GLP-1 for PCOS: weight, periods and fertility. What the trials found

Why GLP-1 medicines are being studied in PCOS

Polycystic ovary syndrome (PCOS) is closely tied to insulin resistance and higher weight. It also brings high androgen levels, which can cause excess hair growth (hirsutism) and acne, and irregular or absent ovulation. In 2026, an international consensus endorsed by 56 organizations renamed it polyendocrine metabolic ovarian syndrome (PMOS). The diagnostic criteria did not change, and it affects about 10% to 13% of reproductive-age women.¹¹ Most research still uses the name PCOS, so we do too.

Because weight loss can improve the hormonal side of PCOS, researchers have tested liraglutide (Saxenda, Victoza), semaglutide (Wegovy, Ozempic), exenatide and, most recently, tirzepatide (Zepbound, Mounjaro).

What the 2023 international PCOS guideline says

The 2023 International Evidence-based Guideline for the Assessment and Management of PCOS is the main reference doctors use. On anti-obesity medicines, it says:¹

• Anti-obesity medicines, including the GLP-1 medicines liraglutide and semaglutide, could be considered, in addition to active lifestyle intervention, for higher weight in adults with PCOS, as per general population guidelines.
• Effective contraception should be ensured when pregnancy is possible, because pregnancy safety data are lacking.
• Decisions should weigh side effects and the likely need for long-term use, given the high risk of weight regain after stopping.
• For fertility, anti-obesity medicines should be used only in research settings until efficacy and safety are established.

Weight loss: the clearest result

Pooling 11 randomized trials, a 2026 meta-analysis found GLP-1 medicines added on to other care lowered BMI by 1.38 kg/m² more than comparison treatments. The authors called this modest short-term weight loss and rated the certainty low.² These trials were short and used older or lower doses than today's obesity regimens, so they likely understate what semaglutide 2.4 mg or higher tirzepatide doses can do. For those, see GLP-1 weight loss results.

Do GLP-1s help periods and testosterone?

Several trials report improvements, though the pooled evidence is still weak.

• Periods. In the Danish liraglutide trial, the bleeding ratio (a measure of how regular cycles are) improved more than with placebo. In the open-label Chinese trials, regular cycles returned in 72.5% on semaglutide plus metformin vs 42.3% on metformin alone, and 53.6% on tirzepatide plus metformin vs 21.4%.^4–6
• Androgens. Liraglutide 3 mg lowered the free androgen index, a blood measure of active testosterone, compared with placebo. Liraglutide 1.8 mg raised SHBG, the protein that binds testosterone, by 7.4 nmol/L.^3,4
• Hirsutism. The 2026 meta-analysis found the evidence on hirsutism, menstrual regularity, glucose and insulin insufficient to draw conclusions. No trial it found measured quality of life or mental health.²

GLP-1 and PCOS fertility

This is the question many readers care most about, and the evidence is the thinnest.

• A 2023 meta-analysis found a higher natural pregnancy rate with GLP-1 medicines (risk ratio 1.72), but no difference in total pregnancy rate or IVF pregnancy rate.⁷
• In the Chinese trials, women stopped the GLP-1 medicine after 16 weeks and continued metformin before trying to conceive. Natural pregnancy rates were 35% vs 15% in the semaglutide trial and 39.3% vs 14.3% in the tirzepatide trial. In the semaglutide trial, total pregnancy rates did not differ significantly.^5,6
• A 2018 pilot trial in 28 women with obesity and PCOS who had not responded to first-line fertility care found pregnancy rates per embryo transfer of 85.7% after liraglutide plus metformin vs 28.6% after metformin alone. It was small and open-label.⁹
• A 2026 health-record study of 48,126 matched women per group linked GLP-1 or tirzepatide use before conception with fewer adverse pregnancy outcomes than metformin (hazard ratio 0.38). That is observational data, and the authors say it does not support use during pregnancy.¹⁰

Metformin vs GLP-1 for PCOS

Metformin is used off-label in PCOS for metabolic features. A 2019 meta-analysis of 8 randomized trials found GLP-1 medicines lowered BMI and improved insulin sensitivity more than metformin, but caused more nausea and headache. The authors rated the evidence moderate to low quality.⁸ The 2023 meta-analysis found GLP-1 medicines alone were not better than metformin at lowering free testosterone, DHEAS or the free androgen index.⁷ Several of the newer trials combined the two.

Side effects in the PCOS trials

Side effects matched what is seen in other people taking these medicines. In the liraglutide 3 mg trial, stomach and bowel side effects, mostly mild to moderate, occurred in 58.2% on liraglutide vs 18.5% on placebo.³ See GLP-1 nausea and stomach side effects for what helps.

Pregnancy: what the labels say

GLP-1 medicines are not FDA-approved for PCOS. The US labels list uses such as weight reduction, type 2 diabetes and heart risk, not PCOS.¹² Because so many women with PCOS hope to become pregnant, the labels' pregnancy sections matter:

• Wegovy (section 8.3) says, for weight reduction or heart risk use, to discontinue at least 2 months before a planned pregnancy because of semaglutide's long half-life. Ozempic (section 8.3) gives the same 2-month instruction.¹²
• Zepbound (section 8.1) says weight loss offers no benefit in pregnancy and to discontinue when a pregnancy is recognized. Section 8.3 says tirzepatide may make birth control pills less effective, and advises a non-oral method or added barrier method for 4 weeks after starting and after each dose increase.¹²

These are the labels' instructions to prescribers, not personal advice. More detail is in GLP-1, birth control and pregnancy.