Botulinum toxin: not just for facelifts anymore

early everyone who suffers from migraine has a theory about what causes it. Women suffer from migraines three times as often than men, suggesting a hormonal cause. Magnesium deficiency or diet can also trigger migraine. Citrus that has been stored too long is effective at inducing migraine. Reports going back centuries suggest that intestinal problems from diet can trigger migraine, hinting at an effect from overgrowth of bad intestinal bacteria.

Many treatments for migraine have unpleasant side effects. Feverfew, a traditional herbal remedy, is an example. After rigorously adhering to the recommended dose, patients may get blood coagulation problems and subconjuctival hemorrhage. Triptans, which block 5HT1B/1D/1F receptors, are the standard treatment. They are expensive: nine bucks for a 25 mg tablet. But though triptans are superior to the ergotamines they replaced, they too can have unpleasant side effects.

Migraines are so painful that patients sometimes look forward to old age, because migraine severity is known to decrease with age. So it was almost a miracle when botulinum toxin was found to prevent chronic migraines.

Botulinum Neurotoxin (BoNT)

Botulinum toxin is a deadly protein produced by anaerobic bacteria. It's the main reason why it's so important not to feed honey to infants or eat any food from cans that show a bulge. BoNT blocks acetylcholine release at the neuromuscular junction, causing paralysis of the respiratory muscles and death by asphyxiation.

There are four formulations of botulinum neurotoxin used on patients: OnabotulinumtoxinA (Botox or A/Ona), AbobotulinumtoxinA (Dysport or A/Abo), IncobotulinumtoxinA (Xeomin or A/Inco), and RimabotulinumtoxinB (Myobloc/Neurobloc or B/Rima). The first three contain different amounts of botulinum toxin type A with different excipients such as hemagglutinin, saccharose, sodium chloride and human albumin.^[1] The last one contains type B, which is a different serotype. There are seven major serotypes (A to G), which have different half-lives and different targets in the brain (see below).

BoNT is approved in the USA for hyperhidrosis, migraine, neurogenic bladder disorder, muscle spasms due to spasticity, tremor, dyskinesia, cerebral palsy, and urinary incontinence (detrusor muscle), and it's being used off-label for achalasia (inability to get food through the esophagus) and sialorrhea (drooling or salivary hypersecretion). And, as we all know, it's approved for cosmetic use.

Botulinum toxin and depression

The latest discovery is that facial injections of BoNT seem to work against depression. There have been several double-blind randomized placebo controlled studies, and though not all of them showed a benefit, some scientists are starting to speculate that maybe there's an effect.

The challenge is separating it from a placebo effect. Patients easily notice they're not getting a placebo. Since it is injected into the glabellar region, which controls forehead wrinkling, one theory is that BoNT makes it impossible to frown by paralyzing the facial muscles; the feedback effect makes people think they're less depressed than before. When BoNT is injected into the orbicularis oculi muscles (for crow's feet), patients actually become more depressed because they can no longer make a Duchenne smile.

Interestingly, if you tap someone's forehead, they will blink. This is called the glabellar reflex. If they continue to blink after the first few taps, it could be a sign of Parkinson's disease.

One way to test BoNT is by treating patients who are actually depressed, and indeed, some case report studies show benefits for bipolar depression. A clinical trial in 2014 found improvement in major depressive disorder, but again many of these studies are compromised by the location (the forehead) where BoNT was injected. A better way is to test BoNT injected somewhere else.

A recent study^[1] based on 12,185,458 reports in a post-marketing surveillance database of patients found that eight other injection sites besides forehead, including injections for blepharospasm (eyelid twitching) and limb spasm, also reduced depression, while injections for incontinence had no statistically significant effect. Here are some of their results:

In this table I calculated Pearson chi-square p-values from the author's data. Only the odds ratio and 95% confidence limits were published, so I estimated the number of patients with depression from a screen grab of the authors' bar chart. At the time of this writing, the publisher's link to the supplementary data (if it exists) goes nowhere, so the p-values are only approximate.

An odds ratio is the percent depressed in the treated group divided by the percent depressed in the control population. A 95% confidence limit is, roughly, considered statistically significant at p<0.05 if it doesn't pass through 1.0.

There are many limitations in this study. Most importantly, patients with depression were excluded, and the database was purely voluntary. Excluding the entire population that suffers from the disease you're trying to treat is about as big a limitation as it's possible to get, which means more research is needed. Finally, no safety data are available, which means we don't know what else happened to the patients, only that they were less depressed.

The authors suggested several possible ways BoNT could be working, but one plausible way is that the BoNT was curing whatever problem they were having and the patients were happy about that. Another possibility is that relieving muscle tension gives patients more energy. Or maybe it could be releasing BDNF, a peptide with powerful antidepressant effects which is known to increase after exercise. Or it could be working as it does for migraine.

How does botulinum toxin treat migraine?

BoNT is a protein, so it cannot cross the blood-brain barrier. Moreover, we know that migraine has nothing to do with muscles. So, how can BoNT affect migraine?

It's generally believed that migraine pain starts in the meningeal vasculature, which is to say small blood vessels in the three membranes that envelop the brain and spinal cord. BoNTs are taken up by afferent pain (nociceptor) nerves in the blood vessels and transported through the nerve back through the trigeminal ganglia to a brain region called the trigeminal nucleus caudalis, or TNC. From the TNC, it's believed that the toxins are transported through the cell membrane to other nearby neurons, maybe by a process called transcytosis.

As you might expect, all this takes a long time: it takes twelve weeks before BoNT starts to have a therapeutic effect, and 56 weeks before it starts really reducing headaches. It works best on chronic migraine, with little or no benefit on episodic migraine. It has no effect on the other symptoms of migraine, such as aura, light sensitivity, and nausea—only the pain. So its action is quite different from that of analgesics and opioids.

Migraine is an example of what is called referred pain, which means that where the patient feels the pain, such as the supraorbital or occipital region of the head, is not where the pain is actually happening.

When BoNT is in the presynaptic part of the nerve terminal, it gets proteolyzed and creates the actual toxin consisting of a heavy chain and a light chain. The heavy chain acts almost like a virus by binding to the cell surface and entering into the endosome, where it's cleaved to release the light chain, which is a zinc-containing protease. This zinc protease reacts with a cell protein complex called SNARE, whose job is to package neurotransmitters into synaptic vesicles at the synapse. Each serotype of BoNT chews up a different protein in the SNARE: type A, C, and E cleave SNAP-25, while type B, D, F, and G cleave synaptobrevin II (aka VAMP 2). C also cleaves syntaxin.^[2] All this gnashing and chewing of proteins prevents the neuron from releasing its neurotransmitters and voilà—three months after the injections—no more pain.

The short message is that BoNT isn't acting on acetylcholine motor neurons as when it paralyzes a muscle, though motor neurons might be involved in some way. The nerves affected by BoNT release glutamate along with peptides such as substance P (a pain and inflammation peptide) and CGRP, or calcitonin gene-related peptide. This is why drug companies are testing antibodies against CGRP as possible treatments. Ubrogepant is a CGRP antagonist approved in the USA. People are also using genetic engineering to create better BoNTs.

Why do we get headaches?

There are many unanswered questions here. BoNT cannot be given systemically: it must be injected into a muscle somewhere. From the above, we now know that BoNT can get into the brain, and it can affect neurotransmitter release, so it makes sense that it might be able to mitigate depression. But a bigger question is: Why do we get headaches at all? How did it improve the survival fitness of our caveman ancestors to be incapacitated with pain for days at a time for no apparent reason and no clue what they could do about it? Or are headaches, as some suggest, caused by our current diet or lifestyle? We're waiting for answers on that question.

1. Makunts T, Wollmer MA, Abagyan R (2020). Postmarketing safety surveillance data reveals antidepressant effects of botulinum toxin across various indications and injection sites. Sci Rep 10, 12851. https://doi.org/10.1038/s41598-020-69773-7

2. Ramachandran R, Yaksh TL (2014). Therapeutic use of botulinum toxin in migraine: mechanisms of action. Br J Pharmacol 171, 4177–4192. DOI:10.1111/bph.12763

sep 01 2020, 5:40 am