Migraine is a highly prevalent and disabling disorder, with an estimated 1 in 6 individuals in the United States affected by migraines, and is one of the leading reasons for emergency department visits.1,2 Despite its high prevalence and impact on disability rates, migraine remains underrecognized and undertreated.3 Migraine attacks are characterized by unilateral headache accompanied by a cluster of other sensory, autonomic, and cognitive symptoms that may include photophobia, phonophobia, allodynia, pain upon movement, and nausea/vomiting; different stages marking the clinical course of migraine include the premonitory, aura, headache, and postdromal phases.4 Migraine can occur with or without aura,3 and depending on the days per month a patient is affected, it is classified as episodic (<15 migraine or headache days) or chronic (at least 15 days, with 8 or more as migraine days).5 Developments over the past few decades have led to migraine’s classification as a complex neurological condition rather than a vascular disorder, with multiple cortical, subcortical, and brainstem regions involved in pain initiation and the broad spectrum of symptoms.2,6,7

In the current neurovascular model for migraine pathogenesis, pain is thought to originate from the trigeminovascular system, with secondary changes in cerebral perfusion; upon activation, trigeminal nociceptive neurons on dural blood vessels trigger the release of plasma proteins and vasoactive substances, including calcitonin gene-related peptide (CGRP), substance P, and neurokinin A.2,3 The release of vasoactive factors triggers vasodilation and dural plasma extravasation, resulting in neurogenic inflammation and thereby producing pain. Subsequently, these pain impulses are transmitted along the trigeminovascular system to the trigeminal nucleus caudalis and eventually to the higher cortical pain centers.8 Among the vasoactive factors invoked in migraine pathogenesis, CGRP has been pinpointed as a key mediator in migraine pathophysiology by promising data from clinical studies.3,9-11

Migraine Pathogenesis and Progression: Role of CGRP

Two isoforms of CGRP are present in humans: α-CGRP, abundant mostly in primary spinal afferents from sensory ganglia, and ß-CGRP, found mainly in the enteric nervous system.12 The α-CGRP isoform (hereafter referred to as CGRP), a product of alternative splicing of calcitonin mRNA, is the predominant calcitonin gene product outside the thyroid and coexists with substance P in sensory nerves and neurons.10 The CGRP peptide family, which includes calcitonin, adrenomedullin, and amylin, has been found in the trigeminal system and central nervous system.10 CGRP family receptors consist of a multicomponent receptor system: a calcitonin receptor-like receptor (CLR) that must be linked with 1 of 3 receptor activity–modifying proteins (RAMPs) to generate a functional CGRP receptor and a receptor component protein; the CGRP receptor is composed of CLR and RAMP1.9

CGRP is a 37-amino acid peptide with an N-terminal disulfide bond and amidated C terminus, both of which are required for CGRP receptor activation.13 Structural studies of the CLR–RAMP1 ectodomain complex indicate that small-molecule antagonists of the CGRP pathway block the peptide-binding cleft at the CLR–RAMP1 interface.14 Given its role in migraine pathogenesis, it is perhaps not surprising that CGRP has been shown to regulate the cardiovascular system, mediate neurogenic inflammation, and modulate nociceptive input.13,15 CGRP is expressed in >50% of the trigeminal neurons and in small- to medium-sized nociceptive nerve fibers (C-fibers) in both central and peripheral branches.3,10 Although CGRP is widely expressed in neurons of the cerebral cortex, the hippocampus, the cerebellum, the thalamic nuclei, the hypothalamic nuclei, and the brainstem nuclei, the cognate receptor components colocalize predominantly in some of the trigeminal neurons, satellite glial cells, and vascular smooth muscle cells.9 However, the CGRP peptide and receptor are rarely co-expressed in the same cells; CGRP is expressed/stored in small-to-mid-sized trigeminal neurons, whereas the CGRP receptor is found in large neurons.9

Upon trigeminal nerve activation, CGRP released into the plasma is rapidly degraded (half-life <10 min), suggesting that CGRP-mediated effects are focused on release site–proximal regions.9 CGRP release promotes non–endothelium-mediated vasodilation.16 CGRP released from terminals of C-fibers can sensitize nearby A-fibers, which express functional CLR–RAMP1 CGRP receptors.17

The role of CGRP in migraine pathogenesis and progression is supported by several lines of evidence. First, CGRP levels are elevated in serum between attacks of episodic and chronic migraine13,18,19; this increase is reversed upon treatment with triptans, the selective 5-hydroxytryptamine receptor agonists that are often used as first-line treatment for migraine because of their ability to mitigate pain by reducing neurogenic inflammation.13,19,20 Second, moderate to severe headaches akin to migraine are induced by injection of the CGRP peptide. Notably, induction of delayed migraine-like headaches occurs in migraineurs, but only initial nonmigraine headache symptoms are reported by nonmigraineurs. The final line of evidence is the observation of relief from both the pain and associated symptoms of migraine in clinical studies of selective CGRP receptor antagonists.13 Taken together, these data establish a central role for the CGRP–CGRP receptor axis in migraine pathophysiology (FIGURE21).

The broad distribution pattern of CGRP and CGRP receptors provides numerous targets where the CGRP pathway could impinge on neuromodulation of light aversion, central sensitization, and
cortical spreading depression, processes involved in migraine generation and progression.13 In the periphery, CGRP promotes dilation of vascular beds. CGRP also contributes to neurogenic inflammation and peripheral sensitization of nociceptive neurons.13 Transgenic studies in mice have established a role for CGRP in light aversion, a pathway that is a surrogate for migraine photophobia.22,23 CGRP can also enhance glutamatergic signaling-driven synaptic transmission, thereby promoting central sensitization.13,18

Current and Emerging Anti-CGRP Agents

The evidence implicating CGRP in migraine pathogenesis has led to increased interest in development of CGRP/CGRP receptor antagonists, with data from the first proof-of-concept phase 2 trial (NCT02198339) published in 2004.24,25 This trial, which evaluated olcegepant, established the clinical validity of CGRP as a therapeutic target in treating acute migraine attacks; olcegepant yielded a response rate of 66% compared with 27% with placebo (P = .001).24 Olcegepant was the first of the gepants, small molecule CGRP receptor antagonists, to demonstrate clinical efficacy in migraine management. Many gepants, including rimegepant, telcagepant, and ubrogepant, have since been evaluated in acute migraine treatment. However, the development of some gepants was terminated because of an inability to administer the formulation orally (olcegepant) or increased hepatotoxicity risk (telcagepant, which forms a liver-toxic metabolite not formed by other gepants).26,27 Investigators are currently evaluating 3 gepants—rimegepant, ubrogepant, and atogepant—in late-phase clinical studies.27

In addition to small-molecule CGRP receptor antagonists, CGRP- or CGRP receptor–directed monoclonal antibodies (mAbs) have also gained momentum in migraine therapy.9,27,28 To date, investigators have evaluated 4 antibody-based agents in clinical studies: eptinezumab, galcanezumab, fremanezumab, and erenumab.9,27,28

Gepants in Migraine Treatment

Investigators are conducting or have conducted phase 3 clinical trials to evaluate several gepants for the treatment of migraine: ubrogepant (NCT02828020, NCT02867709) in the acute setting, atogepant (NCT02848326, NCT03700320) in the preventive setting, and rimegepant (NCT03732638) in the preventive setting. Preliminary data29,30 from 2 pivotal phase 3 trials for rimegepant, an oral CGRP receptor antagonist, showed that with a single dose of rimegepant, 19.2% (P <.03 vs placebo) of 543 patients and 19.6% (P <.001 vs placebo) of 537 patients achieved freedom from pain at 2 hours compared with 14.2% of 541 patients and 12.0% of 535 patients who received placebo, respectively.29 Overall, the study authors concluded that a “single dose of rimegepant, without any rescue medications, was superior to placebo for pain freedom and pain relief at 2 hours post-dosing.” Notably, rimegepant-treated patients showed improvement on measures of functional disability, with a greater proportion of patients achieving normal function, and the safety profile of rimegepant was similar to that of placebo, including liver function assessments.30

Preliminary data for ubrogepant, also orally administered, from phase 3 trials have been reported.31,32 In the ACHIEVE I (UBRMD-01) study, 1327 adult patients were randomly assigned 1:1:1 to placebo or ubrogepant (50 mg or 100 mg) and were treated for a single migraine attack of moderate to severe headache intensity. The topline data showed that compared with placebo (11.8%), pain freedom at 2 hours after the initial dose was achieved by more patients randomized to both the 50-mg (19.2%, P = .0023 vs placebo) and 100-mg ubrogepant arms (21.2%, P = .0003 vs placebo).31 Freedom from the most bothersome symptom at 2 hours was also observed in more patients taking ubrogepant at both 50 mg (38.6%, P = .0023 vs placebo) and 100 mg (37.7%, P = .0023 vs placebo) compared with placebo (27.8%).31 In the ACHIEVE II study, 1686 patients were randomized 1:1:1 to placebo or a lower dose of ubrogepant (25 mg or 50 mg) than in ACHIEVE I. Again, the results showed an improvement in pain freedom at 2 hours (20.7% [P = .0285 vs placebo] in the 25-mg group, 21.8% [P = .0129 vs placebo] in the 50-mg group, and 14.3% in the placebo group). Although the percentage of patients with freedom from the most bothersome symptom at 2 hours was higher for both ubrogepant doses compared with placebo, this difference failed to achieve statistical significance for the 25-mg dose.32 No hepatotoxicity was observed in either trial.

Preliminary phase 2/3 data from a clinical trial of atogepant, another oral gepant, indicated a significant reduction in mean monthly migraine days across all doses (10 mg once daily to 60 mg twice daily) compared with placebo. In addition, no concerns regarding hepatic safety were noted.33 Forthcoming phase 3 results for atogepant, along with results from studies of other gepants in acute and prophylactic treatment, will help further clarify the role of gepants in the migraine treatment paradigm.

CGRP/CGRP Receptor–Targeted mAbs in Migraine Therapy

Therapeutic mAbs targeting CGRP are being developed for the treatment of migraine; of these, 3—galcanezumab (Emgality; Eli Lilly), fremanezumab (Ajovy; Teva), and erenumab (Aimovig; Amgen, Novartis)—have gained FDA approval for migraine treatment (TABLE).27, 34-42

Erenumab, a fully human immunoglobulin (Ig) G2 mAb, was the first CGRP receptor–targeted mAb approved for migraine.34,43 Data from 2 phase 3 studies, ARISE and STRIVE, established its efficacy and safety in migraine prevention. In the ARISE study, which included 577 patients randomized to monthly injections of 70 mg erenumab or placebo, 39.7% (vs 29.5% in placebo, P = .010) achieved a ≥50% reduction in monthly migraine days.44 In STRIVE, 955 patients were randomized to monthly injections of erenumab 70 mg, erenumab 140 mg, or placebo; data showed a ≥50% reduction in monthly migraine days in 43.3% of patients in the 70-mg group (P <.001 vs placebo), 50.0% in the 140-mg group (P <.001 vs placebo), and 26.6% in the placebo group.45 Adverse event profiles for erenumab and placebo were comparable in both studies.

Galcanezumab, which binds both CGRP isoforms, recently gained FDA approval for preventive migraine treatment.28,37 Results from 2 phase 3 clinical trials (EVOLVE-1, 862 patients randomized 1:1:2 to galcanezumab 120 mg, 240 mg, or placebo; and EVOLVE-2, 915 patients randomized 1:1:2 to galcanezumab 120 mg, 240 mg, or placebo) showed a ≥50% reduction in monthly headache days in 62.3% (P <.001 vs placebo) of patients in the 120-mg group, 60.9% (P <.001 vs placebo) in the 240-mg group, and 38.6% in the placebo group (EVOLVE-1).46 In EVOLVE-2, a 59% reduction (P <.001 vs placebo) was observed in the 120-mg group, 57% (P <.001 vs placebo) in the 240-mg group, and 36% in the placebo group.47 Although treatment-emergent antidrug antibodies were seen in a small percentage of patients, no effects on safety or tolerability were noted in either study.27,47

Fremanezumab, a humanized IgG2 mAb that binds both isoforms of CGRP, was also recently FDA approved for preventive treatment of migraine.39,48 In a phase 3 trial (NCT02621931), 376 patients were randomly assigned to fremanezumab quarterly, 379 to monthly, and 375 to placebo. Results showed that 41% (P <.001 vs placebo) and 38% (P <.001 vs placebo) of patients in the monthly and quarterly groups, respectively, achieved a ≥50% reduction in monthly headache days compared with 18% in the placebo group.49 In another phase 3 study, 875 patients with episodic migraine were randomized to monthly low-dose 225-mg injections, quarterly high-dose 675-mg injections, or placebo. Results showed that 47.7% (P <.001 vs placebo) and 44.4% (P <.001 vs placebo) of patients in the monthly and quarterly groups, respectively, achieved a ≥50% reduction in monthly migraine days compared with 27.9% in the placebo group. A small percentage (1.4%) of patients developed antidrug antibodies in the low-dose monthly group.50

Eptinezumab, a humanized IgG1 mAb that binds to both CGRP isoforms,28 was evaluated in 2 completed phase 3 trials, PROMISE 1 (NCT02559895; 888 patients randomized to receive eptinezumab 30 mg, 100 mg, 300 mg or placebo infusions once every 12 weeks)40,51 and PROMISE 2 (NCT02974153; 1072 patients randomized to eptinezumab 100 mg, 300 mg, or placebo),41,52 as well as a completed open-label safety study (PREVAIL/NCT02985398).53 Preliminary data from the PROMISE 1 study showed that for weeks 1 to 12, 49.8% (P = .0085 vs placebo) and 56.3% (P = .0001 vs placebo) of patients in the 100-mg and 300-mg groups, respectively, achieved a ≥50% reduction in monthly migraine days compared with 37.4% of patients in the placebo group.51 For months 6 to 12, 70.7% of patients had a ≥50% reduction in monthly migraine days compared with 58.7% in the placebo group.54 In PROMISE 2, 61% (P <.0001 vs placebo) of patients achieved a ≥50% reduction in monthly migraine days in weeks 1 to 12 compared with 39% in the placebo group.52


The development and approval of mAbs targeting CGRP have shifted the migraine treatment paradigm, driven by high response rates and favorable safety profiles. Although clinical studies of CGRP pathway–targeted mAb-based agents in migraine have yielded ≥50% migraine reduction in a significant proportion of patients, subpopulations of patients achieved even higher migraine reduction (≥75% in some cases), indicating that these therapeutic agents have great potential yet to be explored and clarified in acute and prophylactic migraine treatment.27 Along with the data for gepants, this new class of drugs has provided much-needed options over the triptan-based mainstay modalities, ushering in a new era in migraine management.