Monday, 1 December 2014

Cortisol and cytokines: a diagnostic tag-team for autism?

A quote from the paper by Chang-Jiang Yang and colleagues [1] begins today's post: "The results of ROC [receiver operating characteristic] analysis indicated the cortisol VAR, IL-6 and TNF-α were potential biomarkers in diagnosis of ASD [autism spectrum disorder]."

With many thanks to Natasa for providing me with a copy of this paper, I'd like to discuss these 'joined up' findings a little further. A few pointers to begin with:

Boot the grime of this world in
the crotch, dear
  • Based on quite a bit of previous research looking at the steroid hormone cortisol in relation to autism (see here) alongside an increasingly important body of work talking about cytokines and autism (see here), the authors embarked on assessing the joint role of these compounds as "potential biomarkers in assisting the diagnosis of ASD."
  • Based in China, a small-ish group of participants diagnosed with ASD (n=35) aged around 10 years were recruited alongside an asymptomatic group (n=32) "unrelated to the autistic participants".
  • Salivary cortisol levels were measured (8 of them) at various points of the day from waking up to just before going to sleep. A fasting blood sample was also provided from each participant for analysis of cytokines. The Childhood Autism Rating Scale (CARS) served as the measure of autism severity.
  • Results: the cortisol VAR - "diurnal variation of cortisol" - did show something of a group difference between those with autism and controls exemplified by "reduced diurnal amplitude in... cortisol concentration". Group cortisol concentrations differed particularly at the time of "just before going to sleep", where the ASD group tended to show higher levels than controls.
  • Insofar as those cytokines: "There was a significant difference noted in median plasma concentrations of IL-6 and TNF-α between the ASD and control individuals." Both measures were elevated in the autism samples when taking groups as a whole.
  • Then came the ROC analysis, and based on those values for cortisol VAR and those two cytokines, various measures of sensitivity and specificity were put forward individually. "The combination of three factors had a sensitivity of 91.43% and a specificity of 96.87% (AUC = 0.97)". Those are pretty good values in anyone's book
  • The authors conclude that their study results "may supply a simple clinical approach for aiding the diagnosis of ASD." Importantly however, they make mention of the small participant groups included in their study.

As always, further independent replication of these findings are warranted before anyone goes and gets too excited about the possible implications. That and the fact that 'biomarkers' mentioned in the context of autism perhaps doesn't mean as much as you might think bearing in mind the heterogeneity of the autism spectrum ('the autisms'?) and those all-important comorbidities which I keep going on about. Whether results also translate to other geographic or ethnic populations is another point to be seen.

What I perhaps like best about the Yang paper and results is however the logical simplicity behind their study. As indicated, both areas of cortisol - as part of the HPA axis - and cytokines have something of an important history in autism research [2] which has thus far been seldom looked at together. I acknowledge the paper by Brian Lovell and colleagues [3] (discussed in this post) looking at pro-inflammatory biomarkers and cortisol levels in parents of children with autism or ADHD (attention-deficit hyperactivity disorder) but that was parents, not children with autism.

The obvious next questions after any independent replication are 'why' and what could this mean for potential interventions (if and when required). The 'why' question is probably going to be rather complicated as per the involvement of genetics (see here), epigenetics (see here) and biochemistry (see here) intersecting when it comes to immune function involvement and autism for example. Despite some headlines talking about elements of immune function as being the next frontier for autism research I'd be minded to say that is has been for at least the past 20 years or so, in some circles at least. Oh, and science is also picking up the idea of "an immune-mediated subtype of autism" too [4] perhaps at the centre of any cytokine biomarkers [5].

Insofar as the intervention side of things, well this is where things can get a little more contentious. Accepting that concepts like inflammation are starting to be more readily used in the context of psychiatry (see here), the idea that treating said inflammation might impact on behavioural measures is still something squarely in the research domain at least for now.

But this kind of work does represent an interesting area ripe for further study...

Music to close... and I am the only one transfixed with this 'how to play Heart and Soul' on the piano?


[1] Yang C-J. et al. The roles of cortisol and pro-inflammatory cytokines in assisting the diagnosis of autism spectrum disorder. Research in Autism Spectrum Disorders. 2015; 9: 174-181.

[2] Ashwood P. et al. Elevated plasma cytokines in autism spectrum disorders provide evidence of immune dysfunction and are associated with impaired behavioral outcome. Brain Behav Immun. 2011 Jan;25(1):40-5.

[3] Lovell B. et al. The psychosocial, endocrine and immune consequences of caring for a child with autism or ADHD. Psychoneuroendocrinology. 2012 Apr;37(4):534-42.

[4] McDougle CJ. et al. Toward an immune-mediated subtype of autism spectrum disorder. Brain Research. 2014. November 13.

[5] Rose D. & Ashwood P. Potential cytokine biomarkers in autism spectrum disorders. Biomark Med. 2014 Oct;8(9):1171-1181.

---------- Yang, C., Tan, H., Yang, F., Liu, C., Sang, B., Zhu, X., & Du, Y. (2015). The roles of cortisol and pro-inflammatory cytokines in assisting the diagnosis of autism spectrum disorder Research in Autism Spectrum Disorders, 9, 174-181 DOI: 10.1016/j.rasd.2014.10.012