This week: the PCOS piece we have been promising, now that the reproductive research is finally coming into focus, and a beauty-aisle comparison that deserves a grade.

Let's go. ☕

GLP-1s and PCOS, the reproductive picture

That question keeps showing up around GLP-1s and PCOS: what happens to my cycle, my hormones, my ovulation, my chances of getting pregnant? Until recently, the reproductive evidence was scattered across small trials and observational studies. A new review now pulls much of it into one place. Here's the picture, in four parts.

Why PCOS and GLP-1s keep colliding

Polycystic ovary syndrome affects an estimated 10 to 13% of women of reproductive age, roughly 1 in 10, and it sits right at the intersection of reproductive hormones and metabolism. PCOS is a leading cause of irregular ovulation and infertility, and insulin resistance, obesity and type 2 diabetes are also more common in women with the condition.

The relationship is not as simple as insulin resistance causes weight gain, which causes PCOS. These pieces can feed into one another: insulin resistance and metabolic dysfunction can worsen hormonal abnormalities, while excess weight can further worsen insulin resistance and PCOS symptoms. That overlap helps explain why GLP-1 medications have become part of the conversation.

The international PCOS guideline already says medications including liraglutide and semaglutide can be considered for management of higher weight in adults with PCOS, using the same general principles applied to the wider population. That is not the same thing as FDA approval to treat PCOS itself. And that distinction matters.

What the new review actually found

A scoping review published September 28 gathered 28 studies looking at reproductive and gynecologic outcomes after GLP-1 receptor agonist treatment in women with PCOS. The researchers looked at menstrual cycles, androgen levels, ovulation, pregnancy, ovarian function and metabolic outcomes.

The strongest signal was not fertility. It was menstrual function and hormones. Across the included studies, improvements in menstrual regularity and reductions in amenorrhea were frequently reported. Studies also found improvements in markers of hyperandrogenism, including testosterone, free androgen index and sex hormone-binding globulin. Weight, BMI and insulin-resistance measures improved in nearly all of the included studies.

Ovulation and pregnancy are where the evidence gets thinner. Some studies reported improved ovulation or pregnancy rates, particularly in selected populations and when GLP-1 medications were combined with other treatments. But those outcomes were much less consistently studied, and the authors could not pool the studies into one quantitative answer because the drugs, study designs and outcomes were too different.

And another 2026 systematic review that limited its analysis to randomized trials reached a more cautious conclusion: GLP-1 drugs produced modest short-term weight loss in women with PCOS, but the evidence was still insufficient to draw firm conclusions about menstrual, metabolic or other reproductive outcomes. A separate 2026 meta-analysis focused specifically on liraglutide did find improved menstrual frequency and insulin resistance, but ovulation and pregnancy outcomes could not be pooled because the data were too sparse.

That is probably the fairest summary right now: there is a reproductive signal. There is not yet a reproductive promise. And most importantly, these drugs have not been proven fertility treatments for PCOS.

Off-label is not the same as experimental

Worth saying plainly, because the internet blurs this one. Off-label means an FDA-approved medication is being used for a condition, population, dose or other circumstance that is not included in its approved labeling. The FDA says clinicians generally may prescribe an approved medication for an unapproved use when they judge it medically appropriate. That does not mean the FDA reviewed that particular use and decided it works. The approval applies to the indications the FDA actually evaluated.

There is also an important wrinkle with PCOS. If a woman with PCOS meets Zepbound's approved criteria for obesity or overweight with a weight-related condition and receives it for weight management, the prescription may be completely on-label for weight, even though she also has PCOS. What remains off-label is saying Zepbound is specifically treating her PCOS, restoring her ovulation or improving fertility.

And that is a very different situation from retatrutide. Retatrutide is still investigational. Lilly reported positive Phase 3 results this summer and plans to submit it for FDA review in 2027, but it has not been approved, and the FDA says retatrutide cannot currently be used in compounding under federal law. Anything being sold directly to consumers as “research retatrutide” is not an FDA-approved medication.

Approved drug used off-label. Investigational drug. Gray-market vial claiming to contain that investigational drug. Those are three different conversations.

The part that surprises people: returning fertility

PCOS is one of the most common causes of irregular or absent ovulation. So if menstrual and ovulatory function improve, pregnancy may become possible even after years of unpredictable cycles. That does not mean GLP-1 medications have been proven to restore fertility. It means irregular periods are not reliable contraception.

And pregnancy language differs by drug. For Wegovy used for weight management, the current label says to discontinue it when pregnancy is recognized and to stop semaglutide at least two months before a planned pregnancy because of its long half-life. For Zepbound, the label says to discontinue the medication when pregnancy is recognized; unlike Wegovy, the current U.S. label does not specify a set preconception washout period.

Tirzepatide also comes with another reproductive warning: because Zepbound delays gastric emptying, the label says oral hormonal contraceptives may be less effective. Patients using them are advised to switch to a non-oral method or add a barrier method for four weeks after starting Zepbound and for four weeks after every dose increase. And retatrutide does not have an FDA pregnancy label because it does not have an FDA-approved label at all.

So “my cycle was irregular anyway” is not a contraception strategy.

What to bring to the appointment

• Have my cycles changed since starting treatment, and could that change how we think about ovulation?
• If pregnancy is possible, does my contraception still make sense with this specific medication?
• If pregnancy is the goal, what does the label say about stopping this particular drug?
• Are we measuring success only by weight, or are there PCOS-specific symptoms or metabolic markers we should follow too?

The point is not to turn a weight-management prescription into a fertility treatment. It is to remember that PCOS is reproductive and metabolic. Both deserve to be part of the conversation.

The evidence is growing, and there are legitimate reasons these medications come up in PCOS care, especially when weight and metabolic health are part of the picture. But the studies women need most, large trials designed specifically around PCOS, ovulation, fertility and long-term reproductive outcomes, are still missing. That is the gap.

“This $16 serum is the closest thing to $1,300 peptide injections.”

What the evidence says. The comparison fails on the first question our primer teaches: which door? A topical peptide serum sold for appearance is a cosmetic. With limited exceptions, cosmetics do not need FDA premarket approval, and companies are not required to prove to the FDA that a wrinkle serum works before selling it. They are still responsible for product safety and truthful labeling. An injectable peptide is a different category entirely: depending on the compound, it may be an FDA-approved drug, a legitimately compounded medication or an unapproved research product.

And route matters. Your skin is a barrier, and penetration depends on the peptide and the formulation. For acetyl hexapeptide-8, Argireline, one of the best-known “Botox-like” cosmetic peptides, an in-vitro study using human skin found that most of the applied peptide remained on or near the surface: about 0.22% was measured in the stratum corneum, about 0.01% in the epidermis, and none was detected in the dermis under the study conditions. A later review reached the same broader point: limited permeability through intact skin, though formulation can change how well it penetrates.

That does not mean topical peptides do nothing. A 2026 systematic review of 19 randomized trials found some improvements in skin-aging outcomes, including a modest pooled effect on wrinkles, although the evidence was heterogeneous and much of the wrinkle benefit was driven by oral rather than topical peptide studies. So there is real cosmetic science here. It is just not injectable-peptide science. Sometimes the topical and injectable products even share the name of a peptide. That still does not make the route, exposure, evidence or risk equivalent.

Calling a serum “the closest thing” to an injection is like calling a postcard of Paris the closest thing to the flight. And the direction of the comparison matters: it flatters the serum while quietly turning the injection into the “real” version you graduate to. That is exactly the on-ramp we keep talking about.

The fair version of the claim: some topical peptide products have modest human evidence for cosmetic outcomes such as wrinkles or skin texture. Judge the serum on those terms. Not as a substitute for, or a gateway to, something in a vial.

Worth your inbox

Researching health for yourself is one job. Doing it for an aging parent at the same time is two. GrayBrief is a plain-English education resource for adult children navigating elder care: what to ask, what to watch for, and what nobody tells you until it's urgent. If that's your season of life, it's worth a look.

No. Off-label does not mean experimental. It means an FDA-approved medication is being used for something outside the uses the FDA specifically approved. Clinicians can prescribe approved medications off-label when they believe it is medically appropriate. What it does not mean is that the FDA has reviewed the evidence and approved that particular use.

With PCOS, I would first clarify what your doctor is actually treating. If you qualify for Zepbound under its approved weight-management indication, your treatment may be on-label even though you also have PCOS. What would be off-label is using it specifically to treat PCOS or assuming it has been proven to restore cycles, ovulation or fertility.

I would ask what you and your prescriber are watching besides the number on the scale, whether that is your cycle, metabolic markers or other PCOS symptoms. And if pregnancy is possible, make contraception part of that conversation. PCOS does not mean pregnancy cannot happen, and Zepbound has a specific warning that oral hormonal contraceptives may be less effective for four weeks after starting it and after every dose increase.

Off-label does not mean you are somebody's experiment. It means the evidence for that specific use deserves its own conversation.

Mais, PharmD, Medical Reviewer, Pep in Her Step

• Waking at 3am on a GLP-1: the sleep disruption cluster nobody warned anyone about.
• Pill versus pen: what the oral Wegovy data actually shows.
• The muscle question, answered properly: protein, lifting, and what the scans actually say.