Si-Tian Liu¹, You-Lun Gui², Cui-Hong Lin², Chang-Hai He²

Despite varies endeavors devoted to the development of reversible male contraceptives, no generally accepted regimen has been widely used to date [1-3]. In studies on the effect of androgens either alone or in combination with progestogen or GnRH antagonists [4-8], a consistent finding is that azoospermia is achieved in only 40 %-70 % of Caucasian and a little more than 90 % of Asian men. Another striking finding from these studies is the marked individual differences within and between population groups in the extent of spermatogenetic arrest. The basis for the incomplete suppression of spermatogenesis is unknown. This is the key issue influencing the clinical efficacy, general acceptance and wider applicability of hormonal male contraceptives.

In Caucasian studies, there were no significant differences in physical or hormonal characteristics, pharmacokinetics and pharmacodynamics between the responders and non-responders [9, 10]. However, these observations have not been carried out in Chinese men. We, therefore, retrospectively compared the hormonal patterns in volunteers participating in a contraceptive trial with levonorgestrel (LNG) implants and testosterone undecanoate (TU) injectable at an aim to identify possible factors contributing to the differences in spermatogenic suppression between responders and non-responders in Chinese men.

Each LNG implant contained 75 mg LNG (Da Hua Pharmaceutical Company, Shanghai, China). TU was suspended in tea-seed oil at a concentration of 250 mg/2 mL (Xian Ju Pharmaceutical Company, Zhejiang, China). The same batches of TU and LNG were used throughout the study.

The study consisted of a 2-week control period, an 18-week treatment period and a recovery period of 4-8 weeks. At the beginning of the treatment period, two LNG rods were implanted under the skin of the upper arm. Three weeks after LNG implantation, each subject received TU at a dose of 250 mg i.m. every 4 weeks for 12 weeks. At the end of week 18, the implants were removed. The subjects were followed until the serum T, FSH and LH recovered to pretreatment levels and sperm counts returned to >20 million/mL on two consecutive occasions.

All variables were checked for normal distribution in the Kolmogorov-Smirnov one-sample test for the goodness of fit. Variations between study groups were evaluated by two-way ANOVA for repeated measurements. Variations over time within the study were evaluated by one-way ANOVA for repeated measurements. In the case of an overall P<0.05 in the ANOVA, differences between baseline and the following time points were tested by Tukey's post-hoc test. In the case of a single missing value per time point, the appropriate mean was inserted to allow ANOVA for repeated measurements. When necessary, analysis was performed on logarithmically transformed data. Data were presented as arithmetic meanSEM. Baseline values were computed as the mean of the two visits prior to treatment. For all analyses a two-sided P value of 0.05 was considered significant. Computations were performed using the statistical software package SPSS/PC+ version 10.0.

Table 1. Baseline anthropometrical data and testis volume (meanSEM). a Body mass index, b body surface area.

Figure 1. Sperm density profiles in subjects treated with LNG implant plus TU injection. ^bP<0.05, Group R vs Group N.

Figure 2. Sperm motility profiles in subjects treated with LNG implant plus TU injection. ^bP<0.05 Group R vs Group N.

Figure 3. Sperm vitality profiles in subjects treated with LNG implant plus TU injection. ^bP<0.05, Group R vs Group N.

The serum testosterone levels remained within the normal range in all subjects throughout the study, but they were significantly suppressed after LNG implantation and lower concentrations continued for two weeks. Following TU injection, they reached the peak value at week 1 and returned to the normal range within two weeks after removal of implants. Serum testosterone levels increased and declined at faster rates in Group R than that in Group N following TU injections (P<0.05). However, there were no significant differences in serum testosterone levels at each time point between Groups R and N throughout the study (P>0.05) (Figure 4).

There were no pretreatment differences in serum FSH and LH between the two groups (P>0.05). Both the FSH and LH levels were suppressed to lower concentrations following LNG implantation and TU injections and recovered gradually to the baseline during the post-treatment phase (Figures 5 and 6). At week 14-18, suppression of serum FSH level was more pronounced in Group R than that in Group N (P<0.05) (Figures 5). At week 4, 8, 9 and 12-15, the serum LH concentrations were significantly higher in Group R than those in Group N (P<0.05) (Figure 6).

The release of LNG reached the peak at week 1 after LNG implantation and then gradually decreased and remained at a steady value at week 8 with a mean steady value of 0.24 ng/mL (Figure 7). Serum LNG level was reduced to below the limits of detection after removal of implants. Serum LNG levels in Group R were significantly higher at several points compared with those of Group N (Figure 7).

There were no pretreatment differences in serum T/LH ratio between the two groups (P>0.05). The changes in serum T/LH ratio were similar to those of serum T during the treatment period. At week 3, 5, 7, 16 and 21, the serum T/LH was significantly lower in Group R than that in Group N (P<0.05).

There are positive correlations between serum T and LH (P<0.01, r=0.358), between serum FSH and LH (P<0.01, r=0.436) and between serum FSH and sperm density (P<0.01, r=0.235), sperm motility (P<0.01, r=0.326) and sperm vitality (P<0.01, r=0.219). Sperm counts were correlated positively with sperm motility (P<0.01, r=0.562) and sperm vitality (P<0.01, r=0.373). There were no significant correlation between serum LNG and any other serum hormones or seminal fluid parameters.

Testosterone enanthate (TE) either alone or with progestogen or GnRH antagonist have been considered as the most promising regimen for reversible and effective male contraception [12, 13]; in these two multi-centre trails, azoospermia occurred in approximately 70 % of Caucasian and over 90 % of Asian men. The explanation for the non-uniform induction of azoospermia remains unclear. It has been suggested that low androgen levels in the testis may still support spermatogenesis and the latter could be completely suppressed only by eliminating the residual intratestiscular testosterone [14]. Some authors recently reported that there was a higher 5a-reductase activity in men who failed to achieve azoospermia after weekly injection of TE [15]. It was also suggested that Chinese might have less 5a-reductase activity than Caucasian men [13], but there is no direct evidence supporting this hypothesis.

We observed that in Group R, the sperm motility and vitality decreased significantly when sperm density was reduced during treatment, while in Group N they remained practically unchanged. This is presumably responsible for the higher pregnancy rate in non-responders. No significant differences were observed in the rate of fall of sperm production between the 2 Groups, while in Caucasians treated with TE, a faster rate of fall in sperm production has been found in responders than in non-responders [9]. The possible explanation for the discrepancy might be ethnic differences and/or different regimens used in different studies. This suggests that the rate of spermatogenic suppression may not be dependent upon hormonal status alone but also associated with intrinsic variations in spermatogenic kinetics.

The T levels remained within the normal range throughout the study, which may probably explain the fact that none of the 16 subjects complained of libido decline and sexual disturbances. After LNG implantation, serum T levels were higher in Group N than in Group R, however, following TU injection T increased and declined at faster rates in Group R which might be related to some of the following factors: (1) responders may have lower concentrations of sex hormone binding globulin [9, 10]; (2) lower values of serum T/LH ratio in Group R may indicate that Leydig cell suppression was significantly more pronounced in this group. The extent of endogenous testosterone and Leydig cell inhibition may be the basis for the heterogeneity in spermatogenic suppression. Similar studies have suggested that such differences might result from differences in peripheral androgen metabolism and spermatogenetic potential [15, 16].

The LNG profiles were similar in the two groups during the treatment phase, although the LNG levels were significantly higher in Group R at weeks 4 and 16. LNG levels might contribute to the individual differences in spermatogenic suppression. Studies in women treated with the same LNG implant [17, 18] showed that LH and FSH could not be completely inhibited by the LNG released and ovulation suppression occurred only in some women. In the present study, FSH and LH levels were suppressed after LNG implantation and were reduced to lower concentrations following TU injection, indicating that LNG alone could not inhibit gonadotropins completely but required synergetic action of TU.

The FSH levels were significantly higher and the LH levels slightly lower in Group N than in Group R. This differed from the results in Caucasian and primate studies in which no significant differences in serum LH were observed between responders and non-responders [19,20]. There are significant individual differences in spermatogenic suppression upon administration of exogenous testosterone; the extent of FSH inhibition might be one of the main factors responsible for the pronounced difference in spermatogenic suppression in Chinese men. In accordance with this, there was a positive correlation between FSH and sperm density, motility and vitality.

In conclusion, the blood LNG and T levels, the degree of FSH inhibition, the sensitivity of the pituitary-testis axis to exogenous steroids, the individual spermatogenetic potential and the functional status of Leydig cells may occasion the individual difference in spermatogenic suppression in Chinese men treated with hormonal male contraceptives; these are likely to be the main factors influencing the efficacy of the hormonal approach.

The work was based on the data of on sub project supported by the Contraceptive Research and Development (CONRAD) Program, Eastern Virginia Medical School, U.S.A.