Androgen therapy for low female libido

Androgen physiology, pharmacology and role in sexuality 

Androgens in women are made by the ovaries and the adrenal glands, and are partially regulated by the pituitary gland. The major androgens in women include: dehydroepiandrosterone–sulfate (DHEAS), androstenedione (A), testosterone (T), and dihydrotestosterone (DHT). Some would also include dehydroepiandrosterone (DHEA) as an androgen precursor. An anatomical presentation of the pathways of these androgens is given in Figure 1.

View Large Image Download to PowerPoint

Figure 1. An anatomical presentation of the pathways of androgens in a woman's body. ACTH = adrenocorticotropic hormone, LH = leuteinizing hormone, T = testosterone, A = androgen, DHEAS = dehydroepiandrosterone, DHT = dihydrotestosterone.

Dehydroepiandrosterone–sulfate, DHEA and A are considered pro-androgens, because they need to be converted to T, the most potent androgen, to express their effect. The adrenal zona fasciculata produces 25% of circulating T and is regulated by adrenocorticotropic hormone (ACTH). The ovarian stroma, under the effect of leuteinizing hormone (LH), produces 25% of circulating T. The rest is produced by peripheral tissue conversion of circulating A.11

Testosterone in the plasma exists in three forms: free testosterone (FT), testosterone nonspecifically bound to albumin (AT), and testosterone specifically bound to sex hormone binding globulin (SHBG). The sum of FT and AT is referred to bioavailable testosterone. It is estimated that 66% of T is bound to SHBG and 1%–2% is estimated to be free or biologically available.12 There is enough evidence that the bioavailable fraction rather than total testosterone reflects the clinical milieu. Moreover, the total testosterone level does not accurately reflect the bioavailable T, because the latter is dependent upon the concentration of binding globulins.

Several methods can be used to measure bioavailable testosterone: FT can be measured using equilibrium dialysis to measure the apparent FT concentration (AFTC), which is the gold standard. Alternatively it can be measured by precipitation of SHBG with ammonium sulfate. Both methods are considered time consuming. Faster but less reliable methods are usually used: free androgen index (FAI), which is 100 T/SHBG, and immunoassay using an analog ligand (aFT).13

Plasma levels of total testosterone in women are 0.2–0.7 ng/mL (0.6–2.5 nmol/L). The level fluctuates with the phase of menstrual cycle, being highest at ovulation and lowest during the early follicular phase. It also has circadian variations with a peak in the early morning hours. Testosterone is converted to DHT in target tissues via the enzyme, 5α-reductase. Dihydrotestosterone is the principal ligand of testosterone to the androgen receptors (AR). Those receptors occur in the proximal and distal vagina and the clitoris. Androgens and estrogens regulate their expression. Testosterone can also be aromatized to estradiol (E2) in the central nervous system14 and, along with estrogen modulates cortical centers involved in the interpretation of sexual sensations. Both affect, via the release of neurotransmitters and neuromodulators, the sensitivity of genital organs and the hypothalamic-limbic structures.15

In mice, there is evidence of an indirect effect of androgens on the brain via aromatization to estrogens, whereas in women, even though it is known that androgen and estrogen receptors are present throughout the brain, it is not yet clear whether androgens have a direct effect on the brain or whether the effects occur after the aromatization of androgen to estrogen.16 It has been suggested that free T relates to sexual desire and masturbation in young women.17 In women reporting low libido following menopause, 5α-reductase activity is significantly impaired.18

Serum androgens decline in women for several reasons. Advancing age promotes a decline of the level of circulating androgens (free testosterone, total testosterone, DHEA, and DHEAS). This is caused by the decrease in adrenal and ovarian production of androgen in older women.19 Physiologic menopause leads to decreased ovarian A synthesis, which leads to decreased levels of testosterone produced by peripheral conversion of A.14 Surgical or chemical ovariectomy (by GnRH antagonists, radiotherapy and exogenous estrogens administration) can reduce both testosterone and androstenedione by 50%. Oral contraceptive pills and oral estrogen can also decrease androgen levels by suppressing LH secretion and increasing the level of sex hormone binding globulins (SHBG).20, 21 Glucocorticosteroids can also reduce adrenal androgen production by suppressing ACTH production.22

Androgen formulations

Androgens are available in the following forms: oral methyltestosterone, micronized testosterone, DHEA, topical testosterone ointment, intramuscular testosterone injections, testosterone implants and testosterone patches. All—alone or in combination with estrogen—can used for treatment of low libido in women. However none of these preparations has been approved by the FDA for female hypoactive sexual disorder and thus their use is considered off-label. Unique characteristics of these individual agents follow.

Methyltestosterone 

Methyltestosterone is available in doses of 1.25 or 2.5 mg/day. It is well-absorbed orally and is available in a fixed-dose combination product with esterified estrogens (0.625 mg/day or 1.25 mg/day).9 Its use for FSD have been associated with a significant decrease in high density lipoprotein (HDL) cholesterol level, increased ratio of total cholesterol to HDL cholesterol, but significant decrease in triglyceride levels.10, 35, 36 Although there exist health concerns regarding high dose methyltestosterone on hepatic enzymes and blood pressure, there are little data to support convincingly the detrimental effects of low-dose methyltestosterone on these endpoints.36

Injectable testosterone 

Injectable testosterone is available as testosterone esters (cypionate, propionate, and enanthate) and are administered 25–50 mg intramuscularly every 2–4 weeks. The injections can be painful and the formulations may result in extreme fluctuations in androgen levels.11

Testosterone implants 

Testosterone implants are available in a 50-mg pellet, which are placed alone or in a pair every 3 months, requiring a minor surgical procedure. Depending on the number of implants inserted, they can result in supraphysiological dosing.37

Topical testosterone 

Topical testosterone can be compounded to any strength in an absorbent medium and is applied to the inner thigh and inner arm. Despite its popularity, there are few good studies of this application.38

Testosterone patch 

Testosterone is available in 300-μg patches for men applied once daily. It achieves supraphysiologic plasma levels and, as of this writing, haven't lead to significant changes in the mean levels of lipids, lipoproteins, and other metabolic markers.27, 30 A lower dosage patch is under development for women.

DHEA 

Dehydroepiandrosterone, a popular over-the-counter supplement, is available in 25-mg and 50-mg tablets and can be used in women with Addison's disease. This treatment produces a physiologic serum level of androgens and no significant side effects at these doses.31, 33, 34, 39, 40

It is well known that low libido is associated with the conditions that decrease the physiologic levels of androgens such as oophorectomy,23 adrenal gland dysfunction or hypopituitarism.24 Decreased testosterone levels are associated with loss of libido, fatigue and decreased feelings of well-being.25 In a cross-sectional study, six hundred eighty-four women, ages 50–89 years, were surveyed for hysterectomy and oophorectomy status and had plasma obtained between 1984–1987. They found that total and bioavailable (non-SHBG-bound) testosterone levels were 40–50% reduced in bilaterally oophorectomized postmenopausal women compared to those in intact women26 leading to decreased libido in 30% to 50% of those patients.14, 26, 27

Since our ability to measure free testosterone accurately in the range normally found in women is limited, it is generally considered inappropriate to treat a woman with low level of testosterone unless she complains of low libido.13, 28

Testosterone and low libido 

The role of androgens in female sexuality has been addressed since the 1940s by Greenblatt et al.29 Many studies have been published since the Greenblatt study with mostly inconclusive results. Most of those reports were retrospective, uncontrolled and used unsystematic methods of data collection. Several new, more methodologically sophisticated studies have been recently published and our review focuses mostly on those. The majority of those studies have been done on postmenopausal women (surgically or naturally) receiving or not concomitant estrogen replacement therapy. The sudden and uniform decrease of plasma levels of androgen and estrogen provided in fact a unique opportunity for studying the effect of those steroids on sex behavior.

Sherwin et al found that surgically menopausal women who received supraphysiologic doses of testosterone enanthate administered intramuscularly alone or with estrogen had significantly higher scores for sexual desire, fantasies and arousal compared to women who received estrogen alone. The effect on the frequency of coitus and orgasm was not significant.10

Lobo et al found that in healthy postmenopausal women (natural or surgical > 6 months) methyltestosterone treatment results in improvement in sexual interest or desire partially by decreasing the levels of SHBG, leading to increased bioavailable testosterone, and possibly by a direct effect of methyltestosterone acting as androgen. They also concluded that the combined treatment wasn't associated with a clinically significant change in acne and hirsutism.9

Shifren et al found that the 300-μg testosterone patch significantly increased the scores for thoughts, i.e., desire; frequency of sexual activity, and orgasm compared to placebo in surgically menopausal women.8

SR Davis et al found that the addition of testosterone to estradiol in postmenopausal women resulted in significantly greater improvement compared to estradiol alone for sexual activity (p < 0.03), satisfaction (p < 0.03), pleasure (p < 0.01), orgasm (p < 0.035) and relevancy (p < 0.05). They concluded that the addition of testosterone has therapeutic effect for dimished libido in postmenopausal women.28

S Davis et al found a statistically significant increase in the sexual desire score of the profile of female sexual function (PFSF) compared with the placebo group (p < 0.05) as well as change in other domains of the PFSF including orgasm, sexual arousal, sexual responsiveness, sexual self image and sexual concerns, and significant reduction in personal distress in surgically menopausal women using 300-μg/day testosterone transdermal patch.30

Buster et al found a significant improvement of all seven domains of sexual function, including sexual desire, pleasure, arousal, orgasm, concerns, responsiveness and self image, in surgically menopausal women using 300-mg testosterone.27 Their data agree with other clinical findings that high physiologic/low supraphysiologic doses of testosterone improve several aspects of sexual function in postmenopausal women with FSD—precisely low libido.

The majority of the aforementioned studies used different validated and nonvalidated instruments and scales to measure the effect of androgen on sexual function. The effectiveness and value of those scales to measure low libido, even though used by several studies, is questionable. Also the measurement of the different serum testosterone fraction varied from one study to another, which makes it even harder to compare those studies (see Table 2).

TABLE 2. Testosterone studies

	Investigator	N	Testosterone form	Duration of study	Scale used to measure sexual function	Type of study
	Sherwin et al10	43	Monthly testosterone enanthate injection	8 months	Daily menopausal rating scale (DMRS)	Crossover
	Lobo et al9	218	1.25 mg of methyl-testosterone	16 weeks	Sexual interest questionnaire (SIQ)	Double-blind randomized clinical trial
	Shifren et al8	75	150-mg/300-μg testosterone patches	12 weeks	Brief index of sexual functioning for women (BISF-W)	Randomized clinical trial
	SR Davis et al28	34	50-μg testosterone implant	2 years	Sabbatsberg sexual self-rating scale	Single-blind randomized clinical trial
	S Davis et al30	77	300-μg testosterone transdermal patch	24 weeks	Profile of female sexual function (PFSF)	Double-blind randomized clinical trial
	Buster et al27	533	300-μg testosterone transdermal patch	24 weeks	PFSF and the personal distress scale	Double-blind randomized clinical trial

Dehydroepiandrosterone and low libido 

Patients with Addison's disease usually have a chronic deficiency in DHEA, because their treatment focuses on glucocorticoides and mineralocorticoides and neglects the androgen deficiency. Dehydroepiandrosterone is transformed into potent androgens in various tissues; therefore its replacement may be important in women suffering from Addison's disease.31 In this randomized crossover study, nine women with normal baseline DHEA underwent adrenal suppression of DHEA with dexamethasone to levels found in Addison's disease and were randomly assigned to receive 50 mg or 100 mg oral capsules of DHEA. The administration of 50 mg of DHEA was sufficient to restore the serum concentration of DHT and ADG to nonsupressed baseline levels, whereas 100 mg DHEA was required to fully restore A and T concentrations to baseline. A dose of 100 mg/day induces supraphysiological levels of DHEA in these women. The investigators concluded that a daily dose of 50 mg DHEA is a suitable replacement therapy in females with adrenal androgen deficiency.31

In the randomized clinical trial conducted by Lovas et al, 39 women aged between 18 and 70 years with confirmed primary or secondary adrenal failure were randomized to receive 25 mg of oral DHEA versus placebo for 9 months. They found that 25 mg DHEA was enough to produce physiologic serum levels of androgen in patients with Addison's disease but did not have any significant effect on sexual function.32 The same result was found in the randomized crossover clinical trial done by Hunt et al using a daily 50 mg dose of DHEA for 12 weeks. The treatment corrected the deficiency of DHEA and DHEAS, achieving normal range of those hormones, with T levels at the lower end of the normal range. However, there was no significant difference in any of the indices for sexual interest compared to placebo.33

In contrast, Arlt et al conducted a double-blind, placebo-controlled, crossover study on 24 women with adrenal insufficiency. Women received in a random order 50 mg of oral DHEA daily for 4 months versus placebo. The investigators concluded that replacing DHEA in women with adrenal insufficiency improves sexual desire as a direct effect of DHEA on the nervous system, an increase of peripheral androgen synthesis or both.34

Side effects and safety of androgen therapy 

Side effects of androgen therapy can be divided into two categories: early reversible side effects that encompass acne, skin oiliness, hirsutism and menstrual irregularities; and late irreversible side effects of virilisation including male-pattern baldness, hirsutism, voice changes and hypertrophy of the clitoris. These late side effects have been reported primarily in supraphysilogical doses of androgen, especially those doses used in men. However, the androgen regimens currently used in postmenopausal women haven't been shown to possess high risk in most women, and are felt by many to provide significant benefit, encompassing sexual health.8, 9, 27, 41

However, all patients should be aware of the possibility of side effects whenever androgen replacement is administered. They should be aware that prolonged use of high doses of certain androgens (17α-alkyl androgens like methyltestosterone) have been associated with serious hepatic adverse effects: peliosis hepatis, hepatic neoplasms, cholestatic hepatitis and jaundice. Peliosis hepatis can be a life-threatening or fatal complication. Long-term therapy with testosterone enanthate, which elevates blood levels for prolonged periods, has produced multiple hepatic adenomas. Testosterone is not known to produce these adverse effects. Edema with or without congestive heart failure may be a serious complication in patients with preexisting cardiac, renal or hepatic disease. Diuretic therapy may be required in addition to discontinuation of the drug. In patients with breast cancer, androgen therapy may cause hypercalcemia by stimulating osteolysis. In this case, the drug should be discontinued. Although there are rare reports of hepatocellular carcinoma in patients receiving long-term therapy with androgens in high doses, withdrawal of the drugs did not lead to regression of the tumors in all studied cases. Hemoglobin and hematocrit levels should be checked periodically in patients on long-term androgen therapy to detect the possible side effect of polycythemia. Also liver function, cholesterol and high-density lipoprotein should be checked periodically to rule out any hepatic disorder.42

Finally, further studies need to be done in order to evaluate the long-term side effects of androgen replacement therapy in women.

Conclusion 

Evidence agrees that supraphysiologic doses of testosterone improve the desire/interest and pleasure of sexual experience in surgically postmenopausal women. However, sexual function is complex and depends on more than hormones. It is very possible for postmenopausal women with low androgen levels to have fulfilling sexual lives. The reverse, that is women with healthy levels of hormones having unsatisfactory sexual lives, is also true. Before embarking on any treatment for low libido, it is important to take a thorough sexual and general psychiatric history to exclude mental illness associated with sexual dysfunction, preoccupation with a life crisis and/or core relationship problems and to find out whether the patient's psychological profile is fairly intact. As Levine suggested, it may be useful to conceptualize libido/sexual desire as consisting of drive (biological, mostly hormones), motive (individual and relationship psychology), and wish (cultural) components.43 Further variables affecting libido include age, social situation, and health.43 Thus, though hormones play an important role in sexual desire, they are only part of the entire story. Nevertheless, assessing the effect of testosterone and other androgens in premenopausal women could help improve our understanding of the role of androgens in female sexuality and sexual desire.

The ideal androgen preparation to be used in treating low sexual desire in women would be the one that combines safety and efficacy and would maintain androgen levels reliably within the female normal range. The most effective therapeutic formulation remains to be defined. Sensitive instruments designed specifically to assess sexual function in postmenopausal women are lacking, and the aspects of sexual function measured by the interventions have been unspecified. Few data have been published concerning the long-term effects of androgen substitution in women; therefore the long-term safety of this treatment is still unknown.