Long-held assumptions about calcitonin gene-related peptide (CGRP) as a migraine biomarker are being challenged after researchers found that plasma CGRP levels were lower — not higher — in individuals with migraine than in healthy individuals.

In a large, cross-sectional study of 588 adults with migraine and 147 matched healthy adults, median plasma CGRP levels were significantly lower in the migraine group (125 vs 151 pmol/L; P < .001). CGRP levels also showed no significant differences by migraine subtype, ictal vs interictal status, or preventive medication use.

“Peripheral plasma CGRP does not appear to behave as many in the field have long assumed,” study investigator Hakan Ashina, MD, PhD, associate professor of neurology, Copenhagen University Hospital – Rigshospitalet and Technical University Hospital of Greater Copenhagen in Copenhagen, Denmark, told Medscape Medical News. “CGRP remains key to migraine biology and is a validated therapeutic target, but our data suggest that this does not translate into elevated circulating plasma CGRP in people with migraine.”

The study was published online on June 12 in Neurology.

Inconsistent Findings

CGRP has long been considered a central neuropeptide in migraine biology because of its role in pain signaling within the trigeminovascular system. Early research in the 1990s reported elevated CGRP levels during migraine attacks, giving rise to the widely held belief that plasma CGRP reflects disease activity.

However, those studies were small and relied on older assay methods that are now considered methodologically limited.

More recent studies have produced inconsistent findings, with some reporting elevated CGRP levels during migraine attacks and others showing no difference — or even lower levels — in those with migraine.

As a result, uncertainty has persisted over whether plasma CGRP accurately reflects trigeminovascular activity or can serve as a reliable biomarker of migraine.

To address these discrepancies, investigators conducted a single-center, cross-sectional observational study using data from the Registry for Migraine in Denmark collected between 2020 and 2022. 

The migraine and control groups had a mean age of 43.1 and 41.5 years, respectively, and were predominantly women (88.8% and 85.7%). Participants were matched in a 4:1 ratio on age, sex, BMI, and sample storage duration before analysis.

Among participants with migraine, 64.6% had chronic migraine and 29.3% had migraine with aura, with a median disease duration of 22 years. Nearly half were receiving preventive migraine medication at enrollment, including 12.9% treated with onabotulinumtoxinA. At the time of blood sampling, 45.3% were experiencing a migraine attack, 35.9% were headache-free, and 18.8% had a nonmigraine headache.

Plasma CGRP was measured twice for each sample using a validated high-affinity radioimmunoassay specific for human alpha-CGRP.

The primary outcome was the difference in plasma CGRP concentration between migraine participants and healthy control participants. Secondary analyses assessed differences across clinical subgroups and associations with demographic and clinical variables using nonparametric testing and multivariable regression modeling.

Long-Held Assumptions Challenged

Plasma CGRP levels were lower in participants with migraine than healthy participants, with median values of 125 pmol/L vs 151 pmol/L, (P < .001). Mean values showed a similar pattern, at 131±85 pmol/L in the migraine group compared with 164±67 pmol/L in the control group.

Chronic migraine and episodic migraine did not differ significantly from one another, and both showed similar reductions to control individuals (128 pmol/L vs 118 pmol/L). Migraine with aura and migraine without aura demonstrated similar CGRP levels (125 pmol/L vs 124 pmol/L), with no statistically significant differences after correction for multiple comparisons.

Neither disease state at the time of blood sampling nor preventive treatment influenced plasma CGRP levels. Median CGRP concentrations were similar during ictal and interictal migraine (116 vs 133 pmol/L) and in participants receiving vs not receiving preventive therapy (125 vs 124 pmol/L).

Key limitations of the study included its cross-sectional design, which did not assess changes in plasma CGRP levels over time, and a cohort comprised predominantly of women recruited largely from specialized care, which may limit the generalizability of the findings to broader migraine populations.

The study’s findings do not contradict the therapeutic importance of CGRP, Ashina said. Rather, they suggest that CGRP signaling may be highly localized within meningeal compartments and therefore not be reflected in venous plasma.

“A molecule can be mechanistically important and therapeutically actionable without being a useful peripheral blood biomarker,” he said.

Ashina said the results suggest future biomarker research should move beyond single peripheral analytes and focus on longitudinal sampling strategies and multimodal approaches incorporating biochemical, clinical, and genetic data.

CGRP’s Limited Utility

The findings align with current understanding of CGRP biology and underscore its limited utility as a blood-based biomarker, said Hope O’Brien, MD, adjunct associate professor at Morehouse School of Medicine in Atlanta, who was not involved in the study.

“What we know about CGRP is that it acts locally at perivascular trigeminal terminals, causing vasodilation and facilitating pain transmission in a spatially restricted microenvironment,” O’Brien, who is also the medical director at the Headache Center of Hope in Kenwood, Ohio, told Medscape Medical News.

Although the lower plasma CGRP levels observed in participants with migraine are not easily explained, chronic trigeminovascular activation could potentially deplete peptide stores over time, O’Brien added.

She noted that the study reinforces an important clinical point: “The fact that blocking CGRP works doesn’t mean measuring CGRP in blood tells us who has migraine, how severe it is, or how well treatment is working.”

Disclosure information for study authors is available in the original study publication. O’Brien reported having no relevant financial disclosures.