1976. First trial. Oral progestin and testosterone implants

In the years 1971–1980, encouraged no doubt by the success of female hormonal contraception, several American and Scandinavian teams initiated clinical protocols for male hormonal contraception using steroids (androgens, progestins) [3]. France was not absent from this trend. In 1976, Salat-Baroux and his team [4] carried out the first French trial of male hormonal contraception by combining an oral progestin (R 2323) with testosterone implants. In terms of efficacy, the results were interesting as azoospermia was achieved in 2 to 3 months. The experiment could not be continued because of the development of sexual disturbances (loss of libido, impotence), gynecomastia and weight gain.

Testosterone implants at a dose of 300 mg were insufficient to maintain plasma testosterone at eugonadal levels. Further studies indicated that achievement of eugonadal levels required 400 to 800 mg testosterone implants in combination with progestins administered either orally (desogestrel) [5–7], or as implants (etonorgestrel) [8] or injections (DMPA) [9, 10].

Development of a contraceptive treatment using percutaneous testosterone

A pilot study in six volunteers demonstrated that, in these conditions, the sperm count reached very low values (−90% at 3 months), that luteinizing hormone (LH) and follicle stimulating hormone (FSH) were equally inhibited and that plasma testosterone remained within the normal range [18, 19]. For the first time, satisfactory inhibition of spermatogenesis was achieved without elevation of plasma testosterone and without the injection of high doses of steroids.

In order to better define the effect of the treatment, other subjects were treated with PT alone at the successive doses of 125 mg testosterone for 3 months followed by 250 mg for the next 3 months: although plasma testosterone increased by 30 to 100%, sperm production did not markedly change [20].

Later, the kinetics of inhibition of spermatogenesis, the hormonal profile and the side-effects of the treatment were determined in 35 men and its contraceptive efficacy in 25 couples [21, 22]. Spermatogenesis inhibition was accurately measured: sperm concentration was decreased by 47% at 1 month, by 90% at 2 months and by 98 to 100% at 3 months. At 3 months, 80% of men had a sperm concentration of 1 million/mL (M/mL) or less, which is the accepted threshold of contraceptive efficacy [23]; 19% of men already had a sperm concentration <1 M/mL at 1 month and 39% at 2 months. When treatment was discontinued, spermatogenesis rapidly recovered (73 ± 29.5 days) and two couples who wished to have a child had no difficulty in conceiving.

Above all, during treatment, plasma testosterone remained at a physiological level and was maintained throughout the day. Estradiol level was not increased. FSH and LH were rapidly inhibited. Contraceptive efficacy at a sperm count threshold <1 M/mL corroborated the results obtained in the WHO trials (cf. section 4.): 25 couples used this contraceptive method exclusively for 211 months. One pregnancy occurred, due to the fact that the man had discontinued treatment without informing his partner [22].

The combination of MPA-PT was better tolerated than the testosterone enanthate (TE) injections that were used in the WHO trials: it is significant that not a single man stopped treatment for the reasons described in the WHO trials (cf. section 4.) No laboratory parameters were modified, except for a transient moderate increase in hematocrit. However, it was observed that cutaneous application of an alcohol-based testosterone preparation could result in transfer to the partner, and two couples discontinued treatment for this reason. This adverse effect had already been reported elsewhere [24, 25]. It therefore seemed indispensable to clearly define the rules of administration and/or to develop new pharmaceutical forms.

In parallel, Le Lannou’s team [26], disappointed by the variable efficacy of MPA in the first three men treated, used the same progestin as the team of Guérin and Rollet, norethisterone, at a dose of 5 mg/day. Eight of 12 subjects were azoospermic after 6 months treatment.

The third team, Bouchard and Garcia [27], tested the efficacy of long-acting LHRH agonist in ten volunteers; five men received in addition one low monthly dose of TE (125 mg by intramuscular (IM) injection) and the remaining five received a more physiological dose of testosterone (120 mg/day oral TU). The treatment was ineffective as soon as androgen replacement was sufficient: in the first group, 4 of 5 men became azoospermic but spermatogenesis returned as soon as injected testosterone was increased. In the second group, the treatment was ineffective.

International impact of percutaneous contraception

Following the French studies, several teams from other countries sought to use the percutaneous approach as a means of contraception.

Two studies used a single testosterone patch [29, 30] in combination with oral levonorgestrel (250 then 500 microgr/day) or oral desogestrel (300 microgr/day). The doses of progestin administered were higher than those used in female contraception. In both cases, spermatogenesis was not sufficiently inhibited to ensure effective contraception: in addition, plasma testosterone was unacceptably reduced (−40%).

For this reason, Wang’s team in 2002 [31] increased the dose of testosterone by using two patches, but they prescribed oral levonorgestrel at a lower dose (125 microg/day), similar to that of female contraceptive pills. Inhibition of spermatogenesis was improved, but it was still insufficient: after 3 months treatment, only 15% of subjects had a sperm concentration <1 M/mL. This time, doubling the dose of testosterone maintained plasma testosterone within a physiological range.

Page and colleagues used the same treatment principle (MPA-PT) that had been tested in France, but MPA (depomedroxyprogesterone acetate, DMPA) was given as one injection every 3 months and combined with 100 mg PT/day. They obtained good inhibition of spermatogenesis in 75% of subjects, and sperm concentration was <1 M/ml at 3 months. During treatment, plasma testosterone was increased [32]. Fifty percent of the men who took part in the trial were satisfied with this method and were prepared to use it with their partner [33]. This study also had the merit of showing that use of GnRH antagonists, presented as the male hormonal contraceptive method of the future [34], was not more active than the combination of MPA-PT.

More recently, Wang’s team proposed an “all-in-one” formulation with testosterone and progestin combined in the same gel [35]. The testosterone gel was the same as that used by the French teams. It was combined with nestorone, a new-generation progestin with original properties: it does not bind to the estradiol receptor and its binding affinity with the androgen receptor is 600 times less than that of testosterone, while that of levonorgestrel is 40 to 70% that of testosterone.

Using this combination, 85% of men reached the threshold of contraceptive efficacy at 3 months, with plasma testosterone in the physiological range [35]. These results appeared sufficiently convincing for clinical trials to be launched in the USA in view of commercializing the gel.

Studies dealing with the combination of oral MPA and PT are no exception to the rule according to which some men do not sufficiently respond to hormonal treatments. Among 30 men examined 1, 2 and 3 months after the beginning of treatment (using the threshold value of contraceptive efficacy as < 1 M sperm/mL at 3 months), five men were poor responders while the good responders could be divided into 3 types: rapid (n = 4), intermediate (n = 11) and slow (n = 10) according to whether they achieved less than 1 M sperm/mL at month 1, 2 or 3, respectively (Fig. 1) [19, 22].

The azoospermia observed as soon as the first month of treatment strengthens the observations showing that DMPA-PT treatment is able to have a striking effect on spermiation [46]. Besides, the persistence of spermatogenesis in poor responders has been explained by an increased sensitivity of spermatogenesis to high testosterone levels induced by androgen injections [47]. This was not the case for the combination of oral MPA and PT, as no supraphysiological elevation of blood testosterone was induced by this regimen [19, 22]. On the other hand, it may be supposed that the biological availability of oral MPA, which varies greatly from one individual to another [48], could explain the differences observed in response to the oral MPA and PT treatment.

It had been shown that DHT (125 mg daily) administered alone reduced blood testosterone from 5.0 to 2.9 ng/mL [49]. The same team had demonstrated that the anti-gonadotropic effect of progestin (even when testosterone-derived) did not involve the androgen receptor but the progesterone receptor, whose expression depends on estradiol originating from testosterone aromatization [50].

These results suggest that the treatment failures responsible for low blood testosterone levels (testosterone patches or DHT gel) or using DHT (which cannot be aromatized) may be explained by a blood testosterone level that is not sufficient to promote an anti-gonadotropic effect of progestin.

1986-90 and 1990-94: two multicenter WHO studies. Contraceptive effectiveness of hormonal treatments

In 1986, the WHO (Task Force on Methods for the Regulation of Male Fertility) undertook two international studies aimed at determining the contraceptive efficacy of an androgen, testosterone enanthate (TE), by IM injection once a week for 18 months.

In seven countries, including France (Soufir, CHU Bicêtre-Université Paris Sud), 271 men with normal semen analysis and in a stable relationship with a partner not suspected of infertility were treated according to this protocol. One hundred fifty-seven men became azoospermic and used this contraceptive method exclusively in their couple. During 1486 months of exposure, only one pregnancy occurred, a Pearl index of 0.8, which is similar to that of female contraceptive pills [51]. Treatment with TE did not completely suppress spermatogenesis in 35% of these men: the majority presented with oligozoospermia below 5 M sperm/mL.

This raised a question: what is the lowest sperm concentration required for men to be fertile? A second multicenter study, which began in 1990, established that 3 M sperm/mL appeared to be an acceptable threshold of efficacy [52]. These two protocols also established that less than 2/3 of Europeans have less than 1 M sperm/mL when treated with androgens alone, and that east Asian men are better responders (up to 90%) to these treatments. Efficacy was greater in both Europeans and Asians when androgens were combined with progestins. Spermatogenesis inhibition occurred earlier in Europeans [53, 54].

Various explanations of such ethnic differences have been put forward. Chinese men may have fewer germ cells per Sertoli cell, a higher apoptotic index of germ cells, lower testosterone production with lower plasma testosterone levels, reduced 5 alpha-reductase activity, and LH levels more easily inhibited by testosterone [54].

These results are however not always homogeneous. In a study comparing Europeans from Edinburgh with Chinese men from Shanghai treated with 150 micrograms of desogestrel and a subcutaneous pellet of 400 mg testosterone, treatment seemed more effective in Europeans [6]; a group of Chinese men living in Yunnan were poorer responders to TU probably due to absorption of a local medicinal drink [44]; testosterone or 5 alpha-reductase activity levels did not differ between American Chinese men and American men of Caucasian origin [55].

We may therefore ask whether diet [44] or environment may explain the differences observed between east Asian and European men. The question arises through observations regarding other pharmacological compounds that seemed more active in Chinese than in Caucasian subjects at similar doses [56].

Contraception using the combination MPA-T in the rat: testicular changes and quality of the descendants after contraception

For better understanding of the effects of the combination MPA-T, Soufir’s team developed an animal model. This treatment administered for 55 days (duration of a spermatogenesis cycle) to adult Sprague-Dawley rats induced a massive decrease in intratesticular testosterone and a particular type of spermatogenesis suppression: the spermatogonia divided normally, but spermatocytes and above all round spermatids decreased by half, while elongated spermatids totally disappeared. This demonstrated that meiosis and above all spermiogenesis are the phases of spermatogenesis that are most sensitive to androgen deficiency.

Seventy days after this treatment, the rats’ fertility returned to normal: litter size was not reduced. There were no fetal resorptions indicative of chromosomal aberrations. The new-borns had no malformations: follow-up of their development in collaboration with Auroux and colleagues showed that behavior did not differ from those of untreated controls [57].

Combination of MPA and testosterone: protection of spermatogenesis against cytotoxic agents

Treatment with MPA-T had an unforeseen effect: protection of spermatogenesis against major cytotoxic effects (anticancer drugs, high-dose radiation). The teams of Jégou and Soufir demonstrated this under well-defined conditions (prolonged treatment).

Prior treatment of rats with MPA-T for 55 days protected spermatogenesis against procarbazine-induced damage. This protective effect concerned both the quantity of spermatozoa produced and their genome.

Treatment of the male rats with MPA-T (55 days) before administration of cyclophosphamide prevents the appearance of these behavioral disturbances in the offspring.

Thermal contraception: history and principle

The discovery of the thermal dependence of spermatogenesis in man dates from 1941 [79]. It was confirmed by experimental studies carried out between 1959 [80] and 1968 [81]. Some authors were already suggesting it might be possible to use an increase in scrotal temperature as a male contraceptive method [80–82]. The contraceptive effect of heat in man was in fact only reported 20 years later by Shafik in 1991 [83].

The increase in temperature was either whole-body heating (steam room at 43 °C, sauna at 77–90 °C) [79, 84, 85], or a high-intensity increase in scrotal temperature (38 to 46 °C) for a short period [80, 82, 86–89], or a low-intensity increase (~1 °C) in scrotal temperature throughout the day [90, 91].

The degree of inhibition depended on the level of temperature increase and on its duration. The smaller the range of temperature increase, the longer the daily duration of exposure needed to obtain the same inhibiting effect. Spermatogenesis returned to normal at cessation of temperature elevation.

Development of an original technique for elevation of scrotal temperature

Based on these findings, the aim was to develop a practical technique for application of this method that did not interfere with the users’ daily life.

In parallel, two reassuring experimental studies on the reversibility of this method were published. In the first study, surgically induced cryptorchidism in the adult dog led to alteration of spermatogenesis that was reversible after return of the testicles in the scrotum [99], while in the second study, local cooling of a naturally cryptorchid testicle in pigs initiated and maintained spermatogenesis leading to complete differentiation in numerous seminiferous tubules [100].

Effects of the technique on sperm production and maturation. Successive adaptations

The testicles were maintained in the required position during waking hours, or 15 h/day, for periods of 6 to 49 months.

Mechanisms of the effects induced by elevation in testis temperature

Molecular mechanisms of testicular heat stress induced by different types of external or internal factors have been reviewed in several recent publications (see for example [103–105]).

Induced elevation of testis temperature for contraceptive purposes is aimed at healthy men in their reproductive life. As shown in Tables 2, 3 and 4, the testis temperature reached ranges from supraphysiological to physiological values. Two of the main advantages of using testis temperature as a male contraceptive are that spermatogenesis can be recovered and fertility is preserved; until now, only physiological increases in testis temperature met such criteria, as spermatogenesis and fertility both recovered after 6 to 24 months of 15 to 24 h/day exposure to +2 °C elevation [83, 106].

In a 15 h/day induced increase (2 °C) in testis temperature, the temperature reached is still within the physiological range. This was not sufficient for most men to achieve azoospermia. Despite the high rate of heat-induced apoptosis [107, 108] some cells – the most heat-vulnerable germ cells, i.e. early primary spermatocytes and early round spermatids in humans [109] – did develop into mature sperm containing damaged DNA, as observed in the inhibitory and recovery phases in 5 healthy volunteers [96]. In this last study of a 15 h/day 2 °C increase in testis temperature for 120 days, on the basis of the literature and of their own results the authors suggest that at the spermatocyte stage some cells underwent apoptosis, some appeared as round cells in the semen, a few continued to develop into sperm and others became arrested in a ‘frozen state’ [96]. As spermatogonia continued dividing and differentiating at the testis temperature reached (only scrotal temperature higher than 42 °C affected mitotic proliferation and the number of spermatogonia [109]) several waves accumulated as late spermatogonia B and spermatocytes in the ‘frozen state’. Finally, when heating was stopped, all arrested germ cells restarted their evolutionary process together, giving a sperm output which began to improve as soon as day 33 after cessation of heating [96]. This could explain why total sperm count values reported after cessation of heating were higher than initial values, whatever the method used to elevate testis temperature, as indicated in the last column of Tables 2, 3 and 4.

Contraceptive efficacy

Nine volunteer couples evaluated the contraceptive method developed by Mieusset and colleagues [106]. Three men used the first technique and six the second. The partners of these men discontinued all contraceptive methods after a motile sperm concentration (MSC) of less than 1 M/mL was observed in two successive semen analyses carried out at an interval of 3 weeks. Throughout the duration of the contraceptive period with the first technique, the mean MSC was 1.87 M/mL (range 0 to 7.4) with an MSC below 1 M/mL observed in 41% of sperm analyses performed. Throughout the duration of the contraceptive period with the second technique, azoospermia was observed in 11% of semen analyses and an MSC below 1 M/mL in 86% of analyses.

No pregnancy occurred, except in a single case due to incorrect use of the technique. When temperature increase was discontinued, the MSC returned to the initial values with both techniques [106].

These data obtained between 1985 and 1989 were only published in 1994 [106], after recovery of fertility had been attested. They confirmed the findings of the first study of contraceptive efficacy in men using the method of testicular heating reported by Shafik in 1991 [83]. This researcher used either surgical fixation of the testicles high in the scrotum in 15 men, or the wearing of a cotton sling including two balls that pushed the testes close to the abdomen in 13 other men [83]. In a second study, a polyester sling was used to induce scrotal hyperthermia in 14 men [94]. In both the studies by Shafik, no pregnancies were observed in the 42 couples included in the contraceptive period [83, 94].

In summary, current data from studies evaluating the effect of a moderate increase (1.5 to 2 °C) in testicular temperature induced in men at least during waking hours showed sufficient decrease in the number of sperm and adequate inhibition of their motility to reach the contraceptive threshold. Once this threshold was achieved, contraceptive efficacy was satisfactory in the 50 couples followed over 537 cycles, with occurrence of a single pregnancy due to incorrect use of the technique [83, 94, 106].

Other criteria

What of the other criteria that any contraceptive method must meet: acceptability, reversibility and safety?

Use of the underwear developed by Mieusset and colleagues [92, 93, 96, 106] did not cause any of these complications. Recently, this team has shown that sperm nuclear quality was altered during the inhibition phase of spermatogenesis, but that this was reversible 3 months after cessation of hyperthermia [96]. This finding needs to be taken into account when using contraception, during the inhibition phase and for 3 months after discontinuation.