Wednesday, August 11, 2010

Autism Earthquake? The MRI Adult Autism Brain Scan Biomarker

Every day brings reports of new autism studies, most of which are of little or no obvious value, studies which show that a group of high functioning autism subjects are very intelligent, studies which show that some families of autistic children are relatively affluent, studies which purport to define an autistic "smile" and so on. 

News of a study which claims to have established, using MRI bran scans,  a biomarker for autism disorders is different.  An autism biomarker may  have profound and immediate impact on the very definition of autism as "Socrates" of the New Republic has already mentioned  in  Autism Diagnosis by MRI Brain Scan. Socrates gets right to the point while introducing a quote by one of the study authors Declan Murphy of King's College London:

"In an astonishing turn of events, the study due to be published later today, may throw into chaos the process for revising the DSM V definition of Autism.To date Autism has been defined solely on the basis of behaviour - a situation that is set to change radically.

I agree with Socrates. A reliable biomarker may throw the DSM5 autism revision process into chaos.  In my view the drafters of the New Autism Spectrum Disorder in the DSM5  would be wise to stay their autism revision efforts until it can be determined whether this study receives widespread acceptance in the autism research community. If the study results are deemed reliable as indicating biomarker(s) for autism disorders than the revision process should stop until the full implications of the study are determined. There seems to be little point in revising definitions of autism disorders based solely on behavioral characteristics just as reliable biomarkers are being established and defined.

There could be many significant impacts of a reliable autism biomarker.  The study authors speculate that the biomarker will help autistic people and their families by providing quicker, more cost effective diagnosis. I hope they are right.  Private and public treatment and service providers will undoubtedly move to incorporate the use of the biomarkers into their criteria for determining whether to provide treatment or services.  This could actually result in more waiting list delays for the scans which could  simply be added on to existing eligibility requirements for receipt of  services.

Current autism research conclusions including, possibly, epidemiological based vaccine research conducted during periods of changing definitions of autism in the DSM-IIII and IV, may have to be revisited using the new biomarkers. And of course some of the very high functioning persons currently diagnosed with Autism and Aspergers may lose their diagnostic labels ... and their careers as unelected,  self appointed spokespersons for  all "autistics". Debates about the use of functioning labels and differences of autism severity could become obsolete if clear biological/neurological distinctions are identified by MRI Brain Scans.

If  the study results are confirmed and accepted by the autism research community the autism world may have just felt the first preshock of an autism earthquake that will reshape the autism world. 

Information about the new MRI brain scan autism biomarker from the Medical Research Council:

Autism in adults diagnosed by quick, new brain scan

Tuesday 10 August, 2010

Scientists funded by the Medical Research Council (MRC) have  developed  a pioneering new method of diagnosing autism in adults. For the first time, a quick brain scan that takes just 15 minutes can identify adults with autism with over 90% accuracy. The method could lead to the screening for autism spectrum disorders in children in the future.

In the MRC-funded study, scientists at the Institute of Psychiatry (IoP), King’s College London, used an MRI scanner to take pictures of the brain’s grey matter. A separate imaging technique was then used to reconstruct these scans into 3D images that a computer algorithm can assess for structure, shape and thickness – all intricate measurements that reveal Autism Spectrum Disorder (ASD) at its root. Having developed this process, the computer can quickly pinpoint biological markers, rather than personality traits, to assess whether or not a person has ASD.

ASD is a lifelong and disabling condition caused by abnormalities in brain development. It affects about 1% of the UK population (over half a million people), the majority of these being men (4:1 male to female). Until now, diagnosis has mainly relied on personal accounts from friends or relatives close to the patient – a long and drawn-out process hinged on the reliability of this account and requiring a team of experts to interpret the information.

Dr Christine Ecker, a Lecturer in the Department of Forensic and Neurodevelopmental Sciences from the IoP, who carried out the study said:

“The value of this rapid and accurate tool to diagnose ASD is immense. It could help to alleviate the need for the emotional, time consuming and expensive diagnosis process which ASD patients and families currently have to endure. We now look forward to testing if our methods can also help children.”

Professor Declan Murphy, Professor of Psychiatry and Brain Maturation at the IoP, who led the research said:

“We think that our new method will help people with ASD to be diagnosed more quickly and cost effectively. Most importantly their diagnosis will be based on an objective ‘biomarker’ and not simply on the opinion of a clinician which is formed after an interview. Simply being diagnosed means patients can take the next steps to get help and improve their quality of life. People with autism are affected in different ways; some can lead relatively independent lives while others need specialist support or are so severely affected they cannot communicate their feelings and frustrations at all. Clearly the ethical implications of scanning people who may not suspect they have autism needs to be handled carefully and sensitively as this technique becomes part of clinical practice.”

Professor Christopher Kennard, Chair of the MRC’s Neuroscience and Mental Health funding board said:

“Bringing together the knowledge gained from neuroscience in the laboratory and careful clinical and neuropsychological evaluation in the clinic has been key to the success of this new diagnostic tool. In fact, this approach to research is a crucial theme throughout the MRC’s strategy. We know that an investment like this can dramatically affect the quality of life for patients and their families. The more we understand about the biological basis of autism, the better equipped we will be to find new ways of treating those affected in the future.”

The research studied 20 healthy adults, 20 adults with ASD, and 19 adults with ADHD. All participants were males aged between 20 and 68 years. After first being diagnosed by traditional methods (an IQ test, psychiatric interview, physical examination and blood test), scientists used the newly-developed brain scanning technique as a comparison. The brain scan was highly effective in identifying individuals with autism and may therefore provide a rapid diagnostic instrument, using biological signposts, to detect autism in the future.

The research was undertaken using the A.I.M.S. Consortium (Autism Imaging Multicentre Study), which is funded by the MRC. Support funding was also provided by the Wellcome Trust and National Institute for Health Research.

The paper, ‘Describing The Brain In Autism In Five Dimensions - MRI-Assisted Diagnosis Using A Multi-Parameter Classification Approach’ is published in the Journal of Neuroscience on Wednesday 11 August.


M.J. said...

I haven't seen the study yet that describes this technique but, if the article on the topic at the New Scientist is accurate, this isn't a straight test empirical test but rather a piece of computer software making an interpretation of the results of a MRI.

The article talks about how the analysis was done by a computer program and the program had to be "taught" what a brain with autism looked like. If that is the case, then this system is only as good as the samples that were used to teach it and the system might not be good at labeling autism that looked even a little different.

For example, if it was taught using samples from adult males, it might fail at finding the same patterns in an 18 month old. It could also fail in other adults males whose brain structure was just different enough.

So I am not sure that this test represents a biomarker, at least in the classical sense of the word. But still, this is a step in the right direction.

farmwifetwo said...

I asked the Ped once about an MRI for my son. It's well known, or well assumed, that the wiring in an ASD brain is different from that of a "normal" brain. Actually, I was curious to see if it was "autism" or "damage from prenatal care" NOT that that would change our lives any... but I am curious. Brain damaged people also flap, spin, have verbal apraxia etc etc....

They won't do it here. Unlike the USA where these scans are done regularly and autistic children see neurologists as well... Not done here... Surprised... I wasn't. Do to the difficulty of finding decent medical professionals in the early stages of this journey.

So, even if the marker is found, good luck getting the attitudes to change.

But one day... maybe... I'd like my curiousity settled for once.

Autism Mom Rising said...

Thanks Harold. I hadn't heard. I agree they should wait on the revision. This thing is too complex and needs to be broken into subgroups, etiologies better identified.

My son has has multifocal seizures that cover his entire brain, likely had them from 6 months of age. Once they were identified at age 7 and he was given meds & biofeedback within weeks he started asking and answering questions, giving longer sentences, pretend play. People say the seizures are a "comorbid" but if that is so then why would several core Autistic symptoms drastically improve once they were finally identified and treated? Just tying that into my point that they need to geniunely learn more about subgroups.

Sinead said...

Autism Mom Rising This is exactly why I am not listening to my daughters neurologist who thinks she has autism as well as epilepsy. She has epilepsy. If we solve the epilepsy the behaviors will improve. Saying she has autism just muddies the pool for people who really do have autism.

Jennifer said...

I wonder how this relates to less severe neuro disorders. My daughter has significant SPD - would her brain be different on an MRI as well? And what CAUSES the brain difference? Either way, the earlier people get a proper diagnosis, the earlier intervention can begin. If you read what Brain Balance has to say, there is much evidence that the brain can be changed and the communication improved to reduce or eliminate the symptoms regardless of the cause. I love their website's explanation of functional disconnection syndrome. You can read about it at . I also like Hartley Steiner's explanation of what SPD is - whether co-morbid with autism or not...

Socrates said...

"So far, Ecker's team has only looked at men but there are plans to extend the work to women and children.

"We think this approach will work even better with kids because the brain abnormalities you see in autism develop over the life span and they're most prominent during childhood,"

"If we can get up to 90% accuracy in adults, we think it'll be even better in kids."

From here.

I've no doubt this technique, if it's sound, will soon start teasing out the different Autisms, and it'll certainly weed-out the quirky and a bit socially anxious.

blogzilly said...

Dude, I just want to thank you, which I have never done, for finding all these great pieces of information. I'd give you a very manly high five if I ever saw you on the street.


Suzanne said...

This is so very exciting. Thanks for posting! I would love to get my kids tested for this.

Jenn said...

I find this fascinating, and confusing, all at the same time. I stumbled upon your blog when I was busy google-ing "non-specific changes on mri, autism". So happy to have found you! We recently had an mri scan done on my four year old son, because the diagnosis of autism did not cover all of what he experiences. It has taken almost 2 and a half years to get the pediatrician to agree to refer us for this scan, and where we live in Ontario (not unlike the situation in New Brunswick I am told) we now have to wait upwards of 18 months to see a neurologist to interpret the "non-specific" changes found on the scan. This should not so difficult. I feel as though we have been written off in almost every specialists eyes as soon as we mention the provisional diagnosis of autism. Can anyone else concur?