Medications for women's hair loss: what do antiandrogens and oral minoxidil offer, and what's the risk?

Four drugs are in play, and each has a different regulatory status. Spironolactone is a diuretic also used for heart failure and high blood pressure; for hair loss, it's prescribed off-label.2 Cyproterone acetate is an antiandrogen, meaning a drug that reduces male hormones' effect; the European Medicines Agency (EMA) restricted its 10 mg a day or higher-dose products in 2020.9 Finasteride and dutasteride's summary of product characteristics contraindicates them in women.1112 And low-dose oral minoxidil hasn't been approved by the American drug regulator (FDA) for hair loss.17

This article is about female pattern hair loss, where hair thins along the parting over years. If the shedding started suddenly, over weeks, and is even across the whole scalp, that's a different condition, which I cover in the telogen effluvium article. There's more on telling the two types apart in the women's hair loss article.

I'm not a doctor, I'm training to be a cosmetologist. This article gathers what official labels and trials say about these drugs, so you know what to ask your doctor.

Finasteride and dutasteride block DHT's formation, spironolactone blocks the male hormone's binding to its receptor. DHT (dihydrotestosterone) forms from testosterone via an enzyme called 5-alpha reductase; finasteride blocks this conversion in certain tissues.11 Spironolactone competitively blocks the male hormone from binding to its receptor inside the cell.5

Whether female pattern hair loss is caused by male hormones isn't settled. That's exactly the starting point of Sinclair and colleagues' 2005 trial: it isn't clearly established that female pattern hair loss is of male hormone origin, nor that it responds to oral antiandrogens.4 In Famenini and colleagues' 2015 American data series, a small proportion of women had high testosterone (5 of 61), and those with acne or excess hair growth as well responded better to spironolactone.5 Ask your doctor this: do I have signs of a hormonal difference, and does that need a blood test?

I found one small, placebo-controlled trial, and in it the hair-count difference wasn't statistically significant. Werachattawatchai and colleagues split 48 premenopausal women aged 21 to 45 into two groups in Thailand in 2025: for 24 weeks they got either 100 mg spironolactone or placebo daily, and everyone also used a 3% minoxidil solution twice daily. Hair density and hair thickness rose in both groups. In the trial's primary measure, hair density and thickness, the difference between the two groups wasn't significant (total hair count p=0.086, thickness p=0.206); the number of thick hairs, in the 43 women who completed the trial, rose by an average of 9.48 hairs per square centimetre in the spironolactone group and 5.32 in the placebo group, which approached significance.3

Based on photos, a secondary measure, moderate or marked improvement was seen in 8 women in the spironolactone group (38 percent) and 2 in the placebo group (9 percent), and this difference was significant. The side effect was also more common in the spironolactone group: 9 of 24 women developed irregular periods, 2 of whom dropped out of the trial because of it, versus none on placebo. At week 24, a transient, symptom-free potassium rise was measured in 2 spironolactone and 1 placebo woman, which resolved on repeat testing. The authors list the small sample size as a limitation.3

I found only this one trial against placebo; the other randomised trials compared spironolactone with another treatment, and most of the data series are uncontrolled. Liang and colleagues split 120 premenopausal women into three groups in China in 2022: 5% topical minoxidil alone, minoxidil plus 80–100 mg spironolactone daily, or minoxidil plus microneedling. After 24 weeks, hair density grew more in the spironolactone group than with minoxidil alone, there was no difference in thickness, and this group reported the most side effects. Participants weren't blinded to their treatment, only the assessor didn't know which group each person was in.20 Jha and colleagues (2026) split 204 women in India, double-blind, for 24 weeks, into 100 mg spironolactone daily or 50 mg bicalutamide daily, another antiandrogen used off-label for hair loss similarly to spironolactone.2123

Sinclair and colleagues studied 80 women in 2005: 40 got 200 mg spironolactone daily, 40 got cyproterone acetate, for at least 12 months, without placebo. The two drugs' results didn't differ, so they were pooled: hair regrew in 44 percent, hair density didn't change in 44 percent, and it kept shedding in 12 percent. From this the authors write that 88 percent of women can expect their condition not to worsen, or to improve, and that a placebo-controlled trial is needed for comparison.4

Vexiau and colleagues compared cyproterone acetate against 2% topical minoxidil in 66 women in a 12-month randomised trial in 2002. The cyproterone group got 52 mg cyproterone acetate daily for 20 of 28 days along with ethinylestradiol, which falls within today's restricted range of 10 mg a day or more. In this group, the number of hairs thicker than 40 micrometres decreased on average, while it rose in the group using minoxidil alongside the contraceptive. Acne and excess hair growth improved more in the cyproterone group than with minoxidil; but within the cyproterone group, hair count didn't differ depending on whether other signs of male hormone excess were present.22

Famenini and colleagues pooled two data series in 2015: of 39 women taking spironolactone, 29 (74.3 percent) reported their hair loss stopped or improved; the average dose was 110 mg. This is retrospective, self-reported data, without a control group.5 A 2023 meta-analysis pooling 192 patients across four trials calculated a 56.6 percent improvement rate, but the trials' results varied a lot.6

In my own Europe PMC search, I found only one small trial against placebo, and in it everyone also used minoxidil. What you can take from this: there's encouraging data on spironolactone, but it hasn't been measured on its own, against placebo.

Spironolactone's summary of product characteristics doesn't list hair loss among its indications, and contraindicates the drug in several situations. The British Aldactone label's indications are heart failure, oedema from liver cirrhosis, malignant ascites, nephrotic syndrome, and adrenal aldosterone overproduction. It's contraindicated in acute kidney failure, significant kidney damage, Addison's disease, high blood potassium, and alongside other potassium-sparing diuretics. Per the label, severe potassium rise can develop alongside potassium supplements, a potassium-rich diet, potassium-based salt substitutes, ACE inhibitors, angiotensin II antagonists, or non-steroidal anti-inflammatories.1

In healthy, young women, routine potassium monitoring's benefit was small in one large, retrospective trial, but that trial was about women with acne. Plovanich and colleagues reviewed 1,802 potassium measurements from 974 healthy young women taking spironolactone for acne in 2015: the rate of high potassium was 0.72 percent, essentially the same as such women's baseline rate (0.76 percent). The authors say routine testing is unnecessary in healthy women taking it for acne.7 This wasn't measured in women taking it for hair loss, and the trial excluded people with heart and kidney disease. Ask your doctor this: do I need potassium and kidney function testing, and how often?

Per the label, it can be given in pregnancy only if the expected benefit justifies the risk, and not during breastfeeding. In animal trials, reproductive harm was seen tied to spironolactone's anti-male-hormone effect.1 Per a 2026 review, the concern is feminisation of a male foetus, but this mostly comes from animal trials: the systematic review cited found only 5 cases, and these case reports describe normal genital development. Despite this, the authors still recommend not taking spironolactone when planning pregnancy, at least 1 month before, and during pregnancy.8 The 2025 placebo-controlled trial didn't enrol any woman without adequate contraception.3

The non-clinical section of the American Aldactone label's December 2025 version states that spironolactone was carcinogenic in rat feeding trials.19 A 2022 meta-analysis, pooling data from 7 trials totalling 4,528,332 participants averaging 62 to 72 years old, found no significant link between spironolactone and breast cancer, though the certainty was very low.2

Over meningioma risk: cyproterone acetate at 10 mg a day or more can only be given for hair loss if another treatment hasn't worked. The EMA's pharmacovigilance committee (PRAC) recommended the restriction on 13 February 2020, and the co-ordination group for mutual recognition and decentralised procedures (CMDh) adopted it on 25 March 2020. In the authority's words: products with 10 mg a day or more can only be used for hirsutism (excess hair growth), androgenic hair loss, acne and oily skin if other treatment options, including lower doses, haven't worked; after improvement, the dose should be gradually reduced to the lowest effective one. Meningioma is usually not malignant, but because of its location it can cause a serious problem.9

The risk is rare, and rises with the total amount taken. Per the EMA, it may affect 1 to 10 in 10,000 people, depending on dose and treatment duration; most cases were reported after several years of a daily dose of 25 mg or more. For the small, 1 to 2 mg cyproterone combined tablets, the data don't signal a risk, but as a precaution these also can't be given to anyone who has or has had meningioma.9 A 2026 meta-analysis, pooling five observational trials, found an odds ratio of 12.36 for meningioma with cyproterone acetate, with moderate certainty, and also described the tumour shrinking after stopping the drug. The same analysis found no signal for spironolactone; that doesn't mean the risk is proven to be zero.10

Per the EMA, the doctor should watch for symptoms: vision changes, hearing loss or tinnitus, loss of smell, worsening headache, memory problems, seizures, limb weakness. If meningioma is diagnosed, cyproterone-containing products must be stopped for good.9 In Sinclair's 2005 trial, cyproterone acetate was 50 mg daily, or, before menopause, 100 mg daily for 10 days a month, which falls within today's restricted range of 10 mg a day or more.4 Ask your doctor this: why this drug for me, what came before it, and how do we reduce the dose?

The daily 1 mg finasteride's summary of product characteristics contraindicates the drug in women, and so does dutasteride's. Per the Propecia label, finasteride isn't indicated in women, and is contraindicated over pregnancy risk: in a pregnant woman it may cause abnormal development of a male foetus's external genitalia.11 Avodart's (dutasteride) label contraindicates it in women, children and adolescents, its indication being benign prostate enlargement.12 This article is about women taking it themselves; if your partner takes it, I've written about that in the finasteride and having children article.

A 2026 review recommends stopping finasteride at least 14 days, and dutasteride 5 to 6 months, before trying to conceive, because dutasteride's half-life is roughly 5 weeks.8

After menopause, a placebo-controlled trial found no effect. Price and colleagues treated 137 postmenopausal women aged 41 to 60 for a year with 1 mg daily finasteride or placebo, double-blind, in 2000. After a year there was no difference in hair count change, hair count fell in both groups at the front and middle of the crown, and neither the patients, nor the assessors, nor blinded photo assessment showed improvement.13

A 2021 meta-analysis, pooling 490 women across 9 trials, found that hair density didn't increase with finasteride. But per the assessors' judgement, 81 percent of women responded: 70 percent at doses under 2.5 mg daily, 86 percent at higher doses. The authors say the trials varied a lot, the reliability of the result is weak, and all three randomised trials used 1 mg.14 The picture is therefore mixed: the placebo-controlled measurement found no effect, the more favourable result comes from the assessors' judgement.

I wrote about the effects and side effects measured with finasteride and dutasteride in men in the guide to oral DHT blockers. Ask your doctor this: why would this come up for me after menopause, when its own label contraindicates it in women?

Oral minoxidil is a blood-pressure medication, used off-label for hair loss at low doses. Sinclair's 2018 paper describes both minoxidil and spironolactone as oral blood-pressure medications.18 A 2023 review states that low-dose oral minoxidil has no FDA approval for hair loss. Its most common side effects are excess hair growth and cardiovascular symptoms; women are more prone to the excess hair growth than men; as an individual, dose-independent side effect it also describes pericardial effusion.17

Ramos and colleagues compared 1 mg daily oral minoxidil with a 5% topical solution in a 2020 randomised trial for female pattern hair loss. The paper is a short communication, without an abstract, and I couldn't access its full text; the same research group's 2023 review states that both groups improved, with no difference between the groups, and a mild trend in favour of the oral form.1516 In 2018, Sinclair gave 100 women one capsule daily, containing 0.25 mg minoxidil and 25 mg spironolactone, for 12 months, without a control group: shedding and thinning scores decreased, no one developed high potassium, 8 women had a side effect, and 2 women stopped due to hives.18

I wrote about topical minoxidil in the article on minoxidil. Ask your doctor this: which form, topical or oral, makes sense for me, and if it's the latter, how do we monitor my blood pressure and cardiovascular symptoms?

Prescribing and follow-up are the doctor's job. Worth bringing these with you, because the contraindications and precautions in the above labels ask about them:

• since when and how your hair has been thinning;
• what medication and supplements you take, including potassium;
• whether you have kidney, heart or liver disease;
• whether you're planning a child, breastfeeding, and what contraception you use;
• whether you've ever had a meningioma.

Microneedling is a mechanical stimulus, it doesn't contain medication; based on its mechanism we wouldn't expect a hormonal effect, but this hasn't been measured. It's been studied in women with female pattern hair loss alongside minoxidil, with clinic-performed microneedling, not with our fixed 1 mm at-home device. I wrote up what that study measured, and what it didn't, in the article on female hair loss.