Women's Health · PCOS Research

For years, a woman with polycystic ovary syndrome who brought up her weight at a checkup often left with the same two things: a metformin prescription and some version of "eat less, move more."

The pamphlet was optional.

Metformin earned its place. It's inexpensive, well studied, and helps the body use insulin more effectively, which matters a great deal in a condition where insulin resistance is the rule rather than the exception. For plenty of women, though, the scale didn't seem to get the memo.

That conversation is starting to change. A 2025 narrative review in Endocrine Connections and a large analysis of real-world health records from Epic Research have both looked at what happens when GLP-1 drugs enter the picture. The results are promising, though uneven in places, and certainly worth a closer look.

A Real-World Scoreboard

Clinical trials are tidy by design. Real life is not. That's what makes a 2025 Epic Research analysis useful: it followed 36,674 women ages 18 to 50 with PCOS who started either a GLP-1 drug or metformin between January 2021 and November 2024, using health records from everyday practice rather than a research clinic.

After one year, the gap was hard to miss.

Blood sugar told a similar story. HbA1c, a roughly three-month average of blood sugar, fell a median of 0.5 points in the GLP-1 group and 0.1 points with metformin. And 83.5% of GLP-1 users saw at least some HbA1c improvement, compared with 55.9% of metformin users.

Before anyone declares a winner, a caution. This was an observational study, not a randomized trial. The women weren't assigned a drug at random, so other differences between the two groups could explain part of the gap. The researchers adjusted for age, BMI, other health conditions, and starting HbA1c, and the results still pointed the same way. That's encouraging. It isn't proof.

What the Controlled Trials Add

Randomized trials are smaller, but they let researchers isolate the drug itself. The 2025 review by Monney and colleagues gathered a number of them, and two stand out for women with PCOS.

In a 32-week, placebo-controlled trial of 82 women, Elkind-Hirsch and colleagues found that liraglutide (3 mg daily) reduced weight, BMI, and waist size and improved blood sugar measures. It also lowered the free androgen index, an estimate of how much active testosterone is circulating, and increased how often participants had a period. For a condition defined partly by excess androgens and irregular cycles, that's more than a cosmetic result.

Semaglutide has a smaller but growing record. In a 2023 study of 27 women with PCOS, Carmina and colleagues reported that 0.5 mg weekly for three months lowered body weight, BMI, fasting glucose, insulin, and a standard measure of insulin resistance. According to the review, the same dose reduced body weight in almost 80% of women who hadn't responded to lifestyle programs. (For perspective, the weekly semaglutide dose approved for chronic weight management tops out at 2.4 mg, so this was a fairly modest dose.)

The GLP-1 family is also bigger than the two names most people know.

Metformin Isn't Out of a Job

If this were a boxing match, it would be tempting to stop the fight. It isn't, and the research on combining the two drugs shows why.

According to the review, liraglutide plus metformin produced more weight loss than metformin alone. Liraglutide 3 mg on its own beat the combination on weight, but the pairing was easier to tolerate and had a more favorable effect on androgens. Earlier work points the same way: in a 24-week study of 60 women, Elkind-Hirsch and colleagues found that exenatide plus metformin improved menstrual frequency and ovulation rates more than either drug did by itself.

Not every trial is a highlight reel. In a 12-week study of 82 women, Zheng and colleagues found that exenatide produced weight loss but no better results than metformin on cycles, excess hair growth, or acne. Short trials may simply be too brief to show hormonal changes, though that's a possibility, not a conclusion.

The practical takeaway is less dramatic than a knockout, and more useful. The newer drug may make the older one work better, and the older one may make the newer one easier to live with.

Fertility: Encouraging, and Unfinished

Some of the most-watched findings involve pregnancy. In a Chinese trial of 176 women with PCOS who were trying to conceive, the pregnancy rate at 24 weeks was 43.6% with exenatide, compared with 18.7% with metformin.

The longer view is more modest. When 160 of those women were followed out to 64 weeks, there was no difference between the groups in total pregnancy rate or pregnancy outcomes. One reasonable reading is that the drug may have helped some women conceive sooner rather than more often, though the trial wasn't built to answer that question directly.

A small IVF study adds another wrinkle. Among 28 women with PCOS and infertility, Salamun and colleagues reported a pregnancy rate per embryo transfer of 85.2% with liraglutide plus metformin, versus 28.6% with metformin alone, even though weight loss was similar in both groups. That hints the benefit may not be purely about pounds. With 28 participants, it's a hint, not a verdict.

Timing still matters. The 2023 International Evidence-based Guideline for PCOS advises effective contraception while taking a GLP-1 drug and a washout period before attempting pregnancy. Registry data from women treated for type 2 diabetes or obesity haven't shown a significant rise in adverse outcomes after early pregnancy exposure, but the review describes that data as limited.

The Fine Print on Stopping

Most headlines about GLP-1 drugs focus on starting them. The PCOS research has something to say about stopping, too.

In a 2024 follow-up by Jensterle and colleagues, 25 women with PCOS took semaglutide with metformin for 16 weeks and were then tracked for two years off treatment. On average, they regained about a third of the weight they had lost. The good news is that they kept a net loss. The less-good news is that the other metabolic and hormonal improvements didn't last.

That's why the international guideline recommends shared decision-making that accounts for side effects, limited long-term safety data, and the likelihood that treatment may need to be long term because weight regain after stopping is common. That trade-off is worth weighing before the first dose, not after the last one.

What Researchers Still Don't Know

For all the momentum, the review is candid about the gaps. Many PCOS trials are small, and most follow women for months rather than years. Several big questions remain open:

None of that cancels out the progress. It just means the next round of studies has plenty of work to do.

Featured Product

Tirzepatide / Prescription Weight Management

A Daily, Needle-Free Option

Tirzepatide, which acts on both GLP-1 and GIP receptors, is one of the newer medications researchers are watching. The Wellness Company offers it as a once-daily, mint-flavored tablet that dissolves under the tongue, prescribed only after a licensed provider reviews your medical history.

Shop Now

Findings described here come from the specific trials and analyses cited, and results differ from person to person. The Epic Research findings are observational and show associations, not cause and effect. GLP-1 receptor agonists require a prescription and a medical evaluation. This article is educational and is not a substitute for personalized medical advice, diagnosis, or treatment. Please consult your healthcare provider before beginning, discontinuing, or adjusting any medication.