Thursday, 13 December 2012

Cord blood testosterone and autistic-like traits: no link?

Prof. Andrew Whitehouse has featured quite a bit on this blog. The most recent occasion (aside from this post) was a piece of research on maternal vitamin D status and early adult offspring scores on the Autism Spectrum Quotient (AQ) which concluded no overall link despite some interesting details. Indeed quite a bit of his work derived from the Raine study focuses on recording factors during pregnancy and the very earliest days in the big, wide world and seeing how they might relate to results like those derived from the AQ some years later. Not wholly dissimilar from the ethos behind looking at archive dried blood spots it has to be said.
Cord clamping @ Wikipedia  

In a similar vein, I want to talk about another of his papers* (open-access), this time on perinatal testosterone exposure and once again, early adult offspring AQ scores. And in particular a lack of association between the two variables and some potentially important implications for one of the more widely cited theories of how autism might come about.

I've kinda covered testosterone and autism before on this blog with reference to the finger length ratio (2D:4D), apparently quite an interesting correlate of how much testosterone we might have been swimming around in in-utero, and also more speculatively on a post on PCOS.

Most people thinking testosterone and autism are brought back to the theory posited by Prof. Simon Baron-Cohen - see one of the papers here** - extending the so-called extreme male brain (EMB) theory of autism*** itself extending the systemising/empathising sex differences**** again extending good old Theory of Mind (ToM). It's quite a logical train of thought I must admit, and probably why so many people truly do believe testosterone may show some primary connection to cases.

That being said, regular readers perhaps know that I'm not really one for sweeping generalisations when it comes to autism (see pages 949-951), no matter how much the science of psychology loves to try and compartmentalise conditions like autism. And take my word for it, psychology has been pretty fanatical about compartmentalising autism, its core cognitive features, down the years, almost it seems searching for some kind of scientific closure.

Anyhow, the paper in question is open-access but here are a few bullet points:

  • The study is part of the Raine initiative again, whereby out of 861 children where BioT (free testosterone & albumin-bound testosterone) taken from umbilical cord blood at delivery was available, 707 provided diagnostic data at any of the specified follow-up points through infancy and early adulthood.
  • Testosterone analysis was made by mass spectrometry, so one can't really fault accuracy, and DNA analysis from a small sample of the cohort ensured that cord blood samples were not contaminated by maternal blood so as to rule out reading maternal testosterone levels over offspring.
  • Five of those 707 offspring were eventually diagnosed with an autism spectrum disorder (ASD).
  • Results: "no significant correlations between TT levels and scores on any AQ scale among males (rho range: -.01 to .06) or females (rho value range: -.07 to .01)". TT = total testosterone.
  • Indeed, "no significant association between BioT or TT concentrations and AQ scores among males (rho value range: -.07 to .08) or females (rho value range: -.06 to .12)".
  • When looking at those 5 who were diagnosed with ASD, four "had TT and BioT levels lower than the sex-specific BioT means of the broader cohort, and all cases were within one standard deviation of these means".
  • Not surprisingly the authors conclude that "testosterone concentrations from umbilical cord blood are unrelated to autistic-like traits in the general population".

There have been a few murmurs about this study in cyberspace, its results and implications for the EMB theory; partly genuine scientific interest, partly prejudice it has to be said. I'm going to try and remain objective. Assuming there was no significant sample degradation given that cord samples were taken at parturition and then thawed out apparently some years later for analysis, there are a few other factors to take into account which might have more general implications of the EMB-autism work.

Fair-do to Prof. Whitehouse for not totally poo-pooing the other research literature built up on testosterone and autism based on his findings. Indeed an even more recent study again by Bonnie Auyeung and colleagues***** (including Prof. Baron-Cohen) looking at amniotic fluid foetal testosterone and Q-CHAT results in 18-24 months old found a more positive relationship; bearing in mind a much smaller sample size and the focus on infancy not adulthood.

Back to the Whitehouse study: he for example talks about concentrations of cord testosterone not necessarily being the sole factor of any relationship but rather "individual differences in biological sensitivity to testosterone". That and the fact that the AQ is not a comprehensive autism assessment but rather a screen, and quite a subjective screen by all accounts. Other authors have talked about testosterone within a wider context of other biological markers as per this study by Geier and Geier****** who bring in words like glutathione, cysteine and homocysteine. I think you can see where I'm going with this; similar to the increasingly distant prospect of an autism gene, so an individual biomarker for autism is also looking less and less likely. Then there's heterogeneity (autisms not autism), comorbidity, etc. Indeed comorbidity is something that really needs a lot more study with testosterone-autism in mind.

Finally, I was interested to see that poverty might have been an influence on the AQ scores reported in the recent trial, to quote: "Total AQ scores were significantly higher for those adults whose mother was 
living below the poverty line during pregnancy". Assuming that you share the same enthusiasm for epigenetics and things like the Barker hypothesis as I do, you can perhaps see that this might be an area ripe for some further investigation.

To finish, a bad moon rising... oh yeah, I see it now.


* Whitehouse AJ. et al. Perinatal testosterone exposure and autistic-like traits in the general population: a longitudinal pregnancy-cohort study. J Neurodev Disord. 2012; 4: 25.

** Auyeung B. et al. Fetal testosterone and autistic traits. Br J Psychol. 2009; 100: 1-22.

*** Baron-Cohen S. The extreme male brain theory of autism. Trends Cogn Sci. 2002; 6: 248-254.

**** Baron-Cohen S. et al.  Is there a link between engineering and autism? Autism. 1997; 1: 101-109.

***** Auyeung B. et al. Prenatal versus postnatal sex steroid hormone effects on autistic traits in children at 18 to 24 months of age. Molecular Autism. 2012; 3: 17.

****** Geier DA. & Geier MR. A clinical and laboratory evaluation of methionine cycle-transsulfuration and androgen pathway markers in children with autistic disorders. Horm Res. 2006; 66: 182-188.

---------- Whitehouse AJ, Mattes E, Maybery MT, Dissanayake C, Sawyer M, Jones RM, Pennell CE, Keelan JA, & Hickey M (2012). Perinatal testosterone exposure and autistic-like traits in the general population: a longitudinal pregnancy-cohort study. Journal of neurodevelopmental disorders, 4 (1) PMID: 23110806