PCOS — insulin, androgens & ovulation

What actually decides this outcome, in order of size

The pathway's own framing is correct and this section exists to put numbers under it. What decides how this behaves:

1. Insulin, acting at two places at once. Insulin is a co-gonadotropin at the ovarian theca cell, amplifying LH-driven androgen synthesis through CYP17A1, and it simultaneously suppresses hepatic production of sex hormone binding globulin. More androgen made, less of it bound. That is why a single metabolic lever moves two hormonal numbers and why the free fraction moves further than the total does.
2. Which of three outcomes is actually wanted. Ovulation this year, androgenic symptoms over eighteen months, and cardiometabolic risk over decades are three different goals with three different first-line agents, and the commonest mistake on this page is treating them as one. The 2023 international guideline is organized around exactly that separation Teede 2023.
3. Whether this is the diagnosis at all. Non-classic congenital adrenal hyperplasia, hyperprolactinemia, thyroid disease and androgen-secreting tumors all present with irregular cycles and hyperandrogenism. Each is excluded by a specific test rather than by pattern recognition, and anti-Mullerian hormone has been evaluated as a component of the diagnosis in adults rather than as a screening test on its own Piltonen 2024.
4. The size of the ovulation-induction effect, which is known. Randomizing 750 women to letrozole or clomiphene produced live births in 27.5% against 19.1% (P=0.007) and ovulation in 61.7% of cycles against 48.3% Legro 2014. Neither of those drugs is on this pathway's list, and that is the most useful sentence on this page.
5. Where the metabolic agents actually rank for that outcome. In 626 infertile women randomized to clomiphene, metformin or both, live birth rates were 22.5%, 7.2% and 26.8% respectively, with metformin significantly worse than either comparator (P<0.001) Legro 2007. Metformin is a good drug for insulin resistance and was the worst single choice for a baby, and both halves of that are true.

The order to run these in, and what has to be true first

Confirm the diagnosis, decide which of the three outcomes you are buying for, then treat insulin. The metabolic work is first for symptoms and risk, and it is not first for conception.

1. Exclude the imitators once, on one draw. 17-OH Progesterone in the early follicular phase for non-classic congenital adrenal hyperplasia, Prolactin, TSH (Thyroid-Stimulating Hormone) with Free T4 (Thyroxine), and DHEA-S with Androstenedione because a markedly raised adrenal androgen points somewhere other than the ovary. A very high Total Testosterone is a reason to stop and investigate rather than to start a supplement.
2. Characterize the metabolic half. Fasting Insulin with HbA1c (Hemoglobin A1c), SHBG (Sex Hormone-Binding Globulin) and Free Testosterone alongside Total Testosterone, LH & FSH, and Anti-Müllerian Hormone (AMH) where the diagnosis is uncertain Piltonen 2024. SHBG is the most sensitive single indicator of the insulin lever on this page, because it is made by the liver in inverse proportion to portal insulin.
3. Myo-Inositol first among the supplements, because it has the most pooled data. Inositol phosphoglycans act as second messengers downstream of the insulin receptor, and myo-inositol is converted to D-chiro-inositol by an insulin-dependent epimerase, which is the mechanistic argument for the 40:1 ratios that are sold. Early randomized work showed improved ovarian function and metabolic parameters Gerli 2007, and two systematic reviews reach a favorable conclusion, one of them written to inform the 2023 guideline update Fitz 2024 Greff 2023.
4. Metformin next where insulin resistance is the target rather than pregnancy. It inhibits mitochondrial complex I, raises the AMP/ATP ratio and activates AMPK, which suppresses hepatic gluconeogenesis. Its place in the ovulation ranking is settled and unflattering Legro 2007; its place in the metabolic ranking is not.
5. Berberine and Dihydroberberine argue the same AMPK mechanism from the botanical side. Berberine's oral bioavailability is under 1%, limited by P-glycoprotein efflux and by extensive first-pass metabolism, which is the entire reason the dihydro form exists. It also inhibits CYP3A4, so it is the item here most likely to change the exposure of something already prescribed.
6. NAC, Alpha Lipoic Acid, Chromium, Magnesium, Zinc, Vitamin D and Omega-3 (Fish Oil) are the substrate and redox layer. Alpha-lipoic acid is a mitochondrial cofactor for pyruvate dehydrogenase with an insulin-sensitizing signal of its own; NAC supplies cysteine for glutathione. These are adequacy and adjunct arguments, not primary ones.
7. Saw Palmetto and DIM sit on the androgen and estrogen handling side. Saw palmetto is argued as a 5-alpha-reductase inhibitor, and diindolylmethane shifts estrogen hydroxylation toward the 2-hydroxy pathway; Estrogen metabolism & clearance is where that second claim is examined properly. GLP-1 Support Stack belongs to the metabolic goal and is listed here because weight-independent improvements in insulin sensitivity change every number above.

What gets bought for this that cannot move it

An insulin sensitizer is the wrong first choice if the goal is a pregnancy this year. That is not an opinion: 7.2% live births on metformin against 22.5% on clomiphene, in 626 randomized women Legro 2007, and letrozole beat clomiphene in turn Legro 2014. A reader who spends eighteen months on inositol with conception as the goal has chosen a real mechanism over a tested outcome, and the cost of that choice is measured in cycles.

Androgen suppression and conception are opposing goals that require opposite precautions. The agents used to reduce androgenic symptoms are the ones with the strongest reasons not to be taken during a pregnancy, which means the two outcomes cannot be pursued in the same month without a plan. Deciding which one this year is for is the sequencing question the guideline is built around Teede 2023.

The category that fails structurally here is anything aimed at the ovary rather than at insulin. The follicular arrest is downstream: intrafollicular androgen excess and a high LH-to-FSH drive stall selection of a dominant follicle. Supplements marketed for ovarian support with no insulin mechanism are addressing the last step in a chain whose first step is still running.

And if the cycles are regular and the complaint is metabolic, this is the wrong page. Insulin resistance without oligo-ovulation is Glucose disposal, absorption & the post-meal curve and AMPK activation & cellular fuel sensing. If the goal is egg quality with ovulatory cycles, that is Fertility & egg quality. A raised 17-OH Progesterone is a different diagnosis entirely, and no page on this site is the right one for it.

How you would know it was working, on a real read-out and a real timescale

This is one of the few goals on the site with a binary read-out: either a cycle ovulated or it did not, and a single blood test says which. The prediction: if insulin was the lever, SHBG rises before testosterone falls, and ovulation returns after both have moved.

• Progesterone in the mid-luteal phase, roughly seven days before the expected period rather than on a fixed day 21. A value above about 3 ng/mL is evidence that ovulation occurred in that cycle. On an irregular cycle, day 21 is frequently the wrong day, and a low result drawn on the wrong day is the commonest false negative in this entire goal.
• SHBG (Sex Hormone-Binding Globulin) at 12 weeks, as the earliest metabolic signal. Hepatic SHBG production is suppressed by insulin, so it rises when portal insulin falls, often before Fasting Insulin itself looks convincingly different. A rising SHBG with an unchanged total testosterone still means less free androgen.
• Free Testosterone rather than Total Testosterone at 12 to 16 weeks. The free fraction is what reaches the follicle and the hair follicle, and it moves further than the total because SHBG is moving underneath it. Calculated free testosterone from total and SHBG is more reliable at female concentrations than most direct assays.
• Fasting Insulin with HbA1c (Hemoglobin A1c) at 12 weeks. Insulin is the sensitive number and HbA1c is the slow one; in a young woman with normal glucose, HbA1c can be entirely unremarkable while fasting insulin is the whole finding.
• Androgenic skin and hair changes take two to three cycles longer than the blood does. The terminal hair growth cycle runs months, so hirsutism responds on a hair-cycle clock rather than a hormonal one, and judging an intervention on skin at 8 weeks judges it too early.

What will fool you. Cycle length in this condition is variable by nature, so one regular cycle is not evidence and three consecutive ones start to be. Anti-Müllerian Hormone (AMH) is typically high here and does not fall usefully with treatment, so it is a diagnostic input rather than a response marker Piltonen 2024. And any hormonal contraceptive suppresses the entire panel above, which means the numbers cannot be interpreted while one is being taken.

Sources read for these sections

• Legro RS. Letrozole versus clomiphene for infertility in the polycystic ovary syndrome. New England Journal of Medicine 2014;371(2):119-29 · PMID 25006718
• Legro RS. Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. New England Journal of Medicine 2007;356(6):551-66 · PMID 17287476
• Teede HJ. Recommendations from the 2023 International Evidence-based Guideline for the Assessment and Management of Polycystic Ovary Syndrome. Human Reproduction 2023 · PMID 37580037
• Fitz V, et al. Inositol for Polycystic Ovary Syndrome: a systematic review and meta-analysis to inform the 2023 update of the International Evidence-Based PCOS Guidelines. Journal of Clinical Endocrinology and Metabolism 2024 · PMID 38163998
• Greff D, et al. Inositol is an effective and safe treatment in polycystic ovary syndrome: a systematic review and meta-analysis of randomized controlled trials. Reproductive Biology and Endocrinology 2023 · PMID 36703143
• Gerli S, et al. Randomized, double blind placebo-controlled trial: effects of myo-inositol on ovarian function and metabolic factors in women with PCOS. European Review for Medical and Pharmacological Sciences 2007 · PMID 18074942
• Piltonen TT, et al. Utility of Serum Anti-Mullerian Hormone Measurement as Part of Polycystic Ovary Syndrome Diagnosis. Seminars in Reproductive Medicine 2024 · PMID 38776986