and some of the listeners may know that there are studies called “birth cohorts” where they follow everyone who was born in a certain year and tracked them. They’ve done that in Rotterdam for about eight thousand children. They tested amount of blood when the mother was twenty weeks pregnant and they tested the blood when the baby was born. They sent that blood to Brisbane and we had a hypothesis  that maybe if the mother has low vitamin D and the baby had low vitamin D in early life, that may increase the risk of autistic-related behaviors. So, we had funding from the National Health and Medical Research Council and my colleague, Darryl Eyles set up to test Vitamin D and, lo and behold, our hypothesis was supported. So, we found that in this case those with low Vitamin D – their offspring at age 6 have slightly more autistic-like behaviors. In that study, we didn’t actually measure clinical autism, we measured social communication and as you know, many people with autism are not very good at picking up the needs of the listener and may not be so good with social responsiveness.  

But what was interesting is a very solid finding. We looked at it in many different ways: we looked at it in particular subgroups, ethnic subgroups, we looked at adjusted for genetics, we looked at for adjusted for various thresholds and the finding was very convincing. So, we thought this is interesting; it suggests that it's true. Now, we need more clinical trials and we need more evidence that raises the…we can start to think the unthinkable, that maybe if we can identify women who have low vitamin D during their pregnancy, treat them with a safe simple cheap supplement, we could prevent or lower the risk of their offspring having autism later on. That’s why I think that was such an interesting study to do. 

Donna Lu:  In some cases, for their own health, mothers have to take medications while pregnant they can increase the risk of autism. Professor Sherr comments:

Prof. Sherr:  There are certain epilepsy prevention drugs, anticonvulsants that we know can increase the risk of developing autism, so I would say if you're on a medication and you're planning on having a child to go speak with your doctor to make sure that that's discussed.

Donna Lu:  A few decades ago now a researcher published an article in the Lancet linking autism to the measles mumps and rubella vaccine. We know now that that study was falsified, has been widely discredited, and yet in the community there is still a lingering link in some people's minds between that vaccination and autism. What do we know now in terms of the research?

Prof. Sherr:  Right, so that's a great question. First of all, let me say that I understand why parents are anxious and why they might wrongly but might still link the two together and that's because the MMR vaccine that occurs around fifteen to eighteen months is the time at which autism often becomes the most evident to the families of these young kids and so they see the autism bloom display itself, and they know that the child just got vaccinated and sometimes even when they get vaccinated they'll have a fever or have some other sort of a reaction and so parents want to link the two together. I think that thankfully there are enough experiments that have been done now, and by experiments I mean observations of the community. So, the best studies have been done in Europe where they removed thimiresol, which is a preservative that people thought was the cause of the autism risk in kids who were having a MMR vaccine and they were able to look over time, both with the thimiresol in the vaccine and with it out, and show that if anything the autism rate went up a little bit when the thimiresol was removed – and that was probably more because you know people are diagnosing it more readily. So, there’s really no evidence that the thimiresol, which was the presumed culprit, causes autism. 

Donna Lu:  Are there any other points that you'd like to emphasise to the general public when it comes to understanding autism? 

Prof. Sherr:  So, one thing is that – I just came from a family meeting where a lot of the kids have autism – I would say that kids with autism, a lot of them, have a lot more potential to be happy and to be productive in society than I think we initially give them credit for because they can struggle early on both in the way they interact with their peers, as well as the way they do things like self-soothe, but that with enough attention to their needs a lot of these symptoms can be lessened and their lives can be substantially enhanced because of that. 

Donna Lu:  That's all for this episode. I'm Donna Lu; our podcast is produced by Jessica McGaw. If you enjoyed this episode, let us know on Twitter or Facebook or you can give us a review on iTunes. Thanks for listening.