For weight and insulin resistance, modestly, and on thin evidence. Eight randomized trials compared GLP-1 drugs directly with metformin in PCOS [2]. GLP-1 drugs improved insulin sensitivity and lowered BMI and waist size more. The authors rate that evidence moderate to low quality.
Whether GLP-1 drugs help PCOS at all is a broader question, answered in the main PCOS evidence. This page is about the comparison with the drug most women with PCOS are offered first.
Head to head with metformin
The 2019 meta-analysis pooled eight trials against metformin [2]. Insulin sensitivity improved more on a GLP-1, at a standardized difference of −0.40. BMI fell more, at −1.02, and abdominal girth at −0.45.
Those are standardized differences, not kilograms. Nausea and headache were more common on the GLP-1. The authors call it a possible good choice for PCOS with obesity and insulin resistance. They also call the evidence inconclusive.
The review that informed the 2023 international PCOS guideline was more cautious [3]. Only one head-to-head comparison could be pooled, exenatide against metformin. It found no difference in weight, androgens or metabolic outcomes. Fasting glucose fell slightly more on metformin.
Its verdict was that published data are very limited. The two reviews differ in years searched and in which drugs they included.
On top of metformin, not instead of it
A different design adds a GLP-1 to metformin rather than replacing it. Eight trials with 337 women tested that against metformin alone [4].
Weight fell 1.37 kg more on the combination (95% CI −2.07 to −0.67), and BMI 0.88 kg/m² more. Waist size fell 2.46 cm more. Fasting glucose and insulin resistance improved, and SHBG rose. Adverse events did not differ.
A 1.37 kg difference is small beside what these drugs do in obesity trials, and 337 women is a small base.
What eleven trials could not answer
The 2026 review searched 9,654 records and included 11 randomized trials against several comparators, metformin among them [1]. Added to standard care, GLP-1 drugs cut BMI by 1.38 kg/m² (95% CI −2.39 to −0.38), at low certainty.
For glucose, insulin, hirsutism and menstrual regularity, evidence was insufficient to conclude anything. Those last two are among the commonest reasons women with PCOS seek treatment.
What that means for a buyer
The strongest case for a GLP-1 in PCOS is the weight and the insulin resistance. On cycles and hirsutism, the evidence cannot yet say.
That makes it a weight treatment with a plausible knock-on effect. Most women taking one for PCOS have no diabetes, as the off-label prescribing data show.
When the benefit is less certain, the price matters more. The market is set out in what a GLP-1 costs per month. Watch the gap between advertised and billed prices, and weigh the wider case in whether a GLP-1 is worth the cost.