Trigeminovascular biology and CGRP¶
TL;DR — Migraine headache is best modelled as brain-state change coupled to trigeminovascular nociception, not as primary pathological vasodilation. Trigeminal afferents innervating meninges and vessels converge in the trigeminocervical complex, ascend through thalamic and cortical networks and release CGRP peripherally and centrally (Ashina 2019, PMID 31160203; Iyengar 2019, PMID 30982963). CGRP infusion can provoke migraine-like attacks in susceptible people, and antagonizing the peptide or receptor treats attacks and prevents them; that human provocation-to-therapy chain is unusually strong mechanistic evidence (Christensen 2018, PMID 30409109). Yet CGRP is neither necessary in every attack nor a validated response biomarker, and large antibodies appear to act mainly outside an intact blood–brain barrier (Noseda 2020, PMID 31856583). Hypothalamic and brainstem activity across the premonitory phase suggests that trigeminal activation is one component of a cycling network disorder rather than a single “generator” (Schulte 2016, PMID 27190019).
From vascular theory to trigeminovascular system¶
The older vascular model treated extracranial vasodilation as the proximate cause of throbbing pain. It could not explain premonitory symptoms, sensory amplification, aura without headache or efficacy that dissociates from vasoconstriction. The trigeminovascular hypothesis instead centers sensory afferents that innervate cranial vessels and dura, their cell bodies in the trigeminal ganglion, second-order neurons in trigeminal nucleus caudalis/upper cervical cord, and ascending projections to thalamus and cortex (Ashina 2019, PMID 31160203).
| Level | Main elements | Evidence contribution | Limitation |
|---|---|---|---|
| Meninges/dural vessels | Aδ and C-fiber nociceptors, mast cells, vascular and immune signals | Single-unit recording and pharmacology | Mostly animal preparations |
| Trigeminal ganglion | Sensory neuron somata, satellite glia, CGRP expression | Molecular target localization | Ganglion biology does not specify attack initiation |
| Trigeminocervical complex | Convergence of cranial and cervical inputs | Explains referred neck/cranial pain | Central recordings are invasive animal data |
| Thalamus/cortex | Sensory-discriminative and affective integration | Human imaging and allodynia | Signals are distributed and not migraine-specific |
| Hypothalamus/brainstem | Homeostatic, autonomic and descending-control networks | Premonitory and cycle imaging | No single stable “generator” demonstrated |
Dural nociceptors have mechanical and chemical response properties capable of encoding intracranial tissue disturbance (Strassman 2006, PMID 16492942). Activation and sensitization are distinct: activation increases ongoing firing, whereas sensitization lowers thresholds or increases responses to mechanical stimuli such as vessel pulsation.
CGRP biology¶
CGRP is a 37-amino-acid neuropeptide widely expressed in small- and medium-diameter trigeminal neurons. Canonical signaling involves the calcitonin receptor-like receptor with receptor activity-modifying protein 1 and intracellular cAMP pathways. Its distribution spans peripheral terminals, ganglion and central trigeminal pathways (Raddant 2011, PMID 22123247; Karsan 2015, PMID 25887765).
| Observation | What it supports | What it does not prove |
|---|---|---|
| CGRP is released during some attacks and after trigeminal activation | Participation in attack signaling | That circulating level diagnoses an individual attack |
| Intravenous CGRP provokes delayed migraine-like attacks | Susceptibility of migraine biology to CGRP | That spontaneous attacks always begin with CGRP |
| Gepants treat attacks | Receptor signaling is actionable acutely | Exact anatomical site of action |
| Ligand/receptor antibodies prevent migraine | Sustained peripheral blockade changes attack probability | That central CGRP has no role |
| Nonresponders exist | Pathway heterogeneity or incomplete blockade | A clean CGRP-negative subtype |
CGRP also dilates vessels, but vasodilation is one action rather than the whole disease mechanism. Human attack-release, provocation and antagonist evidence supports an actionable CGRP pathway without reducing migraine to vascular dilation (Raddant 2011, PMID 22123247; Edvinsson 2015, PMID 25790955). It facilitates nociceptive transmission, neuroimmune interactions and peripheral/central modulation. Sex-steroid interactions with CGRP signaling are biologically plausible and supported by experimental literature, but the mechanisms behind the two- to threefold female prevalence difference remain incomplete (Labastida-Ramírez 2019, PMID 29082826).
Human provocation as a causal bridge¶
Provocation models administer endogenous signaling molecules and compare delayed migraine-like attacks with placebo. CGRP and PACAP38 both trigger attacks in subsets of people with migraine, creating a within-person experimental phenotype (Guo 2017, PMID 28260603).
In 13 previous erenumab-trial participants, a randomized crossover CGRP infusion study explored whether provocation susceptibility tracked antibody response; the small sample made the association exploratory rather than a clinical selector (Christensen 2018, PMID 30409109). A larger 139-person study then tested whether CGRP-induced attacks predicted 24-week erenumab effectiveness. Hypersensitivity carried predictive information but was not sufficiently established as a routine biomarker (Al-Khazali 2024, PMID 38859744).
Provocation has three interpretation constraints:
- intravenous exposure is not the same spatial or temporal signal as a spontaneous attack;
- attack definitions and expectation can influence endpoints;
- responders/nonresponders are graded, not perfectly separable groups.
The strongest inference is that CGRP signaling is sufficient to raise attack probability in susceptible systems and that blocking it is therapeutic—not that a plasma concentration is the disease.
Peripheral sensitization¶
Inflammatory mediators, nitric oxide signaling and tissue interactions can sensitize meningeal afferents in animals. Dural mast-cell mediators activate and sensitize nociceptors (Zhang 2007, PMID 17483291); TNF-α sensitization involves local cyclooxygenase and p38 MAP-kinase actions (Zhang 2011, PMID 21036476); and nitroglycerin-related signaling can increase meningeal mechanosensitivity (Zhang 2013, PMID 23447360).
| Model | Endpoint | Translational caution |
|---|---|---|
| Dural inflammatory mediators | Trigeminal firing, mechanical threshold | Supraphysiological mixture and exposed dura |
| Nitroglycerin/glyceryl trinitrate | Delayed allodynia or afferent sensitization | Systemic NO donor has many targets |
| Mast-cell degranulation | Persistent meningeal sensitization | Human migraine is not established as a mast-cell disorder |
| Cortical spreading depolarization | Trigeminal activation in some preparations | Links aura and pain incompletely |
| CGRP infusion | Human migraine-like attack | Provoked susceptibility is not spontaneous initiation |
TRPA1/NOX signaling in trigeminal ganglion neurons mediates glyceryl-trinitrate allodynia in mice, offering a molecular route from oxidative stress to pain behaviour (Marone 2018, PMID 29985973). “Sterile neurogenic inflammation” is a useful umbrella for neuropeptide, vascular and immune interactions, but classic plasma-extravasation programs failed clinically; the term should not imply tissue-destructive inflammation (Ramachandran 2018, PMID 29568973).
Naproxen can inhibit meningeal nociceptor sensitization in animal recording, consistent with a peripheral cyclooxygenase contribution to acute efficacy (Levy 2008, PMID 18333963). Timing matters: once central trigeminovascular neurons are sensitized, purely peripheral interruption may be less complete, although a rigid “treat within 20 minutes” rule cannot be generalized from animal physiology (Burstein 2011, PMID 22348935).
Central sensitization and allodynia¶
Cutaneous allodynia during an attack reflects amplified central processing: normally non-painful scalp or facial stimuli become painful. Second-order trigeminal neurons can enlarge receptive fields and increase responses, and thalamic sensitization may extend allodynia beyond the head (Burstein 2011, PMID 22348935).
Central sensitization is a state description, not a complete cause of migraine. It explains the evolution and persistence of an attack more directly than why an attack began. It also overlaps other pain disorders, limiting specificity as a biomarker (Ashina 2021, PMID 33773610).
Where antibodies act¶
Monoclonal antibodies are large molecules with limited penetration across an intact blood–brain barrier. Fluorescent fremanezumab in rats distributed to dura and sensory/autonomic ganglia but not brain, supporting a mainly peripheral locus (Noseda 2020, PMID 31856583). Reviews converge on accessible sites including dura, trigeminal ganglion and vascular compartments (González-Hernández 2020, PMID 32552657).
Peripheral action can still change central activity by reducing incoming traffic. An fMRI study found treatment-associated changes in trigeminal and hypothalamic responses that differed between galcanezumab responders and nonresponders; this is compatible with downstream network adaptation, not proof that antibody entered brain (Basedau 2022, PMID 35604755).
| Drug type | Target | Likely access | Functional distinction |
|---|---|---|---|
| Erenumab | CGRP receptor | Predominantly peripheral | Blocks receptor signaling from CGRP |
| Fremanezumab/galcanezumab/eptinezumab | CGRP ligand | Predominantly peripheral | Sequesters peptide before receptor activation |
| Gepants | CGRP receptor | Small molecules; peripheral and some central access varies | Rapid, reversible blockade suitable for acute/preventive use |
| Triptans | 5-HT1B/1D | Neural and vascular targets | Inhibit trigeminal transmission/release plus vasoconstrictive pharmacology |
| Lasmiditan | 5-HT1F | Neural target without 5-HT1B vasoconstriction | Inhibits trigeminovascular signaling (de Vries 2020, PMID 32173558) |
Premonitory networks and the “generator” problem¶
Premonitory symptoms—yawning, fatigue, cognitive or mood change, thirst, urinary frequency and neck stiffness—may begin hours to days before pain (Karsan 2018, PMID 30074545). Retrospective prevalence estimates span 7–88%, showing severe ascertainment heterogeneity (Maniyar 2015, PMID 25919990).
Imaging during spontaneous or experimentally triggered premonitory phases implicates hypothalamic, brainstem and sensory networks, but studies are small and timing is difficult (Karsan 2020, PMID 32269547). Continuous scanning over 30 days and three spontaneous attacks found changing hypothalamic connectivity across the cycle and brainstem activation during pain, arguing against one static generator (Schulte 2016, PMID 27190019). The attack is better represented as a dynamic network transition spanning premonitory, aura, pain and postdrome phases (Charles 2013, PMID 23278169).
What the CGRP success does—and does not—settle¶
The translational chain settles that CGRP-pathway signaling is an effective target. It does not settle:
- the upstream event that changes susceptibility on a particular day;
- whether different CGRP therapies share identical response biology;
- why many patients respond partially or not at all;
- how aura and pain couple in attacks where either can occur alone;
- whether long-term blockade changes disease biology or only suppresses attacks.
REFORM is collecting clinical, blood, structural-MRI and functional-MRI variables to predict erenumab response, illustrating the move from target validation to patient selection (Karlsson 2023, PMID 37303034). No selector is yet validated for routine choice.
Open questions¶
- What upstream state makes the trigeminovascular system CGRP-sensitive on one day and quiescent on another? (Schulte 2016, PMID 27190019; Ashina 2021, PMID 33773610)
- Can provoked CGRP hypersensitivity predict net benefit sufficiently well to justify an infusion test? (Christensen 2018, PMID 30409109; Al-Khazali 2024, PMID 38859744)
- Which accessible peripheral compartment—dura, ganglion, vascular wall or immune interface—dominates antibody efficacy in humans? (Noseda 2020, PMID 31856583)
- Are ligand- and receptor-targeted antibodies biologically interchangeable after clinical nonresponse? (Basedau 2022, PMID 35604755)
- Which sensitization findings reproduce in human tissue or non-invasive physiology rather than exposed-dura animal preparations? (Levy 2008, PMID 18333963; Ramachandran 2018, PMID 29568973)
Related pages¶
- Aura and cortical spreading depolarization — cortical events that may couple to trigeminal signaling.
- Acute treatment — gepants, triptans and ditans as pathway probes.
- Preventive treatment — sustained CGRP blockade and clinical response.
- Biomarkers — provocation, blood and imaging candidates.
- Sex, hormones, pregnancy and lactation — hormone–CGRP interactions and reproductive evidence.
References¶
- Ashina M, et al. Migraine: disease characterisation, biomarkers, and precision medicine. Lancet. 2021. PMID 33773610
- Ashina M, et al. Migraine and the trigeminovascular system—40 years and counting. Lancet Neurol. 2019. PMID 31160203
- Iyengar S, et al. CGRP and the trigeminal system in migraine. Headache. 2019. PMID 30982963
- Karsan N, et al. CGRP mechanism antagonists and migraine management. Curr Neurol Neurosci Rep. 2015. PMID 25790955
- Karsan N, et al. Calcitonin gene-related peptide and migraine. Curr Opin Neurol. 2015. PMID 25887765
- Raddant AC, Russo AF. Calcitonin gene-related peptide in migraine: intersection of peripheral inflammation and central modulation. Expert Rev Mol Med. 2011. PMID 22123247
- Labastida-Ramírez A, et al. Gender aspects of CGRP in migraine. Cephalalgia. 2019. PMID 29082826
- Guo S, et al. The role of genetics on migraine induction triggered by CGRP and PACAP38. Dan Med J. 2017. PMID 28260603
- Christensen CE, et al. Migraine induction with CGRP in patients from erenumab trials. J Headache Pain. 2018. PMID 30409109
- Al-Khazali HM, et al. Hypersensitivity to CGRP as a predictive biomarker of migraine prevention with erenumab. Cephalalgia. 2024. PMID 38859744
- Karlsson WK, et al. The Registry for Migraine (REFORM) study. J Headache Pain. 2023. PMID 37303034
- Strassman AM, Levy D. Response properties of dural nociceptors in relation to headache. J Neurophysiol. 2006. PMID 16492942
- Zhang X, et al. Sensitization and activation of intracranial meningeal nociceptors by mast cell mediators. J Pharmacol Exp Ther. 2007. PMID 17483291
- Zhang X, et al. TNF-α induces sensitization of meningeal nociceptors via local COX and p38 MAP kinase actions. Pain. 2011. PMID 21036476
- Zhang X, et al. Vascular ERK mediates migraine-related sensitization of meningeal nociceptors. Ann Neurol. 2013. PMID 23447360
- Marone IM, et al. TRPA1/NOX in trigeminal ganglion neurons mediates glyceryl-trinitrate migraine-related pain in mice. Brain. 2018. PMID 29985973
- Ramachandran R. Neurogenic inflammation and its role in migraine. Semin Immunopathol. 2018. PMID 29568973
- Levy D, et al. Sensitization of meningeal nociceptors: inhibition by naproxen. Eur J Neurosci. 2008. PMID 18333963
- Burstein R, et al. The science of migraine. J Vestib Res. 2011. PMID 22348935
- Noseda R, et al. Fluorescently-labelled fremanezumab distributes to ganglia and dura but not brain in rats with intact BBB. Cephalalgia. 2020. PMID 31856583
- González-Hernández A, et al. Locus of action of CGRPergic monoclonal antibodies: peripheral over central mechanisms. CNS Neurol Disord Drug Targets. 2020. PMID 32552657
- Basedau H, et al. Migraine monoclonal antibodies against CGRP change brain activity depending on ligand or receptor target. eLife. 2022. PMID 35604755
- de Vries T, et al. Pharmacological treatment of migraine: CGRP and 5-HT beyond the triptans. Pharmacol Ther. 2020. PMID 32173558
- Karsan N, Goadsby PJ. The migraine premonitory phase. Continuum. 2018. PMID 30074545
- Karsan N, Goadsby PJ. Imaging the premonitory phase of migraine. Front Neurol. 2020. PMID 32269547
- Maniyar FH, Goadsby PJ. The premonitory phase of migraine—what can we learn from it? Headache. 2015. PMID 25919990
- Charles A. The evolution of a migraine attack. Headache. 2013. PMID 23278169
- Schulte LH, May A. The migraine generator revisited: continuous scanning over 30 days and three spontaneous attacks. Brain. 2016. PMID 27190019