Spontaneous Abortion Risk Higher With Some Migraine Drugs, U.S. Study Shows

— Anti-CGRP monoclonal antibodies tied to 42% greater miscarriage risk

Women exposed to anti-calcitonin gene-related peptide (CGRP) monoclonal antibodies for migraine prevention early in their pregnancies had a higher risk of spontaneous abortion, retrospective U.S. data showed.

Among 7,100 pregnant migraineurs, anti-CGRP monoclonal antibody exposure was associated with a 42% higher risk of spontaneous abortion compared with no exposure (adjusted HR 1.42, 95% CI 1.12-1.80), reported Anick Bérard, PhD, of the Centre Hospitalier Universitaire Sainte-Justine in Montreal, and colleagues.

Women who used propranolol for migraine prevention did not have this risk (adjusted HR 1.02, 95% CI 0.71-1.47), Bérard and co-authors wrote in JAMA Neurologyopens in a new tab or window.

"Our study is the first large-scale study looking at the safety of anti-CGRP during pregnancy for the prevention of chronic migraine," Bérard told MedPage Today.

"Our findings are in line with the current guidelines on migraine preventive therapies, which recommend discontinuation of anti-CGRP at least 5 months before conception and switching to established migraine preventive therapies such as propranolol," she stated.

CGRP is a potent vasodilator thought to play a significant role in migraine. In 2018, erenumabopens in a new tab or window (Aimovig), an injectable treatment targeting CGRP receptors, was the first monoclonal antibody approved for migraine prevention; fremanezumabopens in a new tab or window (Ajovy), galcanezumab (Emgality), and eptinezumab (Vyepti) soon followed.

CGRP is important in placental development, the researchers observed.

"Safety concerns during pregnancy stem from CGRP's role in vascular regulation, implantation, placental development, and uteroplacental blood flow, and have led to recommendationsopens in a new tab or window to discontinue anti-CGRP monoclonal antibodies 5 to 6 months before conception due to their 1-month half-life," they wrote.

New migraine prevention guidelinesopens in a new tab or window from the American Academy of Neurology and the American Headache Society also note that women with migraine have higher risks of preeclampsia, preterm birth, and cerebrovascular and cardiovascular events during pregnancy. Whether migraine prevention during pregnancy decreases these risks is unknown.

Bérard and co-authors used data from 2018 to 2021 in the AM-PREGNANT cohort, which was built from the Merative MarketScan commercial database. The study included 7,119 women with pregestational migraine with a mean age of 31.6 years who had 7,579 singleton pregnancies.

The researchers assessed outcomes for women treated with one of the four approved anti-CGRP monoclonal antibodies -- erenumab, fremanezumab, galcanezumab, or eptinezumab -- or propranolol. Overall, 6,946 pregnancies were not exposed to any of these drugs; 350 were exposed to anti-CGRP monoclonal antibodies, and 283 were exposed to propranolol. Exposures were defined as one or more prescriptions for the drugs and were modeled as time-varying.

The main outcome measure was spontaneous abortion at 20 weeks or less gestation, based on medical claims. A total of 1,862 (24.6%) spontaneous abortions occurred during the study period, including 100 spontaneous abortions (28.6%) during anti-CGRP exposed person-time and 37 (12.8%) during propranolol-exposed person-time.

Women who used anti-CGRP treatments had greater healthcare use, depression, and concomitant medications, Bérard and colleagues said. A bias analysis to account for residual confounding by migraine severity modestly weakened the relationship between anti-CGRP monoclonal antibodies and spontaneous abortion (HR 1.33), they noted.

MarketScan data are based on a commercially insured U.S. population, and findings may not apply to uninsured or publicly insured individuals, the researchers pointed out. Pregnancy intention was unknown and may have contributed to residual confounding.

The study likely excluded very early pregnancy losses as the outcome definition captured only clinically recognized spontaneous abortions recorded in administrative health data. Exposure misclassification also was possible because prescription records do not confirm medication use, Bérard and colleagues acknowledged.