Polycystic ovary syndrome: treatment and what GLP-1s show

What is polycystic ovary syndrome and how many women have it?

Polycystic ovary syndrome (PCOS) is a common endocrine and metabolic disorder of reproductive age. The World Health Organization (WHO) estimates it affects 10-13% of women of reproductive age and that up to 70% do not know they have it¹. WHO lists insulin resistance, type 2 diabetes, obesity, cardiovascular disease and endometrial cancer among its longer-term complications¹.

The name is changing. In 2026 a global consensus process, with 56 organizations and responses from 14,360 people with the condition and health professionals from all world regions, concluded that the term PCOS is inaccurate because it implies pathological ovarian cysts and obscures the endocrine and metabolic features; the new name is polyendocrine metabolic ovarian syndrome (PMOS), with a transition period planned². This article uses PCOS because it is still the term most people search for.

Symptoms and how PCOS is diagnosed

Symptoms differ between people and over time: irregular or absent periods, difficulty conceiving, excess facial or body hair, thinning hair, and acne or oily skin¹. Diagnosis requires two of three features once other causes are excluded: high androgens, irregular or absent periods, and polycystic ovaries on ultrasound; WHO states that ovarian cysts are not required for the diagnosis¹.

The 2023 international guideline, developed with 39 professional and consumer organizations from 71 countries, refined the criteria, simplified the diagnostic algorithm and included anti-Müllerian hormone (AMH) as an alternative to ultrasound in adults only³. A meta-analysis of 82 studies found a pooled sensitivity of 0.79 (95% CI 0.76-0.82) for AMH in diagnosing PCOS in adults⁵. The guideline stresses that the evidence remains of low to moderate quality³.

Standard treatment today: what the guidelines recommend

There is no cure: WHO says treatments can improve quality of life and mentions lifestyle changes, contraceptive pills, androgen receptor blockers and fertility treatments¹. The 2023 guideline maintains the emphasis on healthy lifestyle, emotional wellbeing and quality of life, with awareness of weight stigma, and on evidence-based medical therapy and cheaper, safer fertility management³.

For metabolic care, the guideline recommends that metformin alone be considered in adults with a body mass index (BMI) of 25 kg/m² or more, and says it could be used over the combined oral contraceptive for metabolic indications⁴. For cycle control, metformin may be considered when the contraceptive is contraindicated, not accepted or not tolerated⁴.

For fertility, the guideline recommends letrozole as the first-line pharmacological treatment for ovulation induction in anovulatory women with PCOS, notes that its use is still off-label in many countries and says it should not be given if there is any possibility of a pre-existing pregnancy⁴. The basis is the double-blind trial of 750 women: letrozole produced more cumulative live births than clomiphene (103 of 374 [27.5%] versus 72 of 376 [19.1%]; rate ratio 1.44; 95% CI 1.10-1.87) and more ovulation (61.7% versus 48.3% of cycles); there were no significant differences in pregnancy loss or overall congenital anomalies, although there were 4 major anomalies with letrozole versus 1 with clomiphene⁸.

PCOS also matters outside the gynecology clinic. A meta-analysis of 104 studies and 106,690 pregnancies found higher odds of miscarriage, gestational diabetes, gestational hypertension, pre-eclampsia and cesarean section, which persisted when age and BMI were matched⁶. Another, of 20 studies and 1.06 million women, associated PCOS with a higher risk of composite cardiovascular disease (OR 1.68; 95% CI 1.26-2.23)⁷.

What peptide research has shown

Two peptide families are studied in PCOS: glucagon-like peptide-1 receptor agonists (GLP-1), aimed at weight and insulin resistance, and kisspeptin, a hypothalamic regulator of the reproductive axis. For the general background on GLP-1 drugs see the guide to semaglutide, tirzepatide and retatrutide and, for ovarian stimulation with peptides, infertility and ovarian stimulation.

What the guideline says. The 2023 guideline includes a consensus recommendation (not based on PCOS-specific trials of its own) that anti-obesity medications, including liraglutide and semaglutide, may be considered alongside active lifestyle intervention; it adds practice points to ensure effective contraception when pregnancy is possible, to escalate the dose gradually to reduce gastrointestinal effects, and to decide jointly because of the high risk of weight regain after stopping⁴.

Liraglutide. It is the GLP-1 drug with the most PCOS data. A meta-analysis of 7 trials with 330 women with overweight or obesity found that, compared with controls, it increased menstrual frequency (g = 1.76; 95% CI 0.28-3.24, with high heterogeneity), reduced BMI (g = -0.52) and insulin resistance (HOMA-IR, g = -0.52), and lowered luteinizing hormone and the free androgen index; ovulation and pregnancy could not be pooled for lack of data, and adverse events were mainly mild gastrointestinal symptoms⁹. In an open-label trial of 60 women with overweight or obesity followed for 12 weeks, metformin plus liraglutide reduced weight, BMI and visceral fat more than metformin alone (P < 0.05) and free testosterone more (P = 0.01)¹³.

Semaglutide. A meta-analysis of 8 trials (526 patients) comparing semaglutide with placebo or conventional treatment found a BMI reduction of 2.20 kg/m² (95% CI -2.42 to -1.97)¹². In the analysis of 4 placebo-controlled trials with 176 participants, of whom 23 received semaglutide and 103 liraglutide, GLP-1 drugs reduced BMI (mean difference -2.42), waist circumference (-5.16 cm) and total testosterone (-1.33), with no difference in HOMA-IR; 49 of 112 with safety data had mild effects such as nausea and abdominal pain¹¹.

The overall GLP-1 picture. A more recent systematic review of 11 randomized trials found that adding a GLP-1 drug reduced BMI by 1.38 kg/m² (95% CI -2.39 to -0.38), with low certainty, and that the evidence on glucose, insulin, hirsutism and menstrual regularity was insufficient; it found no studies of quality of life, mental health or cost-effectiveness¹⁰.

Pregnancy. A review of 3 trials (252 participants) comparing GLP-1 therapy, alone or with metformin, against metformin alone found no improvement in overall pregnancy (OR 1.41; 95% CI 0.81-2.45) but an increase in natural pregnancy (OR 3.10; 95% CI 1.74-5.49; moderate certainty) and no effect on pregnancy after in vitro fertilization (OR 1.01); the authors call for larger trials with live birth as the outcome¹⁴. A target-trial emulation of 48,126 patients per group associated pre-conception GLP-1 or dual GLP-1/GIP use with a lower risk of adverse pregnancy outcomes than metformin (HR 0.38; 95% CI 0.34-0.41), but it is observational, cannot evaluate safety during pregnancy and its authors state it does not support use during gestation¹⁹.

Tirzepatide. The sources reviewed contain no PCOS trials. A retrospective study from a UK digital service with 54,114 women, 4,241 of them with self-reported PCOS, found at 10 months a weight change of -19.40% in 40 women with PCOS versus -18.74% in 397 without, not significantly different¹⁸. It is descriptive, with few cases at the endpoint.

Kisspeptin. This is the earliest evidence. A meta-analysis of 12 studies (660 women with PCOS versus 600 controls) found higher circulating kisspeptin in PCOS, but the authors note that all studies described sample collection and analysis poorly¹⁷. In an exploratory pilot, repeated kisspeptin-54 in 12 anovulatory women with PCOS produced a small rise in luteinizing hormone; 2 women grew a dominant follicle and ovulated, 1 had follicle growth without ovulation, another showed desensitization and the rest did not respond¹⁵. The work received industry funding and the authors describe it as proof of principle¹⁵. In another study of 8 women with PCOS, kisspeptin-10 raised luteinizing hormone (from 5.2 to 7.8 IU/L)¹⁶. None of these figures supports a claim of therapeutic efficacy.

Safety: adverse effects and precautions

In PCOS trials, the adverse effects of liraglutide and semaglutide were mainly mild gastrointestinal symptoms⁹¹¹. The Wegovy (semaglutide) label, revised June 2026, lists nausea, diarrhea, vomiting, constipation, abdominal pain, dysesthesia, headache and fatigue among the most common reactions; it carries a boxed warning for thyroid C-cell tumors in rodents, with unknown human relevance, and is contraindicated with a personal or family history of medullary thyroid carcinoma or MEN 2²¹. It says to stop the drug when pregnancy is recognized and at least 2 months before a planned pregnancy, because of its long half-life²¹. PCOS is not among that label's indications²¹.

Weight returns after stopping: withdrawal trials showed a 6.9% gain over 48 weeks after switching from semaglutide to placebo (STEP 4) and 14.0% over 52 weeks after switching from tirzepatide to placebo (SURMOUNT-4)²⁰. For letrozole, the guideline warns it should not be given if there is a possibility of pre-existing pregnancy⁴, and in the 750-woman trial it caused more fatigue and dizziness than clomiphene, which caused more hot flushes⁸.

Summary by agent

Frequently asked questions

How many women have polycystic ovary syndrome? Between 10% and 13% of women of reproductive age according to WHO, and up to 70% do not know they have it¹.

Is PCOS now called PMOS? Yes, that is the name proposed by a global consensus published in 2026: polyendocrine metabolic ovarian syndrome, with a transition period².

Is there an approved peptide for PCOS? Not in the sources reviewed. PCOS is not among the indications on the semaglutide label²¹, and the 2023 guideline mentions them only as anti-obesity medications that may be considered alongside lifestyle⁴.

Do GLP-1 drugs improve fertility in PCOS? The signal is preliminary: in 3 trials with 252 women there were more natural pregnancies than with metformin alone (OR 3.10), but no improvement in overall pregnancy or in vitro fertilization¹⁴.

What is the first-line treatment for ovulation? Letrozole, according to the 2023 guideline, although its use remains off-label in many countries⁴; it beat clomiphene on live births in a 750-woman trial⁸.

Is weight regained after stopping a GLP-1 drug? Commonly: +6.9% over 48 weeks after stopping semaglutide (STEP 4) and +14.0% over 52 weeks after stopping tirzepatide (SURMOUNT-4)²⁰.