Early pregnancy exposure to anti-CGRP antibodies was associated with higher miscarriage risk

Design

Retrospective cohort using US commercial claims, time-varying exposure

Population

7,579 singleton pregnancies in women with pre-existing migraine

Primary outcome

Spontaneous abortion at or before 20 weeks

Effect

Anti-CGRP aHR 1.42 (1.12–1.80); propranolol aHR 1.02 (0.71–1.47)

This retrospective cohort used US commercial insurance claims from 2018 to 2021 to identify 7,579 singleton pregnancies in 7,119 women with migraine diagnosed before pregnancy. Exposure to anti-CGRP monoclonal antibodies (erenumab, fremanezumab, galcanezumab, eptinezumab) or propranolol was modelled as time-varying, and pregnancies were followed to 20 weeks.

There were 1,862 spontaneous abortions overall. Compared with no exposure, anti-CGRP antibody exposure was associated with a higher risk (adjusted HR 1.42, 95% CI 1.12–1.80), and the estimate held after bias adjustment (HR 1.33). Propranolol showed no association (aHR 1.02, 0.71–1.47). The concern is biologically plausible, because CGRP has a role in placental development, and these antibodies have half-lives of about a month, so exposure can continue for months after the last dose.

This is observational claims data, with likely misclassification of both exposure and early pregnancy loss, and women prescribed antibodies may have more severe migraine and more comorbidity. It does not prove harm. But it is enough to make contraception and pregnancy planning an explicit part of starting these drugs, and to favour a preventive with a longer pregnancy record when pregnancy is planned.

• Discuss contraception before starting an anti-CGRP antibody in anyone who could become pregnant.
• When pregnancy is planned, consider switching to a preventive with more pregnancy data, such as propranolol.
• The long half-life means exposure can persist for months after the last dose; plan stopping well ahead.
• If pregnancy occurs on an anti-CGRP antibody, stop it and refer to antenatal care; this finding does not by itself justify ending a pregnancy.
• Report pregnancy exposures to a pregnancy registry where one exists.

Why it matters

It is the first sizeable signal that these drugs may not be benign in early pregnancy.

Don't overread it

This is an association from claims data; it does not establish that the antibodies cause miscarriage.

The statistics, in plain English

An adjusted hazard ratio of 1.42 means about 40% higher risk of miscarriage during exposed time compared with unexposed time, after adjusting for measured factors. The bias analysis estimated how much unmeasured confounding might shift this and still found a raised risk, but it cannot rule out every bias.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.