GLP-1 Drugs for PCOS: What the Trials Actually Show

No GLP-1 receptor agonist carries an FDA-approved indication for polycystic ovary syndrome. Any use for PCOS is off-label, prescribed on the strength of a clinical rationale and a growing but still small body of trial evidence — not an approved use the way weight management or type 2 diabetes are. That distinction matters before anything else here.

Why the rationale exists

Insulin resistance is a core feature of PCOS for most women who have it, independent of body weight, and it's mechanistically linked to the excess androgen production that drives many of the condition's most disruptive symptoms — irregular cycles, acne, excess hair growth. GLP-1 drugs improve insulin sensitivity and reduce weight, which is the direct biological reasoning behind testing them in PCOS specifically, rather than assuming general obesity-drug benefits would transfer automatically.

What a 2024 meta-analysis of the randomized trials found

A 2024 meta-analysis pooled four randomized controlled trials comparing GLP-1 receptor agonists against placebo in women with PCOS and obesity — 176 participants total, mostly on liraglutide (58%) with a smaller group on semaglutide (13%)¹. GLP-1 treatment was associated with statistically significant reductions in waist circumference (mean difference -5.16 cm), body mass index (-2.42), and total testosterone (-1.33), plus a reduction in triglycerides. There was no significant difference in total cholesterol or HOMA-IR, a standard insulin-resistance measure — a genuinely mixed result, not a clean win across every metric tested.

What a 2025 trial found on reproductive outcomes specifically

A 2025 randomized, open-label trial in China went further, testing combined semaglutide and metformin against metformin alone in 100 overweight or obese women with PCOS over 16 weeks, then following pregnancy outcomes through week 40². The combination group lost significantly more weight (6.09 kg versus 2.25 kg with metformin alone) and showed greater improvement in testosterone and inflammatory markers. The reproductive finding is the one worth reading carefully: from weeks 16 to 40, after all participants switched to metformin alone, the combination-therapy group had a significantly higher natural pregnancy rate — 35% versus 15% in the metformin-only group. That's a real, clinically meaningful outcome, not just a metabolic marker, though it comes from one moderate-sized trial at a single country's clinical trial registry rather than a confirmed, replicated finding.

Reading this evidence honestly

Both studies are real randomized trials with real, statistically significant findings — this isn't anecdote dressed up as science. But the numbers behind them are still small by the standards this site otherwise cites: 176 participants across four pooled trials in the meta-analysis, 100 participants in the single reproductive-outcomes trial. Larger, longer, more diverse trials would be needed before "GLP-1 drugs improve fertility in PCOS" could be stated as a settled fact rather than a promising early signal. It's also worth noting semaglutide itself independently delays gastric emptying in women with PCOS and obesity, a mechanism-level finding from a smaller 2023 study that's consistent with — but doesn't add new proof beyond — the GI side-effect pattern covered in GLP-1 side effects.

What this means if PCOS is why you're considering a GLP-1

This is a conversation to have with an OB-GYN or reproductive endocrinologist who can weigh your specific case, not a self-directed off-label decision. If you're pregnant or trying to become pregnant, note that GLP-1 drugs are not studied for use during pregnancy and current labeling directs discontinuing well before a planned pregnancy — see how long GLP-1 drugs stay in your system for the specific washout guidance. For the underlying molecule choice if a GLP-1 is part of your plan, semaglutide vs tirzepatide covers the trial evidence in the general population, and how to choose a GLP-1 provider covers what to vet before starting with any telehealth clinic. The rest of the Desk's sourced evidence is in the research library.