Kalantaridou, Sophia N; Calis, Karim A

The therapeutic use of testosterone in women, although controversial, is becoming more widespread despite the lack of accurate clinical or biochemical findings of androgen deficiency. Androgens have more important biological effects in women, acting both directly via androgen receptors in tissues, such as bone, skin fibroblasts, hair follicles and sebaceous glands; and indirectly via the aromatization of testosterone to estrogen in the ovaries, bone, brain, adipose, and other tissues. 

In premenopausal women, the daily testosterone production is approximately 300 mg, of which approximately one-half is derived from the ovaries and one-half from the adrenal glands. Indeed, in premenopausal women, oophorectomy results in a 50% decrease in serum testosterone levels.

Testosterone deficiency in women may no be readily identified, because the symptoms are nonspecific and the measurement of total and free testosterone is inaccurate with commonly used techniques. It has been suggested that the diagnosis of androgen deficiency in women should be based on three criteria: (1) clinical symptoms of androgen deficiency (diminished sense of well-being; persistent or unexplained fatigue; and sexual function changes such as decreased libido, decreased sexual receptivity and decreased pleasure); (2) the diagnosis should be made only in adequately estrogenized women; and (3) free testosterone levels should be at or below the lowest quartile for healthy women. 

Androgen Production in Women

Androgens are the most abundant circulating sex steroid in men and women. Androgens are precursors in the biosynthesis of estrogens. The major androgens and pre-androgens in women include testosterone, dihydrotestosterone, androstenedione, dehydroepiandrosterone (DHEA), and dehydroepiandrosterone sulphate (DHEAS).

They are secreted by the ovary and the adrenal glands and are interconverted in peripheral tissues such as adipose tissue, muscle, and skin. Ovarian androgen secretion is regulated by the luteinizing hormone, while adrenal androgen secretion is regulated by the adrenocorticotropic hormone (ACTH), along with intraglandular paracrine and autocrine mechanisms. There is no direct evidence for an endocrine feedback loop regulating androgen secretion. 

The ovaries are the primary site of testosterone biosynthesis, whereas an important portion of testosterone secretion is derived via androgen precursors. Although a large percentage of testosterone biosynthesis is derived from adrenal precursors, the adrenal gland does not secrete testosterone directly. Normally, only 1%-2% of total testosterone circulates unbound (free testosterone). The rest is bound by sex-hormone binding-globulin (SHBG) or albumin. Bio-available testosterone includes the fraction of free testosterone and the fraction that is bound to albumin, the latter being able to dissociate from the albumin to enter cells. In the peripheral target tissues, testosterone is 5-hydroxylated to its active metabolite dihydrotestosterone or aromatized to estradiol. 

Testosterone is at its lowest levels in the early follicular phase of the menstrual cycle and rises to a mid-cycle peak; luteal phase levels are higher than those in the early follicular phase. The mid-cycle rise in testosterone is lost in the decade preceding menopause. Androstenedione is secreted by the ovarian stroma and the adrenal zona fasciculata. Similar to testosterone, androstenedione levels show a mid-cycle peak, which is lost in older premenopausal women. Oophorectomy results in a significant decrease of androstenedione. 

DHEA is secreted by the ovarian theca and the adrenal zona reticularis. It is also derived from circulating DHEAS, which is secreted solely by the adrenal zona reticularis, with peak levels achieved between the 20s and 30s. DHEAS levels decline as a function of age, and no significant changes occur during the menstrual cycle. DHEAS, DHEA and androstenedione are considered to be pre-androgens because they require conversion to testosterone to exhibit androgenic effects. The role of DHEA is reviewed elsewhere. 

Causes of Androgen Insufficiency in Women

Pathophysiological states affecting ovarian and adrenal function, along with aging, other disease states and some medications have been associated with androgen deficiency in women. 

Natural aging is a major cause of androgen insufficiency. Total and free testosterone concentrations decrease in pre-menopausal women, such that women in their 40s have about one-half the circulating levels of women in their 20s. A recent large cross-sectional study demonstrated that the majority of testosterone decline occurs in the early reproductive years. Testosterone levels remain relatively stable during the menopausal transition. Thus, testosterone levels decrease as a function of aging rather than menopausal transition. 

Androgen insufficiency of ovarian origin occurs in young women with spontaneous premature ovarian failure, hypogonadotropic hypogonadism, and premenstrual syndrome. Iatrogenic causes of ovarian androgen insufficiency include oophorectomy and ovarian failure due to radiation or chemotherapy. Oophorectomy results in approximately a 50% decrease in circulating concentrations of testosterone. Also, exogenous administration of GnRH agonists (or antagonists), oral estrogens, and oral contraceptives can lead to lower circulating androgens. AIDS wasting syndrome also affects testosterone production by suppressing the gonadotropin axis. 

Oral estrogens produce a marked dose-dependent increase in SHBG levels, thereby decreasing bio-available testosterone. In addition, estrogen therapy decreases testosterone levels by suppressing luteinizing hormone levels. Oral contraceptives decrease free testosterone by increasing SHBG levels and by suppressing ovarian sex steroid production. 

For women taking oral hormone therapy, switching to parenteral therapy may alleviate symptoms of androgen deficiency by increasing bio-available testosterone (as a result of decreased SHBG levels). Similarly, discontinuing oral contraceptives may be effective. 

Androgenic progestogens given orally also lower SHBG levels. Oral androgens such as methyltestosterone produce a marked decrease in SHBG. These effects result from hepatic first-pass interactions, in which the steroids modify the synthesis or clearance of SHBG. In contrast to oral steroid administration, transdermal estradiol and transdermal testosterone exert weaker hepatic effects and have a smaller influence on SHBG levels. Hyperthyroidism and anti-epileptic medications may also increase SHBG concentrations. 

Androgen insufficiency of adrenal origin occurs in women with primary or secondary adrenal insufficiency or with chronic glucocorticoid administration. Glucocorticoid treatment decreases adrenal androgen production by suppressing ACTH. Chronic stressful conditions, such as anorexia nervosa and Cushing's syndrome, may also decrease androgen levels. Androgen insufficiency of combined origins occurs in women with hypopituitarism, whether or not they are on estrogen therapy. Rheumatoid arthritis and systemic lupus erythematosus also have been associated with low androgen concentrations in women. 

Diagnosis of Androgen Insufficiency: Problems and Methodological Issues

Clinical symptoms of androgen deficiency include loss of libido, lack of well-being and persistent fatigue. These symptoms can also be attributed to psychosocial and environmental factors. Signs of androgen deficiency include bone loss and muscular decline. Other potential causes of androgen deficiency symptoms, such as depression, chronic fatigue syndrome, major life stressors, thyroid disease, hyperprolactinemia, and iron or vitamin D deficiency, should be excluded

Given the relative unreliability of most clinically available testosterone assays, clinical factors are more reliable than androgen serum concentrations in identifying potential candidates for androgen therapy. Indeed, the available direct radioimmunoassay methods were developed for the measurement of testosterone levels in men; in women, the direct radioimmunoassays have unacceptably high systematic bias and random variability. The gold standard for the measurement of free testosterone in women is equilibrium dialysis. Most laboratories, however, do not use highly sensitive testosterone assays and do not measure free testosterone levels by equilibrium dialysis. Free and bio-available testosterone can also be calculated from total testosterone and SHBG concentrations using published algorithms. 

Testosterone Therapy

The pharmacodynamics of testosterone in women are not well established. Exogenous testosterone therapy appears to improve libido, mood and well-being, and bone and muscle mass. The therapeutic use of testosterone in women became widespread in the last decade. 

In most of the studies of testosterone in women, testosterone was combined with either estrogen therapy (after surgical menopause) or estrogen plus progestogen therapy (after natural menopause). Therefore, at present, testosterone therapy should only be used with concomitant estrogen therapy.

Effect of Testosterone Therapy on Sexual Function

The majority of studies demonstrating beneficial effects of testosterone on mood and sexual function have been performed in postmenopausal women. 

Hypoactive sexual desire disorder is characterized by chronic or recurrent deficit in or absence of desire for sexual activity that causes personal distress and has been associated with androgen insufficiency. Sexual function is complex and influenced by multiple variables. These include psychosocial factors, education, cultural background, environmental factors, and past experiences and relationships with the partner. According to the on U.S. study, approximately 43% of women have sexual dysfunction, manifested mainly as low sexual desire. 

The relationship on endogenous androgen and sexual function has not been clearly defined, and recent controlled studies have not found a link between endogenous androgens and sexual function. An association between surgical menopause and lack of sexual desire has, however, been suggested; up to 50% of women post-oophorectomy report a decrease in sexual desire soon after surgery. 

In randomized controlled trials of testosterone administration in postmenopausal women, there is evidence that the addition of testosterone to hormone therapy has a beneficial effect on sexual function, mainly sexual desire and frequency of sexual activity. 

In an early study, Berger et al. compared the efficacy of testosterone implants (100 mg) and estradiol (40 mg) versus estradiol implants (40 mg) only in surgically and spontaneously menopausal women. At 6 weeks, the testosterone group had significant improvement in sexual function. Similar findings were also shown in two recent studies using oral preparations (oral methyltestosterone and esterified estrogen versus esterified estrogen alone). Four published randomized controlled studies investigated the effect of transdermal testosterone patches on women experiencing sexual dysfunction after surgically induced menopause. In all of these studies, transdermal testosterone was well tolerated and significantly improved sexual desire and frequency of satisfying sexual activity. 

Only a few studies investigating effects of testosterone therapy have been performed thus far in premenopausal women. Goldstat et al. conducted a randomized controlled cross-over study investigating the effects of transdermal testosterone on well-being, mood, and sexual function in healthy premenopausal women who were not clinically depressed and who considered themselves to have low libido. The study showed restoration of well-being and mood scores and significant improvement in sexual function with active treatment as compared with placebo. 

Effect of Testosterone Therapy on Bone and Muscle Mass

Androgens exert a profound effect on the physiology of bone and muscle in women. Androgen receptors are found in all bone cells (osteoblasts, osteoclasts, and osteocytes) but are predominately expressed in osteoblasts. Androgens modulate the bone remodeling cycle by both direct androgenic activity and biotransformation to estrogens. Evidence suggests that low testosterone concentrations in pre-, peri-, and post-menopausal women are associated with low bone mass. Androgen deficiency may explain why despite taking standard estrogen/progestin hormone replacement therapy, 67% of young women with premature ovarian failure have a bone mineral density that has been associated with an increased risk for hip fracture. Estrogen and androgen therapy increases bone mineral density to a greater degree than does estrogen therapy alone. 

Androgens also increase muscle mass and strength. Interestingly, a recent study suggested that 2-year treatment with nandrolone decanoate reduces the vertebral fracture rate in elderly osteoporotic women (>70 years old).

Testosterone Regimens for Women

Empirically, there are several testosterone regimens that can be used in women, such as methyltestosterone, testosterone undecenoate, subcutaneous implants, intramuscular injection, transdermal patches, gels, and creams. Until recently, pharmacologic preparations for physiologic androgen replacement in women have not been available. Such testosterone formulations are currently undergoing clinical trials, however, and may become commercially available in the near future. 

Oral methyltestosterone is approved by the FDA in combination with esterified estrogen (in doses on 1.25 mg of methyltestosterone plus 0.625 mg of esterified estrogen and 2.5 mg of methyltestosterone plus 1.25 mg of esterified estrogen) for the treatment of moderate to severe menopausal symptoms in those women whose symptoms could not be managed by estrogen alone. In Australia and the UK, testosterone pellets have been used in post-menopausal women. 

Parenteral testosterone preparations (i.s. transdermal patches, gels, creams, and inhaled preparations) may be preferred over oral preparations because they avoid first-pass liver adverse effects. 

Potential Risks and Contraindications

Most studies with testosterone use lasted up to 24 weeks, and long-term trials evaluating safety and efficacy of testosterone therapy in women are lacking. Common adverse effects include hirsutism and acne, which generally reverse with discontinuation of treatment. 

Concerns about potential risks of testosterone in women are derived from effects observed from the use of high doses of testosterone in men, including liver toxicity. Severe adverse androgenic effects include alopecia, voice deepening, and clitoromegaly. Of note, pre-menopausal women should use caution during testosterone administration because of the theoretical risk of virilization of a female fetus. 

Cardiovascular risk. Long-term cardiovascular effects of testosterone therapy are not known. In postmenopausal women, however, transdermal testosterone improves endothelial function, a marker of atherosclerosis. On the other hand, endogenous androgen excess in pre-menopausal women with polycystic ovary syndrome may be associated with increased cardiovascular risk. In addition, a recent study showed that low SHBG levels and increased free androgen index (both features of hyperandrogenism) are strongly associated with increased cardiovascular risk. Therefore, women receiving testosterone therapy should be monitored carefully, and circulating androgen levels should be maintained close to the normal female reproductive range. 

Oral methyltestosterone may have adverse effect on the lipid profile due to first-pass effect on the liver. Parenteral user of testosterone does not seem to adversely affect the lipid profile. 

Breast Cancer Risk. One of the main concerns is the long-term effect of testosterone therapy on the breast. Long-term trials evaluating breast cancer risk in women using testosterone therapy are lacking. A review of studies assessing effects of testosterone therapy did not find an increase in breast cancer risk. Data from studies in post-menopausal women, women with polycystic ovary syndrome, and experimental studies indicate that androgens may have an inhibitory protective role on breast cancer. Nevertheless, a recent prospective study of 677 postmenopausal women who subsequently developed breast cancer and 1309 matched controls suggested that endogenous androgens may be associated with increased breast cancer risk. 

Endometrial cancer risk. Although long-term data regarding the effect of testosterone therapy on the endometrium are lacking, experimental findings show a neutral or inhibitory effect of androgens on endometrial cell proliferation. A study of postmenopausal women are 40-60 years showed no differences in endometrial histology in women receiving esterfied estrogens with or without oral methyltestosterone. Also, women with premature ovary failure continued to have scheduled vaginal bleeding patterns while receiving transdermal testosterone in addition to estrogen/progestogen, with no significant changes in the onset, duration, or amount of vaginal bleeding. 

Contraindications

Contraindications to testosterone therapy include pregnancy and lactation, androgen-dependent neoplasia, hirsutism, acne, and androgenic alopecia. 

Follow-Up of Women Using Testosterone Therapy

Follow-up of women taking testosterone therapy should include evaluation of efficacy and assessment of virilizing features as a consequence of exposure to exogenous testosterone. Measurement of serum testosterone levels should be performed to monitor for supraphysiological levels during therapy. Testosterone levels should be maintained within the high-normal range of reproductive-age women.

Conclusion

Increasing evidence suggests that androgens have important biological roles in women. This raises the possibility that androgen therapy may be necessary for the management of carefully selected women with androgen deficiency. Women with natural or surgically induced menopause may be candidates for testosterone therapy, if they have decreased sexual desire and no other identifiable cause for the decreased libido. Prospective trials with testosterone therapy in young women with androgen insufficiency are needed to evaluate the effectiveness of such therapy in pre-menopausal women. The incidence of adverse effects appears to be low, but studies with adequate follow-up are necessary to assess long-term safety. 

Most of the studies showing clinical improvement with testosterone therapy reported supraphysiological androgen levels. The availability of testosterone regimens specifically designed for women has the potential to maintain testosterone levels within the normal range and help to clarify whether the apparent beneficial effects of testosterone therapy are physiologic or pharmacologic. At present, use of testosterone without concomitant estrogen therapy cannot be recommended, because data on safety and efficacy of testosterone-only therapy in women are lacking.