Science On The Internet – Friend or Foe – Review of an Article for the MS Community

Today an interesting article was brought to my attention, but not entirely. By this I mean it was an OpEd piece that reviewed it on a blog/website, with the title, “Neurontin and Lyrica are Highly Toxic to New Brain Synapses” that was referencing a totally separate article in a highly respected peer-reviewed journal called Cell.

The conversation was with MS patients, real ones, reaching out to each other in the exceptional on-line communities that pop-up to support each other, and where they are able to ask real people about their own experiences. One person asked about others’ experiences with pregabalin (Lyrica) and gabapentin (Neurontin and others). Another smart gentleman posted a link to this OpEd piece, which references another article, “Gabapentin receptor alpha2delta-1 is a neuronal thrombospondin receptor responsible for excitatory CNS synaptogenesis”  (Eroglu et al., 2009).

First, I went and read the source paper for this study. Trust me, it is a hard read, go for it if you are inclined but it is not intended for the layperson. I would caution against making judgments against single source papers but I will give you my “science brain” unbiased view from the snooty point of view as a science critic:

  1. Some anti-epileptics and the like got in trouble for off-label uses. But they don’t mean for what MS patients are using it for: neuropathic pain, they mean it as adjuvant therapy for psychiatric diseases like bipolar disorder. Now, there’s nothing wrong with off-label use, but there are in fact multiple trials that look at it for this use (Postherpetic Neuralgia), not just epileptic disorders.
  2. They only use rats and mice (in vivo), some (limited) in vitro (my preferred – cell lines in a test tube) testing. Not humans. Because guess what, it’s the best we have and it turns out people get all bent out of shape about experimenting on real live humans. Always remember though, you are not a rat, nor is your brain, and these are only to say “ok it sucked for the mousies, let’s put a safety factor on it and infer that it’s gonna happen this way in humans”. There’s no other way, and neurotoxicity studies are notoriously hard and have a TON of variables you just can’t control for, which means, you can never be certain your results are because of the stuff you just fed them, or other totally unrelated reasons (called confounders), and they are a Bench (thanks The Good Place – best show ever!) to account for – note this study does not go into detail about confounders – red flag. Believe me, I wish there was more. Studies I have been involved in have had to use zebra fish and surf clams and it sucks, but it’s all we’ve got. So just remember though, while it’s the best we have, people are not mice.
  3. Re: #2, this study does give me enough pause to say that in the future I would advocate to my patients to discontinue use while pregnant and nursing, while their fetuses and neonates are undergoing neurogenesis (making a brain) and early synaptogenesis (making synapses that are making memories). It is already NOT recommended for use in children 3 years or under. I would really, really go with that.
  4. Synaptogenesis is reduced, not eliminated. In other words, neurons still are being made, just not a lot. It appears that this reduction is what is causing clinically valuable outcomes, like controlling pain and zaps and twitches. And not all the mice responded, actually, only half of them did. “Only 50% of the mice responded strongly to GBP (gabapentin) injections. It is possible that a critical threshold concentration of GBP in the cerebrospinal fluid is required to be effective in blocking synapse formation, which is only achieved in half of the mice”. This means, in English: (a) it’s probably very likely dose dependent, and they never figured out what that dose was, and (b) there is something else going on here, and again, they don’t know what that is.
  5. “Since GBP strongly blocks TSP-induced synapse formation within its therapeutic concentration, it is possible that inhibition of excitatory synapse formation is an important mode of its therapeutic action in epilepsy and pain. Reactive astrocytosis is prominent both in epileptic lesions and in the spinal cord after peripheral nerve injury that leads to neuropathic pain”. This is saying, gabapentin-reducing synapse formation = A GOOD THING, because it’s reducing pain. But not at the expense of turning you into a zombie.
  6. Synaptogenesis in general follows long-term potentiation (which means an increase in strength of nerve impulses along pathways that have been used previously, either short-term or long-term). This happens in the hippocampus, and it’s how long-term memories are formed. This has nothing to do with motor issues, which is what MSers are dealing with. This is most active when we are kids, but in grown-ups, if we had the same level of synaptogenesis, it would actually be bad, because it would interfere with already pre-existing circuits. This is actually a big deal with kids with Autism Spectrum Disorder. They don’t undergo enough “synaptic pruning” and so we can see that they just have these big jumbles of neurons and that their hippocampuses are bigger than normal (Onore, Careaga, & Ashwood, 2012; Sara B. Johnson, Ph.D. & , Robert W. Blum, M.D., Ph.Db, and Jay N. Giedd, 2010; Sullivan, 2015). So what I get out of this is, hey, interesting future research for early-identified children with ASD as a potential therapy. But not a humongous potential problem for grown-ups taking gabapentin. What it also says to me is, if you are having memory issues, going on a drug holiday to see if gabapentin is contributing to them is not a terrible idea to see if it makes a difference, but again, is unlikely to be the whole issue.
  • Chronic pain and epilepsy are both characterized as involving an excess of synaptic connections. Synaptogenesis isn’t stopped completely, but reduced to a level that offers benefit to patients in this category.
  • The point of the study was to see HOW gabapentin worked, aka it’s mechanism.
  • “Wellness Resources” did a bang-up job with their marketing by saying it effectively causes brain death to get people to pay attention to it! Kudos to that, friends. Because “Neurontin and Lyrica are Highly Toxic to New Brain Synapses” is a way more catchy title than the study’s title of “Gabapentin receptor alpha2delta-1 is a neuronal thrombospondin receptor responsible for excitatory CNS synaptogenesis”. 😀
  • A well-done paper. Except for the statistics, which were relatively glossed over, and I would want to see the raw data and crunch it myself. Publish on Science Direct with the raw data attached and let researchers critique it for real and we’ll talk.
  • You can’t make an assessment based on one non-peer reviewed OpEd piece.


Eroglu, C., Allen, N. J., Susman, M. W., O’Rourke, N. A., Park, C. Y., Ozkan, E., … Barres, B. A. (2009). Gabapentin receptor alpha2delta-1 is a neuronal thrombospondin receptor responsible for excitatory CNS synaptogenesis. Cell, 139(2), 380–92.

Onore, C., Careaga, M., & Ashwood, P. (2012). Brain , Behavior , and Immunity The role of immune dysfunction in the pathophysiology of autism. Brain Behavior and Immunity, 26(3), 383–392.

Sara B. Johnson, Ph.D., M. P. H., & , Robert W. Blum, M.D., Ph.Db, and Jay N. Giedd, M. D. (2010). NIH Public Access – Adolescent Maturity and the Brain: The Promise and Pitfalls of Neuroscience Research in Adolescent Health Policy. Journal of Adolescent Health, 45(3), 216–221.

Sullivan, D. L. (2015). Temporality of Risk Factors and the Gender Differential Related to Autism Spectrum Disorder Diagnosis. Walden University. Retrieved from