Wednesday, 19 September 2012

Lenalidomide and autism?

Lenalidomide @ Wikipedia
Of the thousands of medicines which have been or are currently in use, a few really enter the mainstream public consciousness for a variety of different reasons.

That little blue pill which now graces many a man's medicine cabinet worldwide is probably the best example, as any email spam folder can readily confirm.

One drug, suggested to treat morning sickness during pregnancy in the late 1950s, is however synonymous with the darker side of pharmaceutics; illustrative of the failings of appropriate medicines testing: thalidomide.

If you don't know the story of thalidomide this article offers quite a succinct timeline, and how only recently the company responsible for the invention of thalidomide has formally apologised to all those affected by this medicine.

You would perhaps think that after such a torrid history, this drug would and should be consigned to the annals of history, never to be used in medical practice again. You might however want to think again; as thalidomide is re-emerging as a medicine for various conditions including certain cancers, inflammatory diseases and even persistent cough, albeit with a few tweaks and heaps more pharmacovigilance particularly when administered to females of reproductive age.

OK so what's all this got to do with autism? Well, my attention was grabbed by the publication of this proof-of-concept study by Michael Chez and colleagues* (open-access) and the suggestion that lenalidomide, an analogue of thalidomide, might (might) show some promise in cases of autism. I had seen mention of this study, or something like this study, in a previous abstract from IMFAR 2010 (see here). The entry for the study can also be seen here.

The keen eyes out there might recognise the name Dr Michael Chez and the Sutter Institute for some other recent publicity regarding the first FDA-approved stem cell trial for autism (see trial details here) which has created some column inches among mainstream media and bloggers alike (including me). It seems that this research group is intent on creating scientific waves.

Back to the paper, which is open-access, for a summary and few comments:

  • The basic premise of the trial was the recognition of quite a bit of research which has indicated the immune system and various features of immune function might correlate with the presence of autism in some cases. Mention for example of that dastardly cytokine, interleukin-6 (IL-6) is made, but the main protagonist in this particular study is tumor necrosis factor - alpha (TNF-α), another inflammatory cytokine, and a target for lenalidomide (see here). The authors cite the focus of TNF-α as an outcome measure "because other cytokines were not always available in all patients" referring to tested parameters.
  • A daily dose (2.5mgs) of lenalidomide was given for 12 weeks to 7 males (aged 6-12 years) diagnosed with autism and with parental report of regression being associated with symptom onset. Autoimmune "dysfunction" (I assume this means autoimmune disease) was reported in first-degree relatives of all participants, although prospective participants with various diagnosed autoimmune conditions were excluded from the present study.
  • Various measures of behaviour and cognitive functioning were used during the study (baseline - 6 weeks - 12 weeks) including ADOS, CARS and the Receptive and Expressive One-Word Picture Vocabulary Test. This accompanied measurement of both serum and CSF levels of TNF-α (CSF levels collected via that most invasive of methods, lumbar puncture - don't click on this link if you are squeamish). 
  • To ascertain the pharmacokinetics of the drug (drug metabolism), blood draws at 1, 2, 4 and 8 hours after the first dose of the lenalidomide were administered. 
  • Results: well, mixed is probably the best description. Safety-wise, two participants were withdrawn from the study following their developing a rash. Another participant discontinued when their neutrophil count (white blood cells) dipped; albeit transiently. 
  • Four participants registered a drop in TNF-α levels in both serum and CSF at study end. Whilst on the surface of things this might be considered an interesting finding, one has to bear in mind that analysis of CSF levels of the cytokine were accepted anywhere up to 8 months before the baseline testing started.
  • Behaviourally, there was some suggestion of improvement on the various measures included for study, particularly at 6 weeks, but alas none of these changes were statistically significant. Indeed with the attrition (drop-out) rate (n=3) mentioned at study end, I'm not really surprised that nothing was found to be significantly improved. It would have to be an almost spectacular change in behaviour for 4 participants to register a significant change from baseline to study end.
  • Conclusion: some interesting trends from the various data but nothing concrete probably as a function of the small participant group.

The science-y types out there may very well look at this trial and its results and focus on its failings and weaknesses. Whilst not trying to defend the study, I would draw your attention to several keywords in the title like 'pilot study' as a clue to the very preliminary nature of this investigation. Indeed, the authors do mention that this was not a randomised trial (i.e. participants randomly allocated to treatment or not), did not include a placebo group and neither was it double-blind; everyone knew that participants were taking lenalidomide and nothing else. It's more of case series description than a scientific trial.

Before any chatter arises about me somehow endorsing thalidomide derivatives to 'treat' autism by highlighting this study, I just want to say that I'm not. My mind keeps wandering back to the devastating teratogenic effects that thalidomide bestowed, and still I can't help but wonder if this is indeed one of those drugs which given its history perhaps should have been consigned to the great pharmaceutical rubbish tip, never to be used again.

I'll let you form your own opinion on whether this is an area requiring further research with autism in mind or not. Bear in mind however that for some people with certain conditions, lenalidomide has probably had some real benefit for them (see here) and its adoption reflects, quote "the pressing need to develop molecules with enhanced immunomodulatory and antitumor activity". Not that there may not be alternatives as per this study however.


* Chez M. et al. Safety and observations in a pilot study of lenalidomide for treatment in autism. Autism Research & Treatment. 2012; 291601.
DOI: 10.1155/2012/291601

---------- Michael Chez, Renee Low, Carol Parise, & Tammy Donnel (2012). Safety and observations in a pilot study of lenalidomide for treatment in autism Autism Research & Treatment DOI: 10.1155/2012/291601