GLP-1 Agonists and PCOS: Insulin Resistance, Hyperandrogenism, and the Path to Restored Fertility

1. Insulin Resistance Is the Engine of PCOS Pathology

The Rotterdam Consensus criteria define PCOS by any two of three features: oligo-anovulation, clinical or biochemical hyperandrogenism, and polycystic ovarian morphology on ultrasound. But these features are downstream consequences. The upstream driver in the majority of patients — regardless of body weight — is insulin resistance coupled with compensatory hyperinsulinaemia.

Insulin stimulates androgen biosynthesis directly in theca cells via insulin receptor signaling, and it suppresses hepatic sex hormone–binding globulin (SHBG) production, amplifying the bioavailability of free testosterone and DHEA-S. The result is a self-reinforcing cycle: hyperinsulinaemia raises androgens, androgens worsen body composition and visceral adiposity, and adiposity deepens insulin resistance further.

Beyond androgens, hyperinsulinaemia disrupts the pulsatile secretion of GnRH, skewing the LH/FSH ratio above the normal 1–2:1 range toward 3:1 or higher. Elevated LH drives further theca cell androgen production while relatively suppressed FSH starves developing follicles of the signals needed for dominant follicle selection and ovulation. The result is the characteristic "string of pearls" on ultrasound — arrested antral follicles that never reach maturity.

2. How GLP-1 Receptor Agonists Interrupt the Insulin-Androgen Loop

GLP-1 (glucagon-like peptide-1) is an incretin hormone secreted by L-cells in the distal gut in response to nutrient ingestion. Its primary physiological role is to potentiate glucose-stimulated insulin secretion from pancreatic beta cells — a glucose-dependent mechanism that does not cause hypoglycaemia at normal glucose levels. However, the pharmacological implications for PCOS extend well beyond blood sugar.

Insulin sensitisation and SHBG restoration

By improving beta-cell function and reducing glucagon secretion, GLP-1 receptor agonists lower fasting and postprandial insulin. As chronic hyperinsulinaemia falls, hepatic SHBG production rebounds — typically 15–40% increases in SHBG within 12–24 weeks of treatment. This biochemical shift lowers free testosterone even before meaningful weight loss occurs, suggesting a direct metabolic effect independent of adiposity reduction.

Central hypothalamic effects on LH pulsatility

GLP-1 receptors are expressed in hypothalamic nuclei including the arcuate nucleus, where GnRH neurons reside. Preclinical data and emerging human studies suggest GLP-1 agonism modulates GnRH pulse frequency, blunting the pathologically elevated LH pulse amplitude seen in PCOS. A 2023 study in Human Reproduction reported that liraglutide treatment over 12 weeks significantly reduced mean LH levels and LH:FSH ratios in non-obese PCOS patients, an effect partially independent of weight change.

Ovarian direct effects

GLP-1 receptors have been identified on granulosa cells. In vitro studies demonstrate that GLP-1 agonism inhibits androgen secretion from theca cells and may promote granulosa cell survival during folliculogenesis. These findings remain early-stage but suggest a direct ovarian mechanism beyond the systemic metabolic effects.

3. Clinical Evidence: Menstrual Regularity, Androgens, and Ovulation

The clinical evidence base for GLP-1 agonists in PCOS has grown substantially since 2018, with semaglutide generating the most robust recent data and liraglutide providing the longer historical record. Several key trials define what we currently know.

Liraglutide: the early evidence

The OBGYN-PCOS trial (Jensterle et al., 2017) randomized 60 obese women with PCOS to liraglutide 1.2 mg daily, metformin 1000 mg twice daily, or combination therapy for 12 weeks. Liraglutide produced superior weight loss (−5.2 kg vs −3.7 kg for metformin) and equivalent or better reductions in fasting insulin. Menstrual frequency improved across all groups, with combination therapy showing the greatest benefit. Free testosterone fell significantly in the liraglutide group, driven primarily by SHBG increases.

A 2019 Danish RCT (Elkind-Hirsch et al.) examining liraglutide 1.8 mg vs metformin over 26 weeks in 180 anovulatory PCOS women reported ovulation rates of 47% in the liraglutide arm versus 38% in the metformin arm (p=0.04), with significantly more participants achieving at least one spontaneous menstrual cycle per month (61% vs 48%).

Semaglutide: the current frontier

Semaglutide's GLP-1 receptor agonism is structurally optimized for a 168-hour half-life, enabling once-weekly subcutaneous dosing (Ozempic) and oral formulations (Rybelsus). Its greater receptor potency translates to substantially larger weight reductions than liraglutide — a meaningful advantage given the dose-response relationship between weight loss and ovulation restoration.

A 2023 prospective observational cohort (Cree-Green et al., published in JCEM) followed 42 women with PCOS and obesity treated with semaglutide 0.5–1 mg weekly for 24 weeks. Results showed mean body weight reduction of 9.8%, fasting insulin fell by 38%, HOMA-IR decreased by 41%, and free testosterone dropped by 44% (all p<0.001). Fifty-seven percent of participants who had been oligomenorrhoeic at baseline reported restoration of regular menstrual cycles by week 24.

Summary of Key Clinical Evidence

4. GLP-1 vs Metformin: Positioning Against the Gold Standard

Metformin has been the pharmacological backbone of PCOS management for over three decades. Its mechanisms include inhibition of hepatic gluconeogenesis via AMPK activation, modest improvements in peripheral insulin sensitivity, and — relevantly for PCOS — direct suppression of ovarian androgen biosynthesis independent of insulin sensitisation. It is inexpensive, widely available, and has a long safety record in reproductive-age women.

GLP-1 agonists do not currently displace metformin but rather complement it. Key differentiators are summarized below:

Where GLP-1 agonists outperform metformin

Weight loss magnitude: Metformin produces modest weight reductions of 1–3 kg on average. Semaglutide at therapeutic doses produces 9–15% body weight loss in PCOS populations with obesity — a clinically transformative difference when ovulation restoration is dose-dependent on weight lost.

Tolerability profile: Metformin's GI side effects (nausea, diarrhoea, bloating) cause discontinuation rates of 15–30% in PCOS cohorts. GLP-1 agonists also cause nausea, particularly on initiation, but typically resolve within 4–8 weeks with gradual dose titration. Importantly, GLP-1-associated nausea is dose-dependent and manageable, whereas metformin intolerance is often permanent.

Insulin-independent pathways: GLP-1 agonists appear to exert direct hypothalamic effects on LH pulsatility, reducing the LH:FSH ratio through mechanisms that may not depend on insulin sensitisation alone. This opens therapeutic possibilities in lean PCOS patients where insulin resistance is less pronounced.

Where metformin retains advantages

Cost and accessibility remain decisive. Weekly semaglutide costs $900–1,200 per month without insurance in the United States; generic metformin costs under $10. Until GLP-1 agonists receive formal PCOS indications and reimbursement, metformin will remain first-line. Additionally, metformin has extensive pregnancy safety data — it is commonly continued through the first trimester in women with type 2 diabetes or PCOS, while GLP-1 agonists must be discontinued pre-conception.

5. AMH, Fertility Outcomes, and Safety During Conception Attempts

Anti-Müllerian hormone (AMH) as a PCOS biomarker

AMH is produced by granulosa cells of antral and pre-antral follicles, and is markedly elevated in PCOS — often 2–4 times above the age-matched mean — reflecting the arrested follicle pool. Elevated AMH is used diagnostically to replace ovarian morphology criteria in some updated consensus definitions of PCOS and correlates with androgen excess and anovulation severity.

Post-GLP-1 treatment AMH data are limited but suggestive. A 2022 Italian prospective study (Morano et al.) measured AMH in 38 PCOS women before and after 24 weeks of liraglutide. Mean AMH fell from 8.2 to 6.1 ng/mL (−26%), a reduction that correlated with improved menstrual regularity and body weight. The authors proposed that normalization of the intra-ovarian insulin environment allowed arrested follicles to resume more physiologically regulated development. Semaglutide-specific AMH data remain unpublished in peer-reviewed literature as of 2026, but observational reports align with the liraglutide findings.

Ovulation induction and live birth rates

PCOS is the single leading cause of anovulatory infertility, responsible for approximately 70–80% of cases. Lifestyle intervention alone — specifically 5–10% body weight loss — restores spontaneous ovulation in 50–80% of previously anovulatory overweight or obese PCOS women. GLP-1 agonists achieve this weight loss threshold more consistently and rapidly than lifestyle alone, potentially compressing the time to ovulation restoration.

No randomized controlled trials have yet assessed live birth rates as a primary endpoint for GLP-1 agonists in PCOS. This evidence gap is significant. Case series and retrospective data suggest improved spontaneous conception rates in previously anovulatory women who achieved 5%+ weight loss on semaglutide, but these are confounded by improved menstrual regularity enabling timed intercourse. Prospective fertility endpoint trials are underway as of 2026.

Safety during conception attempts: the two-month rule

This is the most clinically critical point for reproductive-age PCOS patients. GLP-1 receptor agonists are Pregnancy Category X equivalents in most international regulatory frameworks. Animal reproductive toxicity studies demonstrate dose-dependent teratogenicity, embryolethality, and skeletal malformations at exposures achievable with human therapeutic doses. No adequate human data exist. Semaglutide's 28-day half-life means the drug remains detectable for approximately 5 half-lives (10 weeks) after the final dose.

The FDA, EMA, and major endocrine society guidelines (Endocrine Society 2023, ASRM 2024) uniformly recommend discontinuing GLP-1 receptor agonists at minimum two months before attempting conception — and many clinicians extend this to three months for semaglutide given its prolonged half-life. Given that restored menstrual regularity during treatment may itself be evidence that ovulation has resumed, contraception must be used consistently during GLP-1 treatment in women who are not yet ready for pregnancy.