Peptide basics

Pregnancy and contraception in GLP-1 trials: what the record says

This page is not medical advice and does not tell anyone what to do about pregnancy, contraception or any medication. It contains no dose, schedule, route or contraceptive method. It reports what the trial documents and approved labeling record about who was excluded and what that left unmeasured.

What the protocol actually says

STEP 1's protocol was posted to ClinicalTrials.gov and can be read in full. Its exclusion criteria run to 31 numbered items. Number 30 covers a female who is pregnant, breastfeeding, intends to become pregnant, or is of childbearing potential and not using a highly effective contraceptive method.¹

Exclusion was not the only mechanism. The protocol's list of criteria for discontinuing trial product includes pregnancy at item 5 and intention of becoming pregnant at item 6. A participant who became pregnant during the trial came off treatment.¹

The protocol also required that whether a female participant was of childbearing potential be recorded in the case report form, and required pregnancy testing for those participants at screening and at points through the trial.¹

None of this appears in the abbreviated criteria on the registry's summary page, which mentions pregnancy only in relation to the trial's extension phase. The protocol document is the record that carries the detail.¹²

The requirement fell on one sex

The protocol's contraception appendix states that no contraception measures were required for male participants, on the stated basis that the risk of harm from transfer of semaglutide in seminal fluid was considered unlikely.¹

Female participants of childbearing potential were eligible only if they agreed to use contraception from a defined list, and the appendix sets out at length how childbearing potential and postmenopausal status were to be established.¹

So one sex faced a documentation requirement, a testing requirement and a contraception requirement as a condition of participating, and the other faced none of the three. That asymmetry is a feature of the evidence base, not a footnote to it.¹

The same pattern in the cardiovascular trial

SELECT's registry summary lists the equivalent criterion in public: a female who is pregnant, breastfeeding, intends to become pregnant, or is of childbearing potential and not using a highly effective contraceptive method was excluded.³

That trial also set a minimum age of 45, so its female participants skewed older, but the criterion was still applied.³

What that leaves unmeasured

The consequence is stated on the approved US label. Its pregnancy section records that available pharmacovigilance data and data from clinical trials in pregnant patients are insufficient to establish a drug-associated risk of major birth defects, miscarriage, or adverse maternal or fetal outcomes.⁴

The label also records that a pregnancy exposure registry exists to monitor outcomes in women exposed during pregnancy, which is the mechanism by which data are collected after approval rather than during development.⁴

The lactation section records that there are no data on the presence of semaglutide or its metabolites in human milk, on effects on a breastfed infant, or on milk production.⁴

The same pattern holds for the other approved weight-management product in this class. The tirzepatide label carries its own pregnancy and lactation sections describing what was and was not observed, because its registration programme was built on trials with equivalent exclusions.⁶

The regulator has said this is a problem

FDA issued draft guidance in April 2018 on the scientific and ethical considerations for including pregnant women in clinical trials, aimed at supporting a more informed and balanced approach to gathering data on drug use during pregnancy through judicious inclusion rather than routine exclusion.⁵

That guidance is a draft and states that it contains non-binding recommendations. Its existence establishes that blanket exclusion is recognised as a source of evidence gaps, not that any particular trial should have done otherwise.⁵

The pattern is not specific to one sponsor. The approved US labeling for tirzepatide carries its own pregnancy and lactation sections, reflecting a registration programme built the same way.

Common questions