Could a future cure for migraines come from a chance discovery in the past?

There may soon be an FDA-approved treatment for the intractable condition

Brianna Bibel


Cold Spring Harbor Laboratory

The first targeted treatment for migraines may soon gain FDA approval – and the molecule it targets was discovered by chance by researchers studying cancer.

The story begins back in the early 1980s, when Michael Rosenfeld, an endocrinologist at the University of California, San Diego, and his PhD student Susan Amara, were studying the production of hormones in thyroid tumors. In the course of their research, they found some of the tumors would spontaneously reduce their production of a hormone called calcitonin, which is involved in bone metabolism. 

That in itself wasn’t strange: Scientists already understood that some hormone-producing tumors would stop producing one hormone, usually accompanied by starting to produce another. What mechanisms the cell used for this change in hormone production is actually what Amara and Rosenfeld were trying to figure out. But oddly, they found the low-calcitonin-producing tumors were still copying the calcitonin gene into RNA – suddenly, they just weren’t making the hormone. Even stranger, those tumors started making large quantities of a different hormone, one the researchers had never seen before.

The storm is a metaphor

Brian Cook / Unsplash

They named the new hormone calcitonin gene-related peptide (CGRP). Weirdly, the genetic instructions for CGRP looked like they were coming from the same gene as calcitonin. Amara and Rosenfeld suspected that a then recently discovered phenomenon called RNA splicing was involved. DNA can be thought of as a “cookbook,” with genes representing different recipes. If a cell wants to make a protein, it must first copy, or transcribe, that protein’s recipe into a molecule called messenger RNA (mRNA) – like copying a recipe onto a note card. Protein-making machinery called ribosomes then translate this recipe to make the new protein.

In between the various steps in each recipe, genes often include notes and guidance about when and where to make their products. This “margin writing” is copied into the initial, primary mRNA, but is then removed in an editing process before the mature mRNA is handed off to the ribosome. This editing process is called RNA splicing, and it was discovered in 1977, around the time Amara and Rosenfeld saw their strange results.

'Alternative splicing'

The main purpose of this mRNA editing process is to remove extra regulatory information – essentially notes in the margins of a recipe – known as introns. These need to be separated from the protein-coding parts, which are called exons. But the mRNA editing process also makes it possible to create multiple products from the same recipe by modifications – for example, removing both exons and introns, like you would omit walnuts if you didn't like nuts in your banana bread. This is called “alternative splicing,” and it’s one of the reasons why human DNA doesn’t require more genes than plants and animals that are usually thought of as “less complex.”

Amara and Rosenfeld had discovered one of the first examples of RNA splicing at work: CGRP and calcitonin were produced by the same gene, but were very different due to mRNA editing. Further research suggested that, as opposed to calcitonin, which is mainly produced in the thyroid, CGRP was mainly found in the brain, especially in a region called the hypothalamus.

A painful metaphor

Clinton Naik / Unsplash

Once they published their findings, other researchers in completely different fields took up the case. Neurologists Lars Edvinsson and Peter Goadsby, for example, initially found CGRP was elevated in the blood of patients experiencing migraines, showing a correlation between CGRP and the severe headaches. But the scientists still needed to show causation – thankfully, a group of selfless volunteers stepped up to the task and allowed themselves to be injected with potentially migraine-inducing hormones in the name of science. Sure enough, the volunteers who received CGRP experienced migraine-like symptoms, while those who received a placebo were unaffected.

Even though exactly how CGRP is connected to migraines is still unclear, evidence for its role has only strengthened over the last few decades. The current idea is that CGRP binds to specific receptors and sensitizes nerves in the head, face, and jaw known as the trigeminal nerves. This helps transmit pain signals, trigger inflammation, and increase blood flow. So scientists began to wonder how to use CGRP or its receptors to develop migraine medication, and pharmaceutical companies began investing in the search.

Antibodies and 'XenoMice'

Initially, researchers looked for chemicals called small molecules, like those in many conventional drugs. While there were a few promising leads, they turned out to be toxic. An alternative strategy pursued antibodies, proteins that recognize and bind to specific sites on other molecules: They’re how your body recognizes molecules as foreign and targets them for destruction. Researchers hoped that they might be able to find an antibody that binds to CGRP receptors, effectively intercepting the migraine-inducing message.

Early attempts at using mouse antibodies in humans failed, because the antibodies themselves registered as foreign and the immune system destroyed them. To overcome this obstacle, scientists in the late 1990s engineered mice called “XenoMice,” whose antibody genes had been replaced with human antibody genes, so that antibodies the mice made would be “humanized.” In the search for a migraine antibody treatment, researchers injected these mice with human CGRP receptors. The XenoMice then produced humanized antibodies, and scientists isolated the ones that bound to the CGRP receptor. One antibody that seemed promising was erenumab.

Not a XenoMouse

Taton Moïse / Unsplash

A benefit of antibodies is that they can be incredibly selective – other important and naturally occurring molecules have receptors very similar to CGRP's, so any potential treatment would have to not block various other receptors. Additionally, antibodies are long-lasting; while injections may not be fun, they’d only have to be administered once a month.

Pharmacutical company Novartis (working with Amgen) recently conducted a 12-week, double-blind controlled Phase III trial of Erenumab. During the trial, patients receiving Erenumab (Aimovig™) had significantly fewer migraine days than patients receiving a placebo, and many patients saw improvement in their ability to function. But despite causing an average reduction in headaches, the “response rate” was fairly low; only 30 percent saw at least a 50 percent reduction in migraine days, meaning more than half of the patients receiving the drug saw no significant benefit compared to a placebo. In the patients where the treatment did work, however, it had a big effect; many saw vast symptom improvement. Novartis just announced the encouraging results, and the treatment is slated for FDA approval next month.

It's unclear whether it will be as a first-line treatment, available to all migraine sufferers, or reserved for more difficult-to-treat cases. And it's too early to know whether long-term use of Erenumab could have side-effects. This is particularly an issue because, as Erenumab is a preventative medicine, it would have to be taken regularly, long-term. The main concern is the risk of blood clotting events, such as ischemic strokes and heart attacks, because CGRP also plays a role in expanding blood vessels. 

Finally, of course, is the question of cost. These drugs are likely to be much more expensive that currently-available treatments, especially until (and if) generic versions become available. However, in addition to Erenumab, which targets the CGRP receptor, there are several other mAbs in clinical trials that target CGRP itself, so competition could potentially bring this price tag down.

Although further research is needed to target who is most likely to benefit and what the risks are, this is an important step in migraine treatment, and an interesting example of the way scientific breakthroughs often rely on previous research—crossing many different fields.

As for Amara and Rosenfeld? Their stories have fruitful endings as well—Rosenfeld is still a professor at UC San Diego, where he continues to study molecular signals and Amara is the Scientific Director of the National Institutes of Mental Health (NIMH), where she runs the Laboratory of Molecular and Cellular Neurobiology.

Comment Peer Commentary

We ask other scientists from our Consortium to respond to articles with commentary from their expert perspective.

Gregory Logan-Graf

Cell Biology

Carnegie Mellon University

I’m glad to see some progress here, but I’ll be waiting for a drug that can be taken as needed. Plenty of people (me included), only get a few debilitating migraines per year, and I’m not sure if getting an expensive monthly injection is an improvement.

But what an interesting story about RNA splicing! I am interested to know if they have any idea why some people are more prone to migraines than others, and why some patients respond better to Erenumab… And why do I seem to only get migraines after I eat cashews?