Migraine CGRP Inhibitors Tied to Modest Cardiovascular Events

Initiating migraine treatment with a calcitonin gene-related peptide (CGRP) inhibitor was associated with a modest increase in cardiovascular event risk in a retrospective U.S. cohort study.

"In theory, CGRP inhibitors may have an adverse cardiovascular risk profile, because CGRP — under normal physiology — may confer benefits with respect to blood pressure reduction and the adaptive response to vascular injury," explained Brian MacGrory, MBBCh, of Duke University in Durham, North Carolina, and co-authors in Neurology. "This study provides Class II evidence that in patients with migraine, initiation of a CGRP inhibitor was associated with a modestly increased risk of a composite of cardiovascular events."

The primary end point was a composite of time to cerebral ischemic stroke, myocardial infarction (MI), revascularization, central retinal artery occlusion (CRAO), or peripheral arterial disease (PAD).

The analysis used claims data from the MarketScan proprietary insurance-based registry for 900,370 migraineurs (median age 41, 77.8% women), excluding those with prevalent cardiovascular disease. Overall, 58,679 people started a CGRP inhibitor: 41,871 started a small-molecule CGRP (rimegepant, ubrogepant, zavegepant, or atogepant) and 16,808 started a monoclonal antibody (erenumab, eptinezumab, fremanezumab, or galcanezumab).

"The study period was 2018–2023, chosen to correspond with the period during which CGRP inhibitors became available in the United States," the authors noted.

The researchers reported:

• Overlap-weighted analysis showed that the CGRP inhibitor treatment group had a higher rate of meeting the primary end point composite, with 8.77 versus 6.76 events/1,000 person years and an adjusted HR of 1.26 (95% CI 1.10–1.45).
• Secondary end points for each component of the composite individually, and for intracranial hemorrhage (ICH), found CGRP inhibitor treatment was associated with a higher rate of ischemic stroke (adjusted HR 1.26, 95% CI 1.07–1.49). No significant association was seen for the other composite components or for ICH.
• There was no association between starting a CGRP antagonist and a falsification end point of humeral fracture, making residual confounding less likely.

Although migraine patients with prevalent cardiovascular disease were excluded from the study, "those initiating a CGRP inhibitor had a substantially higher burden of cardiovascular risk factors (including hypertension, hypercholesterolemia, and diabetes mellitus), a higher likelihood of having migraine with aura, and a higher tendency to overall health care utilization," the authors noted.

"Thus, it is strongly likely that, in real-world practice, clinicians are correctly selecting out participants at higher cardiovascular risk for CGRP inhibitor therapy and, although these parameters were included in deriving propensity scores, it remains probable that unmeasured elements of cardiovascular risk may explain the observed association," they wrote.

Primary study results were similar in stratified analysis comparing abortive versus preventive CGRP inhibitors and when results were analyzed per protocol, as opposed to the main intention-to-treat analysis.

"The low absolute magnitude of this association adds to the body of literature, suggesting that CGRP inhibitors can be deployed safely in people with migraine, even when there are preexisting vascular risk factors," MacGrory and co-authors

In an accompanying editorial, Samuel Terman, MD, of the University of Michigan in Ann Arbor, noted that the 7% CGRP antagonist use rate in the cohort indicates widespread use that is in line with a 2024 American Headache Society position statement on CGRP-targeting therapies and asked, "How should we use these data in practice?"

"Although [the authors] did find a statistically significant increase in cardiovascular events, their interpretation was that the absolute magnitude was small," he wrote. "Thus, this finding alone may not be clinically significant enough to dissuade most patients or clinicians."

Terman proposed that clinicians "set appropriate expectations that 100% migraine reduction is not impossible but is uncommon, that headache reductions are felt to be meaningful although they are still realistically on average within approximately 1-3 fewer acute migraine–specific medication days per month compared with placebo, and that branded biologics present extremely high societal costs despite the availability of many considerably cheaper generic alternatives."

The key message from the study is that, while all medications have potentially rare serious and common minor adverse effects, "weighing the complex pros and cons ultimately requires an individualized risk-benefit discussion regarding how much potential benefit is enough, what side effects are common and severe enough to change decisions for an individual patient, and at what price tag," Terman stated.

A 2023 study of the FDA adverse event reporting database for CGRP monoclonal antibodies concluded that underlying safety signals varied, particularly in the cardiovascular system. A 2025 study comparing CGRP-targeted treatment versus onabotulinum toxin injection found similar cardiovascular risk for the two groups.

Previous analyses that have not shown an association between a specific CGRP-targeting treatment or the class include pooled trial data on galcanezumab and fremanezumab.

The current study excluded patients with prevalent cardiovascular disease, who may have higher cardiovascular risk. Data on pain frequency and intensity were not included, and misclassification in administrative data is possible.

The authors also noted that "in general, migraine is not well captured in administrative claims data because the most commonly used algorithms exhibit high specificity but low sensitivity. However, in this study, we found that over 90% of those beneficiaries who initiated a CGRP inhibitor had a billing claim related to migraine before that prescription fill, likely a function of insurance reimbursement practices for this newer drug class."

This study was funded by the American Heart Association and the Duke University Office of the Provost.

MacGrory was supported by the NIH, the American Heart Association, the Duke Office of Physician Scientist Development, Duke Bass Connections, and the Duke University Office of the Provost.

Terman reported relationships with the NIH, Jazz Pharmaceuticals, and Xenon Pharmaceuticals.

Lusk JB, et al "Calcitonin gene-related peptide inhibitors and cardiovascular events in patients with migraine: a retrospective, observational cohort study" Neurology 2026; DOI: 10.1212/WNL.0000000000214479.

Terman SW "Heart of the matter: calcitonin gene-related peptide antagonists and cardiovascular risk" Neurology 2026; DOI: 10.1212/WNL.0000000000214584.