Evidence explainer

Brain, aging, and sleep health

How CGRP Drives Migraine: From Mechanism to Medicine

A neuropeptide called CGRP sits at the center of the nerve circuit that generates migraine pain. Tracing how researchers proved that link explains why a whole class of migraine medicines now exists.

Fully reviewed by Jasaman (Jasmin) Tojjar, MD, PhD

On this page
  1. The short answer
  2. Key points
  3. How researchers build a drug target
  4. Present at the scene: the nerve circuit of a headache
  5. Cause, not coincidence: the provocation test
  6. The right address: an unusual receptor
  7. Remove it and see: two medicines, one pathway
  8. What the mechanism does and does not promise

The short answer#

A single neuropeptide, calcitonin gene-related peptide (CGRP), turned out to sit at the heart of the nerve circuit that produces migraine pain, and that is why an entire class of migraine medicines now exists. Researchers did not stumble onto a drug and reason backward. They built the case link by link: CGRP is released by the right nerves during an attack, injecting it can bring an attack on, its receptors are located exactly where pain is processed, and blocking it lowers how often attacks happen. When each step is tested rather than assumed, you get a target worth treating.

Key points#

How researchers build a drug target#

It helps to start with what "validated target" actually means, because CGRP is a textbook case. A molecule earns that status only when several independent kinds of evidence point the same direction. Ask four questions in order. Is the molecule present at the scene? Can it cause the problem, not just accompany it? Do its receptors sit where they would need to sit to matter? And does removing it change the outcome? CGRP answers all four, which is what separates a durable target from an interesting hypothesis. The rest of this article walks those four questions in turn.

Present at the scene: the nerve circuit of a headache#

Migraine pain travels along the trigeminal nerve, which carries sensation from the face and from the membranes wrapped around the brain, the meninges, together with the blood vessels running through them. The endings of these sensory fibers do two jobs at once. They report pain inward toward the brainstem, and they release signaling peptides locally at the vessel wall. That combination of sensing and secreting is what specialists call the trigeminovascular system.

CGRP is one of the most plentiful peptides carried in these fibers. In their account of the field in Cephalalgia, Edvinsson and Goadsby describe how the peptide was identified in the 1980s and how investigators soon found it in the blood draining from the head during acute migraine and cluster headache attacks. Put plainly, when a migraine is underway, measurable CGRP is pouring out of the trigeminal system. That single finding shifted the understanding of migraine away from a purely blood-vessel event and toward a disorder of sensory signaling with a specific chemical signature.

Cause, not coincidence: the provocation test#

Finding a molecule at the scene is suggestive but not enough, because plenty of bystanders show up during any biological event. The decisive experiment was to run the process backward. Investigators infused CGRP into people who get migraines and watched what happened. In those prone to migraine, the infusion could set off a delayed headache that looked and behaved like their usual attacks, while people without migraine were far less affected. A molecule that rises during attacks and can also start one is acting like a driver, not a passenger. That is the step that moved CGRP from correlation to plausible cause.

The right address: an unusual receptor#

A signal needs a receiver, and where the receiver sits tells you where a drug could act. CGRP works through an unusual arrangement. A protein called the calcitonin receptor-like receptor (CLR) has to pair with a small partner protein, RAMP1, to assemble the standard CGRP receptor. The analysis by Close and colleagues in Cephalalgia notes that this CLR and RAMP1 pairing appears in the cerebral cortex and in the vasculature, and that CGRP can also act through a second receptor, the amylin-1 receptor, giving the peptide more than one way to send its message. Mapping these receptors onto vessels, nerve endings, and pain-processing centers handed researchers a set of addresses where an intervention might work.

A second front: cortical spreading depression#

Migraine with aura adds a further clue. The aura, the shimmering visual or sensory disturbance that precedes some attacks, is thought to reflect cortical spreading depression, a slow wave of intense neuronal firing followed by a brief lull that sweeps across the surface of the brain. Close and colleagues looked at how this wave connects to CGRP and reported that repeated waves raise CGRP gene activity and peptide levels across brain regions, and that cortical CGRP can be released under the high-potassium conditions the wave creates. Their work also suggested that blocking CGRP can change how the wave spreads. That places CGRP at the peripheral vessel wall and, possibly, at the very cortical events that open an attack with aura. The peptide may have more than one plausible place to act.

Remove it and see: two medicines, one pathway#

The final test of a target is subtraction. If CGRP helps generate migraine, interrupting it should help prevent or stop attacks. Two quite different pharmacological tools were built to run that test, and the fact that both work is what gives the biology its strength.

The first tool is monoclonal antibodies, large engineered proteins that lock onto either CGRP itself or its receptor and neutralize the signal. Because antibodies are big and long-lasting, they are dosed at intervals for prevention, lowering how often attacks arrive. The second tool is gepants, small molecules that block the CGRP receptor and can be swallowed as a pill, which suits them to treating an attack in progress and, for some, to prevention. A review by Cohen and colleagues in BioDrugs summarizes that several CGRP-targeting antibodies and several gepants have reached regulatory approval.

The trial evidence anchors the antibody story. In the STRIVE study published in the New England Journal of Medicine in 2017, Goadsby and colleagues compared the CGRP-receptor antibody erenumab against placebo in episodic migraine and reported fewer monthly migraine days in the treated groups than in the placebo group. Approval followed in the United States in 2018, and erenumab is notable as the first approved antibody aimed at a G-protein-coupled receptor. Here is why two tools matter more than one. An antibody and a small molecule are built on completely different chemistry, yet both reduce migraine by interrupting the same pathway. When independent designs converge on the same result, the underlying biology becomes hard to argue with.

What the mechanism does and does not promise#

The CGRP story is a clean example of a target moving from bench to clinic, but a few limits keep it honest. Blocking CGRP does not help everyone, which tells us migraine runs on circuitry wider than this one peptide. The trials establish a direction and a category-level benefit, not a guarantee for you, and whether, when, and how to use any specific therapy is a clinical decision that depends on the whole individual, other conditions, and preferences. Understanding the mechanism does not hand you a prescription. What it does is explain why this class of treatments exists and what biology it engages, which is exactly the kind of reasoning that makes chronic-disease care rational rather than trial and error.

Sources and further reading

  1. Close et al., Cephalalgia (cortical spreading depression and CGRP)
  2. STRIVE erenumab trial, NEJM 2017
  3. Cohen et al., BioDrugs (CGRP antibodies and gepants)
  4. Edvinsson & Goadsby, discovery of CGRP in migraine, Cephalalgia

Questions and answers

Does high CGRP cause every migraine?

Not in a simple one-to-one way. CGRP is strongly implicated and rises during many attacks, but migraine involves several brain networks. That is one reason CGRP-blocking medicines help many people yet not all, and why research continues into other contributors.

Are CGRP medicines painkillers?

No. They do not dull pain the way ordinary analgesics do. Instead they interrupt a specific signaling pathway thought to help generate the attack, which is why some are used to prevent attacks rather than only to treat one in progress.

Why does it matter that two different drug types work?

Because they share almost nothing chemically. An injectable antibody and an oral small molecule reaching the same result through the same pathway is powerful evidence that the pathway itself, not a quirk of one drug, is driving the effect.