By Aly W. | First published August 29, 2018 | Last modified April 2, 2021
AMAB people including cisgender men and transfeminine people are exposed to high levels of testosterone in utero due to their testes. In rodents, prenatal testosterone exposure irreversibly masculinizes not only the genitals and brain but also the mammary glands, stunting their growth and limiting their developmental potential in response to estrogens later in life. It is unknown if this also occurs to any degree in humans. However, there is some clinical support for the notion that it may occur in us as well. Moreover, transfeminine people appear to tend to have poor or suboptimal breast development relative to cisgender women (Wierckx, Gooren, & T’Sjoen, 2014; de Blok et al., 2018; Reisman, Goldstein, & Safer, 2019), and this phenomenon could, unfortunately, explain this. Alternatively, the phenomenon may not occur in humans, and other factors (e.g., hormonal regimens) might instead be responsible for the lesser breast development in transfeminine people. More clinical research is needed to determine whether or not this phenomenon indeed occurs in humans.
Male mice have testes due to their XY chromosomes, whereas female mice have ovaries due to their XX chromosomes. While in utero, the testes produce high levels of testosterone, whereas the ovaries do not. These high levels of testosterone masculinize the brain and genitals during a critical window in prenatal development, resulting in male-typical structures in male mouse fetuses, whereas the absence of testosterone results in the brain and genitals being feminized in female mouse fetuses. In addition to the brain and genitals, prenatal sexual differentiation occurs in the mammary glands of mice. Testosterone prevents the genesis of the nipples in male mouse fetuses during a critical period of prenatal development. It is unclear if testosterone itself is responsible for this or if testosterone is 5α-reduced into dihydrotestosterone (DHT) in the mammary gland, similarly to in the external genitalia, and DHT is responsible for the inhibition of the development of the mammary gland; findings appear to be mixed. Postnatally, the nipples are absent in male mice and the mammary glands of male mice have diminished responses to estrogens and progestogens; they are only capable of partial, stunted development following exogenous administration of estrogens and progestogens compared to female mice. This appears to be because exposure to testosterone in utero, in addition to prenatal inhibition of mammary gland growth, permanently reduces the expression of the estrogen receptor and aromatase in the mammary glands, limiting their capacity to respond to estrogens and by extension progestogens.
Treatment of pregnant mice with an antiandrogen such as cyproterone acetate (an androgen receptor antagonist), cyanoketone (a 3β-hydroxysteroid dehydrogenase (3β-HSD) inhibitor and hence testicular testosterone biosynthesis inhibitor), or 390 MSD (a 5α-reductase inhibitor and hence DHT biosynthesis inhibitor) during the critical window of mammary gland sexual differentiation results in feminization of the mammary glands of male mouse fetuses such that they are indistinguishable postnatally from those of female mice. Likewise, destruction of the testes of male mouse fetuses via targeted external X-ray irradiation prior to the critical window of mammary gland sexual differentiation results in the same observations. Conversely, treatment of pregnant mice with an androgen (androgen receptor agonist) such as testosterone, methyltestosterone, or DHT during the critical window of mammary gland sexual differentiation results in masculinization of the mammary glands of female mouse fetuses such that they are indistinguishable postnatally from those of male mice. One study reported that testosterone was more potent than DHT in this context. Besides their determination of gonadal type (i.e., testes in males and ovaries in females) and the consequent prenatal hormonal milieu (i.e., high testosterone in males and absence of testosterone in females), karyotype (i.e., XY chromosomes in males and XX chromosomes in females) has no role in the sexual differentiation of the mammary glands in mice.
Although we know that prenatal androgen exposure masculinizes the genitals and brain in both mice and humans, and masculinizes the mammary glands in mice, it is unknown if prenatal androgen exposure similarly masculinizes the mammary glands in humans. There are important species differences between mice and humans concerning prenatal sexual differentiation. As an example, aromatization of androgens into estrogens masculinizes the brain in male mouse fetuses, whereas this does not appear to occur in male human fetuses, in whom testosterone is thought to be solely responsible for brain masculinization (Zuloaga et al., 2008; Motta-Mena & Puts, 2017; Puts & Motta-Mena, 2018). Another notable species differences between mice and humans is of course that male mice do not have nipples whereas male humans do have nipples. It has also been stated that whereas the mammary glands of newborn male mice are smaller and structurally dimorphic relative to those of newborn female mice, the mammary glands of male and female human newborns are virtually indistinguishable in morphology. However, it has been reported as well that subtle differences between the male and female human prenatal mammary glands can be observed. In addition, breast tissue in infant girls is larger than that in infant boys, which has been associated with higher testosterone levels in infant boys (Francis et al., 1990; Schmidt et al., 2002; Jayasinghe et al., 2010).
These findings suggest that the phenomenon may not occur in humans or that it may occur to a lesser and/or characteristically differential extent. Unlike the case of mice however, which can very easily be studied experimentally, it doesn’t seem clear how systematically the mammary glands of male and female human newborns have been compared (e.g., in terms of size, etc.). In addition, although the mammary glands of male and female human newborns might be morphologically very similar on a macroscopic level, there may nonetheless be microscopic differences, such as differences in cellular gene expression (e.g., of the estrogen receptor) and hormonal sensitivity, as in mice. For instance, prenatal androgen exposure could induce epigenetic changes in mammary stem cells, which reside in the ductal system and are responsible for growth of the mammary ductal tree and by extension the entire breast. Moreover, there is some clinical evidence to support the notion that prenatal androgen exposure may indeed be involved in mammary gland sexual differentiation in humans, which I will address in the next section.
In contrast to rodents, the prenatal effects of androgens and antiandrogens on the mammary glands of monkeys via administration to pregnant monkey dams unfortunately do not seem to have been studied.
Congenital adrenal hyperplasia (CAH) due to 3β-HSD2 deficiency is a rare intersex condition in which undervirilization of males occurs due to deficiency of testicular 3β-HSD and hence deficiency of testosterone during prenatal development. Conversely, testicular 3β-HSD and hence testosterone levels seem to be relatively normal postnatally (likely due to compensation by the 3β-HSD1 isoform of the enzyme), so these individuals undergo normal spontaneous puberty. A portion of males with 3β-HSD2 deficiency (approximately 35 to 55%, based on two reviews) develop gynecomastia during puberty, with some of these cases being described as “marked”. Moreover, this can occur in spite of testosterone and estradiol levels that are in the normal male ranges. It has been theorized that this may be due to insufficient exposure of the mammary glands to androgens prenatally such that the breasts are more sensitive to estrogens in puberty and adulthood, as with animal models. However, there is high clinical heterogeneity in the reported cases of CAH due to 3β-HSD2 deficiency, and steroid hormone levels in general are also abnormal in the condition. It is possible that the gynecomastia that occurs in the condition might alternatively be due to aromatization of the excessive amounts of adrenal androgens in the circulation. Another alternative possibility pertains to the fact that the breasts contain all the enzymes needed to transform adrenal androgens into testosterone and DHT, and one of these enzymes is 3β-HSD (although I’m not sure which of the two isoforms) (Labrie et al., 2001); loss of 3β-HSD in the breasts and hence diminished androgenic signaling in this tissue in adulthood could be responsible for the gynecomastia seen in 3β-HSD deficiency.
It has been theorized that gynecomastia in general, in addition to an imbalance in the ratio of androgens to estrogens, may have insufficient prenatal androgen exposure and hence prenatal mammary gland sexual differentiation as an important contributing factor in many cases. In accordance, small clinical studies have reported increased estrogen receptor and aromatase content in the breasts of men with gynecomastia. On the other hand however, it is notable that it is very rare for the breasts of males with gynecomastia to approximate those of females, even with high-dose antiandrogen and/or estrogen therapy, and this may reflect prenatal androgen exposure such that the developmental potential of the breasts is fundamentally limited compared to that of females.
Aromatase excess syndrome (AEXS) is a rare intersex condition in which peripheral aromatase, the enzyme that converts androgens into estrogens, is extremely overactive. As a result of this, both males and females with the condition are hyperestrogenic. In addition, both males and females with AEXS develop premature female-type pubertal maturation at a young age, including gynecomastia in the boys (Photo). This begins shortly following adrenarche (average age about 6 years), when the adrenal glands begin secreting weak androgens. These adrenal androgens are excessively aromatized and induce breast development. However, whereas about 50% of women with AEXS are said to have “breast hypertrophy” (excessively large breasts, although I don’t think it’s meant in the form of the rare disease), males with the condition seem to on average have relatively small (and certainly not hypertrophic) breasts compared to females, even if their gynecomastia is definitely marked (Photos). That said, these males also do have higher testosterone levels than females with the condition. However, their testosterone is excessively aromatized, is relatively low, and notably doesn’t start to be secreted until the age of normal male puberty (~12 years old). If the prenatal androgen exposure hypothesis is true for breast development in humans, it may explain the divergent breast observations in males and females with AEXS.
Complete androgen insensitivity syndrome (CAIS) is a rare intersex condition in which 46,XY individuals (“biological/genetic males”) have a defective and completely non-functional androgen receptor. As a result, these individuals, who have testes due to their Y sex chromosome and produce high testosterone levels in utero, experience no masculinization during prenatal development and are born perfectly phenotypically female. They are highly feminine both physically and behaviorally (as well as female in terms of gender identity and androphilic in terms of sexuoromantic orientation), and as a result, are, appropriately, assigned female at birth. The testes never descend in CAIS women, remaining internal in the abdominal cavity. At puberty, the testes produce high levels of testosterone and result in a male hormonal profile, although of course without any masculinization due to the insensitivity of CAIS women to androgens. The testosterone is aromatized into estradiol, which results in elevated but still relatively low estradiol levels (about 50 pg/mL). Nonetheless, estradiol levels are only about 10 to 55 pg/mL during Tanner stages 1 to 4 in normal pubertal girls (Wiki-Table), so this is sufficient for feminization and breast development in CAIS women (Photo, Photos). Conversely, CAIS women, lacking ovaries, produce no progesterone of importance (<1 ng/mL) (Barbieri, 2013).
In spite of their relatively low levels of estradiol and absence of progesterone, CAIS women show complete breast development at puberty, and actually have breasts that are said to be large on average for women (Barbieri, 2013). One review amusingly described the breasts in CAIS women as “jumbo-sized” in fact (Morris, 1953). In accordance with their lack of progesterone, CAIS women show no lobuloalveolar tissue in the breasts on histological examination, suggesting that progesterone and lobuloalveolar maturation are not importantly involved in external morphological breast development (Morris, 1953). The only fundamental biological difference between men and CAIS women is the presence versus absence of a functional androgen receptor, and this illustrates the powerful role of androgens in opposing feminization and breast development. Moreover, the excellent and even above-average breast development in CAIS women is in marked contrast to the relatively small breasts in most cases of gynecomastia in males, gynecomastia due to AEXS in males, and notably gynecomastia due to high-dose androgen receptor antagonist monotherapy (e.g., with bicalutamide or enzalutamide) in men. The last is described as only mild-to-moderate in 90% of cases. The comparatively large breasts in adult CAIS women may be due to their absence of prenatal androgen-receptor signaling. However, other factors might also or alternatively be involved in the divergent breast findings, such as postnatal androgen-receptor signaling, age, and others.
5α-Reductase type 2 deficiency is a rare intersex condition in which the type 2 isoform of 5α-reductase is absent, partial DHT deficiency is present, and undervirilization of male fetuses occurs. This manifests specifically as partially feminized genitalia at birth and as minimal androgen-dependent body hair growth and scalp hair loss in adulthood. However, masculinization at puberty is otherwise normal. Gynecomastia is said not to occur in males with the condition, which suggests that 5α-reductase type 2 may not be involved in prenatal androgen exposure-mediated mammary gland sexual differentiation, assuming of course that it occurs in humans. The type 1 and 3 isoforms of 5α-reductase remain functional in males with the condition however. Moreover, testosterone may be responsible for mammary gland sexual differentiation due to prenatal androgen exposure rather than DHT, as described previously. Gynecomastia occurs at only low rates of about 1.2 to 3.5% in men treated with 5α-reductase inhibitors like finasteride and dutasteride (Wiki).
An implication of the notion that prenatal androgen exposure inhibits later breast development in humans is that transmasculine people, even if they started hormones before puberty, may be at a higher risk of gynecomastia with testosterone therapy due to the minimal exposure of their breast tissue to androgens in utero. I am unaware of whether this is indeed the case or if we have any data on this issue.
Transfeminine people tend to have poor or suboptimal breast development relative to cisgender women (Wierckx, Gooren, & T’Sjoen, 2014; de Blok et al., 2018; Reisman, Goldstein, & Safer, 2019). It’s possible that exposure of the mammary glands to high levels of androgens prenatally, resulting in lower levels of estrogen receptors in the mammary glands, diminished responsiveness to estrogens, and an ultimately compromised capacity for breast development later in life, is either responsible for or is one of multiple contributing factors to this. This, unfortunately, would likely be irreversible and something that nothing could be done about. However, although sexual differentiation of the mammary glands due to prenatal androgen exposure is known to be the case in mice, and although it is supported by some clinical findings in humans, it is not yet certain whether it is actually the case in humans as well. In addition, there are some data that can be regarded as support against it (e.g., presence of nipples in male humans, reportedly very similar mammary glands in male and female newborns), and other factors (e.g., hormonal regimens) might instead be responsible for the lesser breast development in transfeminine people. In any case, prenatal androgen exposure as a limiting factor in our breast development is nonetheless very plausible, and one that could still prove to be true. More clinical research is needed to confirm or reject the notion that this phenomenon indeed also occurs in humans.
Although this possibility is obviously quite discouraging, I felt that it was necessary to cover it. In future posts I will cover other possibilities for the lesser breast development in transfeminine people that we may be able to do something about.
The document here is a collection of literature excerpts (ordered by date) that are relevant to this topic. They are intended to source and provide elaboration on the discussion above.