Thursday, 8 March 2012

On the broader autism phenotype

I think most people would be satisfied with the description of autism as a spectral condition epitomised by the phrase 'if you've met one person with autism, you've met one person with autism'. The term 'autism spectrum' denotes the heterogeneity apparent and that the presentation of core symptoms varies intra-diagnosis according to symptom domain as well as inter-person across different individuals. This is independent of the fluctuations in presentation across different situations and not including variables like maturation, co-morbidity and any effect from intervention.

A spectrum also implies other things including degrees of severity between two poles; ranging from very severe 'disability' at one end to 'difference' at the other end (as per this description from Lorna Wing) depending on your definition and how you grade severity. In between you have a mix of ability and disability. I should point out that by using the term 'difference' I am not in anyway suggesting that those at that end of the spectrum are in any way less deserving of the appropriate support and services. High functioning - however you describe this - does not always mean 'can function'.

At the less severe end of the spectrum, a cut-off point is generally accepted to exist in some shape or form; in diagnostic terms denoting presentation which somehow surpasses a clinical threshold and in cold hard psychiatric terms implies psychopathology. The rights and wrongs of where the threshold has been currently set still continues to create lively debate. The presence of that diagnostic threshold however does not necessarily denote the 'end of the autism spectrum' from a real-world behavioural perspective. It implies the end of the diagnostic spectrum but not that the subtle presentation of certain autistic traits cannot be present beyond the clinical spectrum. Enter the broader autism phenotype (BAP) as evidence of this extra-diagnostic spectrum of presentation.

I first heard about the BAP quite early on in my research career. Prof. Ann Le Couteur, one of the early proponents of the BAP discussed some of the ins and outs of the concept and in particular based on studies of siblings of children with autism, how issues like speech and language problems and social interactive issues were picked up from time to time which did not quite fulfil autism diagnostic criteria either in measure or severity. I also remember quite a few people talking about the overlap with certain personality traits such as introversion and more recently some work looking at autistic traits in conditions such as feeding disorders.

Much of the early work on BAP was tied into the genetic basis of autism based on twin studies, the strength of which has recently come under scrutiny. With this in mind, a recent paper by Davidson and colleagues* caught my eye with their suggestion ".. that BAP traits occur at low rates in simplex families". The results reported in this recent paper were based on an examination of the Simons Simplex Collection which as the name suggests contains details of samples from around 2700 families with one child diagnosed with an autism spectrum condition. Davidson et al looked at over 1500 of these families with the purpose of trying to further elucidate what the BAP is and how it could be appropriately tested for and measured.

I'm intrigued about the potential for differences in BAP traits according to whether a family has one child with autism or more than one. The obvious issue with this study is its exclusive analysis of simplex families, so nothing to compare against in terms of multiplex families outside of external datasets. That and the reliance on a snapshot of where a family is in terms of autism not necessarily ruling out any elevated risk of autism recurrence in simplex families should for example they have other children. This last point is perhaps a fundamental flaw in any simplex analysis.

Putting these issues aside, the first thing that comes to mind is whether the weighting of genetics vs. environment might be different in simplex families where BAP traits are infrequent. In other words, is there a suggestion that environmental factors might trump genetic factors in such cases compared to other families where BAP traits are more frequent in other family members? It is perhaps not as easy to say one is genetic and the other is environment because such simple arguments have not been borne out by the research data and most (if not all) conditions are likely to be the result of an interplay between nature and nurture. Think epigenetics for example.

If the data from Davidson is accurate however, this might provide a good opportunity to look at factors such as regression, comorbidity and early adverse events to determine any difference in individuals with autism among high and low BAP trait families. At the very least if offers another potential phenotypic distinction which could be added to those already being looked at.

* Davidson J. et al. Expression of the Broad Autism Phenotype in Simplex Autism Families from the Simons Simplex Collection. JADD. March 2012.
DOI: 10.1007/s10803-012-1492-1


  1. Unaffected siblings have been found to have similar results to their ASD diagnosed siblings in comparison to typically developing controls. As reported by Dalton et al 2007: 'The siblings gaze fixations and brain activation patterns during the face processing task were similar to that of the autism group and showed decreased gaze fixation along with diminished fusiform activation compared with the control group. Furthermore, amygdala volume in the siblings was similar to the autism group and was significantly reduced compared with the control group'.

    These findings suggest that what is inherited in these families is not autism as such but rather the broader autism phenotype. The prevelance of severe to moderate BAP features in general population children is about 10% suggesting that the BAP is far too frequently present in the general population to be considered as a causal mechanism rather than a risk factor of small effect. Happe, Arnold and Plomin 2006 on the prevelance of the BAP in general population child "Around 10% of all children showed only social impairment, only communicative difficulties or only rigid and repetitive interests and behavior, and these problems appeared to be at a level of severity comparable to that found in children with diagnosed ASD in our sample'.

    Dalton et al 2007.Gaze-fixation, brain activation, and amygdala volume in unaffected siblings of individuals with autism.

    Happe, Arnold & Plomin 2006. Time to give up on a single explanation for autism.

  2. As far as the twin studies are concerned, twin studies were introduced by Sir Francis Galton more than one hundred years ago long before the discovery of DNA. Researchers continue to believe that you can calculate a heritability estimate based on the assumption that the genetic effect seen in MZ twins are always inherited. Based on early European twin studies there isn't a research paper that does not reference these early twin studies and state that autism is the most heritable of the developmental disorders based on the concordance rate in MZ twins compared to DZ twins. Not so. The concordance rates seen in MZ twins with Downs Syndrome is nearly 100% and the concordance rates in DZ twins are nearly 0% which paradoxically would make Downs Syndrome, not autism the most 'heritable' of all the developmental disorders.
    The 1995 British autism twin study, the most referenced twin study ever published excluded twin pairs where one or both twins had an identifiable medical cause. Three twin pairs were excluded, one pair each with Down's Syndrome, Williams Syndrome and congenital rubella syndrome. None of those syndromes are inherited.
    The classical twin studies does not control for de novo gene mutations and will always overstate the 'heritability' of autism.

  3. Thanks RAJ.

    I agree that the 1995 twin study does perhaps seem a little 'dated' in light of the subsequent research on the complexity of the genetics of autism and the introduction of epigenetics on to the scene.

    I'm glad you referenced Fransesca Happe's paper and the important conclusion that they (and everyone else) have come to on autism heterogeneity and how using behavioural 'markers' as a starting point really hasn't cut the mustard in terms of elucidation of underlying biological processes.

  4. How the broader autism phenotype may be a background genetic effect has been demonstated in Downs Syndrome with or without co-occurring autism. Ghaziuddin compared a group of Downs Syndrome children with or without co-occurring autism. In Downs Syndrome with co-occurring autism there was an excess of first degree relatives who met the description of BAP features compared to first degree relatives in children with Downs Syndrome without co-occurring autism who did not meet the description of BAP features. None of the first degree relatives in Downs Syndrome with co-occurring autism were diagnosed either with Downs Syndrome or autism ( Ghaziuddin M ) (Ghaziuddin1997 ).