GLP-1 + Spironolactone + Metformin for PCOS: The Stacking Protocol

Polycystic ovary syndrome affects roughly one in ten women of reproductive age and sits at the intersection of three treatable drivers: insulin resistance, hyperandrogenism, and obesity. The conventional pharmacology — metformin for insulin sensitivity, spironolactone for hirsutism and acne — produces modest weight effects on its own. Adding a GLP-1 changes the math. Jensterle 2015^[1] showed liraglutide produced superior weight loss to metformin in obese PCOS women; Carmina 2023^[4] documented ~80% menstrual restoration on semaglutide; Salamun 2018^[3] reported higher IVF pregnancy rates after liraglutide pre-conditioning. The 2023 international PCOS guideline (Teede et al^[6]) now formally recommends GLP-1 receptor agonists for weight management in PCOS. This article walks through the published evidence for stacking a GLP-1 on top of spironolactone and metformin, the potassium monitoring that the spiro side requires, and the practical protocol for patients trying to conceive.

The honest summary

• PCOS is a Rotterdam-defined syndrome with four phenotypes. The Rotterdam 2003 consensus^[7] requires two of three: oligo-/anovulation, clinical or biochemical hyperandrogenism, and polycystic ovarian morphology on ultrasound. Phenotypes A through D differ in how many of those features are present and predict the severity of metabolic burden.
• The 2023 international PCOS guideline endorses GLP-1 RAs. Teede 2023^[6] updated the recommendation framework and formally added GLP-1 receptor agonists as an option for weight management in PCOS with overweight or obesity, alongside metformin and lifestyle.
• GLP-1 + metformin is additive on insulin sensitivity. Jensterle 2017^[2] randomized obese PCOS women to low-dose liraglutide plus metformin vs high-dose liraglutide alone and found the combination matched the monotherapy on weight while improving glucose handling.
• Spironolactone stays in the stack for androgens. GLP-1 weight loss does lower free testosterone by roughly 10–20% in pooled meta-analysis (Hollanda Morais 2024^[5]), but the anti-androgen effect of spironolactone on hirsutism and acne is independent and not replaced by weight loss alone. The standard stack maintains spiro 50–100 mg daily.
• Potassium monitoring matters. Spironolactone is a potassium-sparing diuretic; GLP-1 dehydration during titration can be additive. Baseline potassium plus a four-week recheck after starting a GLP-1 is the practical safety floor.

What PCOS actually is and why three drugs make sense

The Rotterdam 2003 criteria^[7] define PCOS as the presence of at least two of: oligo- or anovulation, clinical or biochemical hyperandrogenism (Ferriman-Gallwey score above sex-and-ethnicity-specific cutoffs, elevated total or free testosterone, low SHBG), and polycystic ovarian morphology on transvaginal ultrasound. The four phenotypes (A, B, C, D) correspond to which combination of features is present, and phenotype A — the full triad — carries the highest metabolic burden.

The mechanistic logic behind the three-drug stack is straightforward. Insulin resistance is the upstream driver for the majority of PCOS patients; hyperinsulinemia suppresses hepatic SHBG synthesis (raising free testosterone) and stimulates ovarian theca-cell androgen production. Hyperandrogenism then produces the dermatological phenotype (hirsutism, acne, androgenic alopecia) and contributes to anovulation. Obesity amplifies both insulin resistance and adipose-tissue conversion of androgens. Metformin addresses the insulin side; spironolactone blocks the androgen receptor at the peripheral target tissue; a GLP-1 addresses the obesity side at a magnitude no prior PCOS pharmacotherapy has matched.

What the GLP-1 trials in PCOS actually show

Jensterle 2015^[1] randomized 45 obese PCOS women to liraglutide 1.2 mg, metformin 1,500 mg, or roflumilast for 12 weeks. The liraglutide arm lost significantly more weight than metformin and produced larger improvements in BMI and waist circumference. The trial established that GLP-1 monotherapy beats metformin monotherapy on the weight endpoint — the foundation for adding a GLP-1 on top of existing metformin.

Jensterle 2017^[2] followed up with a 12-week randomized trial of low-dose liraglutide (1.2 mg) plus metformin vs high-dose liraglutide (3.0 mg) monotherapy in obese PCOS women. The combination arm matched the high-dose monotherapy on weight loss and delivered superior glucose handling, supporting the practical clinical strategy of keeping metformin in place when starting a GLP-1.

Salamun 2018^[3] randomized 28 obese PCOS women with prior IVF failure to liraglutide plus metformin or metformin alone for 12 weeks before a new IVF cycle. The liraglutide arm produced significantly higher pregnancy rates (85.7% vs 28.6%) and higher cumulative live-birth rates — the strongest signal in the literature that pre-conception GLP-1 conditioning meaningfully improves fertility outcomes.

Carmina 2023^[4] reported on semaglutide in obese PCOS patients who had failed lifestyle programs. About 80% of treated patients recovered regular menstrual cycles within six months alongside significant weight loss, the strongest published evidence for semaglutide specifically in PCOS.

Hollanda Morais 2024^[5] meta-analyzed the RCT evidence and reported pooled effects on weight, BMI, fasting insulin, HOMA-IR, total testosterone, and SHBG. The direction of every effect favored GLP-1 over comparator, and the magnitudes were clinically meaningful: total testosterone fell roughly 10–20% on GLP-1 alone, with SHBG rising in parallel.

The 2023 international PCOS guideline

Teede 2023^[6] codified the 2023 international evidence-based PCOS guideline. The relevant updates from the 2018 version include: GLP-1 receptor agonists are now a formally endorsed option for weight management in PCOS with overweight or obesity, alongside metformin and lifestyle intervention. Combined oral contraceptives remain first-line for menstrual regulation and hyperandrogenism in patients not pursuing fertility. Spironolactone is endorsed for clinical hyperandrogenism when COC is contraindicated or insufficient. The guideline does not prescribe a single stacking sequence but provides the evidence base that supports the combination protocol below.

Magnitude: total testosterone reduction at six months

Potassium monitoring on spironolactone plus GLP-1

Spironolactone is an aldosterone receptor antagonist; the on-target consequence is reduced renal potassium excretion and a small upward shift in serum potassium. The published hyperkalemia rate at the PCOS dose range (50–100 mg daily) is low in young women with normal renal function, but the rate climbs with ACE inhibitor or ARB co-therapy, with potassium supplementation, and during dehydration.

GLP-1 titration produces nausea, occasional vomiting, and reduced oral intake that together can dehydrate a patient through the first eight weeks. The practical monitoring protocol that covers this risk window is: serum potassium and creatinine at baseline before adding the GLP-1, repeat potassium at week 4 after the first dose escalation, and again at week 12 after reaching the target maintenance dose. Hold spironolactone temporarily if potassium exceeds 5.5 mmol/L; investigate and rechallenge at a lower spiro dose if it recurs.

The oral contraceptive interaction with tirzepatide

Tirzepatide reduces the systemic exposure of oral contraceptives during the dose-escalation period. The Lilly-published pharmacokinetic interaction study showed Cmax reductions of roughly 55–66% and AUC reductions of 20–23% after the first 5 mg dose, attenuating over subsequent doses. The FDA Mounjaro and Zepbound labels recommend either switching to a non-oral contraceptive method (transdermal patch, vaginal ring, intrauterine device, or progestin implant) or adding a barrier method for four weeks after starting tirzepatide and four weeks after each dose escalation. Semaglutide does not carry the same labeled interaction. The PCOS triple stack with tirzepatide therefore typically swaps an oral COC for an IUD or transdermal route.

Fertility, pregnancy, and the washout sequence

Both spironolactone and GLP-1 receptor agonists are contraindicated during pregnancy. Spironolactone crosses the placenta and the anti-androgen mechanism carries a theoretical feminization risk to a male fetus; GLP-1s carry an FDA contraindication with no human-pregnancy adequacy data. For patients pursuing conception, the published sequencing is: discontinue spironolactone before active attempts, discontinue the GLP-1 at least eight weeks before conception (covering the ~five-half-life washout for semaglutide and tirzepatide), and continue metformin through early pregnancy if it has been part of the regimen for ovulation induction. The Salamun protocol^[3] specifically used pre-conception liraglutide followed by washout before IVF.

The practical stacking protocol

1. Baseline workup. Total and free testosterone, SHBG, DHEA-S, fasting insulin, fasting glucose, HbA1c, lipid panel, TSH, prolactin, serum potassium, creatinine, eGFR. Ferriman-Gallwey hirsutism scoring. Transvaginal ultrasound if phenotype is uncertain.
2. Establish metformin first. Titrate to 1,500–2,000 mg daily over four weeks. Confirm GI tolerance and add vitamin B12 monitoring at month 12.
3. Add spironolactone at 50–100 mg/day. Confirm reliable contraception is in place (or that conception is not currently planned). Recheck potassium at week 4.
4. Add the GLP-1. Start at the standard induction dose (semaglutide 0.25 mg weekly or tirzepatide 2.5 mg weekly) and follow the published monthly titration ladder. If on tirzepatide and the contraceptive is an oral COC, switch to an IUD, transdermal patch, or implant before the first dose, or add a barrier method for four weeks after each escalation.
5. Re-check potassium at weeks 4 and 12. Hold spironolactone if K+ exceeds 5.5 mmol/L; resume at a lower dose after investigation. Reinforce hydration during titration.
6. Re-check the metabolic and androgen panel at six months. Expect SHBG to rise, total testosterone to fall, free testosterone to fall proportionally more, HOMA-IR to drop, and hirsutism scores to improve.
7. Plan the washout if pregnancy is the goal. Stop the GLP-1 at least eight weeks before active conception attempts; stop spironolactone before active attempts; continue metformin if part of the ovulation- induction regimen.

Insurance coverage and out-of-pocket math

PCOS is a covered indication for GLP-1 weight management on a number of commercial policies. The Cigna IP0206 policy and the UnitedHealthcare commercial drug guide explicitly include PCOS as a qualifying comorbidity in the prior-authorization criteria for semaglutide and tirzepatide for weight management. Most plans still require a BMI threshold (commonly 27 with a comorbidity or 30 alone), prior failed lifestyle intervention, and ongoing lifestyle counseling.

At retail, the metformin plus spironolactone half of the stack runs roughly $20–30 per month at chain pharmacies with manufacturer or discount coupons. Adding a branded GLP-1 takes the monthly cost to roughly $1,350 with a manufacturer savings card on commercial insurance (Wegovy or Zepbound) or to retail cash prices in the $1,000–1,400 range. Compounded semaglutide or tirzepatide via telehealth providers can run $200–500 monthly during the current personalized-prescription compounding window, but the regulatory environment for compounded GLP-1s is unsettled and patients should confirm sourcing and 503A status before relying on a compounded route.

Related research

• GLP-1 for PCOS — the broader evidence — phenotype-by-phenotype treatment response, including the Salamun fertility data in more depth
• GLP-1 + metformin stacking — the DPP, SUSTAIN-7, and the additive math outside of PCOS
• GLP-1 + spironolactone potassium monitoring — the practical hyperkalemia risk and lab cadence
• GLP-1 pregnancy washout — the eight-week discontinuation window before conception

Important disclaimer. This article is educational and does not constitute medical advice. PCOS management decisions, drug combinations, and pre-conception planning should be individualized with a qualified clinician. Spironolactone requires reliable contraception in patients capable of pregnancy because of the theoretical male-fetus feminization risk. GLP-1 receptor agonists are contraindicated in pregnancy. Potassium monitoring frequency may need to be more aggressive in patients on ACE inhibitors, ARBs, potassium supplements, or with chronic kidney disease. PMIDs were verified live against the PubMed E-utilities API on 2026-05-29.

Last verified: 2026-05-29. Next review: every 12 months, or sooner if a new PCOS guideline update or a large PCOS-specific GLP-1 RCT is published.