Human Reproduction Update Advance Access originally published online on April 5, 2006
Human Reproduction Update 2006 12(4):463-482; doi:10.1093/humupd/dml010
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Male hormonal contraception: concept proven, product in sight?
Department of Obstetrics and Gynaecology, Prince Henry’s Institute of Medical Research, Monash University, Monash Medical Centre, Clayton, Victoria, Australia
1 To whom correspondence should be addressed at: Prince Henry’s Institute of Medical Research, PO Box 5152, Clayton, Victoria 3168, Australia. E-mail: kati.matthiesson{at}princehenrys.org
Submitted on November 18, 2005; resubmitted on January 17, 2006; accepted on February 20, 2006
 |
Abstract |
|---|
TOP
Abstract
Male hormonal contraception in...Current male hormonal contraceptive (MHC) regimens act at various levels within the hypothalamic pituitary testicular axis, principally to induce the withdrawal of the pituitary gonadotrophins and in turn intratesticular androgen production and spermatogenesis. Azoospermia or severe oligozoospermia result from the inhibition of spermatogonial maturation and sperm release (spermiation). All regimens include an androgen to maintain virilization, while in many the suppression of gonadotrophins/spermatogenesis is augmented by the addition of another anti-gonadotrophic agent (progestin, GnRH antagonist). The suppression of sperm concentration to 1 x 106/ml appears to provide comparable contraceptive efficacy to female hormonal methods, but the confidence intervals around these estimates remain relatively large, reflecting the limited number of exposure years reported. Also, inconsistencies in the rapidity and depth of spermatogenic suppression, potential for secondary escape of sperm into the ejaculate and onset of fertility return not readily explainable by analysis of subject serum hormone levels, germ cell number or intratesticular steroidogenesis, are apparent. As such, a better understanding of the endocrine and genetic regulation of spermatogenesis is necessary and may allow for new treatment paradigms. The development of an effective, consumer-friendly male contraceptive remains challenging, as it requires strong translational cooperation not only between basic scientists and clinicians but also between public and private sectors. At present, a prototype MHC product using a long-acting injectable testosterone and depot progestin is well advanced.
Key words: contraception / FSH / LH / spermatogenesis / testosterone
|
Male hormonal contraception in the global contraceptive context |
|---|
The World Health Organization (WHO) has designated the provision of means by which couples can avoid childbearing when too young or too old, too quickly in succession or simply just too frequently, as essential to improving the health of women (Marston and Cleland, 2004
). Effective family planning has only become available within the last 50 years following the development of female hormonal contraception in the 1950s (ESHRE Capri Workshop Group, 2002). However, despite this advance, unwanted pregnancy remains a significant health problem, particularly in the developing world where over half a million women die annually from the complications of pregnancy, childbearing or unsafe abortion each year (WHO, 2004a, 2004b). Furthermore, even in the developed world, a wider range of effective and convenient contraceptive choices are required to facilitate greater usage by couples for whom current options are unsuitable.
Barriers to wider contraceptive utilization include lack of financial support for services or knowledge of their availability, socio-cultural traditions and religious beliefs (Ghazal-Aswad et al., 2002). Unfortunately, these barriers are perhaps greatest in the developing world where the need for easy access to birth control is so critical. Currently, men play a major but often under-recognized role in contraceptive practices through the use of natural family planning, condoms and sterilization. Yet, these contraceptive options are restricted and not particularly attractive, for example the condom has limited acceptance and is not highly effective, whereas vasectomy is very effective but invasive, limited in availability and not readily reversible. Presently, there is no male contraceptive product that satisfies the requirements of high efficacy and reversibility whilst being inexpensive and acceptable.
Recognizing the importance of new options as a means of increasing contraceptive utilization rates, the WHO has identified the development of reversible male methods as a priority (WHO/ PAHO, 2002). Male hormonal contraception (MHC) exploits the classic endocrine feedback loop to suppress spermatogenesis. Pituitary gonadotrophins are suppressed via administration of testosterone or a derivative androgen often given in combination with a second anti-gonadotrophic agent (e.g. progestin or GnRH antagonist). Emerging data indicate that MHC provides a reversible option for men with efficacy equal to that of the female oral contraceptive pill (OCP).
The first large MHC trials in which pregnancy was an endpoint, so-called efficacy trials, were conducted under the auspices of the WHO and reported in the 1990s. These trials involved close to 700 couples and demonstrated that the suppression of sperm concentration to azoospermia (WHO Task Force on Methods for the Regulation of Male Fertility, 1990) or indeed only to severe oligozoospermia (below 3 x 106/ml) (WHO Task Force on Methods for the Regulation of Male Fertility, 1996) was highly effective in preventing pregnancy. Nonetheless, a small percentage of men fail to suppress sperm output adequately even when gonadotrophin levels are non-detectable, suggesting the possibility of gonadotrophin-independent spermatogenesis (McLachlan et al., 2004). Two more recent studies have shown that a long-acting testosterone preparation (Gu et al., 2003) and a testosterone plus progestin combination regimen (Turner et al., 2003) also provide effective, cheap and relatively convenient contraception for wider clinical application. There have also been numerous studies [reviewed recently (Kamischke and Nieschlag, 2004; Grimes et al., 2005)] using sperm concentration as a surrogate endpoint to suggest that MHC can provide adequate suppression of spermatogenesis (<1 x 106/ml) for acceptable efficacy based on existing data sets and the consideration of experts in the field (Sixth Summit Meeting Consensus, 2002). Together, these ‘proof of concept’ trials have fostered the first expressions of interest from the pharmaceutical industry (Bonn, 1999; Hay et al., 2005; Brady et al., 2006).
Ideally, an MHC product would provide (i) equivalent or better efficacy to the female OCP, (ii) universal spermatogenic suppression, (iii) quick onset and offset of action, (iv) easy administration, (v) good tolerability, (vi) favourable short- and long-term side-effect profiles (or even health promoting effects) and (vii) be competitively priced with existing options. Studies to date would support MHC as being likely to satisfy the first and maybe second requirements, but uncertainties remain in regard to the later demands. MHC is most clearly to be seen in the context of a stable relationship, as it does not address the issue of sexually transmitted disease (STD) and in particular HIV prevention. However, taking a more wide-ranging view, MHC may still offer advantages for personal fertility control, given that condoms may not be available and/or used and, even if used, have a reasonably high failure rate. It should also be recognized that the OCP does not address the issue of STD prevention. Future contraceptive planning strategies for men must balance the major public health role of condoms whilst bearing in mind the considerable advantages that an MHC product may offer to some men and in particular those in stable relationships.
One frequently expressed concern is whether men would/could be trusted with contraceptive responsibility and/or whether their partners would accept them taking this role. This view is particularly Western, where the burden of contraception has traditionally fallen to the women. Without a final MHC product and knowledge of the burden of its use or side-effect profile, it is difficult to perform ‘market surveys’ of consumer acceptability. However, research across different countries has shown that between 14 and 83% of men would use an MHC if available, and only a small number of women (<2%) would not trust their partners to take contraceptive responsibility (Brooks, 1998; Glasier et al., 2000; Martin et al., 2000; Weston et al., 2002a, 2002b; Heinemann et al., 2005).
In this review, we aim to (i) give an overview of the hormonal regulation of spermatogenesis, and its interruption by MHC, (ii) provide a summary of MHC trial data with particular focus on efficacy trials, (iii) explore the possible factors involved in the variable response to MHC agents, (iv) speculate on new contraceptive targets and therapies to regulate spermatogenesis and (v) consider issues of contraceptive acceptability and the barriers that must be overcome in bringing an MHC product to market.
|
Physiology of spermatogenesis in relation to the MHC strategy |
|---|
The hypothalamic–pituitary–testicular (HPT) axis is a classic endocrine feedback system that allows for the precise control of circulating hormone levels and optimal sperm output. GnRH is secreted in a pulsatile fashion into the hypothalamic portal circulation, acting on the anterior pituitary to stimulate the release of the gonadotrophins, LH and FSH. In turn, these glycoprotein hormones act respectively on the testicular Leydig and Sertoli cells, of which the former provides adequate testosterone secretion for virilization and sperm production. The resulting testicular hormonal products, including testosterone, estradiol (E2) and inhibin B, provide negative feedback on the hypothalamus and pituitary to regulate GnRH and gonadotrophin secretion (Finkelstein et al., 1991a
, 1991b; Hayes et al., 2001). Figure 1 depicts the HPT axis and sites of trialled and potential MHC agent action.
|
Gonadotrophin dependence of spermatogenesis
The relative roles of FSH and LH in human spermatogenesis were explored in a series of studies in which men underwent both short- and long-term gonadotrophin and spermatogenic suppression via exogenous testosterone administration and then were selectively replaced with either physiological (Matsumoto et al., 1986) or supraphysiological (Bremner et al., 1981; Matsumoto and Bremner, 1985) human chorionic gonadotrophin (as an LH substitute) or FSH (Matsumoto et al., 1983). FSH and LH alone could reinitiate and maintain a degree of spermatogenesis, but normal sperm output required both gonadotrophins. Further work has shown that MHC must induce significant FSH and LH withdrawal to cause spermatogenic suppression (Buchter et al., 1999; Gonzalo et al., 2002).
The relationship between gonadotrophin and sperm concentration suppression is linear to an extent with the inadequate withdrawal of gonadotrophins (i.e. detectable by conventional assay), resulting in failure of spermatogenic suppression (Matsumoto, 1990; Meriggiola et al., 1998; Buchter et al., 1999). However, once gonadotrophin levels are barely or non-detectable, the correlation with sperm concentration disappears suggesting the possibility of gonadotrophin-independent spermatogenesis in some individuals administered MHC agents. Data obtained using ultra-sensitive assay methods suggest that MHC treatment abolishes LH from serum (<0.4% baseline), whilst FSH levels remain at 1–2% baseline levels (Robertson et al., 2001). It is possible that despite undetectable LH some LH bioactivity remains, given that intratesticular iT testosterone levels remain above that in serum (i.e. 50 versus 25 nmol) (McLachlan et al., 2002b; Matthiesson et al., 2005b) but more likely is that there is a low level of LH-independent androgen production (Zhang et al., 2001, 2004). Whether these residual levels of FSH and/or iT testosterone are enough to support a degree of ongoing spermatogenesis in susceptible men is a matter for discussion below.
Gonadotrophin-independent spermatogenesis: lessons from animal models
Animal knockout models featuring disruption of gonadotrophin and androgen action provide further evidence for the possibility of gonadotrophin-independent spermatogenesis. The hypogonadal [hpg (GnRH deficient)] mouse shows reduced testicular size, low intratesticular androgens but a degree of germ cell maturation with development arrested at pachytene spermatocytes. The administration of testosterone to these animals results in a rise of all germ cell numbers and the qualitative completion of spermatogenesis despite undetectable FSH (Singh et al., 1995).
FSH receptor knockout mice display reduced fertility, testicular size, sperm count and motility but a degree of completed spermatogenesis (Dierich et al., 1998). FSH-ß knockout mice exhibit normal fertility with qualitatively normal spermatogenesis despite a 60% reduction in Sertoli cell number and a marked decline in haploid germ cell numbers (Kumar et al., 1997; Wreford et al., 2001). These models of targeted ligand and receptor disruption demonstrate that ongoing spermatogenesis, albeit at a reduced level, is still possible in the absence of FSH action. Furthermore, the reduction in sperm output appears to be a result of a lower Sertoli cell complement consequent on the lack of FSH-stimulated development (Allan et al., 2004).
Mice lacking the LH receptor are infertile and display reduced testicular size and hypoplastic accessory sex organs (Lei et al., 2001; Zhang et al., 2001); however, germ cell development nonetheless occurs but is arrested at the round spermatid (rST) stage. The administration of testosterone results in a resumption of spermatogenesis, presumably because of androgen receptor (AR) activation although sperm number is low and animals remain subfertile (Pakarainen et al., 2005). Qualitatively normal spermatogenesis is achieved in 12-month-old LH receptor knockout mice with very low iT testosterone levels (2% control), whilst the administration of the anti-androgen flutamide results in post-meiotic spermiogenic failure (Zhang et al., 2003). LH-
subunit knockout mice also show a degree of germ cell maturation in the presence of markedly reduced iT testosterone levels, but similar to LH receptor knockout animals, development is arrested at rSTs (Ma et al., 2004).
Spermatogenesis is never completed in the absence of androgen action. The role of androgens has also been explored using complete and selective AR knockout mouse models. Complete AR knockout mice are infertile displaying a feminized phenotype with small, cryptorchid testis lying in the lower abdomen, thereby confounding the interpretation of AR effects on spermatogenesis which is arrested at pachytene spermatocytes (Yeh et al., 2002). Infertility is also seen in selective Sertoli cell AR knockout mice that show normal testicular descent but again spermatogenic arrest at the pachytene spermatocyte stage with few rSTs and evidence of increased apoptosis (Chang et al., 2004; De Gendt et al., 2004). Taken together with the data from the hpg and LH ligand knockout models, a degree of germ cell development is still possible even in the absence of androgen action, but whether the second meiotic division can always be completed is unclear.
Gonadotrophin-independent spermatogenesis in humans
The possibility of species differences for the relative dependence of spermatogenesis on FSH and LH must be considered. To this end, the identification of patients with gonadotrophin deficiency because of congenital GnRH deficiency or mutations in either ligand or receptor genes is informative (Diemer and Desjardins, 1999). Similar to mouse models, patients with LH-ß subunit mutations have shown a degree of ongoing though severely impaired spermatogenesis (Weiss et al., 1992; Valdes-Socin et al., 2004), but in contrast, a patient with an FSH-ß subunit mutation displayed azoospermia (Phillip et al., 1998). Data from men with mutations of the FSH receptor show highly variable sperm densities from very severe oligozoospermia to normal (Tapanainen et al., 1997; Phillip et al., 1998). Also interesting is the description of a hypophysectomized man with an activating mutation of the FSH receptor in whom spermatogenesis and fertility was maintained despite absent FSH and LH (Gromoll et al., 1996). These data show that FSH is not essential for the completion of spermatogenesis (similar to rodents), but it is notable that four of these six men with FSH-ß or receptor mutations had sperm densities below 1 x 106/ml, with one at 6 x 106/ml and one with a normal density, suggesting a relatively greater FSH-dependency of human compared with rodent spermatogenesis.
On the basis of human and animal models, the logical target for MHC development is the maximal suppression of both FSH and LH to minimize residual spermatogenesis. However, it remains unclear to what extent the gonadotrophins must be suppressed, and whether a degree of ongoing spermatogenesis is still possible even in their apparent absence. Certainly, the concept of ‘gonadotrophin independent’ spermatogenesis is worthy of further consideration, particularly in regard to new contraceptive developments not solely dependent on gonadotrophin suppression.
|
Male hormonal contraceptive regimens |
|---|
Overview of MHC regimens and efficacy
MHC regimens require the administration of testosterone or a derivative androgen to suppress the pituitary gonadotrophins, and in turn spermatogenesis, whilst ensuring broadly physiological androgen action so as to maintain virilization and avoid excess androgenic side effects. In addition, a second agent, such as a progestin or a GnRH antagonist, is now usually added to augment spermatogenic suppression. To date, regimens have employed long- and short-acting testosterone preparations given by a variety of routes (oral, transdermal, intramuscular or implants) or derivative androgens such as 7-methyl-19-nortestosterone (MENT) (Nieschlag et al., 2003) together with
- a progestin, including depot medroxyprogesterone acetate (MPA) (Brenner et al., 1977; Frick et al., 1977, 1982; Faundes et al., 1981; Knuth et al., 1989; Wu and Aitken, 1989; Pangkahila, 1991; Handelsman et al., 1996; McLachlan et al., 2002b; Turner et al., 2003; Gu et al., 2004), cyproterone acetate (Meriggiola et al., 1996, 1997, 1998, 2003), desogesterel (Wu et al., 1999; Anawalt et al., 2000; Kinniburgh et al., 2001, 2002; Anderson et al., 2002b), etonogestrel (Anderson et al., 2002a; Brady et al., 2004, 2006; Hay et al., 2005), levonorgestrel (Fogh et al., 1980; Bebb et al., 1996; Anawalt et al., 1999; Kamischke et al., 2000b; Pollanen et al., 2001; Gonzalo et al., 2002) and norethisterone (Kamischke et al., 2000a, 2001, 2002) or
- a GnRH antagonist such as acyline (Matthiesson et al., 2005a), cetrorelix (Behre et al., 1992) and teverolix (Erb et al., 2000).
In 1990 and 1996, two WHO-sponsored trials (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996) were published which established the relationship between sperm concentration and contraceptive efficacy (Table I). The design of these two multi-centre studies was similar in that men were given a regimen of testosterone enanthate (TE) 200 mg IM weekly for 6 months until they achieved either azoospermia (WHO Task Force on Methods for the Regulation of Male Fertility, 1990) or azoo-oligozoospermia (<3 x 106/ml) (WHO Task Force on Methods for the Regulation of Male Fertility, 1996). Following this suppression phase, couples entered into the efficacy phase during which TE treatment continued but all other contraception was stopped for a period of 12 months. Results from the first study showed a conception rate of 0.8 (95% CI, 0.02–4.5) per 100 person-years (WHO Task Force on Methods for the Regulation of Male Fertility, 1990), while azoospermic men, in the second study, achieved a conception rate of 0.0 (95% CI, 0.0–1.6) and oligozoospermic men 8.1 (95% CI 2.2–20.7) per 100 person-years. Overall results from the second study showed a conception rate of 1.4 (95% CI, 0.4–3.7) per 100 person-years; however, the wide CIs in the oligozoospermic group (0.1–3 x 106/ml) indicated a need for more efficacy data in this subgroup. These results were encouraging in that they showed conception rates comparable with typical first-year usage of the female OCP (three per 100 person-years) and superior to that of the only available male reversible alternative, condoms (12 per 100 person-years) (Trussell and Kost, 1987). However, the high testosterone dose resulted in frequent androgenic side effects and combined with the need for weekly administration resulted in a high discontinuation rate (29–42%). More perplexing was the small number of men in both studies (primarily those of non-Asian descent), whose sperm concentration remained above 3 x 106/ml for unclear reasons – the so-called ‘inadequate responders’ as discussed below.
|
In the two subsequent efficacy trials (Table I), longer acting testosterone preparations have been used to provide both physiological serum testosterone levels and more acceptable dosing regimens. A study conducted in China with efficacy phase entry criteria of sperm concentration (<3 x 106/ml) used a loading dose of IM testosterone undecanoate (TU) 1000 mg IM (given in teaseed oil) followed by a monthly 500-mg dose (Gu et al., 2003
) to give a conception rate of 2.3 (95% CI, 0.5–4.2) per 100 couple-years in azo-oligozoospermic men. Only 3% of men failed to achieve azo-oligozoospermia demonstrating the propensity of Asian men to achieve adequate spermatogenic suppression with testosterone-alone preparations. The second efficacy trial in Australian men used testosterone implants 800 mg 4 monthly together with depot MPA 300 IM 3 monthly and an entry criteria of severe oligozoospermia (<1 x 106/ml) (Turner et al., 2003). This study reported no pregnancies (95% CI 0–8.0/annum) in 35.5 person-years of exposure. This was the first efficacy trial of a long-acting androgen/progestin combination and featured an equivalent efficacy to the previous large testosterone-alone regimens, an acceptable dosing schedule and a 3.6% failure rate of spermatogenic suppression in a predominantly Caucasian population. However, it should be recognized that in all these studies the differentiation of azoo-oligozoospermia from extreme oligozoospermia is difficult because of the considerations of counting error and the volume of semen (undiluted or concentrated) actually inspected during evaluation. The detection limit for sperm concentration is about 11 000 per millilitre when derived from analysis of a centrifuged pellet obtained using WHO methods (T Cooper, personal communication), and thus, like other laboratory parameters, the term ‘undetectable’ is more applicable. More sensitive methods to detect extremely low sperm number will inevitably reduce the reported azoospermia rate, but from a contraceptive viewpoint this distinction appears unimportant.
It is apparent that there have only been a small number of contraceptive efficacy studies that must be viewed in the context of the difficulties associated with conducting such trials, primarily their high cost (especially relative to the limited public sector funds available), the demanding nature of the studies for participants and concerns about the consequences of contraceptive failure. There has been a much greater number of studies employing various MHC agent combinations using sperm concentration as a surrogate endpoint with a number of excellent recent reviews published (Anderson and Baird, 2002; Kamischke and Nieschlag, 2004; Grimes et al., 2005). Whilst it is tempting to draw conclusions about the relative efficacy of one agent over another, this must be done cautiously. Most of the reported MHC trials have involved small subject numbers of variable ethnicity, different treatment lengths and assorted MHC agents, dosages and combinations. That being said, it is still possible to make some general statements in regard to MHC treatment.
Androgen preparations: importance of dose, route of administration and duration of action
Testosterone delivery systems must provide an adequate and constant serum testosterone level to ensure virilization while, in conjunction with another anti-gonadotrophic agent, being able to suppress gonadotrophin levels in a sustained fashion as is essential in achieving contraceptive efficacy. Excessive testosterone administration needs to be avoided to minimize side effects and, given the possibility of testosterone back diffusion into the testis and subsequent AR activation, to avoid androgenic support for residual spermatogenesis (Meriggiola et al., 2002; McLachlan et al., 2004). While this latter effect is less clear than is seen in rodent models (Awoniyi et al., 1990), it does indicate the prudence of using minimum androgen.
To date, studies employing oral (Nieschlag et al., 1978) or transdermal (Buchter et al., 1999; Hair et al., 2001; Gonzalo et al., 2002) testosterone preparations illustrate that these routes provide inadequate testosterone delivery for this purpose. Alternatively, parental testosterone preparations deliver a sufficient androgen dose with long-acting testosterone preparations such as TU (Kamischke et al., 2000b, 2001; Gu et al., 2003, 2004; Meriggiola et al., 2003), testosterone implants (Handelsman et al., 1992, 1996) and MENT implants (von Eckardstein et al., 2003), providing equivalent or greater efficacy to short-acting IM TE (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996; Handelsman et al., 1992). Long-acting testosterone preparations also have the added benefit of greater patient acceptability.
Additional agents for suppression of gonadotrophins and spermatogenesis
The failure to induce universal azoospermia [70% of Caucasian men (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996)] over the usual 6-month suppression phase of most MHC studies with testosterone-alone preparations has led to the use of additional agents such as progestins (synthetic progesterone derivatives) and GnRH antagonists in an effort to maximize gonadotrophin and spermatogenic suppression. Importantly, these agents also allow for the use of physiological testosterone doses, thereby reducing androgenic side effects that were prominent in the initial WHO studies using TE.
Whether progesterone has a physiological role in normal men is unknown, as is whether additional effects beyond those of feedback on the HPT axis exist. Male progesterone levels (0.25–0.58 nmol/l) are similar to those of mean follicular phase female levels but much lower than those of luteal phase levels (Zumoff et al., 1990; Muneyyirci-Delale et al., 1999) that see reductions in LH pulse frequency and amplitude because of modulation in GnRH secretion (Filicori et al., 1986). Presumably, progestins also act at the hypothalamic level in men but may also act through progesterone receptors in the pituitary (Heikinheimo et al., 1995), although the presence of this receptor in humans is yet to be confirmed. Further evidence for possible hypothalamic and pituitary sites of progestin action is also provided by a recent study of normal men administered progesterone, resulting in reduced LH pulse amplitude and frequency together with a reduced LH response to GnRH administration (Brady et al., 2003).
Progestins can be administered via a variety of routes (oral, injection or implant) and have both short (hours) and long durations of action (months/years). Multiple studies using androgen/progestin combinations have been reported, which show that progestins (i) augment the induction of gonadotrophin and spermatogenic suppression (Bebb et al., 1996; Handelsman et al., 1996; Anawalt et al., 1999, 2000; Kamischke et al., 2001; Robertson et al., 2001; Gu et al., 2004), (ii) maintain suppression of spermatogenesis (Meriggiola et al., 2003, 2005; Gu et al., 2004) and (iii) allow for a lower dose of delivered androgen (Handelsman et al., 1992, 1996). Overall, these agents are well tolerated with relatively few side effects, the details of which will be discussed in a later section of this review.
GnRH agonist and antagonist analogues have been incorporated in MHC regimens, but thus far, only antagonists have shown acceptable suppression of sperm concentration. GnRH antagonists act at the level of the pituitary GnRH receptor, blocking the pulsatile action of the ten amino acid hypothalamic peptide GnRH. Unlike their agonist counterparts, antagonists do not result in a flare of LH and FSH secretion prior to suppression. The addition of GnRH agonists to testosterone treatment has not resulted in improved spermatogenic suppression to azoospermia (Rabin et al., 1984; Bhasin et al., 1985) and may even potentially blunt the spermatogenic response to androgens (Behre et al., 1997), as such their incorporation into MHC regimens has not been actively pursued.
There have only been a few studies that have reported effects on sperm concentration, the majority using daily administration of the nal-glu antagonist combined with IM TE resulting in an azoospermia rate between 67 and 93% between 6 and 16 weeks (Pavlou et al., 1991; Tom et al., 1992; Bagatell et al., 1993; Swerdloff et al., 1998). A more recent study reported that daily cetrorelix combined with 19-nortestosterone hexyloxyphenylpropionate achieved azoospermia in 100% of subjects by 12 weeks; however, this could not be maintained with androgen-alone administration (Behre et al., 2001). This rate of azoospermia induction is comparable with that achieved in trials using progestins. However, GnRH antagonists have the problem of needing daily subcutaneous administration and are expensive. Recently, a new long-acting GnRH antagonist (acyline) given at twice weekly intervals has shown promise in a small number of subjects producing an azoospermia rate of 67% at 8 weeks (Matthiesson et al., 2005a) and as such is worthy of further consideration. However, there are continued problems with local injection site reactions including erythema and induration (Herbst et al., 2002; Matthiesson et al., 2005a).
|
Issues in the transition of MHC from clinical research to market |
|---|
With the intention for the general use of MHC by young healthy men for extended periods, numerous issues require further consideration including the need for better definitions of efficacy and safety. In parallel, there are issues of practicality and acceptability that must be considered, given different societal backgrounds and patterns of health behaviours. For MHC delivery, an ideal system would maintain virilization and provide adequate spermatogenic suppression with relatively few side effects and would provide rapid, profound and universal spermatogenic suppression. Modern testosterone depot plus progestin regimens already approach these goals with 95% of men achieving probably adequate suppression, and often within 3 months. Existing MHC studies provide information on the kinetics of spermatogenic suppression and recovery, but because of their limited subject number and short duration do not accurately delineate the potential long term adverse (or perhaps beneficial) effects on cardiovascular, prostate and bone health. The deficiencies in the current data sets point to the need for vigilance in future MHC studies.
In combination with a progestin to ensure LH suppression, the subject depends upon exogenous testosterone delivery to ensure eugonadism. The maintenance of physiological testosterone levels is important to avoid androgen deprivation (fatigue, loss of muscle, low libido, mood changes, anaemia, adverse lipid profile changes and bone loss) or excess androgenic effects (acne, fluid retention, weight gain, raised haematocrit, mood changes, gynaecomastia, adverse lipid profile and perhaps prostate enlargement). The administration of oral, transdermal and low dose IM TE has resulted in significant decreases in testosterone levels during treatment, sometimes with symptomatic androgen deficiency and, importantly, may fail to provide the synergistic suppression of gonadotrophins needed for contraceptive action (Nieschlag et al., 1978; Meriggiola et al., 1997; Wu et al., 1999; Anawalt et al., 2000; Hair et al., 2001; Gonzalo et al., 2002). Conversely, high dose IM testosterone (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996) and higher dose testosterone implants have resulted in significant elevations of levels in some instances to well outside the normal range (Handelsman et al., 1992; McLachlan et al., 2002b).
Relatively low doses of testosterone appear able to maintain virilization and contraceptive effectiveness, provided attention is paid to continuous uninterrupted delivery (Anderson et al., 2002a). Long-acting testosterone preparations that can maintain physiological testosterone levels are available and ideal for this purpose. The current depot testosterone preparations, TU IM or testosterone implants, provide the best serum testosterone profiles and require administration every 8 (Meriggiola et al., 2003, 2005) or 16 weeks (Turner et al., 2003), respectively. Testosterone implants require some expertise to administer, and extrusions (10%), infection, bruising and haematoma (Handelsman et al., 1990) limit their widespread usage. On the contrary, TU IM is a promising approach undergoing current evaluation (Kamischke et al., 2000b, 2001; Gu et al., 2003, 2004; Meriggiola et al., 2005).
Cardiovascular and haematological effects
A major issue in chronic treatment of male populations is the potential for adverse cardiovascular effects. The administration of combined MHC regimens containing progestins has resulted in a 12–28% reduction in high-density lipoprotein (HDL) cholesterol levels (Bebb et al., 1996; Wu et al., 1999; Anawalt et al., 2000; Kamischke et al., 2001, 2002; Gu et al., 2003; Hay et al., 2005). HDL is thought to protect against atherosclerosis via antioxidant and anti-inflammatory properties and to promote the removal of cholesterol from atherosclerotic lesions. In prospective studies, low HDL has been linked to increased cardiovascular risk; however, subjects often displayed concomitant raised triglycerides and small low-density lipoproteins (Adult Treatment Panel, 2002). Thus, given that much of the data pertain to this triad of lipid abnormalities, it is not clear that the isolated lowering of HDL cholesterol with MHC treatment in healthy young men will be predictably deleterious. Also reassuring is the reduction in time to myocardial ischaemia seen in older men with chronic stable angina with short-term transdermal testosterone administration (English et al., 2000). However, as MHC trials to date have been of short-term duration and the pathogenesis of coronary disease is long term, this highlights the need for careful ongoing surveillance.
MHC treatment also affects the haemostatic system with administration of TE resulting in initial activation returning to baseline with continued treatment (Anderson et al., 1995) and TU alone resulting in anti-thrombotic effects but the addition of a progestin ameliorating this potential benefit (Zitzmann et al., 2002). More recently, medroxyprogesterone acetate (MPA) has been shown to increase cardiovascular hyperactivity in male monkeys probably via an increase in thromboxane prostanoid receptor expression in the vasculature with a proposed predisposition to cardiovascular hyperactivity-mediated myocardial ischaemia (Mishra et al., 2005). The significance of this finding for MHC users is unclear but cannot be disregarded, given the recently reported results of the HERS (Grady et al., 2002) and WHI trials (Rossouw et al., 2002) in menopausal women using MPA. Decreases in haemoglobin, haematocrit and red blood cells have been reported with cyproterone acetate administration, presumably because of its anti-androgenic action on the bone marrow (Meriggiola et al., 1996, 1998, 2003).
Reassuringly, there appears to be little prostatic effect in the short term with MHC administration. A number of trials using both testosterone alone and combined with a progestin or GnRH antagonist have shown no affect of treatment on either prostate specific antigen or prostate volume (Wallace et al., 1993b; Kamischke et al., 2000b, 2001, 2002; Behre et al., 2001; Meriggiola et al., 2005). While these results are encouraging, they must be interpreted cautiously given the small number of young, healthy study participants treated for relatively short periods of time. At present, the effect of MHC administration on intraprostatic steroidogenesis and in particular the relative ratios of testosterone metabolites, dihydrotestosterone (DHT) and E2 remain to be tested. Initial data from the Prostate Cancer Prevention Trial in which older men received the 5 reductase inhibitor, finasteride (5 mg daily) or placebo for 7 years showed that finasteride reduced the occurrence of prostate cancer but increased the risk (6.4 compared with 5.1%) of high-grade tumours (Gleason grades 7, 8, 9 or 10) (Thompson et al., 2003). However, a more recent analysis has raised the questions of excess risk assignment and overdetection bias in the finasteride-treated group (Klein et al., 2005). Taken together, these analyses indicate that a chemo-preventative role for 5 reductase inhibitors and their place in MHC regimens will need to be carefully considered.
Also worthy of further consideration, although not well characterized, are the potential adverse changes in mood with MHC treatment. Results from the two WHO efficacy trials using high-dose testosterone showed relatively little effect of treatment on mood and behaviour with only 2% of men discontinuing because of a change in libido, increased aggression or depression (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996). A less troubling effect noted in combination regimens using progestins has been that of increased sweating (Kamischke et al., 2001; Hay et al., 2005;). In women, it is recognized that the OCP generally stabilized mood across the menstrual cycle but that there are a small subset of women who experience negative effects on mood (Oinonen and Mazmanian, 2002). It is likely that with wider clinical application of MHC, there will be a small number of men who also experience deleterious mood effects.
MHC regimens using androgens alone or in combination with progestins have in general shown a small increase in body weight of less than 5% (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996; Kamischke et al., 2000b, 2001, 2002). However, there are relatively few studies looking at the effects of MHC or levels of androgen exposure on body compositional changes in healthy young men.
In a 20-week study of eugonadal men, administered TE (25–600 mg IM weekly), a dose-dependent increase in fat-free mass together with a negative correlation between serum testosterone and fat mass was found (Bhasin et al., 2001). Another study (Herbst et al., 2003) showed no change in abdominal fat mass and an increase in lean mass in men given 8 weeks of TE and LNG, whilst men given TE alone had a decrease in abdominal fat mass (at 8 weeks) and trended towards an increase in lean mass (at 4 weeks). Men given levonorgestrel (LNG) alone had an increase in abdominal fat mass and no change in lean mass. These results suggest that in the shorter term, LNG may negate the effect of testosterone on fat mass and be additive to its effect on lean mass. However, a second longer term study of men given testosterone implants and etonogestrel for 48 weeks saw no changes in body compositional data from baseline (Brady et al., 2004). Thus, whether there are significant changes in body composition and potential cardiovascular risk with long-term MHC treatment remains unclear, as does the impact of variable androgen exposure.
Kinetics of spermatogenic suppression and recovery
Given the kinetics of spermatogenesis and sperm transport, it is not surprising that the ‘on and off’ rates of MHC are measured for weeks and months. There is clear variability between individuals in regard to patterns of spermatogenic suppression and recovery. Defining these time frames is difficult, as studies have used different MHC regimens, thresholds of response (azoo-oligozoospermia versus oligozoospermia) and statistical analyses (median versus mean). It is recognized that the initial decline in sperm concentration (seen at around 15–18 days) is consistent with a rapid disruption in spermiation (Garrett et al., 2005). Most men then appear to take somewhere between 6 and 12 weeks to achieve sufficiently suppressed sperm concentrations (<1 x 106/ml) to afford likely contraceptive efficacy (Kamischke et al., 2000b; Kinniburgh et al., 2001; Anderson et al., 2002a; Brady et al., 2004; Ly et al., 2005; Meriggiola et al., 2005). However, even at this point, there remain some men with the potential to suppress given more time (i.e. beyond 6 months) (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996). This raises a key practical question whether and how frequently sperm concentration should be monitored during suppression. It must also be contrasted with female hormonal methods that are rapid (one month) and do not require confirmatory testing.
The recovery of baseline sperm concentrations is also slow with the majority of men needing 9–16 weeks to recover sperm concentrations to baseline or >20 x 106/ml (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996; Meriggiola et al., 1996; Anawalt et al., 2000; Anderson et al., 2002b; Turner et al., 2003; Brady et al., 2004; Hay et al., 2005; Ly et al., 2005). It is interesting to note that this problem of delayed recovery (up to 1 year) in some men is particularly seen with the use of TU, depot MPA and norethisterone enanthate (Turner et al., 2003; Meriggiola et al., 2005). This may in part relate to the relatively large volume of distribution of these long-acting agents such that the actual time of recovery phase onset cannot be well defined and may vary between men. Other possible reasons for delayed recovery include unrecognized spermatogenic disorders (despite a sperm concentration of greater than 20 x 106/ml), and the effects of prolonged suppression as supported by the observation that serum inhibin, a marker of seminiferous tubule function, continues to fall through the second 6 months of MHC treatment and shows minimal recovery by 4 months (Brady et al., 2004).
While there is currently no reason for concern, future studies will need to establish whether there is any reduction in fertility following MHC administration through the careful definition of recovery sperm concentration (baseline versus an arbitrary threshold such as >20 x 106/ml) and prolonged subject follow-up. Clinically, men interested in using MHC will need to be informed of the potential for a protracted delay in fertility return. However, it should also be remembered that pregnancies have occurred in the recovery phase of MHC trials despite low sperm densities, reflecting the relative fecundity of the populations involved and in keeping with the experience of fertility treatment in hypogonadotrophic men (Burris et al., 1988).
Despite the relatively slow ‘on and off’ rates, it is still likely that MHC will provide considerable advantages to some couples. It would seem that those men in a stable relationship, where quick changes in fertility status may be less important, would be ideal candidates for these products (e.g. post-natal). It must also be remembered that no contraceptive is perfect for all couples and that any new option will be welcomed given that currently available male options are so limited.
|
Inadequate spermatogenic suppression |
|---|
The non-uniform induction and maintenance of azoo-oligozoospermia or severe oligozoospermia is a significant issue in MHC development. This is particularly true of androgen-alone preparations that induce azoospermia in 70% of Caucasian men compared with near-universal response in Asian men (WHO Task Force on Methods for the Regulation of Male Fertility, 1990, 1996). Undoubtedly, the prospective identification of contraceptive response would be helpful in tailoring treatment for patients. However, previous work exploring baseline and treatment characteristics in MHC trial subjects with heterogenous spermatogenic suppression has only identified a small number of differences (Table II). In combined MHC studies, the ‘non suppression’ issue (about 3–5% of subjects) makes testing to establish the presence of severe oligozoospermia or azoospermia mandatory during the next phase of MHC development as, without this approach, the inclusion of such men will introduce a predictable and substantial background failure rate that is unlikely to be accepted.
|
Persistent spermatogenesis despite fully suppressed gonadotrophin levels
MHC treatment seeks to achieve maximal gonadotrophin suppression and thereby spermatogenic suppression. While the data broadly support this simple proposition, it is perplexing that many studies have found individuals who display ongoing spermatogenesis despite FSH and LH withdrawal and conversely others with spermatogenic suppression in the presence of detectable FSH and LH.
The comparison of treatment phase gonadotrophin levels in azoospermic and oligozoospermic men has provided conflicting results with a number of studies failing to show differences (Wallace et al., 1993a; Handelsman et al., 1995; Anderson and Wu, 1996; Amory et al., 2001) but two more recent reports (Anderson et al., 2002a; Meriggiola et al., 2002) finding higher treatment gonadotrophin levels in oligozoospermic men. A more consistent finding has been that of higher pre- and post-treatment gonadotrophin levels (Wallace et al., 1993a; Handelsman et al., 1995; Handelsman, 1995; Amory et al., 2001) in men becoming azoospermic with MHC treatment. While it is necessary to markedly suppress gonadotrophins to arrest spermatogenesis, once below this threshold, the relationship between gonadotrophins and spermatogenesis becomes less straightforward. As discussed above, this raises the concept of gonadotrophin-independent residual spermatogenesis in some men because of unexplained mechanisms and suggests that focusing on new and better ways to suppress gonadotrophins is unlikely to yield dividends in the ‘non-suppressed’ minority.
Intratesticular testosterone and spermatogenesis: human and animal models
No differences in baseline or treatment levels of serum or seminal plasma testosterone levels have been reported between men achieving azoospermia or oligozoospermia (Wallace et al., 1993a; Handelsman et al., 1995; Anderson and Wu, 1996; Anderson et al., 1997; Amory et al., 2001). Intuitively, the fall in iT androgen action consequent on LH suppression would seem key to MHC action, but the iT testosterone level is not readily assessed. A high level of testosterone in the order of 1000–2000 nmol/l is maintained in human testis under physiological conditions, some 40- to 100-fold higher than circulating serum levels (Morse et al., 1973; Huhtaniemi et al., 1987; McLachlan et al., 2002b; Coviello et al., 2004, 2005; Matthiesson et al., 2005b). This same high testicular to serum gradient also exists for the androgenic metabolites of testosterone [DHT 26 versus 4.7 nmol/l, 3-adiol (81 versus 8.1 nmol/l) and 3-adiol (158 versus 4.8 nmol/l)] and for E2 (11000 versus 27 pmol/l) (Matthiesson et al., 2005b). The administration of MHC results in a marked suppression of urinary epitestosterone (10% baseline) (Brady et al., 2004) and iT testosterone levels (2% baseline) (Morse et al., 1973; McLachlan et al., 2002b; Coviello et al., 2004; Matthiesson et al., 2005b) with the androgenic metabolites (DHT, 3- and 3ß-adiol) and E2 falling to a much lesser extent (Matthiesson et al., 2005b).
Previous work in rodents and monkeys has examined the relationship between iT steroids and spermatogenesis. Given testosterone or DHT implants, the hpg mouse shows completion of meiosis and initiation of qualitatively normal spermatogenesis (Singh et al., 1995). Following testosterone implant removal, elongated spermatids are lost by 3 weeks and testicular weight regresses (Handelsman et al., 1999). Interestingly, subsequent replacement of the implants shows that a lower dose of testosterone is required for restoration of spermatogenesis than that required for initiation, suggesting different androgen-dependent genetic switches for these two processes (Handelsman et al., 1999). In rats, the administration of testosterone/E2 implants results in selective LH/iT testosterone withdrawal with relative preservation of FSH. Examination of the testis in this model shows a marked reduction in the progression of round through to elongated spermatids, reflecting a spermiogenic lesion (Awoniyi et al., 1990; McLachlan et al., 1994). Spermiogenesis is returned to normal levels by high-dose testosterone that results in supraphysiological serum levels but iT testosterone levels 10–40% of normal (Awoniyi et al., 1989, 1990; McLachlan et al., 1994; O’Donnell et al., 1994), thus reflecting the androgen dependence of the later stages of spermatogenesis to iT testosterone restoration and the fact that high testicular levels of testosterone are likely to be a consequence of local production, rather than a necessity for normal spermatogenesis.
In contrast, monkeys administered exogenous testosterone show marked reductions in gonadotrophin levels, yet a less marked reduction in iT testosterone levels (15% baseline) (Narula et al., 2002). Stereological evaluation of germ cell development shows a general fall in all germ cell subtypes but predominantly two lesions: (i) maturation of type A to B spermatogonia and (ii) spermiation (the process of sperm release from the seminiferous epithelium) (O’Donnell et al., 2001a). This is very similar to the pattern seen in men given an MHC regimen of androgen alone or in combination with a progestin but in whom profound falls in iT testosterone (2% baseline) are seen (Zhengwei et al., 1998; McLachlan et al., 2002b; Matthiesson et al., 2005b).
Interestingly, these human studi