Wednesday, 7 September 2011

Biological phenotypes of autism

Heterogeneity is a cornerstone of autism spectrum conditions. I don't like to use the snowflake metaphor because, lets face it, everyone is a snowflake (unique) whether autistic or not as a result of our genes, experiences and environment.

This heterogeneity is however a pretty major stumbling block when it comes to research into autism spectrum conditions and trying to determine anything approaching universal constants outside of the 'what-you-see' diagnostic label. Indeed forget universal constants, how about commonalities between sub-groups of people on the autism spectrum? Blogging now for just over six months, nothing has yet convinced me of anything universal uniting all cases of autism despite some indication of 'similarities' within possible sub-groups.

With all this mind, what would represent a step forward? How about being able to identify 'biological strains' of autism based on a few objective criteria outside of just behaviour? By strains, I don't mean to suggest that autism is a disease or akin to bacteria or anything like that, but rather potentially made up of various biological differences that may be common to more than one person on the spectrum.

Step forward an interesting news piece on the Autism Phenome Project and a presentation delivered very, very recently by Prof. David Amaral, the study lead at the 2011 Asia Pacific Autism Conference (APAC). I should at this point also thank Veronica Rousseau over at the Autism Researchers Link group on LinkedIn for bringing this to my attention.

At the moment there is no paper citation for me or you to look through. Just news and a speaker profile with some information about the talk in question. But the news, if accurate, is potentially quite exciting: two groups of children with autism, one characterised by regression and macrocephaly, one characterised by immune system problems. As Amaral says "The ultimate goal is when a child comes into the clinic, rather than saying you just have autism, to be able to say you have autism type A, or type B, or type C". Whilst this is exciting news for sure, a few parts of the newspaper piece did make me cringe a little. Suggesting for example that there might be different types of autism like there are different types of cancer whilst probably not intended in any negative light, slightly dilutes the message to be transmitted.

The primary implication from this work in progress if accurate and replicated, being that outcome can eventually be plotted against phenotype, so support and services tailored in the areas likely to be required for each phenotype. My obsession with the health comorbidity risks such as diabetes and coeliac disease might also eventually figure in this area. Interesting that at the same conference there is some chatter about the longer-term outcome of autism in what is a growing area of research interest. Intervention-wise, a brave new world is born, as person-specific intervention might come about through phenotype-specific markers.

We wait to see the formal publication of this work and where it takes us next.