Polycystic ovary syndrome is, for a large proportion of the women who have it, a disorder of insulin. The irregular cycles, the androgen excess, the difficulty losing weight and the difficulty conceiving all sit downstream of a metabolic problem that conventional advice has treated badly for decades.
So a drug class that improves insulin sensitivity and produces fifteen percent weight loss was always going to matter here. What has been less well handled is the consequence.
Why they work here at all
Insulin resistance drives a specific chain of events in PCOS. High circulating insulin stimulates ovarian androgen production and suppresses hepatic production of sex hormone binding globulin, which raises free testosterone further. The result is the familiar cluster: irregular or absent ovulation, acne, hirsutism, and weight that behaves differently from other people’s.
Weight loss of five to ten percent has been shown to restore ovulatory cycles in a substantial fraction of women with PCOS. That has been in every guideline for years. The difficulty was never that the advice was wrong — it was that the advice was extremely hard to follow in the presence of the exact metabolic environment causing the problem.
These drugs make the recommended intervention achievable for a lot of people it was previously not achievable for. Most of the PCOS benefit is that, rather than anything ovary-specific.
What actually improves
Reported and observed benefits in women with PCOS on this drug class:
- Cycle regularity. Often the first thing to change, sometimes within a few months.
- Ovulation. Returns in a meaningful proportion of women who were anovulatory.
- Insulin resistance markers. Fasting insulin and HOMA-IR improve, generally in proportion to weight lost.
- Androgen symptoms. Acne and hirsutism improve more slowly, over many months, and less reliably than the metabolic markers.
- Weight. The obvious one, and for many women the first sustained loss of their adult life.
What does not change is the underlying diagnosis. PCOS is not cured by weight loss, and the features tend to return when the weight does — which makes the maintenance question more consequential here than usual.
The fertility problem
This is the part that generates the most email, and the least clinical preparation.
A woman who has been told for a decade that she is unlikely to conceive without assistance stops ovulating irregularly and starts ovulating regularly. She has not been using contraception, because there was no perceived need. The pregnancy is discovered at eight or ten weeks — well into a period during which she was injecting a drug that should be stopped before conception.
There is no strong human evidence of harm at this point, and the reassurance most women in this position need is that the animal signal is not a demonstrated human one. But it is not a situation anyone should be in by accident.
5–10%
Weight loss associated with restored ovulation in PCOS
International PCOS Guideline, 2023
2 months
Recommended gap between stopping semaglutide and conception
4 weeks
Contraception precaution after starting or increasing tirzepatide
The contraception detail that catches people out
There are two separate issues here and they get conflated constantly.
The fertility issue applies to every drug in the class: restored ovulation means restored fertility, so if you do not want to conceive, you need contraception you are actually using.
The absorption issue is specific to tirzepatide. Mounjaro and Zepbound labelling advises that women on oral contraceptives switch to a non-oral method, or add a barrier method, for four weeks after starting the drug and for four weeks after each dose escalation — because slowed gastric emptying can reduce how much of the pill is absorbed. Semaglutide carries no such warning.
Given that tirzepatide has six dose steps, that is potentially six separate four-week windows in which an oral contraceptive is less reliable than it appears. The full detail is in the four-week rule, and it is the single most under-communicated instruction in this drug class.
A drug that treats the cause of your infertility will treat the cause of your infertility. It is remarkable how often that arrives as a surprise.
If you are trying to conceive
Then the GLP-1 is a bridge, not a treatment.
The sequence most reproductive endocrinologists work to: lose the weight, improve the metabolic picture, then stop the drug, allow the recommended washout, and try. Semaglutide’s half-life of about a week means five weeks to clear and a two-month margin recommended on the label. Tirzepatide clears faster, and one month is advised.
Weight regain during that window is common and is worth planning around rather than being ambushed by. This is a conversation to have with a clinician who does fertility work, not one to improvise.
The honest state of the evidence
Nearly everything above is inferred from weight-loss trials in general populations plus a modest and growing PCOS-specific literature. Dedicated randomised trials of GLP-1s in PCOS exist but are small, short, and mostly report metabolic endpoints rather than live births.
That is not a reason to avoid the drugs here — the mechanism is sound and the observed effects are consistent. It is a reason to be honest that this remains off-label prescribing built on a reasonable inference, and to be correspondingly careful about the parts of it that have obvious consequences.