Link Search Menu Expand Document

Summary of Side Effects and Health Risks of Cyproterone Acetate (CPA)

By Aly W. | First published December 31, 2018 | Last modified October 5, 2020


This article is a summary and discussion of the content on the following Wikipedia page:

Note in particular the tables for published case reports of liver toxicity, prolactinoma, and meningioma.

The following is a summary of some of the most important adverse effects relevant to transfeminine people.

Liver Loxicity

It seems that CPA has a dose-dependent risk of elevated liver enzymes, increasing from around 3% at doses of 1 to 20 mg/day and to 20 to 30% at doses of 50 to 300 mg/day. In spite of this however, cases of liver damage and liver cancer have, with only a single exception (50 mg/day), been observed exclusively at doses of 100 to 300 mg/day. Moreover, although hundreds of cases of liver toxicity have been reported, the vast majority of cases have been in aged men with prostate cancer. As such, age and dosage would appear to be major risk factors for CPA-induced liver toxicity. An instance of CPA-induced liver toxicity has never been reported in a transgender woman, although there are a number of cases in cisgender women—including one case in a young cisgender woman taking CPA for hirsutism. There are only three known cases of death with CPA due to liver toxicity in younger people. In two of these cases, very high doses of CPA were given to young children for years. And in one of the cases, it was a CPA-containing birth control pill and the occurrence was very likely just coincidental. In any case, the incidence of liver toxicity with CPA may be considerably decreased via the use of liver function tests to ensure that liver enzymes stay within normal ranges. As such, monitoring of liver enzymes is highly recommended even if the risk may seem low. That said, the risk of liver problems with CPA is likely to be extremely small at low doses, for instance 20 mg/day or less.

Blood Clots

CPA has an important risk of blood clots, both alone and in combination with an estrogen. The risk is likely dose-dependent. It seems to be pretty similar for doses across a range of 25 to 300 mg/day, but is likely to be significantly lower at very low doses (e.g., the 2 mg/day dosage used in birth control pills). Nonetheless, even low doses appear to significantly increase the risk of blood clots. In any case, blood clots are primarily only a concern in individuals with specific risk factors. These include advanced age, obesity, smoking, cancer, concomitant use of high-dose, oral, and/or non-bioidentical estrogens, and genetic predispositions to blood clots. Alarming rates of blood clots (5%) have been observed in transgender women treated with high-dose oral ethinylestradiol and high-dose CPA, particularly in transgender women over 40 years of age (12%). Conversely, the risk is far lower, albeit still somewhat concerning, when high-dose CPA is combined instead with transdermal estradiol (only one case seen; 0.7% incidence). In combination with transdermal estradiol or another parenteral estradiol route at reasonable estradiol doses, the risk is likely to be even lower with low doses of CPA, for instance 10 mg/day or less. Under these circumstances, the absolute risk may prove to be very small, as the risk of blood clots normally is.

Brain Tumors

CPA has an important risk of benign brain tumors—specifically, prolactinomas and meningiomas. A total of 13 cases of prolactinomas have been reported. All cases that provided the CPA dosage used high doses (50 to 200 mg/day) except for one (2 mg/day)—and this case notably used high-dose estrogens (35 μg/day oral EE plus 10 mg/week i.m. E2-EN). The risk of prolactinoma seems to be strongly dependent on the dosage of CPA and of the estrogen. Using low doses of CPA and reasonable doses of estrogens is likely to greatly minimize the risk. Prolactinomas have taken anywhere from 6 months to 15 years to develop. A study of about 2,500 transgender women observed an incidence of about 0.35% or 1 in 284—with a mean treatment interval of about 7.5 years and a mean CPA dosage of about 85 mg/day. A study of about 300 transgender women using high-dose oral EE and high-dose CPA (e.g., 100 mg/day) observed an incidence of high prolactin levels—a potential surrogate of a growing prolactinoma—of 15%. In half of those cases (about 7.5% of the whole sample), elevated prolactin levels persisted in spite of medication discontinuation or dosage reduction, and in about a third of those cases (roughly 2% of the whole sample), pituitary enlargement—suggestive of a prolactinoma—was observed on a brain scan. Prolactinomas can cause central nervous system side effects and can require surgery. In any case, the absolute risk is obviously small, and the risk may be comparatively quite low at CPA doses of 10 mg/day or less, provided that the estrogen dosage is also reasonable.

There have been many more case reports of meningioma with CPA; about 50 so far. Moreover, in contrast to prolactinoma, the CPA dosages have been wide-ranging and highly variable, from 10 mg/day to 200 mg/day, with a considerable number of reported cases at lower doses (10 to 25 mg/day). And meningiomas are far less dependent on estrogens than are prolactinomas, with numerous cases occurring with CPA alone. Despite the fact that case reports of meningioma are more numerous than prolactinoma however, the incidence seems to be about the same; a study of about 2,500 transgender women observed an incidence of meningiomas of about 0.31% or 1 in 320—with a mean treatment interval of about 19 years and a mean CPA dosage of about 47 mg/day. The reason for the excess number of published case reports of meningiomas relative to prolactinomas might be that meningiomas may generally be more severe in symptomatology than prolactinomas. Whereas prolactinomas occur only in the pituitary gland, meningiomas can occur anywhere in the meninges, and hence anywhere on the outer portion of the brain—frontal, temporal, parietal, occipital lobe or elsewhere. Meningiomas can produce central nervous system side effects and can require surgery to correct. However, frequently mere discontinuation of CPA can result in considerable regression, which can obviate the necessity of surgery. Although less dosage-dependent than prolactinomas, the risk of meningiomas may be significantly lower at very low doses of CPA (e.g., less than 10 mg/day).


Low-dose (e.g., 2 mg/day) cyproterone acetate, for instance in birth control pills, has no risk of depression. Moreover, the risk of depression with higher doses of CPA may be considerably less than previously thought. This is on the basis of clinical studies that were in some cases rigorous and well-designed, checking carefully for changes in mood. The severity of depression with CPA may also be much less than previously thought. The risk of depression with CPA may be isolated mainly to when it is used without an estrogen, such as in men with prostate cancer. Only small, uncontrolled, and poor quality clinical studies have observed high incidences of depression with CPA. But many of these studies were in women with severe hirsutism, who had high baseline rates of emotional disturbances due to the nature of their condition. And in at least some cases these studies lumped “depression” and “fatigue” into a single “mood changes” group—fatigue at high doses notably being a more well-established side effect of CPA. Other research has likewise found very low rates of mood changes with medroxyprogesterone acetate, a progestin that is closely related to CPA (Wiki).

It is possible that a nocebo effect (i.e., a psychogenic effect based on pre-existing knowledge and expectations) may be occurring with transgender women when it comes to CPA and depression. It is notable that about 75% of the mood-lifting effect of antidepressants is estimated to be due to the placebo effect (Mora, Nostoriuc, & Rief, 2011), so this is well within the realm of possibility. (Another possibility is misattribution, as transgender people have high baseline rates of depression.) A nocebo effect for depression with finasteride has likewise been suggested (Kuhl & Wiegratz, 2017).

Vitamin B12 Deficiency

This is best explained and summarized by the Wikipedia passage itself:

Both low-dose (2 mg/day) and high-dose CPA combined with an estrogen have been associated with vitamin B12 deficiency in women in some small studies.[60][61][11] Vitamin B12 deficiency in turn has been associated with depression, anxiety, irritability, and fatigue due to depletion of central monoamine neurotransmitters.[62][63] For this reason, it has been suggested that low vitamin B12 levels might be involved in the side effect of depression that is sometimes associated with CPA.[36] Serum vitamin B12 monitoring and supplementation as necessary may be recommended during CPA therapy.[11][60][61]

However, these have been small and low-quality clinical studies, and it appears clear that low-dose CPA is not associated with an increased rate of depression. More research is necessary to better characterize and establish any vitamin B12 deficiency that may occur with CPA. Also, more research would be required before vitamin B12 deficiency could be properly implicated in any mood changes associated with CPA.

Glucocorticoid Effects

This section is actually based on information here and elsewhere. CPA is only a partial glucocorticoid with weak affinity for the glucocorticoid receptor, and does not appear to have clinically significant glucocorticoid effects except at very high doses (e.g., 200 to 300 mg/day) in small children. Accordingly, adrenal insufficiency and associated withdrawal effects/risks are likely only of significant concern at very high doses in small children likewise. As such, glucocorticoid effects with CPA under typical circumstances can be considered essentially negligible, especially at low doses.

Summary and Conclusions

High-dose CPA, especially in combination with oral, non-bioidentical, and/or high-dose estrogens, as well as in people with specific risk factors, has significant risks of rare but dangerous and even life-threatening adverse effects. However, the absolute risks of these adverse effects are small, and the risks are dose-dependent. The use of CPA at very low doses and in combination with safe estrogen therapy is likely to significantly minimize most or all of the risks of CPA. Moreover, as has been discussed in other places, the testosterone-suppressing effect of CPA is likely to be maximal at a dosage as low as 5 to 10 mg/day. As such, higher doses of CPA are not necessarily required, and may simply be unnecessarily adding to risk as well as costs. If one really desires additional androgen blockade on top of testosterone suppression, it would probably be a much better idea to just add an androgen receptor antagonist like bicalutamide or spironolactone on top of very-low-dose CPA instead of using high doses of CPA. As for the depression and vitamin B12 deficiency risks of CPA, they may be of less significance/importance than previously believed, particularly at very low doses of CPA—in which no increase in depression risk may occur (per 2 mg/day CPA).