A Collection of Studies on Spironolactone and Levels of Testosterone and Other Hormones in Men and Transfeminine People
By Aly W. | First published December 19, 2018 | Last modified January 4, 2021
Spironolactone is an antiandrogen that acts as an androgen receptor antagonist but is also known to be a weak androgen synthesis inhibitor (Wiki). A widespread notion in the transfeminine medical literature as well as the transfeminine community is that spironolactone decreases testosterone levels and that this is a major part of its mechanism of action as an antiandrogen. In my exploration of the literature however, I’ve found that the evidence for such claims is poor and there is a great deal of data to the contrary. The purpose of this article is to review the available studies on spironolactone and testosterone levels in people assigned male at birth and to determine whether and to what extent spironolactone decreases testosterone levels in such individuals.
The companion document to this article can be found here. It is a work-in-progress table and collection of studies on spironolactone and hormone levels in cisgender men and transfeminine people.
Many small studies conducted in the 1970s and 1980s assessed the influence of spironolactone on levels of testosterone and other hormones in cisgender men (see the document for sources). In addition, a retrospective chart review by Jerilynn Prior and colleagues described spironolactone and testosterone levels in transfeminine people in the mid-to-late 1980s (Prior et al., 1986; Prior, Vigna, & Watson, 1989). Decades later, additional studies assessed spironolactone and its influence on testosterone levels in transfeminine people (Leinung, 2014; Cheung et al., 2018; Leinung et al., 2018; Liang et al., 2018; Angus et al., 2019; Jain, Kwan, & Forcier, 2019).
The evidence on spironolactone and changes in testosterone levels is mixed. There were 26 individual studies in total (excluding those with the related medication canrenone or potassium canrenoate). In 15 studies (58%), there was no change in testosterone levels; in 6 studies (23%), testosterone levels were decreased; in 1 study (4%), testosterone levels were increased; and in 4 studies (15%), no conclusions could be drawn (not reported, confounded, or other methodological issues). Most of the studies were small, but some of the more recent studies in transfeminine people were larger.
In many of the studies that found no change, this was the case even at very high doses of spironolactone for prolonged durations. For example, one study of 400 mg/day spironolactone for 6 months in young men found no change in gonadotropin, testosterone, or estradiol levels. Notably, this was likewise the case in a couple quite recent and relatively large studies that specifically assessed spironolactone in combination with estradiol in transfeminine people (Leinung, 2014; Leinung et al., 2018). However, a more recent study contradicted these two studies in that it found significantly lower testosterone levels with the combination of spironolactone and estradiol than with estradiol alone (with reportedly no significant differences between the two groups in terms of estradiol dose or levels) (Cheung et al., 2018; Angus et al., 2019). The reason for these highly discrepant results is unclear.
The fact that testosterone levels have usually been unchanged with spironolactone but sometimes are decreased and are almost never increased suggests that spironolactone can be a clinically significant testosterone synthesis inhibitor but that it is only a weakly effective one with highly inconsistent and heterogeneous effects on testosterone levels. In any case, the conflicting findings even in recent studies warrant more research with better study designs.
The use of spironolactone in transfeminine people and the notion that it decreases testosterone levels in our population originated from publications in the mid-to-late 1980s by Prior and colleagues (Prior et al., 1986; Prior, Vigna, & Watson, 1989). In the study, transfeminine people, who were either on high-dose oral estrogen therapy with inadequate testosterone suppression or were pre-hormone therapy, were put on lower-dose oral estrogen therapy in combination with 200 to 600 mg/day spironolactone and 10 mg/day cyclic or continuous oral medroxyprogesterone acetate. The authors reported that despite the lower oral estrogen dosage, testosterone levels significantly decreased, from 169 ng/dL to 87 ng/dL in those who had already been on hormone therapy and from 642 ng/dL to 49 ng/dL in those who had not previously been on hormone therapy.
Prior and her colleagues interpreted their results as demonstrating that spironolactone decreased testosterone levels. However, there is a major confounding factor present that precludes such a conclusion. Medroxyprogesterone acetate, which was used concomitantly, is a progestogen, and progestogens are antigonadotropins that are able to lower testosterone levels themselves (Aly W., 2018; Aly W., 2019). Indeed, studies of medroxyprogesterone acetate have shown it to dose-dependently lower testosterone levels in cisgender men (Wiki). Moreover, medroxyprogesterone acetate has frequently been used in transfeminine people for such purposes, and a recent study showed that the same dosage of medroxyprogesterone acetate used by Prior and colleagues decreased testosterone levels by a considerable degree when added to estradiol and spironolactone therapy (Aly W., 2019). It is highly possible that medroxyprogesterone acetate was responsible for the lower testosterone levels rather than spironolactone. Strangely, Prior and colleagues concluded that spironolactone was responsible for the decreased testosterone levels even though they themselves mentioned in their papers that medroxyprogesterone acetate was included to help suppress testosterone levels (Prior et al., 1986; Prior, Vigna, & Watson, 1989).
It’s notable that Prior’s papers and their methodology and results were presented very confusingly and unclearly. This may have obscured to other researchers the potential role of medroxyprogesterone acetate and that spironolactone wasn’t necessarily causative. The findings of Prior and colleagues were of course in contradiction with the many small studies conducted in the 1970s and 1980s that overall found no influence of spironolactone on testosterone levels in cisgender men.
Surprisingly, no other studies after those of Prior appear to have assessed the influence of spironolactone on testosterone levels in transfeminine people until the mid-2010s. These studies, published more than 25 years after Prior’s, have had mixed findings, with two studies finding no significant influence of spironolactone on testosterone levels (Leinung, 2014; Leinung et al., 2018) but one study observing lower testosterone levels with spironolactone (~300 ng/dL for estradiol alone and ~50 ng/dL for estradiol plus spironolactone) (Cheung et al., 2018; Angus et al., 2019).
The limited influence of spironolactone on testosterone levels is in marked contrast to estrogens like estradiol and progestogens like cyproterone acetate. These agents will essentially always and substantially suppress testosterone levels, provided of course that sufficient doses are used and/or levels are reached (Aly W., 2018). Many transfeminine people don’t realize the capacity of estrogens to suppress testosterone and mistakenly assume that the antiandrogen—often being spironolactone—is mostly or fully responsible for the decrease in their testosterone levels.
In accordance with the limited influence of spironolactone on testosterone levels, many studies have found inadequate testosterone suppression with estradiol plus spironolactone regimens in transfeminine people (Bonzagni et al., 2014; Leinung, 2014; Leinung et al., 2018; Liang et al., 2018; Jain, Kwan, & Forcier, 2019; Sofer et al., 2020). However, there are also studies that have found better testosterone suppression—both with estrogen plus spironolactone and with physiological-dose estradiol alone (Prior et al., 1986; Prior, Vigna, & Watson, 1989; Reardon et al., 2013; Spratt et al., 2014; Cheung et al., 2018; Stewart et al., 2018; Angus et al., 2019; Pappas et al., 2020). It’s unclear why findings differ between studies, but different blood-testing methodologies may in part be responsible.
Although spironolactone doesn’t appear to be very effective for decreasing testosterone levels, it is a competitive antagonist of the androgen receptor in addition to being a weak androgen synthesis inhibitor and hence directly blocks androgens from mediating their biological effects in the body. This is likely to be its main mechanism of action as an antiandrogen and may account for most or all of its therapeutic antiandrogenic efficacy. In accordance, spironolactone is effective in the treatment of androgen-dependent conditions like acne and hirsutism in cisgender women despite not affecting testosterone levels in women in many studies (Aly W., 2020), and spironolactone is effective in the treatment of testotoxicosis (gonadotropin-independent precocious puberty) in boys in spite of having been described as not decreasing testosterone levels in these individuals as well (Holland, 1991). It is also supported by the dose-dependent incidences of breast pain and gynecomastia with spironolactone in cisgender men in spite of testosterone levels being unchanged in many studies.
In any case, spironolactone is also a relatively weak androgen receptor antagonist (Wiki):
Spironolactone has been found to have effectiveness in the treatment of hirsutism in women at a dosage of as low as 50 mg/day. The effectiveness of spironolactone in the treatment of hirsutism in women has been found to be near-significantly greater at a dosage of 100 mg/day relative to a dosage of 200 mg/day (19% ± 8% and 30% ± 3% reduction in hair shaft diameter, respectively; p = 0.07). Levels of free testosterone were unchanged, suggesting that the antiandrogenic efficacy of spironolactone was due exclusively to direct AR blockade. In addition, other studies have found that 100 mg/day spironolactone is significantly or near-significantly inferior to 500 mg/day flutamide in improving symptoms of acne and hirsutism. One study compared placebo and dosages of spironolactone of 50, 100, 150, and 200 mg/day in the treatment of acne in women and observed progressive increases in response rates up to the 200 mg/day dosage. These findings suggest that the antiandrogenic effectiveness of spironolactone is not maximal below a dosage of 200 mg/day, and are in accordance with the typical dosage range of spironolactone of 50 to 200 mg/day in women.
Put another way, spironolactone at typical doses seems best suited for handling female-range levels of testosterone rather than the higher levels that often persist in transfeminine people. Accordingly, animal studies have found cyproterone acetate, flutamide, and bicalutamide to be far more effective androgen recetor antagonists than spironolactone (Bonne & Ranaud, 1974; Hecker, Hasan, & Neumann, 1980; Sivelle, Underwood, & Jelly, 1982; Snyder, Winneker, & Batzold, 1989; Wiki-Table; Yamasaki et al., 2004). Due to its relatively weak strength as an androgen receptor antagonist and its limited effectiveness in lowering testosterone levels, spironolactone is likely a less effective antiandrogen than other options like cyproterone acetate, bicalutamide, and GnRH agonists/antagonists.
Taken together, based on the available evidence, spironolactone doesn’t appear to consistently decrease testosterone levels in people assigned male at birth and instead usually leaves them unchanged. Rather than by decreasing testosterone levels, spironolactone seems to act mainly as an androgen receptor antagonist. In this regard, aside from its antimineralocorticoid activity and other off-target endocrine actions, spironolactone is similar to selective androgen receptor antagonists like bicalutamide. In any case, the available findings on spironolactone and testosterone levels are conflicting, and more research is needed to fully settle this issue. In addition to its weak and inconsistent effects on testosterone levels, the androgen receptor antagonism of spironolactone is relatively weak and the medication is likely a less effective antiandrogen than other options.