Appendix C

Commissioned Paper: Physiological Sex Differences and the Effect of Gender-Affirming Hormone Therapy

Daniel J. Slack,¹ Jiby Yohannan,¹ Derek Chen,² Nithya Krishnamurthy,² Joshua D. Safer^1,3

¹Division of Endocrinology, Department of Medicine, Icahn School of Medicine at Mount Sinai, ²Icahn School of Medicine at Mount Sinai, ³Center for Transgender Medicine and Surgery, Mount Sinai Health System

Prepared for the National Academies of Sciences, Engineering, and Medicine Committee on Sex and Gender Identification and Implications for Disability Evaluation

Correspondence: Dr. Daniel J. Slack, Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, New York, NY, USA 10029-6574; TEL: 212-241-6500; Daniel.slack@mssm.edu

Sex differences in normal physiology and disease pathophysiology are widespread throughout the human body. These differences range from readily apparent anatomical dimorphisms to subtle variations in signal transduction at the molecular level. They may have implications regarding the interpretation of serologic biomarkers and imaging, the diagnosis of disease, and the appropriate selection of therapeutics. Circulating endogenous sex hormones, particularly estrogen and testosterone, are the main driving factor behind some, but not all, of these differences. The study of human disease associated with states of sex hormone excess and deficiency provides a crucial window into the differential effect of sex hormones between the sexes.

Examining the clinical benefits and risks associated with the use of exogenous hormone therapy to replace hormone deficiencies or treat disease has been similarly informative.

Transgender and gender diverse (TGD) individuals have a gender identity that differs from the sex recorded at their birth based on observation of external genitalia. They may seek gender-affirming therapies, particularly gender-affirming surgery (GAS) and gender-affirming hormone therapy (GAHT).

Feminizing GAHT generally involves the use of exogenous estrogens, with or without adjunctive agents that also have antiandrogenic properties, to develop female secondary sex characteristics while suppressing or minimizing male secondary sex characteristics. Conversely, masculinizing GAHT generally involves the use of exogenous testosterone to develop male secondary sex characteristics. This narrative review takes a systems-based approach to synthesize the extant literature on physiologic sex differences, how gender-affirming care for TGD individuals may affect those sex differences, and the potential clinical implications of those changes.

BONE HEALTH AND BODY COMPOSITION

Sex Differences in Bone Health and Body Composition

An individual’s bone mass at any point in adulthood is representative of their peak bone mass minus that which has been subsequently lost. Peak bone mass is achieved, on average, in the late teenage years and early twenties in cisgender females and by the second decade in cisgender males. Peak bone mass is primarily determined by heredity, with nutritional factors, physical activity, and circulating hormones playing a secondary role (Cosman et al., 2014). Adolescence is crucial for skeletal development, as the bone gained during adolescence accounts for up to 60 percent of peak bone mass (Wojtys, 2020). Estrogen and growth hormone (GH) are essential in the pubertal growth spurt, irrespective of sex. Estrogen stimulates insulin-like growth factor 1 (IGF-1) production by increasing the secretion of GH. IGF-1 is the most critical factor in promoting the maturation of chondrocytes and osteoblasts, which leads to epiphysial fusion (Finlayson et al., 2016). This is clinically evident in individuals with estrogen-receptor deficiency or aromatase deficiency, the latter a condition resulting in impaired conversion of testosterone to estrogen. These individuals exhibit diminished bone age, decreased bone density, and continued growth into the third decade. In those with aromatase deficiency, exogenous estrogen administration has been shown to advance bone age and increase bone density (Bilezikian et al., 1998).

There is less consensus regarding testosterone’s role in achieving peak bone mass and maintaining bone health. Although puberty starts earlier in cisgender

females, it lasts longer in cisgender males who exhibit increased periosteal bone formation and more significant marrow cavity enlargement, resulting in larger bones and greater bending strength (Duan et al., 2003; Rothman and Iwamoto, 2019). It is difficult to say how much of this observed dimorphism is due to a direct effect of testosterone on the bone, partly because the data from animal models do not agree with that from human studies. For example, while androgen-receptor knockout mice have reduced periosteal bone formation, patients with complete androgen insensitivity syndrome with nonfunctional androgen receptors have mostly normal bone mineral density (BMD) (Vanderschueren et al., 2014). Indirectly, testosterone contributes to observed sex differences in BMD through its effects on body composition. Whereas prepubertal individuals start with equal lean body mass, skeletal mass, and body fat, at maturity, cisgender males have approximately 1.5 times the lean body mass, skeletal mass, and muscle mass of cisgender females. In contrast, cisgender females have twice as much body fat as cisgender males (Loomba-Albrecht and Styne, 2009). Androgens, including testosterone, are known to increase muscle mass and decrease body fat, thus contributing to the observed sexual differentiation in body composition throughout puberty (Almeida et al., 2017).

In addition to attaining peak bone mass, estrogens also play a crucial role in maintaining bone health throughout adulthood. It is thought that, at least in part, this is due to estrogen suppression of interleukin 6 (IL-6) secretion, a cytokine that stimulates the proliferation of osteoclast precursors, thus accelerating bone turnover (Florencio-Silva et al., 2015). The impact of the loss of endogenous estrogen on bone health is demonstrated in cisgender females experiencing menopause, whereby an increase in daily calcium loss, reflecting an increase in bone resorption over formation, results in, on average, a loss of one standard deviation of BMD and a two-to-threefold increase in fracture risk after 10 years (Black and Rosen, 2016). Hormone replacement therapy (HRT) using conjugated equine estrogens has been shown to increase BMD and reduce the risk of hip and vertebral fractures by about 35 percent (Karim et al., 2011). Low-dose conjugated estrogens and estradiol also increase BMD, although their antifracture effect remains unknown (Tella and Gallagher, 2014).

In the adult cisgender male population, too, estrogen is critical to maintaining bone health. This was examined in a study in which goserelin acetate, a gonadotropin-releasing hormone (GnRH) agonist, was administered to cause hypogonadism. Then, exogenous testosterone was added back to one group in varying doses. At the same time, a second group received add-back testosterone in the same doses with the addition of an aromatase inhibitor. In the group with add-back testosterone only, BMD trended down with lower testosterone levels but did not differ significantly from controls, whereas in the group that also received aromatase inhibition, BMD declined by 1–2 percent in all groups, independent of the testosterone level (Finkelstein et al., 2016).

Our understanding of the differential impact of sex steroids on cortical versus trabecular bone continues to evolve. It was previously thought that the effect of estrogen was predominately on trabecular bone. Still, more recent data reveal that trabecular bone loss occurs in the third decade, before sex steroid deficiency in most individuals (Khosla et al., 2011). Cortical bone, on the other hand, maintains relative stability throughout adulthood before decreasing linearly in older adults. However, cortical bone is lower in postmenopausal cisgender females compared with age-matched cisgender males; this is potentially because of the rapid period of bone loss during menopause (Drake and Khosla, 2013). It seems likely that varying concentrations of estrogens and androgens in aging adults play a key role in apparent sex differences in bone density (Rothman and Iwamoto, 2019). When studying bone loss in later life, it is vital to consider and control for the progressive deficits in renal and intestinal function that occur in normal aging and may impair calcium homeostasis, irrespective of the effects of sex hormones.

GAHT, Bone Health, and Body Composition

Low bone mass has been observed with relatively high prevalence in transfeminine individuals (Fighera et al., 2018; T’Sjoen et al., 2009). These findings were previously thought to be related to inadequate estrogen treatment in the context of suppressed or absent gonadal sex hormone secretion, but multiple studies have now shown that transfeminine individuals tend to have lower bone mass and lower Z-scores compared with age-matched cisgender males even before any GAHT or GAS (Van Caenegem et al., 2013, 2015a). In addition to reproducing these findings, a more recent study found that transmasculine individuals also had low bone mass before GAHT, particularly at the femur, compared with cisgender female controls, but without significant differences in BMD Z-scores as determined by dual-energy X-ray absorptiometry (DEXA) (Ceolin et al., 2023). In that study, researchers aimed to investigate driving factors of low bone mass before treatment in TGD individuals, finding that TGD individuals had lower vitamin D values compared with their cisgender counterparts, that transfeminine individuals had more total fat mass while transmasculine individuals had more total lean mass, and that transfeminine individuals were more likely to be active smokers and to spend time indoors. Thus, it appears likely that body composition and lifestyle factors contribute to low BMD in TGD adults before initiation of GAHT.

Many studies report a positive change in BMD as early as 1 year following initiation of estrogen in transfeminine individuals (Singh-Ospina et al., 2017; Wiepjes et al., 2017). Long-term studies are sparse, but in a cohort study of transfeminine individuals followed for over 10 years, reassessment

of bone density following 10 years of GAHT found a significant increase in Z-score despite no change in BMD at the lumbar spine (Wiepjes et al., 2019). In this study, a positive association between serum estradiol and lumbar spine BMD was noted such that those in the highest tertile of estradiol (mean 121 pg/mL) had a significant increase in lumbar spine BMD while those in the lowest tertile (mean 32 pg/mL) had a significant decrease in lumbar spine BMD. No relationship between level of luteinizing hormone (LH) or degree of testosterone suppression was observed. Only one study appears to have been powered to evaluate fracture outcomes in TGD individuals. In this study, fracture risk was found to be higher in older transfeminine individuals (age ≥50 years) than in age-matched reference cisgender males but not cisgender females. In contrast, in younger transfeminine individuals, fracture risk was increased compared with age-matched reference cisgender females but not cisgender males (Wiepjes et al., 2020). In a prospective cohort study of TGD adults newly initiated on GAHT, including both masculinizing and feminizing regimens, serum bone turnover markers (BTMs) decreased after a year of treatment, with modest negative correlations between changes in BTMs and changes in BMD noted (Vlot et al., 2019). Taken together, these data support the concept that the administration of exogenous estrogen in transfeminine individuals can increase bone density and potentially decrease fracture risk, even in the setting of testosterone suppression.

It is also essential to consider the impact of body composition changes that occur with GAHT and may influence bone health. Before GAHT, transfeminine individuals differ from age-matched cisgender males on measures of body composition, with lower muscle mass, strength, and lean body mass (Van Caenegem et al., 2015a). Feminizing GAHT increases fat mass and decreases lean mass, shifting body composition to mirror the average natal female body more closely (Alvares et al., 2021). It appears probable, then, that the high prevalence of low bone mass observed in transfeminine individuals may be due, in part, to low lean body mass and that some of the improvement in BMD seen with initiation of feminizing GAHT may be related to the appreciable increase in fat mass that occurs. Indeed, in a study of transfeminine individuals on feminizing GAHT for variable amounts of time, investigators found a positive correlation between appendicular lean mass and total fat mass with BMD, explaining 14.9 percent of the observed variation on lumbar spine BMD and 20.6 percent of the variation in total femur BMD (Fighera et al., 2018).

Recent data show that transmasculine individuals may have low bone mass at baseline compared with age-matched cisgender women before initiation of GAHT, mirroring trends seen in transfeminine individuals (Ceolin et al., 2023). Importantly, however, consensus has not been reached on this point; prior, smaller studies have reported similar BMD and body composition measures between transmasculine individuals and age-matched

cisgender females (Van Caenegem et al., 2012, 2015b). Unlike transfeminine individuals receiving exogenous estrogen as part of feminizing GAHT, gains in BMD with testosterone treatment in transmasculine individuals are reported with less consistency. In one study, BMD measures did not change after a year of testosterone treatment, despite increases in bone formation and resorption markers; the authors proposed that the latter effect may reflect an anabolic effect of testosterone treatment rather than bone loss (Van Caenegem et al., 2015b). In a similar, more extensive cohort study assessing transmasculine individuals at baseline and following 1 year of testosterone therapy, increases in the total hip and lumbar spine but not femoral neck BMD were observed, with the increase in lumbar spine BMD found to be much more prominent in those who initiated treatment at an older age (≥50 years) (Wiepjes et al., 2017).

Longer-term data have become available more recently, showing that BMD was similar but that the lumbar spine Z-score had increased after 10 years of testosterone treatment. Again, this change was driven by those in the oldest age group when they initiated treatment (≥40 years) (Wiepjes et al., 2019). Similarly, in a study examining trends in BTMs in individuals newly initiated on GAHT, a difference was noted between younger transmasculine individuals (≤50 years), who experienced an increase in BTMs, whereas the older individuals did not (Vlot et al., 2019). This apparent divergence has led the authors to postulate a role for increased estradiol action via aromatization. However, a definitive relationship between serum estrogen or testosterone and BMD in transmasculine individuals has not been elucidated. Despite the lack of consensus regarding increases in BMD in transmasculine individuals with the use of GAHT, it is reassuring that there does not appear to be an appreciable decline in BMD despite a relative reduction in estrogen levels. Additionally, more recent data powered to evaluate fracture outcomes do not point to an increased risk of fracture in younger transmasculine individuals, although the risk in older transmasculine individuals remains unknown (Wiepjes et al., 2020).

As with estrogen, it is essential to consider the effects of testosterone use as masculinizing GAHT on measures of body composition that may influence BMD. As previously noted, prior to GAHT, transmasculine individuals have similar bone and body composition compared with age-matched cisgender female controls (Van Caenegem et al., 2015b). Several studies have reported changes in body composition following the initiation of testosterone treatment, including an increase in muscle mass and grip strength, a higher waist-to-hip ratio, and a decrease in fat mass, having the net effect of shifting the composition and contour of the body toward that of the average natal male body (Van Caenegem et al., 2012, 2015b). It seems then that acquiring lean mass in place of fat mass with testosterone helps stabilize BMD in transmasculine individuals. It remains unclear how

much of the effect of testosterone on BMD is via such indirect mechanisms versus via a potential direct effect on the bone. In one of the studies, statistical models were used to compare transmasculine individuals before and after GAHT with cisgender female controls. They found a positive association with bone size and endosteal circumference at the radius after adjusting for grip strength, suggesting a direct effect of testosterone on the bone (Van Caenegem et al., 2012).

Research investigating the potential differential impact of various hormonal preparations and routes of administration (ROAs) of GAHT on bone health in TGD individuals has been limited. While data are mixed, it seems unlikely there is an actual differential impact of various sex hormone ROAs on BMD (Cirrincione and Narla, 2021). Further longitudinal research will help reach consensus and may stratify BTMs by ROA to investigate whether there are differences in how various preparations influence bone metabolism. Additionally, there is a need to investigate whether the use of adjunctive antiandrogens influences bone health in transfeminine individuals, as there are currently no data in this area.

Puberty is a critical period for the growth and development of the musculoskeletal system, regulated in part by circulating endogenous sex hormones. In TGD adolescents, GnRH agonists may be used to delay puberty to allow time to determine gender-affirming treatment goals without the development of unwanted secondary sexual characteristics. Through continuous stimulation, GnRH agonists inhibit the pulsatile secretion of gonadotropin, which markedly reduces the production of gonadal sex hormones and results in the cessation of pubertal development (Panagiotakopoulos, 2018). Absence of sex hormones may result in harmful effects on the bone that may be reversible. Experience using these medications to care for individuals with conditions such as central precocious puberty (CPP) and prostate cancer has been informative in elucidating potential effects on bone in TGD youth. Individuals with CPP experience early pubertal development resulting from premature hypothalamic-pituitary-gonadal axis activation. In these individuals, GnRH agonists are administered to delay puberty, generally for several years, to allow the adolescents to go through puberty at a more typical age. Studies investigating the long-term impact on BMD in patients with CPP treated with GnRH agonists have found that, while BMD levels decrease during GnRH agonist treatment, the effects are reversible and bone mass was sufficiently preserved after treatment (De Sanctis et al., 2019). Additionally, BTMs were found to be elevated before treatment, decreased while on GnRH agonists, and stabilized after treatment (van der Sluis et al., 2002). GnRH agonists are also used as a form of androgen deprivation therapy (ADT) for the treatment of prostate cancer. In this population, ADT is associated with a substantial decrease in BMD throughout the skeletal system at multiple sites, with the magnitude of BMD loss

increasing with the duration of therapy and significantly increasing one’s risk of fragility fracture (Pietzak and Mucksavage, 2016). Reassuringly, there does appear to be some reversibility in ADT-induced bone loss for those who discontinue the medication (Wang et al., 2017). Additionally, higher levels of BTMs were observed on ADT and normalized when treatment was withdrawn, in line with the observations pertaining to BMD.

The extant literature examining the effect of GnRH agonist use on bone health in TGD youth is more limited in scope, particularly regarding long-term outcomes and fracture risk, owing to the relatively recent introduction of pediatric GAHT. Before treatment with GnRH agonists or GAHT, TGD youth tend to have lower BMD compared with reference standards for sex designated at birth, mirroring observations in TGD adults (Lee et al., 2020). As with other conditions, TGD youth have been shown to experience a decrease in BMD Z-scores after initiation of GnRH agonists (Lee et al., 2020). Recently, reassuring longitudinal data have become available. In a cohort study of TGD individuals treated with GnRH agonists during adolescence followed by GAHT with an average follow-up duration of 15 years, Z-scores caught up with pretreatment levels on all accounts except for the lumbar spine in transfeminine individuals, which may have been related to low estradiol concentrations in this group (van der Loos et al., 2023). It appears, then, that the effect of pubertal suppression on peak bone mass is more delayed attainment rather than attenuation.

Previously, only one study was cited widely, wherein a small number of TGD youth treated with GnRH agonists (average age of initiation: 15 years) followed by GAHT (average age of initiation: 16.5 years) were evaluated. Individuals had not entirely made up their bone loss by age 22, as Z-scores were still lower than baseline despite a slight increase in BMD (Klink et al., 2015). The findings correlated with trends in BTMs, which were shown to decrease while using a GnRH agonist, in line with the observations in patients with CPP, and continued to decrease despite initiation of GAHT (Vlot et al., 2017). The study was criticized because treatment occurred later than would now be standard and with lower doses of GAHT.

It is important to note that some TGD youth are now starting puberty blockade at earlier ages than in years past, and it will be essential to delineate how the timing of pubertal blockade can impact skeletal development differentially. This concept was examined in a prospective study in which the investigators considered the timing of GnRH agonist initiation (i.e., early vs. late puberty) in their analysis, finding that, during GAHT, the increase in bone mineral apparent density Z-scores was most pronounced for the early pubertal TGD youth (Schagen et al., 2020). In that study, BTMs decreased on GnRH agonists in all but the late-pubertal transmasculine individuals who started with lower BTMs that did not change. BTMs were again shown to decrease on GAHT for all groups except, interestingly,

in the early-pubertal transfeminine individuals who had an initial increase in BTMs in the first year of treatment before they began to decline, possibly because this group had the most growth potential.

DEXA is the most frequently employed imaging technique to measure BMD and diagnose low bone mass and osteoporosis. In its 2019 guidance on the interpretation of DEXA in TGD adults, the International Society of Clinical Densitometry (ISCD) recommended using BMD Z-scores concordant with gender identity in transgender adults and Z-scores concordant with sex recorded at birth in nonbinary adults (Rosen et al., 2019). Although the Endocrine Society Clinical Practice Guidelines recommend assessment of BMD by DEXA for TGD youth treated with GnRH agonist monotherapy and following BMD by DEXA until peak bone mass is attained, there is no official guidance regarding the interpretation of BMD by DEXA in TGD youth (Hembree et al., 2017).

There is now literature showing that hip bone geometry matches gender identity curves in TGD individuals who initiated GnRH agonists in early puberty. In contrast, those who initiated GnRH agonists in mid- to late puberty had skeletal trajectories that more closely mirrored the reference curve of their sex recorded at birth (van der Loos et al., 2021). In a recent prospective study investigating different methods of interpreting BMD Z-scores by DEXA in early pubertal TGD youth, the authors concluded that it might be helpful to utilize skeletal age in addition to both sex reference standards to interpret BMD before and during GnRH agonist monotherapy and to utilize reference standards of the affirmed gender when GnRH agonist therapy is stopped and either GAHT is started or endogenous puberty is allowed to proceed (Lee et al., 2022).

Key Points:

1. Endogenous estrogen is crucial in attaining peak bone mass and maintaining bone health throughout adulthood. Endogenous testosterone contributes to sex differences in BMD primarily through its effects on body composition.
2. Low bone mass has been observed with relatively high prevalence in TGD individuals, both transfeminine and transmasculine, before GAHT, in part because of body composition and lifestyle factors.
3. BMD increases, and fracture risk likely decreases, with exogenous estrogen use in transfeminine individuals. BMD does not appear to increase, but does not appreciably decline, with exogenous testosterone use in transmasculine individuals.
4. Feminizing GAHT increases fat mass and decreases lean mass, shifting body composition and contour toward that of the natal female body. Masculinizing GAHT increases lean mass and reduces fat mass, moving body composition and contour toward that of the natal male body.

5. Differences in ROA and dose of GAHT do not appear to affect BMD.
6. TGD youth may exhibit a decrease in BMD and levels of BTMs with the initiation of GnRH agonists. BMD then increases with initiation of GAHT, particularly in early pubertal TGD youth, although whether peak bone mass is attenuated is not entirely clear.
7. TGD youth who initiate GnRH agonists in early puberty may have skeletal trajectories that match their gender identity curves, whereas those who start GnRH agonists later in puberty more closely match the reference curve of cisgender individuals with the same sex recorded at birth.
8. For interpretation of DEXA in TGD adults, the ISCD recommends using BMD Z-scores concordant with gender identity in transgender adults and Z-scores concordant with sex recorded at birth in nonbinary adults.
9. While limited data are available to guide the interpretation of DEXA in TGD youth, it may be helpful to utilize skeletal age in addition to both sex reference standards before and during GnRH agonist monotherapy and then to use reference standards for their affirmed gender once GnRH agonists are stopped and GAHT is started, or endogenous puberty is allowed to resume.

CARDIOVASCULAR SYSTEM

Sex Differences in the Cardiovascular System

The cardiovascular system serves as the channel through which blood is circulated throughout the body. Physiologic differences in the cardiovascular system between cisgender males and cisgender females have been well studied. While the size of the heart between the two sexes before puberty is similar, after puberty, the hearts of cisgender males have a greater degree of myocyte hypertrophy than the hearts of cisgender females (Prabhavathi et al., 2014). When adjusted for exercise ability, the hearts of cisgender females tend to beat faster than those of cisgender males, which is thought to be a compensatory mechanism resulting in equivalent cardiac output between the sexes (Ramaekers et al., 1998). Cardiovascular disease generally occurs later in cisgender females, and some studies suggest that cisgender females have a better prognosis than their cisgender male counterparts for diseases such as heart failure, myocardial ischemia, and hypertrophic cardiomyopathy (Deswal and Bozkurt, 2006; Vaccarino et al., 1999). Cisgender females also have lower blood pressure, on average, than that of cisgender males (St Pierre et al., 2022).

Sex hormones play a vital role in the physiology of the cardiovascular system. Circulating estrogen is cardioprotective. This is evidenced by

premenopausal cisgender females having a decreased incidence of cardiovascular disease compared with cisgender males. In contrast, postmenopausal cisgender females, who have relative estrogen deficiency, tend to have similar or even higher rates of cardiovascular disease compared with cisgender males (Armeni and Lambrinoudaki, 2022). This may partly be due to estrogen’s vasodilatory effects on endothelial cells lining the blood vessels (Chakrabarti et al., 2014). The estrogen receptor is present in estrogen receptor (ER)α, ERβ, and G-protein-coupled estrogen receptor (GPER) forms. These receptors are present in both sexes and found throughout the body, including the cardiovascular system. Stimulation of ERα results in the activation of the endothelial nitric oxide synthase (eNOS) pathway, leading to vasodilation and prevention of smooth muscle proliferation. ERβ stimulation helps prevent cardiac fibrosis (dos Santos et al., 2014). Murine models have shown that mice without ERβ receptors experience systolic and diastolic blood pressure elevations (Zhu et al., 2002). GPER regulates vascular tone and protects the heart from reperfusion injury (dos Santos et al., 2014). Estrogen also has protective effects on myocardial contractility, likely related to estrogen’s role in modulating calcium homeostasis in the heart (Jiao et al., 2020). Indeed, a separate study using female rats found that gonadectomy was associated with decreased myocardial contractility, effects that were partially reversible with administration of exogenous estrogen (Scheuer et al., 1987).

Testosterone receptors are also present throughout the cardiovascular system in both sexes. While some studies have linked testosterone to adverse effects on myocardial remodeling following infarction, several studies have shown that physiologic levels of testosterone may inhibit the formation of atherosclerosis, prevent dyslipidemia, and decrease overall inflammation (Herring et al., 2013; Nahrendorf et al., 2003). Testosterone’s effects on the vasculature are concentration dependent. At lower concentrations, testosterone exhibits vasorelaxation effects by modulating the calcium and potassium channels of smooth muscle cells as well as interacting with the eNOS, cyclic guanosine monophosphate, and cyclic adenosine monophosphate pathways (Deenadayalu et al., 2012; English et al., 2002; Jones et al., 2003; Montano et al., 2008). An antiarrhythmic effect of testosterone has also been observed in a study that showed a lower incidence of early after-depolarizations of cardiac myocytes in rabbits with physiologic testosterone levels compared with those with testosterone deficiency (Pham et al., 2002). Additionally, testosterone has been shown to shorten the length of the QTc interval and thereby decrease the duration of the cardiac myocyte action potential, likely because of the modulation of potassium channels (Brouillette et al., 2005; Herring et al., 2013).

Given the known effects of endogenous sex hormones on the function of the cardiovascular system, it is essential to identify the impact that

exogenous sex hormones may have on this system. In the cisgender population, the use of exogenous estrogen as HRT has been associated with increases in blood pressure and the risk of venous thromboembolism (VTE) (Harvey et al., 2015; Martinez et al., 2020). The Women’s Health Initiative, the first and largest randomized controlled trial examining the risk of VTE with the use of exogenous estrogen in cisgender female individuals, started a larger conversation on the cardiovascular risks posed by HRT, particularly in postmenopausal cisgender females over the age of 60 (Howard and Rossouw, 2013). In this study, researchers found evidence of an increased risk in peri- and postmenopausal cisgender females using combined estrogen and progesterone HRT. The VTE risk was about twice as high for those taking HRT compared with those on placebo (Cushman et al., 2004). A retrospective analysis of cisgender women in the United Kingdom who had a history of VTE also found the use of HRT to increase one’s risk of VTE, with those on combined estrogen/progesterone preparations being at higher risk than those on conjugated equine estrogen preparations (Vinogradova et al., 2019). Several studies have found no increased risk of VTE with transdermal preparations of exogenous estrogen (Mohammed et al., 2015; Olié et al., 2010). In a large study of exogenous estrogen and thromboembolic risk in a cohort of postmenopausal cisgender female individuals aged 45–70, those using oral preparations of estrogen had an approximately four-fold increased risk of VTE. In contrast, those on transdermal preparations did not have an increased risk (Scarabin et al., 2003).

On the other hand, testosterone replacement therapy (TRT) for hypogonadal cisgender male individuals does not appear to increase the risk of major adverse cardiovascular events when used appropriately and with routine monitoring (Jones et al., 2011; Lincoff et al., 2023). Supraphysiologic levels of testosterone resulting from overreplacement, however, do increase the risk of cardiovascular morbidity by decreasing high-density lipoprotein (HDL) levels and reducing insulin sensitivity (Bhasin and Herbst, 2003). The use of GnRH agonists as ADT in cisgender male individuals for the treatment of prostate cancer has been associated with an increase in adverse cardiovascular outcomes, including stroke, VTE, and myocardial infarction (MI) (Greiman and Keane, 2017; Zhao et al., 2014). It should be noted that spironolactone, a commonly used mineralocorticoid antagonist with antiandrogen properties, has known cardiac remodeling and blood pressure effects and is regularly used in the treatment of hypertension and heart failure (Kosmas et al., 2018).

GAHT and the Cardiovascular System

While extensive literature examines the effects of HRT on the cardiovascular system in cisgender individuals, available data examining these

effects with the use of GAHT in the TGD population remain sparse. A recent study that included 2,671 TGD individuals detected an increased risk of cardiovascular disease compared with cisgender controls; the investigators speculated that GAHT may contribute to some of this elevated risk (Glintborg et al., 2022). More specifically, when analyzing the cohort of TGD individuals on GAHT, transmasculine individuals were found to have higher rates of cardiovascular disease than cisgender males, but this effect was not seen when comparing transfeminine individuals with cisgender females. This increased risk was attributed, in part, to the changes in one’s lipid profile that may occur with masculinizing GAHT. Socioeconomic status, known to be a social determinant of health, was not found to be a confounding risk factor in this study (Glintborg et al., 2022; Harper and Lynch, 2007).

Some studies examining cardiovascular risk in transfeminine individuals on GAHT, however, have also detected higher rates of cardiovascular disease compared with their cisgender counterparts. In a large Dutch study, transfeminine individuals died more frequently of cardiovascular disease than did cisgender individuals (de Blok et el., 2021). Of note, the study did not attribute this finding to the use of feminizing GAHT because the highest risk for mortality was determined to be from non-hormone-related causes of death. In a separate U.S.-based study, transfeminine individuals receiving feminizing GAHT were found to have an increased incidence of MI (Getahun et al., 2018). Despite the favorable changes that feminizing GAHT may have on one’s lipid profile, some researchers have noted that trans women still have a higher cardiovascular risk than cisgender controls (Cocchetti et al., 2021; Masumori and Nakatsuka, 2023). It is therefore essential to better understand the mechanisms by which GAHT may confer these risks.

Cardiovascular disease is intimately linked to metabolic dysregulation, with dyslipidemia being a validated biomarker for cardiovascular morbidity and mortality. As previously noted, most studies have shown that feminizing GAHT may lead to a more favorable lipid profile, raising HDL, although it is sometimes accompanied by a rise in triglycerides (Leemaqz et al., 2023; Maraka et al., 2017; Streed et al., 2021). The literature surrounding the effect of GAHT on lipids is explored in greater detail in the metabolism section of this review. In terms of blood pressure, studies have suggested that feminizing GAHT may decrease systemic blood pressure, mirroring findings in the cisgender population (Sharula et al., 2012; Streed et al., 2021).

Another primary cardiovascular clinical outcome that has been studied in patients receiving feminizing GAHT has been that of VTE. Feminizing GAHT has been associated with a shift toward a more procoagulable serologic milieu (Cocchetti et al., 2022; Scheres et al., 2021). A large study in the United States showed an increased incidence of VTE and ischemic

stroke in transfeminine individuals receiving feminizing GAHT compared with cisgender men and women (Getahun et al., 2018). The effect of the ROA of exogenous estrogen on VTE risk has been explored, with studies finding that transdermal preparations may not increase one’s risk of VTE, possibly because of bypassing the first-pass effect through the liver (Ott et al., 2010; Slack and Safer, 2021). Thus, the risks and benefits of providing feminizing GAHT should be considered in the context of the individual’s overall cardiometabolic health, and there is a need for more concrete scoring tools to help assess risk in these patients (Arrington-Sanders et al., 2023).

Compared with feminizing GAHT, there is less consensus regarding the effect of masculinizing GAHT on measures of cardiovascular health. Exogenous testosterone, like estrogen, may influence one’s lipid profile, thereby potentially modulating one’s risk of cardiovascular disease. Unlike estrogen, however, testosterone may result in a shift toward a less favorable lipid profile, increasing triglyceride and low-density lipoprotein (LDL) levels while decreasing HDL levels (Leemaqz et al., 2023; Maraka et al., 2017; van Velzen et al., 2019). These changes do not appear to be affected by variations in the ROA of testosterone. A meta-analysis has also shown that testosterone therapy can increase blood pressure (Masumori and Nakatsuka, 2023). However, consistent replication of this finding is lacking, and the clinical significance of this increase, if present, remains unknown (Elamin et al., 2010; Emi et al., 2008). The literature is conflicting concerning the effect of masculinizing GAHT on adverse cardiovascular outcomes (Martinez et al., 2020). Both a U.S.-based study and a Belgian study showed no significant difference in the risk of ischemic stroke or MI in transmasculine individuals compared with cisgender controls (Getahun et al., 2018; Wierckx et al., 2013). Other studies have reported higher rates of MI in transmasculine individuals compared with cisgender controls (Alzahrani et al., 2019; Nota et al., 2019).

Testosterone’s association with erythrocytosis and the prevalence of adverse outcomes associated with erythrocytosis are explored further in the hematology section of this review. Despite these hematologic changes, studies have not shown an increased incidence of VTE in transmasculine individuals receiving masculinizing GAHT (Getahun et al., 2018; Nota et al., 2019; Wierckx et al., 2013).

Few studies have investigated the effect of GnRH agonist therapy on the cardiovascular system. The use of GnRH agonists in TGD youth may increase overall fat mass and decrease lean body mass, potentially impacting overall cardiovascular risk (Schagen et al., 2016). Given the rising use of these agents and their use at younger ages than in the past, more research is needed to determine whether there is an effect on long-term cardiovascular outcomes.

Adjunct antiandrogens, including cyproterone acetate (CPA) and spironolactone, and their effects on lipid profiles have been studied in TGD individuals. In a Canadian study comparing the effects of these two agents on transfeminine individuals’ lipid profiles, spironolactone was associated with a higher HDL than CPA (Fung et al., 2016). This study did not find a significant difference in the total cholesterol, LDL, or triglyceride levels between these two agents. Other studies have shown that CPA may decrease HDL levels compared with baseline (Ott et al., 2011; Wierckx et al., 2014a).

Key Points:

1. Cisgender females have lower blood pressure, on average, and generally experience cardiovascular disease later than cisgender males.
2. Estrogen appears to have a cardioprotective role, exhibiting vasodilatory effects on the vascular system and protective effects on myocardial contractility.
3. Physiologic testosterone levels may be cardioprotective, inhibiting atherosclerosis, decreasing inflammation, and causing vasorelaxation.
4. Feminizing GAHT appears to decrease systemic blood pressure, whereas masculinizing GAHT may increase blood pressure.
5. The association between GAHT and cardiovascular morbidity and mortality remains unclear. Transfeminine individuals using exogenous estrogen may have an increased incidence of MI and ischemic stroke, although other cardiometabolic risk factors may cloud the effect of GAHT. There is even less consensus regarding the risk associated with exogenous testosterone use in transmasculine individuals, with most studies failing to demonstrate increased risk.
6. Differences in the ROA, preparation, and dose of both feminizing and masculinizing GAHT do not appear to affect biochemical or clinical measures of cardiovascular health, except transdermal estrogen, which does not seem to portend an increased risk of VTE.

HEMATOLOGIC SYSTEM

Sex Differences in the Hematologic System

The hematologic system consists of blood and bone marrow—the tissue found in the center of bones involved in the production of the cellular elements of blood, including red blood cells, white blood cells, and platelets. Sex differences in the proportions of the cellular components of blood have been established (Bain and England, 1975; Murphy et al., 2010; Segal and Moliterno, 2006). For example, many studies have noted that iron-replete cisgender females have a 12 percent lower mean hemoglobin level than age- and race-matched cisgender males (Murphy, 2014). There are multiple theories to

explain this sexual dimorphism, with some proposing a variable effect of erythropoietin, a hormone that stimulates red blood cell production; other studies suggest that there are no differences in serum erythropoietin concentrations between the sexes (Tilling et al., 2013). Instead, this variation may be explained by the contrasting effects of estrogen and testosterone on the kidney vasculature, which affects the red cell mass (Murphy, 2014). Other studies have examined polymorphisms of the erythropoietin gene or its receptor (EPOR) as the cause of the dimorphism. EPOR alleles EPORA1 and EPORA10 were noted more often in cisgender females than in cisgender males, while the EPOR5 allele was more frequent in cisgender males (Zeng et al., 2001).

Recently, Cui and colleagues (2023) postulated that the latexin/microRNAThrombospondin 1 (Thbs1) signaling pathway may account for sexual dimorphism in the hematopoietic system. In a murine model of male hematopoietic stem cells (HSCs), the researchers found downregulation of Thbs1, which decreases latexin’s ability to enhance apoptosis and thus decreases self-renewal of HSCs (Cui et al., 2023). Furthermore, testosterone is associated with the inhibition of hepcidin, an iron-regulating molecule produced in the liver. In turn, this reduces the inhibitory hold of hepcidin on the incorporation of iron into red blood cells, thereby promoting hematopoiesis (Bachman et al., 2010; Guo et al., 2013).

Sexual dimorphism is also seen in white blood cells and platelets (Bain, 1996). Regarding leukocytes, cisgender males of all ages have a higher white blood cell count than cisgender females (Chen et al., 2017). Cross-sectional data from 46,879 individuals revealed that premenopausal cisgender females have higher neutrophil percentages, lower lymphocyte percentages, and a higher neutrophil-to-lymphocyte ratio than cisgender males; however, this is reversed after the age of 50 (Chen et al., 2017). Studies suggest that estradiol decreases lymphocyte production in the bone marrow while increasing the time for neutrophil apoptosis, accounting for the variation with age (Medina et al., 2000; Molloy et al., 2003). Additionally, cisgender females have higher cluster of differentiation (CD)4 T cells and higher CD4-to-CD8 T cell ratios and mount a stronger adaptive immune response than do cisgender males (Klein and Flanagan, 2016). Castrated male mice show a similar CD4-to-CD8 pattern as female mice, suggesting this may be an androgen-mediated effect (Roden et al., 2004). The clinical implications of these observations are explored in greater detail in the immunology section of this review. Concerning platelets, there are differences between the sexes, with cisgender females having higher platelet counts than cisgender males and showing increased sensitivity to platelet agonists (Godwin et al., 2022; Segal and Moliterno, 2006; Sloan et al., 2015).

GAHT and the Hematologic System

GAHT can influence platelet numbers and activation. One study found that platelet levels increased in transfeminine individuals significantly after

1 year of GAHT and continued to increase through the fifth year of therapy (Allen et al., 2021). Another study, looking at 48 transfeminine individuals on estradiol patches and CPA and 47 transmasculine individuals on testosterone gel, showed a significant increase at 1 year in platelet activation factors in transfeminine individuals compared with baseline. At the same time, there were no changes in transmasculine individuals (Schutte et al., 2022). Another study confirmed that there is a shift toward a more hypercoagulable serologic milieu in transfeminine individuals after a year of GAHT. Again, there was no change for transmasculine individuals (Scheres et al., 2021).

The clinical significance of GAHT’s effects on the hematologic system is mainly due to the concern for erythrocytosis in transmasculine individuals on testosterone and concern for thrombosis with both masculinizing and feminizing GAHT (Hembree et al., 2017; Safer and Tangpricha, 2019). A recent study of 6,670 transmasculine individuals, the largest to date to examine the prevalence of erythrocytosis in this population, found that only 8.4 percent had a hematocrit level of greater than 50 percent and less than 1 percent had a hematocrit level greater than 54 percent, a level at which therapy would be discontinued and phlebotomy considered (Krishnamurthy et al., 2023). Historically, smaller studies reported significant variation in the incidence of erythrocytosis, although the majority of studies did not demonstrate a higher risk of thromboembolic events for transmasculine individuals (Bunderen et al., 2022; Irwig, 2017; Madsen et al., 2021; Scheres et al., 2021). A longitudinal study by Madsen and colleagues (2021) involving 1,073 TGD individuals on masculinizing GAHT found that 11 percent experienced erythrocytosis, defined as a hematocrit level greater than 50 percent on at least two laboratory assessments. Additionally, 2.2 percent of the cohort had at least one hematocrit level greater than 54 percent, the level at which it is recommended that phlebotomy be initiated (Madsen et al., 2021). In a separate study of 519 transmasculine individuals receiving testosterone therapy, the authors found a 20 percent incidence of erythrocytosis, with 42 percent requiring a dosage reduction and 4.8 percent requiring phlebotomy (Oakes et al., 2021). A 0.9 percent incidence of thromboembolic events, including one ischemic stroke, was also reported. Notably, 80 percent of those with a thromboembolic event had erythrocytosis at any time; however, none had erythrocytosis at the time of diagnosis (Oakes et al., 2021). Other studies have reported a rate of erythrocytosis in transmasculine individuals seven times greater than that of matched cisgender males and 83 times higher than that of matched cisgender females (Antun et al., 2020).

Reports on the effect of the ROA of testosterone, as it portends to the risk of erythrocytosis, are conflicting. In the study of 6,670 individuals previously mentioned, investigators found a lower mean hematocrit level in those treated with transdermal testosterone than in those treated with intramuscular (IM) formulations but concluded that the levels of increase in hematocrit were

unlikely to be clinically meaningful (Krishnamurthy et al., 2023). Earlier studies have had varied results. In one study, transmasculine individuals on testosterone undecanoate had a lower incidence of erythrocytosis than those on testosterone esters or gel (DeLoughery, 2022). In another, those on injectable testosterone enanthate had a 25 percent incidence of erythrocytosis, while those on testosterone undecanoate had a 17 percent incidence, and those on transdermal testosterone had no incidence at all (Nolan et al., 2021). In contrast, a study including 140 trans men found no variations in testosterone formulations, but there was a notable 10 percent phlebotomy rate and a high prevalence of 33 percent for erythrocytosis (Perez-Luis et al., 2019).

The contrasting effects of estrogen and testosterone on the hematologic system are seen in a study comparing 424 transmasculine and 559 transfeminine individuals with matched cisgender controls at three large integrated U.S. health care centers. Transmasculine individuals had increased hematocrit levels, closer to reference ranges for cisgender males, while transfeminine individuals had a drop in hematocrit levels, moving closer to reference ranges for cisgender females (Antun et al., 2020). Similarly, the European Network for the Investigation of Gender Incongruence cohort studied 340 transfeminine individuals treated with oral estradiol and adjunct androgen blockers and 265 transmasculine individuals treated with testosterone formulations. At year 1 of the 3-year follow-up, the average hematocrit levels in transmasculine individuals had increased from 41 to 46 percent. Conversely, at 3 months follow-up, the hematocrit of transfeminine individuals had decreased from 45 to 41 percent and remained stable after that (Defreyne et al., 2018). Studies in adolescents and young adults have also shown similar changes to hematocrit on GAHT (Olson-Kennedy et al., 2018). Of note, these studies show a low incidence of erythrocytosis in those on testosterone therapy, ranging from 7 to 11 percent (Antun et al., 2020; Defreyne et al., 2018; Krishnamurthy et al., 2023).

While prospective longitudinal studies are needed to confirm the relative safety of GAHT concerning hematologic parameters, it is important to note that the effects of estrogen on hematocrit, white blood cells, and platelets, while affected in the short term, soon stabilize at a new baseline. Similarly, novel research including thousands of transmasculine individuals shows a relatively low rate of erythrocytosis and an even smaller rate of clinically significant elevations in hematocrit at the level where intervention would be recommended.

Key Points:

1. Sex hormones contribute to sexual differentiation of the hematology system, resulting in cisgender males having a higher red cell mass and cisgender females having higher platelet counts and altered white blood cell concentrations, leading to a more robust adaptive immune response.

2. Exogenous testosterone use increases red cell mass, resulting in hematocrit levels in transmasculine individuals that are closer to the normal range of cisgender males. In contrast, exogenous estrogen use, through its suppression of testosterone, shifts the hematocrit of transfeminine individuals closer to levels of cisgender females.
3. The risk of erythrocytosis with the use of exogenous testosterone in transmasculine individuals remains low. It is further decreased with the use of transdermal compared with injectable testosterone, with most studies failing to observe an increased risk of thromboembolic events.
4. Exogenous estrogen may increase platelet activation factors and lead to a more prothrombotic serologic milieu in transfeminine individuals.

IMMUNE SYSTEM

Sex Differences in the Immune System

The immune system is the body’s mechanism for eliminating foreign material, pathogens, and aberrant cell bodies. It is divided into two parts—innate and adaptive—that contain different cell types. The innate immune system is the body’s first-line defense mechanism and consists of monocytes, which differentiate into dendritic cells and macrophages, natural killer cells, neutrophils, basophils, and eosinophils. The adaptive immune system provides “memory” immunity. It consists of T and B lymphocytes that differentiate after exposure to new pathogens. These pathogen-specific lymphocytes can mount more rapid responses during repeat infections. While these two systems are common to both sexes, differences in their function and efficiency exist between the sexes, suggesting sexual dimorphism in immunity. It is well established that cisgender females have more robust innate and adaptive immune systems than cisgender males, thereby mounting stronger responses to bacterial and viral infections (Dias et al., 2022; Jacobsen and Klein, 2021; Shepherd et al., 2021). Indeed, in murine models of the H1N1 virus, female mice showed higher levels of antigen-specific B cells and higher titers of antibodies (Fink et al., 2018; Wilkinson et al., 2022). This difference in immunity may explain why, among HIV-positive patients, viral loads are usually lower in cisgender females than in cisgender males (Klein and Flanagan, 2016; Ziegler and Altfeld, 2016). However, more robust immune responses also predispose cisgender females to autoimmune conditions such as systemic lupus erythematosus and multiple sclerosis (Angum et al., 2020; Jacobsen and Klein, 2021). Conversely, while cisgender males experience lower rates of autoimmune conditions, they are more susceptible to pathogenic infections and have higher rates of cancer (Shepherd et al., 2021; Trigunaite et al., 2015). The biological basis

for these differences is multifaceted, attributable to genetic and hormonal differences between the sexes (Klein and Flanagan, 2016). This review will focus on how sex steroid hormones modulate the immune system.

Estrogen is found in both sexes but is generally higher in cisgender females, particularly during pregnancy and the days preceding ovulation. It acts on several types of estrogen receptors—most notably ERα and ERβ—that are expressed differentially according to age and sex (Dias et al., 2022). Therefore, immunological differences related to estrogen are attributed to variations in hormone and receptor levels (Klein and Flanagan, 2016). Extant studies suggest that estrogen has bipotential effects on immune cells. Physiological levels of estrogen are associated with increased production of proinflammatory cytokines and a Th1-type response. Supraphysiological levels are associated with anti-inflammatory responses, such as increased production of anti-inflammatory cytokines and a shift toward a Th2-type response (Kovats, 2015; Straub, 2007). Indeed, an in vitro experiment with human-derived monocytes showed maximal and decreased IL-1 activity with physiological and supraphysiological estrogen levels, respectively (Polan et al., 1988). The effects of estrogen vary according to its concentration, but a complete lack of the hormone leads to dampened immune responses. This was observed in a murine model wherein ovariectomized mice experienced increased mortality from Helicobacter pylori infection, and estrogen replacement in these mice was associated with improved outcomes (Ohtani et al., 2007). In addition to conferring protection against infections, estrogen may increase the risk of developing autoimmune conditions (Klein and Flanagan, 2016). One explanation for this is the inverse association between estrogen signaling and the expression of regulatory T cells (Treg), which are important for preventing autoimmunity. Prior research has shown that deletion of ERα in T cells is associated with higher levels of Tregs (Mohammad et al., 2018).

Circulating testosterone is also present in both sexes, generally in much higher concentrations in cisgender males. It exerts its effect on the immune system through androgen receptors. Testosterone is associated with anti-inflammatory effects, observed in both animal experiments and clinical settings. In vivo exposure of mice to testosterone has been associated with decreased natural killer (NK) cell activity and lowered macrophage expression of toll-like receptor 4 (TLR4) (Hou and Zheng, 1988; Rettew et al., 2008). In female mice, exposure to testosterone has been associated with decreased secretion of interferon-gamma (IFNγ) by NK cells (Lotter et al., 2013). Clinically, cisgender male individuals who have rheumatoid arthritis, an autoimmune condition, have lower amounts of serum testosterone, and those who have androgen deficiencies have higher levels of proinflammatory cytokines (Liu et al., 2023; Malkin et al., 2004). It is notable, however, that androgens have differential effects on immune

cells depending on sex. For example, in an in vitro study examining the effects of dihydrotestosterone (DHT) on human monocytes, DHT was associated with diminished IL-6 and tumor necrosis factor alpha (TNFα) responses in male-derived monocytes, but this effect was not seen in female-derived monocytes (de Bree et al., 2018). Finally, testosterone has also been shown to increase the expression of the FOXP3 transcription factor, which is a positive regulator of Tregs (Walecki et al., 2015). This partially explains the lower rates of autoimmune diseases found among cisgender males. While much is known about the biological underpinnings of sexual dimorphism in immunity, less is known about how hormone therapy used for gender affirmation among TGD individuals affects their immune systems. It is reassuring, however, that existing literature suggests hormone therapy among TGD individuals likely has similar immunological effects as do endogenous hormones among cisgender individuals.

GAHT and the Immune System

Extant literature on immunity in TGD individuals suggests that exogenous estrogen and gender-affirming procedures that affect androgen-producing organs may have proinflammatory effects. For example, oral estrogen has been associated with increased inflammatory markers among transfeminine individuals (Wilson et al., 2009). However, the hormone’s proinflammatory effects were absent in individuals utilizing transdermal estrogen. It is unclear whether this difference is attributed to the ROA, differences in hormone levels, or both. It is also unclear whether this difference has any clinical significance. In a case report of new-onset lupus nephritis in a transfeminine individual who received a gonadectomy, the authors reported that the patient—who had been on exogenous estrogen and antiandrogen therapy for 9 years before their surgery—developed symptoms suggestive of an autoimmune pathology approximately 1 year after their surgery (Pontes et al., 2018). After further workup, they were confirmed to have lupus nephritis. The authors noted that this case echoed results from a study on systemic lupus erythematosus in murine models. Specifically, male mice that were castrated before estrogen administration developed more severe disease outcomes than those not castrated (Roubinian et al., 1978). This is likely attributable to decreased testosterone levels in the former group. In all, these studies suggest that the effects of gender-affirming estrogen therapy on a TGD individual’s immune system depend on its dosage, ROA, and whether the patient has undergone any surgical procedure that affects androgen production.

Current studies on gender-affirming testosterone therapy and immunity suggest that the hormone may also have immunosuppressive effects on TGD individuals. One study showed that transmasculine individuals on testosterone had significantly higher levels of Treg cells than those of cisgender

females (Robinson et al., 2022). Upon further investigation, the authors noted significant transcriptome changes related to Treg cell expression. These results suggest that, in theory, TGD individuals using testosterone may have dampened immune systems. However, the clinical significance of these results is unclear, as more research is needed to understand whether this differential Treg cell expression among TGD individuals will necessarily lead to adverse immunological outcomes.

The impact of hormone therapy on TGD patients’ immune systems necessitates more attention. Based on the current literature, clinicians prescribing estrogen therapy for TGD patients may wish to monitor for autoimmune diseases, especially if the patients have had their gonads removed. Similarly, clinicians may want to watch for signs of immunosuppression among TGD patients receiving testosterone therapy. However, given the lack of data on this topic and the profound known mental and physical benefits of gender-affirming hormone therapy, immunological considerations should not preclude clinicians from prescribing these medications to TGD patients.

Key Points:

1. Sexual differentiation of the immune system results in cisgender females having more robust innate and adaptive immune systems, leading to decreased susceptibility to pathogenic infections and a greater propensity for autoimmunity compared with cisgender males.
2. Estrogen’s effect on immune cells depends on its concentration, with deficiency leading to dampened immune response, physiologic levels associated with a proinflammatory state, and supraphysiologic levels associated with anti-inflammatory responses. Conversely, testosterone is associated with anti-inflammatory effects irrespective of concentration.
3. Exogenous estrogen is associated with increased markers of inflammation in transfeminine individuals. However, this may depend on the first-pass effect through the liver, as these changes are not seen with transdermal estrogen.
4. Exogenous testosterone use in transmasculine individuals may result in a dampened immune response, although an association with adverse clinical outcomes has yet to be observed.

METABOLISM

Sex Differences in Metabolism

Metabolism refers to the chemical reactions in the body that allow for energy production. Carbohydrates and lipids are the primary sources of energy in the body, with protein serving as a source of energy during states

of starvation. Sexual dimorphism in metabolism begins at conception, as male and female reproductive cells differ vastly in their use of energy. The male gamete, sperm, requires energy for motility to reach the female gamete, the oocyte, which dedicates its mitochondria to the zygote to nurture life after fertilization (Mauvais-Jarvis, 2015). Sex differences in the production and use of energy continue throughout development and into adulthood.

Glucose, a carbohydrate, serves as the primary source of energy in the human body, and complex mechanisms are in place to maintain glucose homeostasis. There is evidence that sexual dimorphism exists in this fundamental process. For example, a study showed that fasting and postprandial glucose were higher in elderly cisgender females than in elderly cisgender males, despite endogenous glucose production rates not differing between the two groups (Basu et al., 2006). This differentiation may be due, in part, to the direct effects of both sex hormones on glucose metabolism. ERαdeficient mice are more insulin resistant, as evidenced by their higher hepatic glucose production (Mauvais-Jarvis et al., 2013). Additionally, estrogen has been shown to affect glucose uptake in skeletal cells by modulating the level of GLUT4, a transporter that regulates insulin-stimulated glucose uptake (Campbell and Febbraio, 2022; Mauvais-Jarvis et al., 2013). Similarly, testosterone increases GLUT4-mediated glucose uptake and decreases insulin resistance by increasing lipolysis and stimulating mitochondrial function (Grossmann, 2014; Mitsuhashi et al., 2016). States of impaired glucose metabolism, such as diabetes mellitus, provide an opportunity to explore the potential clinical effects of sex hormones on glycemic control.

Type 2 diabetes mellitus (T2DM) is particularly important from an epidemiologic standpoint as the prevalence of this already common disease is expected to increase over the next decade (Saeedi et al., 2019). Differences in the rate of diagnoses, complications, and even response to diabetes medications exist between the sexes (Kautzky-Willer et al., 2023). Estrogen is thought to be protective against T2DM, as demonstrated by the increased risk of T2DM in cisgender females who have primary ovarian insufficiency and the apparent ability to delay or prevent the development of T2DM when exogenous estrogen is given as HRT for postmenopausal cisgender females (Anagnostis et al., 2019; Mauvais-Javis et al., 2017). Testosterone deficiency in cisgender males is also a risk factor for T2DM, while testosterone excess is reported to be associated with the development of T2DM in cisgender females (Kautzky-Willer et al., 2023). It should be noted, however, that the latter conclusions were drawn from studies including cisgender females with polycystic ovarian syndrome (PCOS), individuals who are at higher risk for metabolic syndrome at baseline.

In addition to carbohydrate metabolism, there are known sex differences in fat metabolism and distribution. Cisgender males tend to distribute fat in the abdominal region (android distribution), while cisgender females generally

distribute fat in the gluteal-femoral region (gynoid distribution). Generally, cisgender females tend to have higher amounts of body fat compared with cisgender males throughout their lifespans, but cisgender males have higher amounts of visceral adipose tissue (Blaak, 2001; Karastergiou et al., 2012). These differences become prominent during puberty, suggesting that the pubertal surge of sex hormones contributes to this observed dimorphism (Hattori et al., 2004; Maynard et al., 2001). Indeed, adipocytes have estrogen and androgen receptors, and sex hormones likely play a role in lipolysis, lipoprotein lipase expression, and leptin secretion (Karastergiou et al., 2012). Additionally, conditions that result in variations in the concentration of sex hormones have also been shown to modify fat distribution patterns. For example, menopause and PCOS, conditions of estrogen deficiency and androgen excess, respectively, often result in a shift toward a more android fat distribution in cisgender females (Carmina et al., 2007; Toth et al., 2000).

Notably, the sexual dimorphism of fat distribution may also have clinical implications. The gynoid distribution has been found to be protective for diabetes, cardiovascular risk, and overall mortality (Carey et al., 1997; Folsom et al., 1993; Lapidus et al., 1984). It has also been shown that if cisgender females develop a more android distribution of fat, they may develop metabolic complications at a rate similar to that of cisgender males (Jensen, 2008). The increased visceral adiposity seen in cisgender males has been linked to abnormal metabolic parameters, including elevated triglyceride and postprandial insulin levels (Muscogiuri et al., 2023).

Obesity, a chronic disease resulting in excessive body fat and associated with a wide variety of cardiometabolic complications, is a global epidemic that manifests differently between the sexes. Globally, obesity rates are higher in cisgender females than in cisgender males (Cooper et al., 2021). Animal models and clinical research explain why this might be. In addition to shifting body fat toward a more android distribution, states of estrogen deficiency, such as in cisgender females experiencing menopause, are associated with increased body weight and adiposity (Lovejoy et al., 2009). Mechanistically, this may be due to modifications in one’s central regulation of hunger and baseline energy expenditure (Clegg et al., 2007; Lovejoy et al., 2008). In contrast, murine models have shown that when exposed to a high-fat diet, male mice generate more proinflammatory markers than female mice (Singer et al., 2015).

Carbohydrate Metabolism

Several studies have found differences in the prevalence of T2DM in TGD individuals compared with their cisgender counterparts. These include a Belgian study that observed a higher prevalence of T2DM in TGD individuals irrespective of gender and a U.S.-based study that observed a higher prevalence of T2DM in transfeminine compared with cisgender female but

not cisgender male individuals (Islam et al., 2022; Wierckx et al., 2014a). In the latter study, no differences were observed between transmasculine and cisgender individuals, whether cisgender males or cisgender females. Importantly, investigators did not attribute these findings solely to GAHT use, given the complex and multifactorial pathogenesis of T2DM. It is also important to note that the differences between studies may be attributed, in part, to differences in baseline population prevalence. Indeed, other studies, including a Dutch study comparing the prevalence of T2DM in the TGD population with that of the cisgender population, have not observed any differences (van Velzen et al., 2022). Importantly, these studies did not conduct a subgroup analysis to determine if there is a difference in T2DM prevalence based on ROA, preparation, or dosage of GAHT.

Given the potential for confounders when studying T2DM at the population level, translational science can be informative. Several studies have investigated the relationship between GAHT and biochemical markers of insulin resistance. An early study of 31 TGD individuals suggested that GAHT for both transfeminine and transmasculine individuals leads to insulin resistance by demonstrating decreased glucose utilization during a hyperinsulinemic–euglycemic clamp after 4 months of GAHT (Polderman et al., 1994). Another study of 37 TGD individuals showed increased fasting insulin and insulin resistance as measured by glucose utilization in transfeminine subjects but not in transmasculine individuals (Elbers et al., 2003). These findings were confirmed in a more recent study in which transfeminine individuals were noted to have higher fasting insulin levels than their cisgender counterparts, suggesting a propensity for insulin resistance in transfeminine individuals (Deischinger et al., 2022). The Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), a method of assessing beta cell function and insulin resistance using fasting glucose and insulin levels, has also been studied in the TGD population. A higher HOMA-IR is indicative of more insulin resistance (Bonora et al., 2002). More recent studies have not shown that masculinizing GAHT causes a significant difference in the HOMA-IR when compared with cisgender males (Gava et al., 2018; Shadid et al., 2019). In contrast, patients receiving feminizing GAHT have been shown to have higher HOMA-IR indices and fasting insulin levels compared with cisgender females (Colizzi et al., 2015; Shadid et al., 2020). Taken together, the extant literature suggests that feminizing GAHT may impair glucose metabolism, as indicated by trends in qualified biomarkers. However, available data are in discordance, and the clinical implications of these trends remain unknown.

Lipid Metabolism

In a meta-analysis of 26 studies investigating the effect of GAHT on body composition and fat distribution, individuals on masculinizing GAHT

gained lean mass and lost fat mass. In contrast, those on feminizing GAHT gained fat mass and lost lean mass (Spanos et al., 2020). Importantly, while overall fat mass is lost with masculinizing GAHT, an increase in visceral abdominal fat has been observed (Elbers et al., 1997). A U.S.-based study also found that among those receiving GAHT, body mass index (BMI) increased in transfeminine individuals but not in transmasculine individuals. However, psychosocial and lifestyle factors likely contributed to this finding (Suppakitjanusant et al., 2020). More recent studies have noted that the size of adipose cells increases in transfeminine individuals and decreases in transmasculine individuals (Shah et al., 2022).

GAHT-driven changes in body composition are explored in more detail in the bone and body composition section of this review, but it is important to note here that these changes appear to occur in tandem with changes to the lipid profile. Both feminizing and masculinizing GAHT have been associated with changes in lipid profiles. While feminizing GAHT has been linked to an increase in HDL, masculinizing GAHT has shown the opposite effect, and both regimens have been linked to a rise in triglyceride levels (Elbers et al., 2003; Leemaqz et al., 2023; Maraka et al., 2017). For feminizing GAHT, this effect may be modified by the ROA, as a meta-analysis showed that oral estrogen increased triglyceride levels, while transdermal estrogen decreased triglyceride levels (Maraka et al., 2017). There were no comparisons made between ROAs for masculinizing GAHT. Notably, some researchers have observed that changes in lipid profiles were not associated with changes in markers of insulin resistance (Elbers et al., 2003). While the cardiovascular implications of GAHT are explored elsewhere in this review, more research is needed to investigate whether the observed changes in lipid profile and fat distribution are associated with cardiovascular morbidity and mortality rates and whether the ROA, preparation, or dosage of GAHT affects lipid parameters.

Key Points:

1. Sex hormones play a central role in the sexual dimorphism observed in human metabolism, with estrogen being linked to modulating energy expenditure and central control of food intake, and both estrogen and testosterone contributing to insulin sensitivity and glucose uptake.
2. Estrogen deficiency in cisgender females and testosterone deficiency in cisgender males have been associated with increased rates of T2DM. Conversely, syndromes of testosterone excess in cisgender females are associated with increased rates of T2DM.
3. Sex hormones contribute to sex differences in body fat distribution, with the gynoid distribution in cisgender females being protective against cardiometabolic disease.

4. Biomarkers of impaired glucose metabolism may increase with feminizing GAHT, and T2DM may be more common in transfeminine individuals compared with cisgender females. However, the explanation is likely multifactorial, and the contribution of feminizing GAHT remains unknown.
5. Masculinizing GAHT results in a gain of lean mass and loss of fat mass, whereas feminizing GAHT results in a loss of lean mass and an increase of fat mass.
6. Exogenous estrogen use in transfeminine individuals may contribute to a more favorable lipid profile by raising HDL. In contrast, exogenous testosterone use in transmasculine individuals may contribute to a less favorable lipid profile by raising triglycerides and LDL while decreasing HDL. Importantly, given the discordance in the literature, consensus has not been reached in this area.

GASTROINTENSTINAL SYSTEM

Sex Differences in the Gastrointestinal System

The gastrointestinal (GI) system consists of the alimentary tract and the accessory organs necessary to break down and absorb nutrients. The alimentary tract starts at the mouth; includes the esophagus, stomach, small intestine, and large intestine; and ends at the anus. The accessory organs of the GI system include the salivary glands, gallbladder, pancreas, and liver. There are noted sexual dimorphisms of the GI system. For example, cisgender males and cisgender females have different gut microbiomes, and differences in the microbiome are thought to occur during puberty (Sisk-Hackworth, 2023). Sexual dimorphisms are also found in the pancreas. For example, glucose-stimulated insulin secretion is higher in cisgender females than in cisgender males, which may be attributed to sex-specific regulation of pancreatic beta cells (McEwan et al., 2021).

The liver is a vital target organ for both sex hormones. Estrogen and testosterone are thought to have a role in fat and glucose metabolism in the liver. For example, lower levels of both testosterone and estrogen are risk factors for nonalcoholic fatty liver disease (NAFLD) (Kasarinaite et al., 2023). As such, HRT may play a role in improving NAFLD. Indeed, one study demonstrated that postmenopausal cisgender females receiving exogenous estrogen had lower levels of detectable liver enzymes in the blood. These changes were attributed to a reduction in fat content in the liver (McKenzie et al., 2006). Similarly, cisgender males receiving exogenous IM testosterone experienced a reduction in hepatic fat as measured by MRI (Apostolov et al., 2022).

Exogenous estrogen affects the production of liver proteins, such as angiotensinogen and coagulation factors (von Schoultz, 2009). The increase

in production of procoagulant factors is one of the reasons that exogenous oral estrogen, which undergoes first-pass metabolism in the liver, is associated with VTE (Abou-Ismail et al., 2020). Oral estrogen and the use of anabolic androgenic steroids have also been associated with hepatic neoplasms, such as hepatic adenomas and hepatocellular carcinoma (Khalid et al., 2023; Lizardi-Cervera et al., 2006). CPA, an antiandrogen agent, is also associated with hepatotoxicity (Chitturi and Farrell, 2013).

Exogenous estrogen is a risk factor for gallstone formation. Studies have found that cisgender females using exogenous estrogen, both oral and transdermal, had higher rates of gallbladder disease, including cholelithiasis and cholecystitis (Cirillo et al., 2005; Uhler et al., 1998). This may be due to estrogen slowing gallbladder motility and increasing the formation of cholesterol crystals (Uhler et al., 1998). Exogenous testosterone has been shown to decrease gallbladder motility in animal models, although the clinical significance of this has yet to be determined (Kline and Karpinski, 2008). In a large Swedish study, estrogen used as HRT in cisgender females was associated with an increased risk of pancreatitis, possibly due to estrogen’s effect of increasing triglyceride production in the liver and decreasing pancreatic enzyme secretion (Oskarsson et al., 2014). This risk was not further stratified by ROA, dosage, or preparation.

GAHT and the Gastrointestinal System

Several studies have investigated the effects of GAHT on liver enzymes with discordant findings. A U.S.-based study found that masculinizing GAHT increased the levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST), enzymes found in high levels in the liver that can serve as surrogates for liver damage; feminizing GAHT had no apparent effect on liver enzyme levels (Hashemi et al., 2021). This study controlled for factors such as BMI and alcohol use. A large Dutch study found that feminizing GAHT decreased ALT and AST levels, while masculinizing GAHT increased ALT and AST levels (Boekhout-Berends et al., 2023). However, a limitation of both studies was that the findings were not further stratified by hormone ROA, preparation, or dosage. Another Belgian study found that IM testosterone increased ALT and AST levels, oral estrogen decreased ALT and AST levels, and transdermal estrogen decreased AST levels but had no significant impact on ALT levels (Wierckx et al., 2014a). A smaller study found that both masculinizing and feminizing GAHT did not have any significant effect on liver enzyme levels (Fernandez and Tannock, 2016). A sizeable European-based study noted no substantial evidence of liver injury in TGD individuals on either masculinizing or feminizing GAHT, despite reported changes in liver enzyme levels (Stangl et al., 2021). The clinical significance of these biochemical changes therefore remains unknown.

In a case report of a transfeminine individual on oral estrogen who developed gallstone-induced pancreatitis, the authors proposed that the use of estrogen for feminizing GAHT was linked to gallstone development (Freirer et al., 2021). Several reports have associated the use of estrogen for feminizing GAHT with hypertriglyceridemia-induced pancreatitis as well (Shipley et al., 2020; Tirthani et al., 2021). There is a need for more extensive clinical studies investigating whether transfeminine individuals using feminizing GAHT experience higher rates of pancreatic and gallbladder disease. In the extant literature, case studies report the use of testosterone for masculinizing GAHT and hepatobiliary malignancies, namely hepatocellular carcinoma and cholangiocarcinoma (Pothuri et al., 2023). Larger studies are needed to better understand whether there is a casual association between masculinizing GAHT and hepatobiliary malignancies.

Key Points:

1. Sex differences are present throughout the GI system, particularly regarding the gut microbiome and the liver.
2. Estrogen deficiency in cisgender females and testosterone deficiency in cisgender males are associated with increased risk of NAFLD, findings that are potentially reversible with hormone replacement.
3. Exogenous estrogen has been associated with gallbladder disease and pancreatitis, and both exogenous estrogen and testosterone have been associated with hepatic neoplasms in the cisgender population, findings that, outside of case reports, have not been demonstrated in the TGD population.
4. Masculinizing GAHT may increase levels of hepatocellular liver enzymes in transmasculine individuals, whereas feminizing GAHT may either decrease these levels or have a neutral effect in transfeminine individuals. The clinical significance of these changes is unknown.

ENDOCRINE SYSTEM

Sex Differences in the Endocrine System

The endocrine system comprises glands that produce hormones to influence various bodily functions and is regulated by a complex series of hormonal feedback mechanisms. Aside from the hypothalamic-pituitary-gonadal (HPG) axis, which is most obviously influenced by sex hormones being that they are predominately produced in the gonads, estrogen and testosterone are also known to affect the GH, thyroid, and adrenal axes.

Pituitary

The pituitary gland, often referred to as the “master” gland, produces various hormones that primarily regulate the activity of other hormone-secreting glands throughout the body.

Pituitary glands tend to be larger in cisgender females than in cisgender males (Pecina et al., 2017). This may be due to differences in the hormonal milieu throughout adolescence that influence the growth of the pituitary gland itself (MacMaster et al., 2007). Sex hormones have well-known, suppressive effects on gonadotropic cells in the pituitary gland by providing feedback inhibition to the HPG axis.

Research has shown that sex hormones may also impact other pituitary cell types differentially. Somatotrophs, which produce GH, are stimulated by testosterone, thought to be via the local aromatization of estrogen (Birzniece et al., 2010). In one case report, a patient with Klinefelter syndrome who was receiving exogenous testosterone was found to have a somatotrophinoma (Fang et al., 2016). While these observations are provocative, further research is needed to delineate the association between a commonly used therapy and an exceedingly rare disease. Estrogen has also been associated with pituitary cell proliferation. In one study, estrogen treatment was found to have a suppressive effect on ERβ, a receptor that inhibits pituitary cell proliferation. This downregulation of ERβ subsequently reduced the expression of PTEN, a tumor-suppressor gene, in pituitary cells (Perez et al., 2018). Similarly, estrogen can stimulate the production and secretion of prolactin, in part by causing lactotroph cells to proliferate, while also inhibiting dopamine-releasing cells, which typically downregulate the production of prolactin (Cunha et al., 2015).

Growth Hormone

In concert with the sex hormones, GH is critical to normal growth, development, and metabolism. Regarding metabolism, GH can increase lipolysis, circulating glucose levels, and protein synthesis (Møller and Jørgensen, 2009). Deficiencies in either GH, estrogen, or testosterone all result in a similar phenotype of insulin resistance, low muscle mass, and fat deposition in the liver (Fernández-Pérez et al., 2016). There is sexual dimorphism in the secretion of GH itself that is influenced by the presence of estrogen and testosterone. Male rats have been found to have more infrequent pulses of GH with higher amplitudes than female rats, and females tend to have a higher baseline level of GH than males (Mode and Gustafsson, 2006). This influences the downstream pathways of IGF-1 secretion, which is regulated by the JAK2-STAT5 system. Male pattern GH secretion tends to stimulate increased expression of growth-promoting pathways (such as IGF-1), leading

to more overall body growth in mice. Mouse models have also shown that STAT5b is linked to male body growth, specifically, while STAT5a affects the growth of both sexes, indicating that estrogen is essential for growth and development irrespective of sex (Fernández-Pérez et al., 2016; Lichanska and Waters, 2008).

Studies have shown that the use of exogenous sex hormones impacts GH. Oral estrogen has been shown to increase GH binding protein and inhibit IGF-1 messenger RNA (mRNA) expression in the liver. Due to decreased feedback inhibition secondary to the decline in IGF-1 levels, concentrations of GH then increase. Additionally, oral estrogen reduces levels of the IGF-binding protein complex, reducing circulating IGF-1 levels (Kam et al., 2000). These effects are not seen with transdermal estrogen, probably due to avoiding the first-pass effect through the liver (Leung et al., 2004; Weissberger et al., 1991). In cisgender males with hypogonadism, exogenous testosterone appears to have a synergistic effect with GH on protein metabolism, resulting in increased protein synthesis and decreased breakdown (Gibney et al., 2005). It is also worth noting that the liver, the target organ of GH, is also influenced by sex hormones, with hepatocytes having receptors for both estrogens and androgens. Sexual dimorphism in hepatocyte gene expression has been observed, and differential concentrations of sex hormones between the sexes also affect how the liver responds to GH (Fernández-Pérez et al., 2016).

Thyroid

Thyroid hormone plays a role in the concentrations of sex hormones by indirectly increasing levels of sex hormone–binding globulin (Selva and Hammond, 2009). As is therefore expected, thyroid dysfunction has been associated with abnormalities in the concentrations of sex hormones (Bates et al., 2020; Gabrielson et al., 2019).

All three estrogen receptor forms have been found in human thyroid tissue. It is thought that ERα plays a role in cell proliferation while ERβ promotes apoptosis (Santin and Furlanetto, 2011). This finding has prompted researchers to investigate the role of these receptors in neoplastic thyroid cells and whether neoplastic thyroid cells may have a higher ratio of ERα expression than ERβ (Chen et al., 2008; Kumar et al., 2010; Rajoria et al., 2010). Indeed, one study found increased ERα expression compared with ERβ in papillary thyroid cancer cells (Qiu et al., 2019). Androgen receptors are also found in both normal and malignant thyroid tissue. Thus, like estrogen, researchers have postulated that testosterone may play a role in abnormal thyroid cell proliferation, although further research is needed (Magri et al., 2012; Stanley et al., 2012). Oral estrogen also increases thyroid-binding globulin (TBG) levels because of first-pass metabolism

through the liver (Fernández-Pérez et al., 2016). In euthyroid individuals, this does not affect the steady state of thyroid hormone; however, in hypothyroid individuals, the use of oral estrogen may change the necessary doses of thyroid hormone replacement (Mazer, 2004).

Adrenal

Sexual dimorphism of both the overall size of the adrenal glands and the size of the different adrenal cortex zones have been reported in the literature (Bielohbuby et al., 2007). In murine models, female mice have also been shown to have higher adrenal cell turnover than males, potentially related to circulating gonadal androgens (Grabek et al., 2019). Both androgen and estrogen receptors are found throughout adrenal tissue, and there is evidence that the gonadal hormones play a direct role in adrenal steroidogenesis. This concept has been particularly well studied as it pertains to glucocorticoid production. Estrogen stimulates steroidogenesis in the adrenal glands, whereas testosterone may inhibit said production (Nowak et al., 1995). Clinically, many diseases of the adrenal cortex, such as hypercortisolism due to cortisol-secreting adrenal adenomas and adrenocortical carcinomas, affect females disproportionately (Lyraki and Schedl, 2021). Given the purported effects of estrogens and androgens on the adrenal gland, it seems plausible that varying concentrations of sex hormones between the sexes may contribute to the pathogenesis of adrenal disease. Oral exogenous estrogen increases cortisol-binding globulin (CBG) and total cortisol levels. However, free cortisol levels are generally unchanged or, as in one study, lower (Brien, 1975; Qureshi et al., 2007). This effect is not seen with the use of transdermal estrogen (Qureshi et al., 2007). Spironolactone, a mineralocorticoid antagonist with antiandrogen properties, can occasionally result in excessive mineralocorticoid antagonism, resulting in electrolyte abnormalities and dehydration, which requires providers to be vigilant for side effects.

GAHT and the Endocrine System

Pituitary

Several case studies and case series have reported on occurrences of pituitary adenomas, both secretory and nonsecretory, in the TGD population (Cunha et al., 2015; Kleinschmidt-DeMasters, 2020; Nota et al., 2018; Roerink et al., 2014). Historically, exogenous estrogen has been associated with the development of hyperprolactinemia in transfeminine individuals, prompting the Endocrine Society to recommend periodic monitoring of prolactin levels in this population (Hembree et al., 2017). However, studies

that have observed an association between exogenous estrogen use and rising prolactin included patients who were also using CPA as an adjunctive antiandrogen, indicating that this agent may be stimulating prolactin levels (Cunha et al., 2015; Nota et al., 2018; Raven et al., 2021). Indeed, observational data in patients using exogenous estrogen and spironolactone for feminizing GAHT have generally not been associated with elevated prolactin levels (Bisson et al., 2018; Fung et al., 2016).

In a study investigating the prevalence of pituitary adenomas in TGD individuals, three adenomas were identified among 1,373 transfeminine individuals—one prolactinoma and two somatotrophinomas (Nota et al., 2018). The prevalence of prolactinoma was not higher than the expected rate. Conversely, the two cases of somatotrophinoma would represent a significant increase from the expected prevalence, given the rarity of these tumors. As previously noted, it has been proposed that testosterone may have a stimulatory effect on somatotrophs (Nota et al., 2018; Roerink et al., 2014).

Growth Hormone

The influence of GAHT on the GH axis has mainly been assessed in the context of pubertal growth and the use of GnRH agonists. As previously noted, GH and sex hormones work in tandem to support one’s growth and development. The contribution of sex hormones is clear in TGD youth who receive GnRH agonists and experience delayed skeletal growth and less change to body composition despite having an intact GH axis (Roberts and Carswell, 2021). The effect of GnRH agonists on growth and body composition is discussed in greater detail in the bone and body composition section of this review. Researchers have shown that, in transfeminine individuals who used GnRH agonists and were later transitioned to feminizing GAHT, there may be a dose-dependent effect on height, such that higher doses of estrogen may help achieve height patterns similar to cisgender females (Boogers et al., 2022; Hembree et al., 2017). The use of aromatase inhibitors or nonaromatizable agents (DHT or oxandrolone) has been considered in specific populations, such as in patients with Turner syndrome, to help augment gain in height (Keenan et al., 1993; Wickman et al., 2001; Wit and Oostdijk, 2015). There is not yet strong enough evidence to support the use of these agents in transmasculine patients to help achieve taller heights, although this is an area of active research (Grimstad et al., 2021).

Thyroid

As previously noted, sex hormones influence the levels of TBG in the body. One study suggests that the prevalence of hypothyroidism in the TGD population is higher than that in the cisgender population. However, these

findings were observed in a cohort of just 54 TGD patients (Christensen et al., 2021). To date, no large study has evaluated how GAHT influences thyroid function. New evidence is emerging that those receiving feminizing GAHT have a higher prevalence of thyroid cancer. However, how much of this prevalence may be due to GAHT itself remains unknown (Christensen et al., 2023). No studies have investigated whether masculinizing GAHT affects the incidence of thyroid cancer.

Adrenal

In a study investigating whether GAHT affected cortisol levels in TGD individuals, 18 TGD individuals underwent an adrenocorticotropic hormone (ACTH) stimulation test before and after initiation of GAHT (Sofer et al., 2024). Total basal serum cortisol was noted to be increased in transfeminine individuals following initiation of GAHT, as expected, because of exogenous estrogen increasing levels of CBG. In both transmasculine and cisgender male individuals, however, ACTH-stimulated serum cortisol levels decreased. A separate study similarly noted an overall reduction in cortisol levels after starting GAHT (Colizzi et al., 2013). The literature is somewhat discordant, however, with one study showing that feminizing GAHT increased ACTH and cortisol secretion while masculinizing GAHT decreased ACTH and cortisol secretion (Fuss et al., 2019).

The link between adrenal androgens and feminizing GAHT has also been investigated. In one study, transfeminine individuals using estradiol and CPA for feminizing GAHT had adrenal androgens measured at baseline and up to 4 years of follow-up. They found that levels of adrenal androgens were suppressed after 3 months and remained suppressed through 4 years. This suppression persisted with the cessation of CPA for patients who received a gonadectomy (Cocchetti et al., 2022). This finding may be due to the LH suppression by CPA and estradiol. CPA has also been shown in some cases to cause cortisol suppression in adults and has been studied alone and in conjunction with estrogen treatment in transfeminine patients. A prior study found that the use of CPA alone does not cause cortisol suppression in transfeminine individuals (de Vries et al., 1986).

Key Points:

1. Differences in the hormonal milieu of developing adolescents result in differences in the size of the pituitary gland, with cisgender females having a larger pituitary gland than cisgender males upon reaching adulthood.
2. Somatotrophinoma has been observed in two transmasculine individuals using exogenous testosterone, while hyperprolactinemia has been associated with exogenous estrogen combined with CPA in

2. transfeminine individuals. The risk of prolactin elevation may exist only for those using CPA as an adjunctive antiandrogen agent.
3. Exogenous oral estrogen may increase TBG, increase CBG, and decrease IGF-1, therefore increasing GH. These effects are not observed with exogenous estrogen given via alternative ROAs.
4. Estrogen may increase steroidogenesis in the adrenal glands, whereas testosterone inhibits steroidogenesis production. Despite some discordance in the literature, GAHT appears to decrease ACTH-stimulated serum cortisol levels.
5. In addition to suppressing the production of gonadal sex steroids, feminizing GAHT appears to suppress adrenal androgens as well.

NERVOUS SYSTEM

Sex Differences in the Nervous System

Unlike many other organ systems in the human body, well-defined anatomical sexual dimorphisms in the nervous system are uncommon, with sex differences depending on a complex interplay between genetic, epigenetic, and hormone-related factors. Historically, this has been a challenging area to study, given the profound influence of environment and experiences on the human brain and behavior. However, in the decades since neuroscience was firmly established as an independent field, an explosion of research in the discipline, from animal models to functional imaging in humans, has dramatically increased understanding of sex differences in the brain.

In the mid-twentieth century, the work of Geoffrey Harris (1948) revealed that the hypothalamus controls the release of hormones from the adenohypophysis, establishing a connection between the brain and the endocrine system and setting the framework for what would become the science of neuroendocrinology (de Vries et al., 1986). Subsequent researchers identified peptide-releasing factors in hypothalamic tissue and, eventually, receptor mechanisms in the hypothalamus and pituitary gland, with structural and mechanical similarities to those observed in other peripheral tissues, including those for sex hormones (Schally et al., 1973; Stumpf and Sar, 1976). Indeed, it is now well established that the hypothalamus is a critical component of the HPG axis, responsible for regulating reproductive function, and is subject to feedback inhibition by circulating sex hormones. The intricacies of this system are discussed in more detail elsewhere in this review. These early findings inspired researchers to search for sex hormone receptors elsewhere in the central nervous system.

To examine the potential impact of sex hormones on the human brain, it is essential to recognize that the adult brain is no longer regarded as a static and unchanging organ. Building upon several decades of research on

brain plasticity in animal models, more recent evidence demonstrates that the human hippocampus, a brain region involved in learning and memory, experiences significant neurogenesis throughout adulthood (Spalding et al., 2013). Aside from sex differences that may arise from the action of circulating sex hormones, differences may occur from contributions of genes on X and Y chromosomes or mitochondrial DNA. As noted, sexual dimorphisms in the brain are uncommon. It is reasonable to suspect that, when they do exist, they may be due to these genetic influences. Perhaps the most celebrated example in the mammalian brain is the sexually dimorphic nucleus (SDN) of the preoptic region, called the INAH-3 in humans, which is three to five times larger in male rats than in females (Gorski et al., 1978). Similarly, the nucleus is larger in male humans than in females, although its function remains a topic of debate (Allen et al., 1989). Generally, sex differences in the human brain are far more subtle.

Receptors for sex hormones have now been identified in many regions of the nervous system. After the hypothalamus, the hippocampus was the next brain region in which estrogen receptors were identified (McEwen and Milner, 2007). Research has found evidence for estrogen-induced synapse formation and maturation involving multiple cell types and signaling pathways in the hippocampus (McEwen and Milner, 2017). In fact, the brain appears to have the capability to locally generate estrogens, either by aromatization of androgen precursors or directly from cholesterol (Hajszan et al., 2008). Interestingly, progesterone treatment following estrogen-induced synapse formation was found to downregulate hippocampal spine synapses in rats; however, the mechanism for doing so remains unknown (Woolley and McEwen, 1993). Although less well described, androgens have also been shown to induce hippocampal spine synapses in rats, and, like estrogen, the brain may also be able to generate the androgen DHT locally (Leranth et al., 2004; Okamoto et al., 2012). Similarly, estrogen-regulated spine synapse formation and turnover have been shown to occur in the prefrontal cortex, a region of the brain implicated in executive functions such as planning, decision making, moderating social behavior, and expressing one’s personality (Hao et al., 2007). Animal models have demonstrated that neurons projecting from the amygdala to the prefrontal cortex undergo dendritic expansion in females dependent on the presence of circulating estrogens, with ovariectomized females failing to show these changes (Shansky et al., 2010).

Given the inherent dichotomy between the concentration of circulating androgens and estrogens in cisgender males and that of cisgender females, differential activation of sex hormone receptors in the central nervous system results in sex-specific changes on the molecular and cellular levels. Additional research involving animal models and human subjects has endeavored to connect these findings to tangible sex differences in behavior.

Perhaps one of the more robust behavioral sex differences in humans is sexual partner preference. In one study, male rats neonatally treated with 1,4,6-androstatriene-3,17-dione (ATD), which blocks the aromatization of testosterone into estradiol, experienced either a loss of sexual preference or a reversal toward preference for mounting other male rats (Bakker et al., 1993). Additionally, the size of the SDN, which sex hormones can manipulate, has been shown to correlate with partner preference in rodents, sheep, and humans (Balthazart, 2016). Also, indirect measures of sex hormone exposure in utero correlate with partner preference in adult humans (Roselli and Balthazart, 2011). It is important to note that the observations in humans have generated considerable controversy. Indeed, given the vast array of variables that impact human brain development and behavior throughout the lifespan, it is challenging to make inferences from nonhuman mammalian models or isolated observations in humans. However, the preponderance of evidence does support a hormonal contribution to sexual preference in humans, and it is reasonable to conclude that humans, as with other mammalian species, exhibit sexual differentiation of the brain.

Increasingly, neuroimaging techniques such as functional MRI (fMRI) and diffusion tensor imaging (DTI) have allowed for new inferences to be made regarding how men and women differ in their response to stimuli and how brain connectivity varies between sexes. fMRI studies have suggested sex differences in distributed brain activation during phonological and spatial processing. During phonological processing, where cisgender females tend to perform better than cisgender males, brain activation in cisgender males has been shown to be predominately left lateralized, whereas the pattern of activation in cisgender females is more diffuse (Shaywitz et al., 1995). Conversely, cisgender females showed less right-lateralized brain activation than cisgender males for spatial tasks, where cisgender males tend to perform better than cisgender females (Gur et al., 2000). In one study of 949 youths modeling the structural connectome using DTI, the authors concluded that the brains of cisgender male individuals are structured to facilitate connectivity between perception and coordinated action. In contrast, the brains of cisgender females are structured to facilitate communication between analytical and intuitive processing modes (Ingalhalikar et al., 2014). These studies, however, have also generated controversy, with some expressing concerns related to implicit bias; reverse inference; and technical considerations, such as the role of brain size and movement in the scanner (McCarthy, 2016).

Investigating sex differences in the human brain through the lens of certain health conditions has provided additional insight. In their studies of cisgender girls with congenital adrenal hyperplasia (CAH), a condition that results in prenatal androgen exposure due to a genetic anomaly, researchers found that cisgender girls with CAH were less responsive than

unaffected cisgender girls to information that particular objects (i.e., toys) are “for girls” (Hines et al., 2016). This research suggests that prenatal sex hormone exposure may influence subsequent behavior in part by impacting an individual’s sensitivity to socializing cues, rather than inducing permanent changes in the brain. The menopausal transition also provides an opportunity to observe how changes in the concentration of circulating sex hormones may impact the central nervous system.

Independent of normal aging, modest declines in delayed verbal recall have been shown to occur early in the menopausal transition, with immediate recall declining late in the transition (Epperson et al., 2013). Importantly, abrupt loss of estrogen with oophorectomy appears to have a much more profound impact on cognition than the gradual decline of estrogen that occurs with natural menopause (Burger et al., 1999). Data are conflicting on whether initiation of HRT for menopausal cisgender female individuals impacts global cognitive function. Still, there is some consensus that the later that hormone replacement is initiated relative to the age of menopause, the less likely it is to have a beneficial impact on scales of cognitive function (Hara et al., 2015). Interestingly, animal studies using ovariectomized female nonhuman primates have shown that cyclical unopposed estrogen treatments improved synaptic health and cognitive performance, while chronic estrogen or combination estrogen–progesterone did not (Hara et al., 2015).

GAHT and the Nervous System

Given what we know about the role of hormones in modulating sex differences in the central nervous system, it is reasonable to assume that the provisioning of GAHT may impact the brain’s structure and function. Indeed, as with cisgender individuals, advances in functional imaging have provided insight into the potential influence of GAHT on brain structure and activity in TGD individuals. While the available data are heterogeneous, studies generally indicate an anabolic and anticatabolic effect of testosterone on brain volumes. In contrast, estradiol and adjunct antiandrogen treatment seem to have the opposite effect (Kranz et al., 2020). In a recent study, investigators examined hypothalamic volume before and after 4 months of GAHT in TGD individuals. They found significant volume reductions only in those using estrogen and adjunctive antiandrogens (Konadu et al., 2022). Task-based fMRI studies involving participants on GAHT are expanding but remain scarce. A recent study leveraging fMRI to investigate how GAHT in transmasculine individuals may influence emotional perception found that, while the neural pattern in transmasculine individuals was like that of cisgender females before GAHT, after 6–10 months of receiving testosterone, the neural pattern resembled the pattern of cisgender

males (Kiyar et al., 2022). While, to the authors’ knowledge, no similar fMRI study exists for transfeminine individuals before and after initiation of GAHT, investigators examining the impact of GAHT following gonadectomy in transfeminine individuals concluded that lower serum estradiol levels were associated with cortical thickening. In contrast, higher levels were associated with cortical thinning, although no appreciable relationship to cognitive function was observed (Schneider et al., 2020).

Researchers have employed additional techniques, including genomic sequencing and positron emission tomography (PET), to investigate the effect of GAHT on sexual differentiation of the brain. In one study, investigators sought to examine whether methylation of region III of the estrogen receptor alpha (ESR1) promoter—the promoter of a gene implicated in brain sex differences—is involved in the biological basis of gender incongruence (Fernández et al., 2020). They found that, before GAHT, TGD individuals showed a characteristic methylation profile that differed from both cisgender males and cisgender females and that, following GAHT, the methylation patterns became more like that of their cisgender counterparts who share their gender identity. Research using PET indicates that GAHT may influence serotonergic neurotransmission, with investigators showing that serum testosterone levels are positively correlated with serotonin reuptake transporter (SERT) binding in transmasculine individuals, and, although SERT levels decline after 4 months of estradiol and adjunct antiandrogen therapy in transfeminine individuals, estradiol levels are also positively correlated with SERT binding (Kranz et al., 2015).

Although hypotheses related to behavior and cognition may stem from examining the effect of GAHT on the structure and function of the nervous system, studies employing cognitive and psychological testing have been even more informative. In general, the extant literature supports an enhancing role of postpubertal GAHT on visuospatial ability in transmasculine individuals. There is less consensus regarding the effect of GAHT on cognitive scales in transfeminine individuals. Moreover, the evidence does not support an adverse impact of GAHT on cognitive function for TGD individuals (Karalexi et al., 2020). In terms of psychological health, GAHT appears to have an overwhelmingly positive effect, reducing symptoms of anxiety and depression, lowering social distress, and improving quality of life and self-esteem in TGD individuals (Nguyen et al., 2018). Prior research in this area was reinforced by a recent prospective cohort study of TGD individuals, in which investigators found a significant reduction in symptoms of depression and a nonsignificant decrease in symptoms of anxiety after 18 months of GAHT (Aldridge et al., 2021). More research is needed to investigate the impact of GAHT on executive function, which would provide more insight into how GAHT might affect the day-to-day functioning of TGD people.

Outside of the improvement in symptoms of depression and anxiety, there is, to the authors’ knowledge, no known correlation between GAHT and diseases of the nervous system in the United States. Historically, the use of exogenous sex hormones has been implicated in the development of meningiomas, a class of common intracranial tumors. Indeed, the use of high-dose CPA in transfeminine individuals appears to be associated with an increased risk of meningioma (Millward et al., 2022). This medication is often used as part of GAHT in Europe but is not commercially available in the United States. Of all the agents commonly used as part of GAHT in this country, none have been definitively associated with an increased risk of meningioma in the TGD population. It is worth noting that evidence from cisgender females does not seem to support an increased risk of meningiomas with the use of oral contraceptive pills, and data are conflicting concerning HRT in menopausal patients (Hage et al., 2022). The risk of thromboembolic disease with the use of GAHT, a topic of great interest and controversy, which may impact the cerebrovascular system in the case of ischemic stroke, is discussed in the cardiovascular section of this review. Similarly, hyperprolactinemia, a disease state of the adenohypophysis, has been associated with adjunctive CPA as part of feminizing GAHT and is discussed in the endocrine system section of this review.

In summarizing the extant literature on GAHT and its effect on the brain, there appears to be a consensus that GAHT has a positive impact on the psychological health of TGD patients and, albeit with less consensus, a neutral effect on cognitive function, with some investigators noting an enhancement in visuospatial ability in transmasculine individuals using testosterone. Further research could stratify these potential effects by the dose, ROA, and duration of GAHT. Aside from data reporting an increased incidence of meningioma in users of CPA, no association has been made between the use of GAHT and neurologic disease. Importantly, however, data from cisgender females have shown a potential protective effect on cognition when HRT is initiated closer to the time of menopause (Hara et al., 2015). This surrogate data may help inform providers caring for TGD individuals who could consider prompt initiation or resumption of GAHT following GAS.

Key Points:

1. Research involving mammalian animal models supports a hormonal contribution to sexual differentiation of the brain.
2. Functional imaging in humans suggests that the brains of cisgender males are structured to facilitate more connectivity between perception and coordinated action. In contrast, the brains of cisgender females are designed to facilitate more communication between analytical and intuitive processing modes.

3. Exogenous testosterone appears to have an anabolic and anticatabolic effect on brain volumes, whereas exogenous estradiol and antiandrogens appear to have the opposite effect.
4. Exogenous testosterone may enhance visuospatial ability in transmasculine individuals, whereas exogenous estrogen does not clearly affect cognitive ability in transfeminine individuals.
5. GAHT, whether feminizing or masculinizing, does not appear to have an adverse impact on cognitive function in TGD individuals and has an overwhelmingly positive effect on psychological health.

PULMONARY SYSTEM

Sex Differences in the Pulmonary System

The pulmonary system is a network of tissues and organs crucial for respiration; it includes the airways, lungs, and blood vessels. Variations between the sexes have been established regarding morphology, lung volumes, airway sizes, and responses to exercise (Bellemare et al., 2003; LoMauro and Aliverti, 2021; Townsend et al., 2012). Cisgender males have larger luminal airways, higher total lung volumes and capacity, and a larger number of alveoli in their respiratory tracts. However, relative lung volumes and capacities are the same between the sexes. The ratio of functional residual capacity to total lung capacity remains consistent between cisgender males and cisgender females, indicating that elastic recoil properties are independent of sex (Dominelli and Molgat-Seon, 2022; LoMauro and Aliverti, 2018).

Regarding exercise, cisgender females exhibit lower minute ventilation (Ve) and tidal volume (Vt) during maximal exertion than do cisgender males. However, the extent of increase in both Ve and Vt after exercise is similar in both sexes. Cisgender females allocate a greater proportion of maximal ventilation, or VO2, to respiratory muscles than do cisgender males. Additionally, they experience a higher work of breathing than cisgender males, influenced by factors such as the prismatic shape of the natal female rib cage, body mass, and hemoglobin concentration (Dominelli et al., 2019; Torres-Tamayo et al., 2018). Recent evidence also suggests that cisgender females have a higher incidence of exercise-induced arterial hypoxemia compared with cisgender males (Ansdell et al., 2020; Dominelli et al., 2019). The heightened oxygen expenditure in cisgender females during breathing suggests that a substantial portion of their oxygen intake and cardiac output is allocated to respiratory muscles, impacting their exercise performance (Dominelli et al., 2015).

Sex hormones may also influence pulmonary pathophysiology. One study found that TRT may slow down the progression of chronic obstructive pulmonary disease (COPD) in cisgender males, as measured by a

relative decrease in respiratory hospitalizations over time. Additionally, both estrogen and androgen receptors are present in bronchial airways, and some evidence suggests that, in patients with asthma, elevated levels of bronchial androgen receptors and higher serum androgen levels are linked to reduced hyperreactivity; fewer symptoms; and lower fractional exhaled nitric oxide, a byproduct of inflammation and biomarker for asthma (Han et al., 2020; Mikkonen et al., 2010; Millas and Duarte Barros, 2021; Zein et al., 2021). Fractional exhaled nitric oxide level is positively correlated to progesterone levels and inversely correlated to 17β-estradiol levels (Mandhane et al., 2009). This might explain why asthma exacerbations are seen more frequently in menstrual cycle phases corresponding to high progesterone levels (Chowdhury et al., 2021). Indeed, a murine model demonstrated that female sex hormones enhance the proinflammatory response in the lung when exposed to allergens (Fuentes and Silveyra, 2018). Asthma prevalence shifts toward cisgender females in adulthood and suggests a sex difference in airway reactivity after puberty (Shah and Newcomb, 2018). Furthermore, up to 40 percent of cisgender females with asthma report premenstrual worsening of symptoms, and 20 percent of female asthmatics experience an exacerbation during pregnancy, supporting the role of female sex hormones as proinflammatory in the lung (Rao et al., 2013; Tan and Thomson, 2000).

GAHT and the Pulmonary System

A recent cross-sectional study compared the cardiopulmonary capacity of 15 nonathlete transfeminine individuals who received GAHT for a median of 14.4 years, with 14 cisgender males and 13 cisgender females. The volume of oxygen at peak flow and the mean expiratory volumes for transfeminine individuals on GAHT were found to be higher than that of cisgender females and lower than that of cisgender males. Importantly, relative cardiopulmonary capacity adjusted for fat-free mass was unchanged between the groups (Alvares et al., 2022). Hemoglobin concentration can also affect exercise performance and aerobic capacity, and research has shown that hemoglobin is reduced by 11–14 percent in transfeminine individuals on GAHT (Wiik et al., 2020). Still, other factors contribute to the VO2, including total blood volume and cardiac contractility, which may lead to differences in the aerobic capacity of transfeminine individuals compared with cisgender females (Hilton and Lundberg, 2021).

A recent review attempted to address the question of athletic advantage for TGD individuals. Researchers concluded that in nonathletic transfeminine individuals, after a year of feminizing GAHT, there was a roughly 30 percent increase in fat mass and a 5 percent drop in muscle mass. After 3 years, the percentage of muscle mass had steadily reduced. Although

transfeminine individuals still have higher absolute lean mass, their relative percentages of fat mass, lean mass, hemoglobin, and VO2 peak, when corrected for weight, were like those of cisgender females. No benefit was seen in the physical performance of transfeminine individuals as determined by running time after 2 years of GAHT (Cheung et al., 2024).

From a practical standpoint, spirometry is used to diagnose obstructive and restrictive lung disease. A study of 17 TGD people out of a cohort of 303 aimed to identify the appropriate gender reference to use for the interpretation of spirometry in TGD individuals (Foer et al., 2021). Of the 17 participants, 15 had complete pulmonary function tests available (5 were transmasculine, 8 were transfeminine, and 2 identified as gender nonbinary). The FEV₁ (forced expiratory volume in the first minute) and FVC (forced vital capacity) values were interpreted differently when the gender reference not matching the individual’s gender identity was used. These results imply that applying cisgender male projected FEV₁ and FVC values for a female-sized body may give a false diagnosis of restriction, while the opposite may conceal a real restriction. That said, the authors found that obstructive lung disease diagnosis was not impacted by gender reference (Foer et al., 2021). Another study investigating the use of gender identity, rather than sex recorded at birth, for the analysis of spirometry values found that for transfeminine individuals, the percent predicted FEV₁ and FVC was higher if cisgender male range values were used, while for transmasculine individuals, the percent predicted for FEV₁ and FVC was lower if cisgender male range values were used (Haynes and Stumbo, 2018). The authors concluded that spirometry interpretation was affected for a total of 45 percent of cisgender males and 70 percent of cisgender females when employing gender identity as the reference (Haynes and Stumbo, 2018).

Sex steroids likely play a role in the pathogenesis and symptomatology of chronic lung diseases. For example, HRT use in a large group of Danish postmenopausal cisgender female individuals was associated with a 63 percent heightened risk of new-onset asthma (Hansen et al., 2021). Additionally, other studies have shown that higher serum testosterone levels are associated with a lower risk of asthma in both sexes (Bulkhi et al., 2020; Han et al., 2020). A small case series on TGD individuals revealed the resolution of obstructive sleep apnea (OSA) in a transfeminine individual on estrogen and a new diagnosis of OSA in several transmasculine individuals who started on testosterone (Robertson et al., 2019). Some studies in cisgender males have shown TRT to be a risk factor for the development of OSA (La Vignera et al., 2020; Payne et al., 2021). Contrary to this observation, in hypogonadal cisgender males, studies have shown testosterone deficiency to be a risk factor for OSA (Gambineri et al., 2003; Kim and Cho, 2019). It is therefore recommended to monitor sleep function in individuals prescribed testosterone.

A 2018 cross-sectional study revealed that 29.6 percent of Medicare beneficiaries who identified as TGD had a diagnosis of asthma, compared with 13.6 percent of cisgender individuals. Of note, the risk was highest for transfeminine individuals but was also significantly higher than the cisgender cohort for transmasculine individuals (Dragon et al., 2017). Similarly, the prevalence of COPD among transgender beneficiaries was 27.3 versus 20.8 percent in cisgender beneficiaries. Cisgender males have 1.5 times the risk of developing COPD compared with cisgender females, even without a smoking history, and it is speculated that this may be related to sex hormones (Pinkerton et al., 2015). Further research is needed to elucidate the role of sex steroid hormones on the development or progression of reactive airway disease.

In conclusion, sexual dimorphism exists in the respiratory system; those recorded female at birth have smaller airways and different responses to exercise. There is minimal research regarding the role of GAHT in the progression of chronic lung disease. Nascent research indicates that the relative cardiopulmonary capacity adjusted for fat-free mass was not significantly changed with GAHT. Finally, spirometry tests need to be adapted to take gender identity into account to diagnose restrictive lung disease more accurately in TGD individuals.

Key Points:

1. Sex hormone receptors are present throughout the respiratory system and contribute to sexual dimorphism in the structure and function of the lungs. However, relative lung volumes and capacities are the same between sexes.
2. Estrogen and progesterone may exhibit proinflammatory effects in the lung, whereas androgens are associated with reduced airway hyperreactivity.
3. While TGD individuals on GAHT may exhibit peak flow and mean expiratory volumes that differ from their cisgender counterparts, relative cardiopulmonary capacity adjusted for fat-free mass remains unchanged and no apparent benefit to physical performance has been observed.
4. The reference curves of one’s gender identity should be considered when interpreting spirometry in TGD individuals to avoid falsely diagnosing or, conversely, missing a diagnosis of restrictive lung disease.
5. Exogenous testosterone has been associated with an increased risk of OSA in transmasculine individuals, so it is prudent to monitor for changes in sleep pattern and quality in these patients.
6. Asthma appears to occur with an increased prevalence in TGD individuals compared with the cisgender population. However, this is likely multifactorial, and additional research is needed to elucidate the role of sex hormones in the pathogenesis of asthma.

RENAL SYSTEM

Sex Differences in the Renal System

The renal system is responsible for the excretion of metabolic waste and maintenance of acid–base and electrolyte balance, and contributes to stable blood pressure. It is well established that, in general, blood pressure runs higher in cisgender males than in cisgender females at similar ages until menopause, at which point blood pressure in cisgender females may increase to levels even higher than those seen in cisgender males (Sandberg and Ji, 2012). Researchers have proposed sexual dimorphism in renal structure and physiology as a potential contributing factor to the observed differences in blood pressure. Indeed, one may assume that some degree of sexual dimorphism must be present to maintain homeostasis in the face of fluid and electrolyte flux that occurs when females experience pregnancy and lactation, whereas males have virtually static renal function throughout adulthood.

Experimental animal models have identified sexually dimorphic patterns of renal transporter expression and salt handling. For example, female rats have lower sodium and water transporters and lower fractional reabsorption in the proximal nephron coupled with more abundant transporters in the distal nephron than do male rats (Veiras et al., 2017). These differences, if present in humans, would have significant implications for renal function and could be responsible for the adaptations in fluid retention required during pregnancy and lactation, as well as the “female advantage” regarding blood pressure observed in premenopausal cisgender females. Mouse models have also provided evidence for sexual dimorphism in renal ammonia metabolism, an essential process for maintaining acid–base homeostasis, associated with fundamental structural differences and differences in the expression of proteins involved in renal ammonia and transport (Harris et al., 2018).

Much of the observed sexual dimorphism in renal structure and function is likely driven by inherent sex chromosome differences. There is evidence to suggest that gonadal hormones play a role in mediating these differences. Androgen receptors are expressed exclusively throughout the proximal tubule in the kidneys of both sexes, whereas estrogen receptors are detected in both proximal and distal tubules (Ransick et al., 2019). In a study using male mice, orchiectomy was found to decrease kidney and proximal tubule size while increasing ammonia excretion, effects that were all reversible with testosterone replacement (Harris et al., 2020). Data from animal studies indicate that estrogen may have renoprotective effects, decreasing glomerulosclerosis and risk of ischemia-reperfusion injury (Elliot et al., 2003; Hutchens et al., 2012). Additionally, estrogen contributes to vasodilation by increasing nitric oxide activity, which may increase renal

blood flow and, thus, glomerular filtration rate (GFR) (Issa et al., 2015). Testosterone, on the other hand, has been associated with worsening albuminuria and glomerulosclerosis in mouse models (Long et al., 2013). If the effects of sex steroids on the renal system in mice can be extrapolated to humans, then one might expect a decline in renal function with the loss of estrogen in cisgender females. Indeed, in a population-based cohort study of cisgender females who underwent bilateral oophorectomy before age 50, a higher risk of developing chronic kidney disease (CKD) was noted even after adjusting for multiple chronic conditions and other possible confounders (Kattah et al., 2018).

Serum creatinine is a commonly used biomarker to estimate one’s GFR. Creatinine is a chemical waste product of creatine that is formed during the digestion of protein or the breakdown of skeletal muscle tissue and cleared by the renal system. Therefore, serum creatinine levels outside the normal range may indicate a change in kidney function, diet, or medications.

Epidemiologic research has shown that cisgender males have higher mean serum creatinine, findings that have prompted the development of sex-distinct reference ranges for what is considered normal serum creatinine (Jones et al., 1998). This is likely due, at least in part, to sex differences in body composition—cisgender males tend to have more lean mass, whereas cisgender females have more fat mass (Bredella, 2017).

GAHT and the Renal System

GAHT may influence body composition, with testosterone-based masculinizing GAHT increasing lean mass and estrogen-based feminizing GAHT decreasing lean mass (Klaver et al., 2017). The influence of GAHT on body composition is explored in more detail in this review’s bone and body composition section. Because one’s serum creatinine level is partly a reflection of skeletal muscle breakdown and GAHT may either increase or decrease lean mass, it is crucial to understand the impact of GAHT on serum creatinine to allow for accurate interpretation of these laboratory values.

Researchers have explored the relationship between GAHT and measures of renal function in a systematic review and meta-analysis of nine studies involving a total of 488 transmasculine individuals and 593 transfeminine individuals (Krupka et al., 2022). In transmasculine individuals, serum creatinine increased significantly after 12 months of GAHT, whereas transfeminine individuals experienced a nonsignificant decline in serum creatinine. Two studies included in the meta-analysis reported on 24-hour urine creatinine excretion, with similar changes again noted in transmasculine but not transfeminine individuals at 12 months. These findings are

supported by a subsequent retrospective chart review of 108 adult TGD patients initiated on GAHT at the Mayo Clinic (Maheshwari et al., 2022). In that study, creatinine levels were found to be significantly decreased in transfeminine and increased in transmasculine individuals as early as 3 months and to have reached a new baseline around 6 months, with changes persisting at 12 months. To investigate the effect of hormonal preparations and dose on serum creatinine, those on feminizing GAHT were further stratified by ROA, dose, and serum estradiol level. No significant differences were observed.

However, the authors noted more prominent serum creatinine level changes in those transfeminine individuals who achieved a serum estradiol level of more than 100 pg/mL and who were on higher doses of estradiol therapy. This analysis was not performed in those on masculinizing GAHT, as there were too few in the study.

Sexual dimorphism exists regarding the structure and function of the renal system, and some of these differences are likely mediated by gonadal sex steroids, with estrogen appearing to be somewhat renoprotective. Researchers have detected significant changes in creatinine levels with the use of GAHT. Still, it remains unclear whether these changes represent alterations in renal function or are surrogate markers for changes in body composition. To answer this question, further research needs to examine the effect of GAHT on other biomarkers of renal function, such as cystatin C and measured GFR. There is also a need to explore this effect in TGD individuals with CKD and pediatric TGD individuals, as there is currently a dearth of research involving these populations whose age, etiology of kidney disease, and other factors may modify the effect. Lastly, additional research is needed to determine whether various doses, formulations, and durations—as well as the use of antiandrogen adjunctive agents—modify the impact of both feminizing and masculinizing GAHT on renal function.

Until more robust data are available, providers caring for transgender individuals should be mindful of the potential impact of GAHT on biomarkers of renal function and be wary of relying on the estimated GFR in isolation, particularly in those with borderline renal function, before the initiation of GAHT. A modifier for “female sex” exists to account for expected sex differences in creatinine when estimating GFR. However, this must be interpreted in the context of one’s gender-affirming care to avoid potential harms associated with an imprecise estimated GFR. Based on the extant literature, it would seem reasonable to reassess serum creatinine 6–12 months after initiating GAHT to establish a new baseline and to use the individual’s gender identity when calculating estimated GFR. In situations of ambiguity when estimating GFR, a measured GFR may be appropriate (Mohottige and Tuot, 2022).

Key Points:

1. Cisgender males have higher mean serum creatinine than cisgender females.
2. Exogenous testosterone may result in an increase in serum creatinine in transmasculine individuals, whereas exogenous estrogen may result in a decrease in serum creatinine in transfeminine individuals.
3. The ROA and dose of exogenous estrogen used as part of feminizing GAHT in transfeminine individuals do not appear to have a significant effect on the magnitude of change in serum creatinine.
4. Clinicians should consider reassessing serum creatinine 6–12 months after initiating GAHT to establish a new baseline and use an individual’s gender identity when calculating estimated GFR.

INTEGUMENTARY SYSTEM

Sex Differences in the Integumentary System

The skin, the largest organ in the body, serves a crucial role as the protective barrier between an individual and their environment. Androgens and estrogens are known to affect the pilosebaceous unit of the skin, with receptors for both expressed in sebocytes and the hair follicle dermal papilla (Choudhry et al., 1992; Hasselquist et al., 1980). There are notable sex differences in the structure and function of the skin, in part driven by sex hormones known to influence skin thickness, hair quality and distribution, and sebum production by sebaceous glands. Additionally, sex hormones drive sex differences in immunology, with potential dermatologic implications, discussed in more detail in the immune system section of this review.

Cisgender males have thicker skin than cisgender females and, with aging, cisgender females experience more thinning than cisgender males (Aubert et al., 1985). Evidence from human and animal research suggests that estrogen plays a crucial role in maintaining skin thickness. Menopause is associated with a marked decrease in skin thickness in cisgender females, and ovariectomy has been associated with skin thinning. In contrast, exogenous estrogen administration has been shown to thicken the skin (Bolognia et al., 1989; Punnonen, 1971). Similar findings have been observed in studies of gonadectomized mice, wherein estrogen seems to have a role in regulating epidermal thickness, while androgens regulate dermal thickness (Azzi et al., 2005).

The role of sex hormones in modulating hair growth is complex, involving an interplay among estrogens, androgens, and potentially progesterone.

Androgenetic alopecia is the most common form of hair loss, irrespective of sex. In cisgender males, it usually involves progressive hair thinning in the frontal and temporal areas of the scalp, beginning after puberty and continuing throughout adult life. This process is driven by DHT, a sex hormone that results from the conversion of testosterone by the 5a-reductase enzyme found in the pilosebaceous unit, resulting in a transformation of thick terminal hair follicles into thin villus-like hair follicles on the scalp while stimulating hair growth on the face and body (Makrantonaki and Zouboulis, 2009). In contrast, androgenetic alopecia in cisgender females generally begins after age 30, involves the frontal and parietal scalp, and is not clearly related to androgens, as most patients have normal serum levels (Olsen, 2001).

Sebum production by sebaceous glands plays a role in the development of acne, which may be increased by androgens and decreased by estrogens (Strauss et al., 1962). Male mice have larger sebaceous glands than those of female mice, and gonadectomy has been shown to cause atrophy of sebaceous glands in male mice and growth in female mice, suggesting a role for sex hormone stimulation in the maintenance of sebaceous glands (Azzi et al., 2005). Clinical observations in humans, too, support the role of androgens in the development of acne, as conditions causing androgen excess—including PCOS, adrenal/ovarian tumors, and CAH—can all cause acne (Chen et al., 2011). Conversely, acne generally does not occur before adrenarche, when levels of dehydroepiandrosterone sulfate rise, and neither acne nor sebum production occurs in cisgender males with androgen insensitivity (Imperato-McGinley et al., 1993). Estrogens and antiandrogens, on the other hand, have been shown to decrease sebum production, resulting in an improvement in acne (Lemay and Poulin, 2002).

Beyond structure, sex hormones appear to influence the function of the skin by changing the epidermal permeability barrier, with implications for wound healing. Animal studies have been informative here as well, as estrogen appears to accelerate barrier development in fetal rat skin while testosterone slows it, with male rats having slower barrier formation than females (Hanley et al., 1996).

While sex differences in the epidermal permeability barrier have not been demonstrated in humans, abnormal wound healing predominates in cisgender males over cisgender females, particularly if they are of advanced age (Fimmel and Zouboulis, 2005). Treatment with topical estrogens has been shown to accelerate wound healing in both sexes (Ashcroft et al., 1999). Interestingly, however, the effect is attenuated in cisgender males compared with their cisgender female counterparts, possibly due to an antagonistic effect of testosterone (Fimmel and Zouboulis, 2005).

GAHT and the Integumentary System

For the most part, the dermatologic effects of GAHT may be readily anticipated, given what we know about how testosterone and estrogen affect the skin in all forms. In transmasculine individuals using testosterone, rapid increases in facial and back acne have been observed within the first 6 months of treatment before stabilizing somewhat after 2 years of therapy (Cocchetti et al., 2022; Rutnin et al., 2023; Wierckx et al., 2014b). Testosterone esters have been shown to dramatically increase the severity of acne compared with other preparations (Cocchetti et al., 2022). Testosterone also appears to reliably increase the growth rate and density of facial and body hair within 6–12 months of treatment, remaining steady after 2 years (Cocchetti et al., 2022; Rutnin et al., 2023; Wierckx et al., 2014b). While all testosterone preparations effectively increase hair distribution, IM injections have shown superior outcomes compared with transdermal administration (Cocchetti et al., 2022). Androgenetic alopecia can be expected to occur with testosterone therapy in a subset of transmasculine individuals, ranging in severity, with studies reporting one-third to one-half developing a mild form and one-fourth to one-third developing a moderate to severe form (Rutnin et al., 2023; Wierckx et al., 2014b). When this occurs, it generally happens within 2–5 years of GAHT initiation (Moreno-Arrones et al., 2017).

In transfeminine individuals using estrogen, improvement in facial and back acne is seen in almost all individuals, usually within 3–6 months of initiation, and is accentuated with the addition of an antiandrogen (Radi et al., 2022; Rutnin et al., 2023). Melasma, a common skin condition that generally presents as symmetric hyperpigmentation, has been associated with circulating estrogen in cisgender females, who experience this condition at about twice the rate of cisgender males (Goandal et al., 2022). In one study, melasma was found in one-third of transfeminine individuals on GAHT, a proportion like that seen in cisgender female individuals (Handel et al., 2014; Rutnin et al., 2023). Feminizing GAHT also appears to change facial and body hair patterns, decreasing terminal hair growth rate and density, generally within 6 months, and remaining mostly constant after that (Cocchetti et al., 2022; Giltay and Gooren, 2000; Rutnin et al., 2023). Importantly, a large proportion of individuals continue to experience some level of hirsutism on long-term feminizing GAHT, prompting many to seek cosmetic treatment (Cocchetti et al., 2022). Consensus has not been reached regarding the likelihood of androgenetic alopecia with feminizing GAHT. One study noted an increase in the proportion of individuals with androgenetic alopecia from 0 percent at baseline to 16 percent after 2 years of feminizing GAHT (Rutnin et al., 2023). This is a surprising finding, as estrogens have been postulated to extend the anagen phase of the hair cycle,

aiding hair growth during pregnancy, for example (Desai et al., 2021). It may be that this observation represents a distinct variant of androgenetic alopecia, one that results from an increased estrogen-to-androgen ratio and that has been observed in cisgender men with hypotestosteronemia (Kerkemeyer et al., 2021). The effect of variations in the preparation and ROA of feminizing GAHT on dermatologic changes has not been assessed.

Key Points:

1. Sex hormones drive sex differences in skin thickness, hair quality and distribution, and sebum production.
2. Use of exogenous testosterone in transmasculine individuals increases facial and back acne, increases the growth rate and density of facial and body hair, and may contribute to androgenetic alopecia in a subset of individuals.
3. Use of exogenous estrogen in transfeminine individuals improves facial and back acne, decreases terminal hair growth rate and density, and may increase one’s likelihood of melasma.
4. Variations in dose, preparation, and ROA of GAHT have an unclear effect on the integumentary system, although IM testosterone appears to increase hair distribution more effectively than does transdermal administration.

FERTILITY

Sex Differences in Fertility

Fertility is defined as the capacity of humans to produce offspring. Accordingly, infertility is defined as the inability of two individuals, one cisgender male and one cisgender female, to conceive after 1 year of unprotected intercourse. Successful fertilization involves a sperm cell meeting an ovum, leading to the formation of a zygote. Infertility affects approximately 15 percent of couples in the United States and can be attributed to factors related to one or both sexes (Leslie et al., 2023). A study conducted by the World Health Organization (WHO) found that in 35 percent of infertility cases, both male and female factors were involved (Walker and Tobler, 2023). Since the average human lifespan has increased and rates of infertility increase with age, many tests and gamete quality metrics have been developed to address fertility issues.

Semen analysis is used to assess the reproductive capabilities of cisgender males. This involves measuring total sperm count, total progressive motility, total motility (progressive and nonprogressive), and sperm morphology. The WHO established minimum threshold values that are associated with a higher likelihood of male fertility (Sunder and Leslie, 2022).

Abnormalities in these metrics are attributed to a multitude of factors, including those related to sex hormones. Spermatogenesis is stimulated by testosterone, and the absence of this hormone prevents spermatogonium from developing into spermatids. However, the effects of low testosterone levels on fertility are currently still equivocal.

In one retrospective cohort study of infertile cisgender males, no statistically significant association was found between lower testosterone levels and abnormal sperm parameters (Di Guardo et al., 2020). The authors noted that the absolute number of abnormalities was higher in the low testosterone group, and the nonsignificance could be attributed to a lack of statistical power. Interestingly, while low testosterone may adversely affect reproductive capabilities, studies suggest that TRT may also lead to infertility (Patel et al., 2019). Spermatogenesis is carried out by Sertoli cells, and their proper functioning requires adequate levels of follicle-stimulating hormone (FSH) in addition to testosterone. Exogenous testosterone may hinder this process, as it can negatively feed back on the anterior pituitary, preventing the production of FSH. Indeed, studies have shown that testosterone therapy may have weak contraceptive effects, as patients on therapy produce sperm with altered, albeit reversible, morphologies (Patel et al., 2019). While testosterone is the most abundant sex hormone in cisgender males, estrogen can also influence their fertility; however, much of the existing knowledge is speculative and observational.

In murine models, inhibition of the aromatase gene was associated with infertility (Robertson et al., 1999). Authors noted that aromatase knockout mice exhibited germ cells that could not mature into spermatids. Additionally, they reported an increased frequency of germ cell apoptosis and abnormalities during acrosomal development. While estrogen may be necessary for spermatogenesis, high levels of the hormone have also been associated with infertility. One study of infertile men at a Chinese academic hospital showed inverse associations between estrogen levels and sperm parameters (Luo et al., 2021). Specifically, high serum estrogen levels were associated with low sperm concentration, lower rates of progressively motile sperm, and lower rates of normal sperm morphology. The authors also reported that patients with normal sperm morphology had lower concentrations of estrogen and ERα. Overall, while the relationship between estrogen and fertility in cisgender males needs to be further elucidated, current studies suggest that estrogen is essential for proper reproductive functioning. Still, higher levels of the hormone may lead to infertility.

Among cisgender females, fertility is assessed by characterizing an individual’s ovarian reserves. This entails testing serum levels of anti-Müllerian hormone (AMH) and FSH and counting the number of antral follicles (Deadmond et al., 2000). These tests provide information about the number of remaining follicles but do not indicate oocyte quality. Females are

born with a set number of follicles and their ovarian reserves decrease with age. It is likely that the oocyte quality also decreases with age. Indeed, in a clinical study of infertile cisgender females, the authors reported that while both groups had diminished ovarian reserves, younger women had higher-quality oocytes that led to more successful cycles of in vitro fertilization (IVF) and intracytoplasmic sperm injection (Chang et al., 2018). While age is a significant factor in determining fertility in cisgender female individuals, sex hormones also have notable influences. For example, adequate estrogen levels are essential for proper progression through the menstrual cycle and endometrial thickening for successful embryo implantation. Indeed, low estrogen levels have been associated with endometrial thinning, which can be reversed with estrogen replacement therapy (Liu et al., 2015). Murine models also highlight the importance of estrogen, as low estrogen levels have been associated with abnormal oogenesis (Liu et al., 2022). While estrogen is present at higher levels than testosterone in cisgender females, the latter hormone is still important for fertility.

Elevated testosterone levels have historically been associated with infertility in cisgender females. An early research study examining testosterone levels among cisgender female partners of infertile couples showed that the cohort’s testosterone levels were much higher than that of the general cisgender female population (Steinberger et al., 1979). The authors reported that higher testosterone levels were associated with prolonged follicular phase, as well as higher rates of amenorrhea and anovulation. Many of the participants were determined to have hyperandrogenism, and all were treated with prednisone, a glucocorticoid that suppresses androgen levels. In this study, treatment led to adequate androgen suppression among 80 percent of the participants and approximately half of those who had sufficient suppression were able to achieve conception (Steinberger et al., 1979).

One might conclude from this study that high testosterone levels adversely affect fertility in cisgender female individuals. An explanation for this association is testosterone’s inhibitory effect on GnRH release, which is important for FSH production in the anterior pituitary. Indeed, an in vitro study showed that high testosterone levels arrested follicular development and that supplementation with FSH reversed this effect (Liu et al., 2015). Overall, these studies suggest that high levels of testosterone may contribute to infertility in cisgender females. It is worth noting, however, that a complete lack of testosterone would also adversely affect their fertility.

Animal studies highlight the important role of testosterone in folliculogenesis. An in vitro study of follicles from fetal calves and baboons showed that testosterone could promote differentiation of primary follicles into secondary follicles (Yang and Fortune, 2006). When the researchers added an androgen receptor (AR) blocker, this differentiation was not seen, which suggests that testosterone’s effect on folliculogenesis is mediated by

ARs localized in follicles. Indeed, the authors reported that an immunohistochemistry study of ARs showed increased immunoreactivity as follicles matured (Yang and Fortune, 2006). Another study corroborated this, showing increased AR mRNA expression during early follicular development in bovine models (Hampton et al., 2004). Finally, IVF studies also highlight the therapeutic capabilities of testosterone in treating infertility, as testosterone therapy has been associated with fertility success among cisgender females who initially had poor ovarian responses while undergoing IVF (Noventa et al., 2019; Saharkhiz et al., 2018).

GAHT and Fertility

Extensive research has increased knowledge on how estrogen and testosterone affect the fertility of cisgender males and cisgender females; this gives us insight into how GAHT may affect the fertility of TGD individuals.

The use of feminizing GAHT among transfeminine individuals has been associated with poor sperm parameters. In two cohort studies involving transfeminine individuals who provided semen samples, authors noted that participants who were using hormone therapy had higher rates of abnormalities such as lower sperm motility, concentrations, and total sperm count (Adeleye et al., 2019; Rodriguez-Wallberg, 2021). An explanation for these results is exogenous estrogen’s inhibitory effect on GnRH production. It is important to note that other studies have found a positive association between behavioral factors, such as tucking and wearing tight undergarments, and abnormal sperm parameters (de Nie et al., 2022). Indeed, Rodriguez-Wallberg and colleagues (2021) found that, while transfeminine individuals who were not on hormone therapy had lower rates of abnormal sperm parameters than those who used exogenous estrogen, they still had higher rates of abnormalities than did cisgender populations. While many studies suggest that estrogen adversely affects the fertility of transfeminine individuals, other research suggests that its effects on fertility are multifaceted and may be reversible.

A cross-sectional study examining testes samples from transfeminine individuals who had been on hormone therapy for more than a year showed that approximately 80 percent of samples had germ cells and 33 percent exhibited signs of spermatogenesis (Jiang et al., 2019). Interestingly, authors noted that those who were older were more likely to have germ cells in their semen samples, which suggests that initiation of hormone therapy at an earlier age may impact fertility negatively. Next, two studies showed that spermatogenesis can be restored in many cases after cessation of estrogen therapy and that the semen quality can be adequate for successful fertilization (de Nie, 2022; Yau and Safer, 2023). It should be noted that these are preliminary findings, and more research is needed to understand the factors that predict success in fertility restoration.

Considering these data, clinicians should encourage early sperm banking for transfeminine individuals before initiation of hormone therapy, when feasible. Additionally, clinicians who are helping transfeminine individuals restore fertility should consider temporarily withholding estrogen therapy and encourage their patients to avoid behaviors that may negatively affect their fertility.

Among transmasculine individuals who use testosterone therapy, much of the current research suggests that they can still be fertile after prolonged testosterone usage. One study showed that despite long-term testosterone usage, TGD participants without PCOS had AMH levels that mirrored those of the general non-TGD population (Yaish et al., 2021). This suggests that, while exogenous testosterone may disrupt the menstrual cycle, it does not affect the number of follicles available for retrieval. The viability and quality of oocytes in transmasculine individuals are also likely unaffected, as previous studies have shown that oocytes recovered from individuals on testosterone can be successfully fertilized and developed into embryos that can be implanted (Greenwald et al., 2022; Leung et al., 2019). Overall, these studies suggest testosterone is not detrimental to oocytes, and fertility can be achieved in many transmasculine individuals utilizing testosterone, especially with the advancement of assisted reproductive technology.

Key Points:

1. Absolute deficiency of endogenous testosterone in cisgender males results in infertility due to the cessation of spermatogenesis, although a milder deficiency may not affect fertility. Exogenous testosterone use in cisgender males may also hinder spermatogenesis through negative feedback on the pituitary, reducing the production of FSH.
2. While some endogenous estrogen is likely necessary for fertility in cisgender male individuals, high levels may lead to infertility.
3. In cisgender female individuals, estrogen is critical for proper progression through the menstrual cycle and endometrial thickening for successful embryo implantation. Relative estrogen deficiencies are associated with abnormal oogenesis.
4. While some endogenous testosterone is likely necessary for fertility in cisgender female individuals, possibly because of the role this hormone plays in normal folliculogenesis, excess levels of testosterone are associated with infertility.
5. In transfeminine individuals, feminizing GAHT results in decreased sperm count and motility, likely because of exogenous estrogen’s inhibition of GnRH production. Therefore, early sperm banking, ideally before initiation of GAHT, is recommended. These effects appear most pronounced for those who initiate GAHT earlier,

5. and they appear to be at least partially reversible when estrogen is withheld.
6. Transmasculine individuals often remain fertile with prolonged masculinizing GAHT, suggesting that while exogenous testosterone may disrupt the menstrual cycle, its effect on the follicles themselves is less pronounced.

SEXUAL FUNCTION

Sex Differences in Sexual Function

Sexual function is multidimensional. The International Index of Erectile Function and the Female Sexual Function Index are validated questionnaires frequently employed to measure sexual function in the cisgender male and cisgender female populations, respectively (Lukacs, 2001; Rosen, 2000). Both surveys include general questions to assess one’s libido, arousal, and satisfaction while also aiming to measure certain anatomical aspects of sexual function, such as erectile performance in cisgender men and lubrication in cisgender women. It is estimated that sexual dysfunction affects 31 percent of cisgender male and 43 percent of cisgender female individuals (Rosen, 2000). While biopsychosocial factors influence healthy sexual functioning, this review will focus on the impact of sex hormones—testosterone and estrogen—on sexual function.

Many studies have shown that testosterone is essential for proper sexual functioning in cisgender males. In a survey of hypogonadism in middle-aged and elderly cisgender males, researchers showed that low levels of testosterone were associated with low libido and erectile dysfunction (Wu et al., 2010). Sexual dysfunction was also present in individuals with chromosomal abnormalities associated with hypogonadism. Indeed, in a meta-analysis investigating the association between Klinefelter syndrome and sexual dysfunction, authors reported that lower serum testosterone levels were associated with low libido (Barbonetti et al., 2021). It is worth noting that low endogenous testosterone was not associated with erectile dysfunction in this study.

Current treatment recommendations for sexual dysfunction depend on patients’ symptoms. In patients with erectile dysfunction, for instance, treatment depends on the pathology underlying their condition. Individuals with milder forms of erectile dysfunction and those experiencing decreased libido with demonstrated low serum testosterone levels would likely benefit from testosterone replacement therapy. Conversely, those with more severe erectile dysfunction and those who have erectile dysfunction in the context of vasculopathy, often with intact libido, might be more suitable for treatment with phosphodiesterase type 5 inhibitors, which increase blood flow

to the penis, along with lifestyle modifications (Corona et al., 2017; Rizk et al., 2017). In addition to testosterone, current studies suggest that estrogen may influence sexual function in cisgender males.

In two studies that compared treatment outcomes of bicalutamide, an antiandrogen, with scrotal castration among patients with advanced prostate cancer, researchers reported that patients in the bicalutamide group had higher sexual interest and functioning. Given that castration leads to lower endogenous estrogen and the use of antiandrogen does not have as profound an effect on estrogen levels, these findings suggest that estrogen may promote sexual function in the context of depleted testosterone levels (Kacker et al., 2012). In cisgender males with normal levels of testosterone, however, blocking estrogen receptor signaling was not associated with any change in sexual function (Gooren, 1985). Additionally, in another study that induced hypogonadal changes in healthy cisgender males using GnRH antagonists, authors reported similar improvements in sexual functioning in groups receiving TRT, irrespective of whether they received testolactone, an aromatase inhibitor (Bagatell et al., 1994). However, these results contrast with murine model studies that suggest that estrogen plays a vital role in sexual behavior. Indeed, the authors reported that the knockout of estrogen receptors or aromatase enzymes was associated with decreased sexual behavior in male mice (Brooks et al., 2020; Ogawa et al., 2000). Taken together, these studies suggest that in the context of profoundly suppressed testosterone levels, estrogen may alleviate sexual dysfunction (Kacker et al., 2012). However, under conditions of normal serum testosterone, estrogen may not significantly impact sexual function.

Studies of sexual functioning in postmenopausal cisgender females highlight the connection between estrogen deficiency and sexual dysfunction. The importance of estrogen in supporting normal sexual function is suggested by the presence of estrogen receptors throughout urogenital tissues (Simon, 2011). Indeed, declining estrogen levels have been associated with vulvovaginal atrophy, decreased vaginal lubrication, lowered vaginal blood flow, and reduced orgasmic functions (Clayton, 2003; Simon, 2011). Estrogen deficiency leads to vaginal atrophy through decreased collagen levels and muscular bundle thinning (Da Silva Lara et al., 2009). Additionally, since vaginal lubrication depends on sufficient genital engorgement and estrogen deficiency leads to delayed vasocongestion in the genitalia, individuals with low endogenous estrogen commonly report vaginal dryness, itching, and pain from intercourse due to lack of lubrication (Simon, 2011). These symptoms, however, are usually reversible through estrogen treatment. Indeed, a study of 169 healthy postmenopausal cisgender females found that oral and vaginal estradiol were able to enhance sexual desire and orgasm, increase vaginal lubrication, and decrease vaginal irritation (Cayan et al., 2008). In addition to estrogen, testosterone also influences female sexual functioning, specifically by influencing libido levels.

Androgen deficiency is common among postmenopausal cisgender females and has been associated with decreased energy, low libido, and fatigue in this population (Khera, 2015). Additionally, lack of androgen is implicated in sexual dysfunction symptoms related to genitourinary tissues, including vaginal atrophy and decreased lubrication (Maseroli and Vignozzi, 2020). These symptoms may improve with TRT, as was evident in a study of postmenopausal cisgender females with low serum testosterone levels, in which investigators reported that transdermal testosterone treatment was associated with an improvement in sexual function (Davis et al., 2006). The magnitude of improvement was the same regardless of whether participants were treated with an aromatase inhibitor, implying that androgens likely directly affect sexual function through AR signaling in different tissues.

GAHT and Sexual Function

While sexual dysfunction is a common problem among transfeminine individuals, the influence of estrogen therapy may be less than what has been assumed historically. In a longitudinal study of TGD individuals’ sexual desire after hormone therapy, authors reported decreases in sexual desire inventory (SDI) scores among transfeminine individuals in the first 3 months after initiation of feminizing GAHT (Defreyne et al., 2020). This score eventually stabilized and then began increasing after 1 year. After 3 years, participants’ SDI scores were higher than their baseline scores, suggesting that hormone therapy may indirectly improve sexual desire among transfeminine individuals. It is also possible that estrogen therapy may be protective of sexual dysfunction attributed to adjunct antiandrogens, echoing results from aforementioned studies investigating the effects of estrogen on biological men with prostate cancer (Bales and Chodak, 1996; Tyrrell et al., 1998). Finally, while estrogen therapy may lead to a reduction in nocturnal erections, studies have shown that estrogen likely does not affect sexually stimulated erections (Bettocchi et al., 2004; Kwan et al., 1985).

Among transmasculine individuals, testosterone therapy has been associated with increased sexual desire. Indeed, one study showed that TGD individuals experienced increases in SDI scores shortly after hormone initiation (Defreyne et al., 2020). On average, scores increased until 12 months after initiation and decreased after that. SDI scores were still above baseline at 3 years, but the difference was not statistically significant. Despite these observations, some have postulated that the use of exogenous testosterone in transmasculine individuals may be associated with decreased estrogen levels in the vagina, resulting in vulvovaginal atrophy and decreased lubrication, thereby increasing the risk of discomfort with sexual activity (Tordoff et al., 2023). More research is needed to better delineate these effects and

how they may change over time. In the meantime, it is reasonable to offer topical estrogen, applied to the vaginal tissues, for transmasculine patients who may be experiencing the bothersome effects of decreased lubrication.

Key Points:

1. Sexual dysfunction in cisgender males is associated with testosterone deficiency and, in specific clinical contexts, estrogen deficiency. These effects are partially reversible with TRT.
2. Sexual dysfunction in cisgender females is associated with both estrogen and testosterone deficiencies. These effects are partially reversible with estrogen replacement therapy.
3. Sexual desire in transfeminine individuals appears to decrease with initiation of feminizing GAHT but increases above baseline after several years of therapy.
4. Sexual desire in transmasculine individuals appears to increase shortly after initiation of masculinizing GAHT. Still, the effects are transient, and sexual desire seems to regress to baseline after several years of therapy.

CONCLUSION

Sex differences in the human body are widespread and complex, with circulating sex hormones playing a pivotal role in developing and maintaining dimorphisms throughout the lifespan. Like endogenous sex hormones, exogenous sex hormones have the potential to modify one’s underlying physiology, exhibiting protective effects against disease in some cases and increasing the propensity for disease in others. Gender-affirming care for TGD individuals regularly involves the provisioning of GAHT, often for many years, and clinicians must understand how these treatments may affect a patient’s underlying physiology, predisposition to pathophysiology, and key biomarkers. While there is a need for larger-scale prospective clinical research in this area, the current body of literature is mostly reassuring. TGD individuals who desire gender-affirming care are likely to derive significant benefits from these therapies, and the potential adverse outcomes associated with such therapies may, for the most part, be readily mitigated with appropriate risk stratification and close monitoring.

REFERENCES