Thursday, 5 December 2013

Autism Phenotypes

Hardly a week goes by without somebody mentioning to me a wonder treatment or even “cure” for autism; the latest one being the GAPS diet.

I think all such reports are worthy of investigation, but many lead to nowhere.

Why is this?

·       Medical science has failed to adequately define autism, so we are not all talking about the same autism

·        Many people putting forward theories have not read even the most basic (and not contested) autism research.  Some are even, apparently, qualified “doctors”.
Autism Phenotypes

What is not disputed is that autism has many sub-types (phenotypes). Researchers tell us 10-20% of cases referred to as autism have a known genetic defect (Fragile X, SLOS, Timothy syndrome etc.).  80% do not have a known genetic marker/cause.

Autism can be subdivided into regressive (when a child loses speech and other learnt skills) and non-regressive (early onset).  Even this can be a subjective judgment, since it effectively relies on parents to determine it, after the event.
Then you have cases of autism which clearly have nothing to do with Kanner’s classic version.  In this blog I showed how even cerebral malaria in a child can lead to the onset of autism.  This clearly is a case of brain damage caused by malaria; but to the observer, months later, it would probably be classed as regressive autism or childhood disintegrative disorder.
Testing for Autism
Researchers and doctors keep repeating that there is no test for autism.  This is not strictly true, but it does explain why so many different conditions are all lumped together as “autism”.

In fact, if you read the research closely, you will see that there are many tests for autism; although they may not be perfect.
The only way to know for sure that it is genuine autism is to examine the brain itself.  The only way to do this 100% accurately is via post-mortem analysis of the brain.  Recently, non-invasive methods have been developed to confirm the same findings of brain malformation that occurred prior to birth.

So the kind of autism that relates to tissue held in brain banks is best understood.  But what kind of autism would that be?  Well, it refers mainly to children and young adults who died prematurely.  They died from things like seizures or drowning.  What does that tell us?  This tells us that these people were most likely severely affected by autism.  The mild, social difficulties, type of autism is, fortunately, hardly likely to make it to the brain tissue bank.
If the person interpreting the MRI of a child’s brain knows what to look for, they may very well be able to identify this type of autism.  The expert here is Eric Courchesne.
A similar approach can followed using Electroencephalography (EEG) to identify autism; but it would be smart to cross check this with Eric.

Regressive vs. Early-Onset
Then you have the difference between regressive and non-regressive autism.  Here again, from my Dean’s List of researchers, we look at Paul Ashwood’s research to see that kids with regressive autism have HIGHER levels of inflammatory markers in their blood.  These include cytokines like interleukin 6, which can be inexpensively measured in most laboratories.  This tells us that perhaps regressive autism is an entirely different condition from non-regressive/early onset autism.  As I would expect, increasing cytokine levels were associated with more impaired communication and aberrant behaviors. 
Lab Testing
We have seen earlier in this blog that some very expensive lab tests exist for autism, but their usefulness and integrity is highly disputed.  There are, of course, many hundreds of other tests that are entirely validated by medical science.  Many of these tests are cheap and available all over the world.

Hormonal Screening
We know from the research that about 30% of people with autism have high blood serotonin. A standard lab test is required.
We know that many have high levels of insulin-like growth factor (IGF-1).  A standard lab test is required.
Thyroid hormone levels and in particular a blunted response of TSH to TRH (i.e. central hypothyroidism) can help define further phenotypes.

The TRH test is now not widely used, but TSH, FT3 and FT4 are cheap tests.
Growth Hormone (GH) is also implicated in autism, along with IGF-1; there is a lab test to measure pituitaryfunction to see how well GH is being produced.

By screening for hormonal dysfunction, it would be possible to identify phenotypes that would most likely benefit from therapies targeting those defects, like NNZ-25266.

Pancreatic Dysfunction
It is reported by Joan Fallon, of Curemark, that 50+% of kids diagnosed with “US autism” seem to have a pancreatic dysfunction.  This can be tested for by measuring fecal chymotrypsin level.  The test measures how well your pancreas is working, and is a standard test for people with cystic fibrosis.  Since the US diagnoses far more kids with autism than other countries, it seems highly plausible that “US autism” includes many more phenotypes than, say, “French autism”.

I was quoted about $8 for a chymotrypsin test.

Ion-Channel Diseases (Channelopathies)
Many diseases like Parkinson’s disease, Spinocerebellar Ataxia and Timothy Syndrome are caused by faulty calcium ion-channels.

The Bumetanide autism therapy, undergoing trials in Europe, is based on another channelopathy, this time a faulty chloride transporter NKCC1.
It is clear from reports I have received, that Bumetanide therapy is totally ineffective in some children with ASD, but in other children, like my son, it is effective.
So some types of autism have certain channelopathies and other types have different ones or, quite possibly, none at all.  

My conclusion today is pure conjecture.  I imagine that possibly as few as a quarter of cases of “US autism” are actually “real” autism, that is with all the brain damage/malformation that is identified in those post mortem brain studies and which forms the basis of 90% of autism research.

The other three quarters may be something entirely different, just like the case of the mosquito that bit the child, produced cerebral malaria and then later the full symptoms of autism.  Within the three quarters may be food allergies, digestive enzyme deficiencies, gut disorders, mastocytosis, blood brain barrier defects, undefined calcium ion-channel diseases etc.
This would account for those occasional amazing “recoveries” and the apparent success, in some cases, of diets like GAPS.  Sadly, diet is unlikely to 100% fix brain damage.  If you are lucky enough to totally “recover”, you cannot have had brain damage in the first place.  It is evident that in some phenotypes of autism, diet can reduce autistic behaviours.  This can only be proved in trials, if biomarkers are established for that specific phenotype.
Most likely the only biological thing all these “autisms” have in common is oxidative stress and neuroinflammation; but only a non-medical scientist, like me, can say such a thing.




  1. I want to share a link with you to the ion channel foundation of berne switzerland.. it is my understanding they are looking for collaborators researching ion channels.. it is my understanding also that they are very well funded.. and looking for projects to fund.. I am thinking about sharing my project with them to see if they would like to work on it as I am not very well. This is Andersen Tawil girl.. Some very respectable researchers in ion channelopathies in the group..

  2. Thanks for this.
    I am myself back on the channelopathy theme. They are most likely all inter-related. It appears that a well known sodium channelopathy, affecting autism may indeed be treatable. In mice it works. The channel is Nav1.1, the gene is SCN1A and the drug is called Clonazepan/Rivotril. I know that Andersen-Tawil is a potassium channelopathy. Sodium, potassium and indeed calcium and magnesium are all inter-related. Magnesium can block calcium channels for example. Have you ever tried Verapamil? It is a very common calcium channel blocker that blocks some potassium channels. It will raise K+ levels in the blood, but if you have a potassium channelopathy it will likely have other effects.

    1. Hi Peter,

      I'm fascinated by your blog and the amount of research that you do. I have a question for you regarding the chloride channelopathy. Can a calcium channelopathy cause or trigger a chloride channelopathy?

    2. It is known that a sodium channelopathy can trigger a second sodium channelopathy. Ion channels are interrelated so it is plausible that a calcium channelopathy could cause a chloride channelopathy, but I doubt anyone has proved it. The understanding of this area of science is in its infancy.

  3. Wonderful stuff here. Thanks for going to the trouble to do the writing.


Post a comment