"Androgens were not associated with autistic traits at 12 months of age"

EARLI - the Early Autism Risk Longitudinal Investigation study - has been mentioned on this blog before (see here) with the aim of the initiative to "examine possible environmental risk factors for autism and study whether there is any interplay between environmental factors and genetic susceptibility."

In this post I'm bringing the paper by Bo Park and colleagues [1] (open-access) to your attention and the observation(s) that umbilical cord blood levels of testosterone and other related androgens were seemingly not associated with autistic traits at 12 and 36 months of age in their cohort. Such findings represent yet another biological research blow (see here) to facets of the Extreme Male Brain (EMB) theory of autism and the suggestion that "ASD [autism spectrum disorder] is an extreme presentation of a typical male cognitive profile where the drive to “systemize” is stronger than the drive to empathize."

So, looking at cord blood samples from 137 children recruited on to EARLI - "a high autism-risk cohort following pregnant mothers with an older child diagnosed with an ASD (autistic disorder, Asperger syndrome, or pervasive developmental disorder not otherwise specified)" - researchers looked at whether measures of various androgens might correlate with scores on the Autism Observation Scales for Infants (AOSI) and Social Responsiveness Scale (SRS). Said schedules were administered at 12 months and 36 months respectively and various potentially confounding variables were taken into account when it came to looking at any associations. It's also worth pointing out that the technology of choice when it came to those measures of cord blood levels of androgens was an old favourite of this blog: liquid chromatography-tandem mass spectrometry (LC-MS/MS).

Results: well as per the title of this post, and after adjustment for potentially confounding variables - "maternal age, gestational age, and cesarean delivery" - there wasn't a great deal to see in terms of levels of androgens and the presence of autistic traits. Indeed, the title of this post only tells half the story as testosterone was also found not to be associated with the SRS scores at 36 months too. The authors do note that: "Male infants (n=75) showed significantly higher umbilical cord testosterone levels and greater social deficits at 36 months of age" than females, but after adjustment for confounders this observation was left wanting. They also talk about some interesting observations about when a child had an older female sibling diagnosed with autism - "androgen levels and autistic traits may depend on sex of the older affected sibling" - but I'm not so sure about the strength of such findings and whether other mechanisms might also be at work. I might reiterate that autistic traits were the name of the research game in this study not a diagnosis of autism.

As mentioned, the Park findings represent another setback for the generalisability of the role of androgens (prenatal and beyond) in relation to autism and/or autistic traits. I guess that in these days of the plural 'autisms' (see here) it's perhaps not entirely unexpected that grand theories of autism seem doomed to fail when put up to scientific scrutiny. Indeed someone recently has talked about this [2]. I still however remain interested in the discussions around the EMB theory of autism, and although this and other research has not been entirely kind to the hypothesis, it is still perhaps deserving of further study in order to see who it may be most relevant to in these days of plural autisms and subgroupings...

To close, isn't this why Twitter was invented?

----------

[1] Park BY. et al. Umbilical cord blood androgen levels and ASD-related phenotypes at 12 and 36 months in an enriched risk cohort study. Molecular Autism. 2017; 8: 3.

[2] Müller R-A. & Amaral DG. Editorial: Time to give up on Autism Spectrum Disorder? Autism Res. 2017. Jan 27.

----------

Park, B., Lee, B., Burstyn, I., Tabb, L., Keelan, J., Whitehouse, A., Croen, L., Fallin, M., Hertz-Picciotto, I., Montgomery, O., & Newschaffer, C. (2017). Umbilical cord blood androgen levels and ASD-related phenotypes at 12 and 36 months in an enriched risk cohort study Molecular Autism, 8 (1) DOI: 10.1186/s13229-017-0118-z

Maternal diagnosis of PCOS increases the risk of offspring autism?

"We found that maternal diagnosis of PCOS [polycystic ovary syndrome] increased the risk of ASD [autism spectrum disorder], even after adjusting for potential confounders. Obesity among women with PCOS appeared to further increase the risk of ASD in the offspring. The risk associated with PCOS was similar in males and females."

Those were some of the main findings reported in the paper by Kyriaki (Sunday) Kosidou and colleagues [1] (open-access) who conducted a case-control study in Sweden looking at "23 748 ASD cases and 208 796 controls, matched by birth month and year, sex and region of birth." PCOS - an endocrine disorder characterised by polycystic ovaries, irregular periods and elevated levels of male hormones (androgens) - "was classified according to any lifetime recorded diagnosis (ICD-8: 256.90, ICD-9: 256E and ICD-10: E28.2)." The frequency of maternal PCOS alongside various other potentially confounding variables were analysed across the groups. Some media about the study can be seen here.

As per the opening paragraph, there did appear to be something to see when looking at rates of maternal PCOS where offspring autism was mentioned. One hundred and sixty-nine mothers of children with autism were found to have had a diagnosis of PCOS (0.7%) compared with 837 control mothers (0.4%). The difference (0.7% vs 0.4%) was deemed statistically significant (<0.001). That being said, and as with many factors when it comes to autism, differences were also noted across quite a few other parameters too such that 'maternal essential hypertension' was more frequently reported in mothers of children with autism compared with controls (1% vs. 0.8% respectively). Rates of diabetes mellitus were also significantly elevated (autism: 1.2% vs. controls: 0.8%). I'll come back to these findings shortly.

A few more details might also be useful to mention including: "The odds of offspring ASD were further increased among mothers with both PCOS and obesity, a condition common to PCOS that is related to more severe hyperandrogenemia." That and that fact that: "Risk estimates did not differ between sexes" and we have some potentially important results that "awaits confirmation, and exploration of potentially underlying mechanisms."

This is not the first time that the words 'PCOS' and 'autism' have graced this blog in the same entry (see here). That post from a few years back did quite a bit of speculating in terms of how mention of androgens and autism might link back to various discussions down the years talking about 'extreme male brains' and the like (bearing in mind that sweeping generalisations are not required [2]). It certainly appears that the Kosidou findings might tap into research chatter about things like testosterone and autism albeit with some provisos (see here and see here for example) and taking into account other recent findings [3].

Going back to the data on maternal hypertension and/or maternal diabetes as also showing some possible 'connection' to offspring autism, I've covered these variables before based on other independent peer-reviewed findings (see here and see here respectively). Maternal diabetes, specifically gestational diabetes, is something where the research evidence is increasingly converging in terms of an elevated risk of offspring autism. Mechanisms of effect are however, a little in short supply outside of talk about more autoimmune mediated diabetes (type 1 diabetes) and autism (see here). Maternal obesity and offspring autism has similarly figured in the peer-reviewed literature (see here).

Reiterating that there is more to do in this area of autism research, I do find the Kosidou results to be interesting. That preferential screening of offspring autism might be offered to women diagnosed with PCOS could be one consideration from such results; another might be for further research to focus on how PCOS and "elevated steroidopathic symptoms" might also feature among some people with an autism diagnosis or autistic traits in mind [4]. That being said, I must re-state that the overall numbers of diagnosed cases of PCOS in mums to children with autism was quite small...

Music: Steve - Emergency Art Rate.

----------

[1] Kosidou K. et al. Maternal polycystic ovary syndrome and the risk of autism spectrum disorders in the offspring: a population-based nationwide study in Sweden. Molecular Psychiatry. 2015. 8 December.

[2] Joel D. et al. Sex beyond the genitalia: The human brain mosaic. Proc Natl Acad Sci U S A. 2015 Nov 30. pii: 201509654.

[3] Bilgiç A. et al. Autistic Trait, Empathy, and Attention-Deficit/Hyperactivity Symptoms in Women with Idiopathic Hirsutism. Int J Trichology. 2015 Jul-Sep;7(3):113-118.

[4] Pohl A. et al. Uncovering steroidopathy in women with autism: a latent class analysis. Mol Autism. 2014 Apr 9;5:27.

----------

Kosidou, K., Dalman, C., Widman, L., Arver, S., Lee, B., Magnusson, C., & Gardner, R. (2015). Maternal polycystic ovary syndrome and the risk of autism spectrum disorders in the offspring: a population-based nationwide study in Sweden Molecular Psychiatry DOI: 10.1038/mp.2015.183