Women with polycystic ovary syndrome are among the most motivated buyers in this market, and they are usually buying for reasons that have little to do with a number on a scale. The systematic review below is the best summary of what has actually been tested, and its most useful sentence is about what has not — a shape this site meets often enough that counting what is absent is half of what it does.
What eleven trials found
Researchers screened 9,654 studies, read 365 in full, and included 11 randomized controlled trials [1]. Added to standard care, GLP-1 drugs reduced BMI by 1.38 kg/m² against control (95% CI −2.39 to −0.38), a result the review rates as low certainty.
LDL cholesterol and triglycerides showed no difference, also at low certainty. For glucose, insulin, hirsutism and menstrual regularity the evidence was rated insufficient to draw a conclusion at all — and those last two are among the commonest reasons a woman with PCOS seeks treatment in the first place.
What to do with a low-certainty result
Treat it as a reason to try rather than a reason to expect. A BMI reduction of that size is real and modest, the certainty rating means further research would likely change the estimate, and the outcomes a reader may care about most have not been established either way.
That is a different situation from the weight indication, where the trials are large and the effect is not in doubt — the contrast is set out in is a GLP-1 worth what it costs. Buying on evidence from one indication for a different one is the commonest way a reader overestimates what they are getting.
The practical version
If PCOS is the reason, the honest framing is that you are buying a weight intervention with a plausible knock-on effect rather than a treatment for the condition. That may still be the right purchase. It does mean the price matters more, not less, because the benefit is less certain — and the price is the part this site can actually establish, in what a GLP-1 actually costs and in the gap between what is shown and what is billed.