How CGRP Antagonist Drugs Treat and Prevent Migraines

CGRP antagonists are a class of drugs designed to block a molecule called calcitonin gene-related peptide, one of the central chemical drivers of migraine attacks. Since the first approval in 2018, these treatments have reshaped migraine care by offering relief to people who had run out of options with older preventives. They come in two broad forms: small-molecule pills and nasal sprays known as gepants, and injectable monoclonal antibodies that target either the CGRP molecule itself or its receptor. The science behind them is more layered than a single mechanism story, and their reach is starting to extend beyond migraine.

How CGRP Fuels Migraine Pain

CGRP is a small signaling protein found in high concentrations in the trigeminal nerve system, the network of nerves responsible for sensation in your face and head. During a migraine, trigeminal nerve endings release bursts of CGRP. Once released, it triggers a chain reaction: blood vessels in the brain’s protective membranes dilate, surrounding tissues become inflamed, and the nerves themselves grow increasingly sensitive to stimulation. This sensitization means that normal signals, like blood pulsing through an artery, start registering as pain.

The process does not stay local. CGRP released inside the trigeminal ganglion (the cluster of nerve cell bodies near the base of the skull) interacts with neighboring neurons and the support cells surrounding them, amplifying peripheral sensitization and eventually feeding exaggerated pain signals to second-order neurons deeper in the brainstem.¹ That shift from peripheral to central sensitization is what makes a migraine progressively harder to treat the longer an attack goes on. Blocking CGRP early, either at the receptor or by mopping up the molecule before it can bind, interrupts this cascade at its source.²

Animal research has added nuance. In a chronic migraine model, the CGRP receptor antagonist olcegepant not only reduced pain behavior but also dialed down gene expression of CGRP itself, inflammatory cytokines, and certain ion channels in the brainstem, spinal cord, and trigeminal ganglia.³ In other words, CGRP antagonists may do more than just sit at the door and block signals. They appear to quiet some of the downstream inflammatory machinery that keeps the pain cycle spinning.

Gepants and Monoclonal Antibodies

There are two fundamentally different engineering approaches to blocking CGRP, and understanding the distinction matters because it affects how you take the drug, how fast it works, and what it is best suited for.

Gepants are small synthetic molecules that fit into the CGRP receptor like a key in a lock, preventing CGRP from binding. The first generation of gepants, developed in the mid-2000s, proved the concept that blocking CGRP could stop migraine attacks. But those early compounds had problems: some could not be taken by mouth, and others caused liver toxicity.⁴ The second and third generation of gepants overcame those issues. Ubrogepant and rimegepant were approved for acute migraine treatment, and rimegepant later became the first drug approved for both acute and preventive migraine use.⁵ Atogepant was developed specifically for prevention, and zavegepant (initially studied as an intranasal spray) joined the approved lineup as well.⁶ Because gepants are small molecules that clear the body relatively quickly, they work well as on-demand treatments you take when a migraine starts.

Monoclonal antibodies take the opposite approach in terms of duration. These are large engineered proteins, given by injection (either self-administered monthly or infused quarterly), that linger in the bloodstream for weeks. Some target the CGRP molecule itself (fremanezumab, galcanezumab, eptinezumab), while one targets the CGRP receptor (erenumab). That distinction is not just academic. Erenumab, because it sits on the receptor, also blocks a related receptor called AMY1 that the ligand-targeting antibodies leave alone.⁷ This may explain why some patients who fail one type respond when switched to the other.

One Drug That Treats and Prevents

The traditional migraine medicine cabinet has always been split: triptans and pain relievers for stopping an attack, beta-blockers or antiepileptics for preventing one. Patients often needed prescriptions from both categories, each with its own side-effect burden. CGRP antagonists blurred that line. A single 75-milligram oral dose of rimegepant can stop an acute attack, and that same drug taken every other day works as a preventive.⁸ This dual-use profile simplifies things for both patients and clinicians, especially for people whose migraine pattern straddles the line between episodic and chronic.

Monoclonal antibodies are used purely for prevention, given their long half-lives. A person injects erenumab or fremanezumab once a month (or in some cases quarterly) to reduce how often attacks happen. The newer gepants like atogepant are taken daily as preventives, while ubrogepant is used only for acute treatment. So within the CGRP antagonist family, you now have a full toolkit: fast-acting pills for when an attack hits, and long-duration injectable or daily-pill options to keep attacks from happening.

Effectiveness in Clinical Trials and Real-World Use

Across randomized trials, CGRP monoclonal antibodies roughly doubled a patient’s odds of cutting their monthly migraine days in half compared to a placebo injection.⁹ That “50% responder rate” is the standard benchmark in migraine prevention research, and the consistency across trials has been striking. Compared head-to-head with older preventive drugs like topiramate or amitriptyline, the antibodies showed efficacy that was at least comparable, with a much more favorable side-effect profile.¹⁰ That last point is underappreciated: older migraine preventives often cause weight gain, cognitive fog, or fatigue severe enough that people stop taking them. CGRP antibodies rarely cause those issues.

For people with treatment-resistant chronic migraine, defined loosely as migraine that has not responded to several prior preventive medications, CGRP antibodies still work. Trials in this harder-to-treat population showed reductions in monthly migraine days, lower acute medication use, and in some cases, reversion from chronic to episodic migraine.¹³ A network meta-analysis comparing different antibodies in patients who had failed prior preventives found that ligand-targeting antibodies (those that grab CGRP itself rather than blocking the receptor) had an edge over the receptor-targeting antibody erenumab in reducing monthly migraine days and achieving 50% response rates in this specific population.¹⁴

Safety, Side Effects, and the Constipation Question

One of the biggest selling points of CGRP antagonists is what they do not do. Unlike triptans, which constrict blood vessels, gepants and monoclonal antibodies have no vasoconstrictive activity. This makes them suitable for people with cardiovascular disease, a group that has historically been underserved in migraine treatment because triptans are contraindicated for them.¹⁵ A large study comparing CGRP antibody users to people receiving Botox injections for migraine found no increased risk of cardiovascular events, hypertensive crises, or Raynaud phenomenon in the CGRP antibody group.¹⁶

The most talked-about side effect is constipation. CGRP plays a physiological role in the gut: it helps drive the muscle contractions that push food along and stimulates fluid secretion into the intestinal lumen. Block CGRP systemically, and those functions slow down. Real-world surveys suggest constipation may affect more than half of patients taking CGRP monoclonal antibodies.¹⁷ Clinical trial rates tend to be lower, but the real-world numbers are a better gauge of what patients actually experience. For most people it is manageable with dietary changes or over-the-counter remedies, but for some it is bothersome enough to drive discontinuation. Injection-site reactions with the antibodies and nausea with the gepants round out the common side effects, though neither tends to be severe.

Another practical advantage: gepants appear not to cause medication-overuse headache, the paradoxical worsening of headache that plagues people who take triptans or traditional pain relievers too frequently.¹⁸ This is a meaningful real-world benefit, because medication-overuse headache is one of the most common complications in migraine management and one of the hardest to untangle once it develops.

Pregnancy, Breastfeeding, and Limited Data

This is where the confidence level drops. Migraine disproportionately affects women of reproductive age, so the question of safety during pregnancy comes up constantly. CGRP antibodies are large proteins with long half-lives, and most regulatory agencies recommend stopping them before a planned pregnancy. The evidence base is thin by necessity: pregnant people are excluded from clinical trials.

The best data available comes from VigiBase, the World Health Organization’s global pharmacovigilance database. An analysis of reports involving CGRP antibody exposure during pregnancy found no specific patterns of maternal, fetal, or neonatal toxicity. Spontaneous abortion rates were not disproportionately higher with CGRP antibodies compared to the broader database or to triptans.¹⁹ That is reassuring, but pharmacovigilance databases have significant limitations: reporting is voluntary, outcomes can be incomplete, and the total numbers remain small. No one should interpret this as a green light.

For breastfeeding, data on gepants is beginning to emerge. A phase 1 study of atogepant in lactating women found that only a tiny fraction of the drug made it into breast milk, about 0.015% of the dose. The relative infant dose was well below the threshold typically considered safe.²⁰ The study was small, and larger confirmation is needed, but it offers the first quantitative reassurance for a population that has historically had to choose between treating migraine and breastfeeding.

Predicting Who Will Respond

Not everyone who starts a CGRP antagonist gets meaningful relief. Roughly a third of patients in prevention trials did not reach the 50% reduction threshold even on active drug. Researchers have been hunting for biomarkers that might predict who will benefit. Real-world data suggest that certain migraine features correlate with a better response: unilateral pain (pain on one side of the head), a history of responding to triptans, and the presence of daily headache or medication overuse. These features may be linked to higher baseline levels of circulating CGRP, which would make intuitive sense since people whose migraines are more CGRP-driven should benefit more from blocking it.²¹

When someone fails one CGRP antibody, switching between a ligand-targeting and a receptor-targeting antibody sometimes works. The mechanism is plausible: since the two types interact differently with receptor subtypes and downstream signaling pathways, a person whose biology does not respond to one approach may respond to the other.²² This switch strategy is increasingly common in clinical practice, though the evidence remains at the level of controlled cohort studies rather than large randomized trials.

Beyond Migraine

CGRP does not only matter in migraine. Researchers have tested CGRP antagonists in several other conditions where the same peptide appears to play a role, with mixed results so far.

The most dramatic data come from cluster headache, a condition sometimes called the worst pain a human can experience. A trial of galcanezumab in episodic cluster headache found a significant reduction in weekly attack frequency: about 8.7 fewer attacks per week in the treatment group versus 5.2 with placebo, and 71% of treated patients achieved at least a 50% reduction by week three.²³ Galcanezumab subsequently received FDA approval for episodic cluster headache prevention. However, a follow-up trial in chronic cluster headache, the more relentless form, failed to meet its primary endpoint.²⁴ Why the drug would work in episodic but not chronic cluster headache is not fully understood and remains an active area of investigation.

Rosacea is a more surprising candidate. The persistent facial redness and flushing in rosacea involve vasodilation, and CGRP is a potent vasodilator in facial skin. A nonrandomized controlled trial of erenumab found that blocking the CGRP receptor reduced erythema and flushing in rosacea patients, suggesting the CGRP pathway plays a meaningful role in the condition.²⁵ The trial was small and unblinded, so this is early-stage evidence, but it opens an entirely different therapeutic avenue for a condition that has few targeted treatments.

CGRP also appears to contribute to pain sensitization in neuropathic and inflammatory pain conditions beyond the head. The peptide is upregulated in damaged nerves and contributes to central sensitization that amplifies pain signals.²⁶ Researchers have even developed microneedle patches that deliver a CGRP antagonist peptide through the skin for localized neuropathic pain, though this work remains preclinical.²⁷ Whether systemic CGRP antagonists would help broader pain conditions in humans is an open question, and the answer is complicated by the fact that CGRP has protective functions in the gut, cardiovascular system, and wound healing that could be disrupted by widespread, long-term blockade.

The Cost Problem

For all their clinical advantages, CGRP antagonists are expensive. Monthly antibody injections and gepant prescriptions carry price tags that far exceed older generic preventives like propranolol or amitriptyline. An analysis by the Institute for Clinical and Economic Review found that while CGRP inhibitors deliver genuine health benefits for patients who have failed other preventive therapies, they are likely to exceed commonly cited cost-effectiveness thresholds.²⁸ In practical terms, this means insurers often require step therapy: you have to try and fail cheaper medications before getting coverage for a CGRP antagonist. That process can take months and is a genuine barrier for people who might benefit most from starting these drugs earlier.

The arrival of multiple branded competitors has not done much to bring prices down, though it has expanded the menu of options. Whether biosimilar versions of the monoclonal antibodies or generic versions of the gepants will eventually change the cost equation remains to be seen as patents expire in the coming years. For now, the gap between clinical promise and real-world accessibility is one of the biggest unresolved tensions in migraine treatment.