Tirzepatide shows larger drops in pill progestin levels than semaglutide, review finds

A review of how incretin-based diabetes and weight drugs interact with birth control pills finds the two drug classes do not behave the same way, with tirzepatide (Mounjaro, Zepbound) producing larger changes in one contraceptive hormone than semaglutide or dulaglutide ^[1].

The paper ran in the May 2026 supplement of Obstetrics & Gynecology for the American College of Obstetricians and Gynecologists' annual meeting. Its authors followed PRISMA reporting standards and searched PubMed, EMBASE, Web of Science and ClinicalTrials.gov, adding FDA reviews and package inserts to the search. Out of 372 records, 10 studies covering 305 people qualified: seven randomized crossover trials and three FDA reviews. Healthy premenopausal volunteers made up most of the sample, and four of the studies looked at postmenopausal women ^[1].

Investigators tracked three pharmacokinetic measures for the hormones inside combined oral contraceptives: total exposure (area under the curve, or AUC), peak blood level (Cmax), and the time needed to reach that peak (Tmax) ^[1].

Ethinyl estradiol, the estrogen used in most combined pills, stayed steady across the studies. The progestin side of the pill was different. With semaglutide (Ozempic, Wegovy, Rybelsus) and dulaglutide, levonorgestrel AUC barely moved, landing somewhere between an 8% decrease and a 5% increase. Tirzepatide, lixisenatide and exenatide taken right before the contraceptive cut levonorgestrel peak concentration by 45% to 66% and total exposure by 18% to 23% ^[1].

Pooling the timing data, the researchers found peak levels arrived later on average, by 1.16 hours (95% CI, 0.10-2.22). Drugs that clear the body quickly caused the biggest shifts when taken ahead of the contraceptive dose. The authors also describe a theoretical model built on slower stomach emptying, which predicts that tirzepatide's delay in reaching peak concentration would persist longer than delays seen with the other agents, even as the body adapts to repeated dosing, a wear-off effect called tachyphylaxis ^[1].

Beside the review, the abstract describes one pregnancy. A 39-year-old woman with obesity, identified as G4P2012, conceived after four months on weekly semaglutide while taking a monophasic ethinyl estradiol/levonorgestrel pill. The authors call it the first documented combined oral contraceptive failure with semaglutide ^[1]. A single case cannot establish that a drug caused a pregnancy.

Why it matters for patients

The review essentially splits the combined pill in two. Estrogen exposure appeared to survive a GLP-1 or dual GLP-1/GIP drug unchanged, while the progestin can show up later and at lower levels, and the size of that shift depends on which incretin drug is involved ^[1].

The authors conclude these pharmacokinetic changes are unlikely to undermine combined pill effectiveness in typical use. They add that counseling should cover when the two medications are taken relative to each other, backup contraception during higher-risk periods, and caution around progestin-only pills and emergency contraception ^[1].

Only combined oral contraceptives were included in the reviewed studies. Effects on progestin-only pills, implants, injections, rings or emergency methods were not measured, so the review provides no numbers for those. The participant pool skewed toward healthy premenopausal volunteers rather than people with obesity or diabetes, so results in other groups have not been established here.

What has not been reported

The abstract does not say how many participants took each drug, whether any pregnancies occurred among the 305 people in the studies, or which specific pill formulations were used beyond ethinyl estradiol and levonorgestrel. Whether the measured lab changes translate into more unintended pregnancies in ordinary use has not been reported.

The findings are available so far as a meeting abstract. A full-length peer-reviewed paper with complete methods and per-study results has not been published, and no timeline for labeling changes or follow-up studies has been disclosed ^[1].