An Ideal Hormonal Contraceptive at Lactation?

Human Reproduction vol.14 no.8 pp.1993–1997, 1999 Parenteral administration of progestin Nestorone® to lactating cynomolgus monkeys: an ideal hormonal contraceptive at lactation?

O.Heikinheimo1,2,6, S.Ranta1, A.Moo-Young3, Introduction 1 4,5 P.La¨hteenma¨ki and K.Gordon In the worldwide scheme of family planning, extension of the 1Steroid Research Laboratory, Institute of Biomedicine, interval between pregnancies would result in major improve- 2Department of Obstetrics and Gynecology, University of Helsinki, ment in maternal and child health (Thapa et al., 1988). Finland, 3The Population Council, Center for Biomedical Research, Lactation itself provides a reasonable contraceptive effect 4 New York, The Jones Institute for Reproductive Medicine, Eastern initially. However, the duration of lactational amenorrhoea Virginia Medical School, Norfolk, and 5Balance Pharmaceuticals Inc., Santa Monica, USA varies greatly (WHO, 1998), and even in fully nursing women ovulatory cycles are resumed in 26% at 6 months post-partum 6To whom correspondence should be addressed at: Steroid Research ´ Laboratory, Institute of Biomedicine, PO Box 8, FIN-00014, (Dıaz et al., 1988). Pregnancy and childbirth require women University of Helsinki, Finland to seek medical attention, thus providing an opportunity for contraceptive counselling and introduction. Thus, safe and ® Nestorone (NES) progestin is highly effective for contra- effective contraceptives which can be initiated during the post- ception following parenteral administration, but ineffective partum period and lactation are important assets in fertility after oral ingestion due to rapid first-pass metabolism. control. Thus, NES might be ideal for lactational contraception; Even though lactational contraception with progestin-only possible NES in milk should be metabolized by the nursing contraceptives has been endorsed by international family infant. We evaluated the distribution of NES, its endocrine planning organizations, hormonal methods are often viewed effects and infant weight gain in five cynomolgus monkeys cautiously during lactation. All the contraceptive progestins and their nursing infants. Nestorone® implants, releasing studied to date can be detected in the milk (Dı´az and Croxatto, ~40 g NES/day in vitro, were placed s.c. in the mothers µ 1993), thus concerns about the possible ill-effects on the 3–4 months following delivery, where they remained in situ suckling infant remain. for 4 weeks. Sampling (blood daily from the mother; milk An ideal molecule for lactational hormonal contraceptive and blood from the infant at 3 day intervals) was initiated would be one with high contraceptive efficacy in the mother, at 2 weeks prior to insertion, and continued for 2 weeks yet it would be without effects on lactation or the nursing following removal of the implant. NES, oestradiol, proges- infant. Nestorone® (NES, previously known as ST-1435) is terone and prolactin were measured by radioimmunoassays an orally inactive progestin, currently in phase II clinical and the infants were weighed weekly. The (mean ⍨ SD) contraceptive trials using implants, vaginal rings, and maternal serum and milk concentrations of NES were 337 intracervical or transdermal administration (Kurunma¨ki et al., ⍨ 90 and 586 ⍨ 301 pmol/l during the use of the implants. 1984; Laurikka-Routti et al., 1990; Haukkamaa et al., The ratio of milk/serum NES was 1.68 ⍨ 0.12 (mean 1991). NES has high binding affinity to the human ⍨ SE), and the serum and milk concentrations were progesterone receptor (La¨hteenma¨ki, 1986), yet due to rapid significantly correlated (r ϭ 0.75, P < 0.001). NES was first-pass metabolism, oral administration of NES is ineffective not detectable (<13 pmol/l) in any infant serum samples. Concentrations of prolactin (mean ⍨ SD) were 41.1 ⍨ 32, (Coutinho et al., 1981; Heikinheimo et al., 1994). However, 26.7 ⍨ 7.6 and 26.3 ⍨ 9.5 ng/ml before, during and after parenterally-administered NES is highly effective for contra- the use of the implants respectively. The (mean ⍨ SE) ception, and ovulation is inhibited with very low serum infant weight increased from 643 ⍨ 54 g 1 week prior to concentrations of NES (Coutinho et al., 1981; La¨hteenma¨ki insertion to 713 ⍨ 54 g 1 week following removal. These et al., 1982). data confirm that NES in milk is rapidly metabolized by NES, therefore, has several key features which may make the suckling infant. Therefore, NES appears to be an ideal it an optimal hormonal contraceptive to be used during hormonal contraceptive for use during lactation. lactation. While parenteral administration of NES to nursing Key words: distribution/lactation/pharmacokinetics/progestin- mothers would guarantee effective contraception, the NES only contraception/radioimmunoassay passed to the suckling infant via milk would be rapidly metabolized by the infant’s liver. The purpose of the present study was to examine the distribution as well as the hormonal effects of lactational Nestorone® is a registered trademark of The Population Council, use of NES in nursing cynomolgus monkeys and their New York, NY, USA infants.

© European Society of Human Reproduction and Embryology 1993 O.Heikinheimo et al.

Materials and methods Animals Five nursing cynomolgus monkeys (Macaca fascicularis) and their infants were used for the present study. The mother–infant couples were housed in individual cages under a 12 h:12 h light–dark schedule at 20–23°C; the mother was fed monkey chow (Agway, Elizabeth City, NJ, USA) twice per day with water available ad libitum. Blood samples were drawn by femoral venipuncture under ketamine-induced anaesthesia (10 mg/kg i.m.). Before enrolling the primates into the study, chronic jugular venous catheters were inserted in two of the adult monkeys under ketamine anaesthesia (20 mg/kg i.m.) supplemented with xylazine (1 mg/kg i.m.) to allow serial blood sampling for studies on prolactin secretion. The catheters ended in s.c. ports capped with silicone-coated latex diaphragms which allowed transcutaneous access to the venous system Figure 1. Concentrations (mean Ϯ SE) of Nestorone® (NES) in the via small (21–27)-gauge needles inserted into the ports. mothers’ serum (u), milk (r) and infants’ serum (d) during lactational use of NES implants. The ratio of milk/serum NES was Study protocol 1.68 Ϯ 0.12 (mean Ϯ SE); the concentrations of serum and milk Blood sampling (daily from the mother, every third day from the NES were significantly correlated (r ϭ 0.75, P Ͻ 0.001). NES was infant) was initiated at approximately 90 days following delivery. not detectable in any of the infant samples studied. The insert Pre-treatment samples were collected for 2 weeks, after which a shows the disappearance of NES following removal of the implant, 1 cm Nestorone® implant was inserted s.c. The implants were used the calculated t1/2 (ϮSD) of NES was 1.4 (Ϯ0.4) h. for a total of 4 weeks, after which the implants were removed, and blood sampling continued for an additional 2 weeks. Based on in- vitro testing, the implants released ~40 µg NES/day. The infants were Serum oestradiol and progesterone were measured using commer- weighed weekly during the duration of the study. cially available radioimmunoassays from ICN Biomedicals Inc. (Costa To study more closely the dynamic variations in prolactin secretion Mesa, CA, USA). Serum prolactin was measured by specific radio- during NES administration, two of the monkeys had indwelling immunoassays developed for primate work, as described in detail jugular cannulae inserted prior to initiation of the study. Frequent elsewhere (Gordon et al., 1992). CV were calculated from pools of blood sampling (every 15 min between 8:00 and 14:00) was performed standard serum. Intra-assay and interassay CV were 11 and 18%; 10 at 2 weeks before insertion (pre-treatment), at 2 weeks after insertion and 15%; 10 and 12%; for the oestradiol, progesterone and prolactin (treatment) and at 2 weeks after (post-treatment) removal of the at the average hormone concentrations of 195 pmol/l, 4 nmol/l and Nestorone® implant. Due to blockade of the system, sample collection 10 µg/l respectively. The detection limits were 70 pmol/l, 1 nmol/l was successful in only one of these primates. On the preceding and 3 µg/l respectively. afternoon, the animals were fitted with special vests allowing lactation To study the possibility that NES is metabolized while being passed and mobile steel tethers, which protected the catheters passing from to milk, a fractionation of diethyl ether extracts of milk and serum the s.c. port to the back of the cage and into an adjacent room, where samples was carried out using high performance liquid chromato- the blood samples (2 ml) were drawn. graphy (HPLC) followed by detection of the immunoreactive NES metabolites with NES–radioimmunoassay. The HPLC–NES–radio- Animal welfare compliance immunoassay system for the studies on the NES metabolism has This study was approved by the Institutional Animal Care and Use been described previously (Heikinheimo et al., 1994). The milk Committee of the Eastern Virginia Medical School. The facilities of samples were processed for HPLC as for radioimmunoassay using the division of Animal Resources are fully accredited by the American ether extraction and Sep-Pak® C18 Cartridges. Association for the Accreditation of Laboratory Animal Care. Data analysis Radioimmunoassays A P value ഛ 0.05 was considered significant. One way analysis of Serum concentrations of NES were measured as previously variance (ANOVA) was used to evaluate the changes in serum 125 described (La¨hteenma¨ki et al., 1981) using I-labelled NES as tracer concentrations of prolactin. and polyclonal NES antibody, raised against NES-3-(O-carboxyme- thyl)oxime–BSA conjugate in rabbits. The intra- and interassay coefficients of variation (CV) of serum NES–radioimmunoassay were Results 5 and 7% respectively. The practical detection limit was 13 pmol/l. The improved method for the radioimmunoassay measurement For the analysis of milk concentrations of NES, the milk samples of NES in milk was evaluated using blank primate milk spiked were extracted with petroleum ether (b.p. 40–60°C), evaporated until ® with unlabelled NES. The practical detection limit of the milk dry, redissolved into petroleum ether and applied on Sep-Pak C18 NES–radioimmunoassay was 52 pmol/l; recovery of the added Cartridges (Waters Corp., Milford, MA, USA) as described by the NES was 68%. The intra-assay CV was 12% and the interassay manufacturer. The Sep-Pak® C18 Cartridges were washed with petroleum ether, after which NES was eluted using methanol. There- CV 11–13%, when calculated using primate milk spiked with after, the samples were evaporated, and redissolved into PBS buffer 135–270 pmol/l of unlabelled NES. used in the NES–radioimmunoassay. In the milk assays, 3H-NES was Concentrations (mean Ϯ SE) of NES in maternal serum, used as a tracer; otherwise the radioimmunoassay was performed as milk and infant serum are shown in Figure 1. The mean described previously (La¨hteenma¨ki et al., 1981). (Ϯ SD) maternal serum concentrations of NES were 337 1994 Nestorone® progestin during lactation

Figure 2. The identity of the NES in primate milk was ascertained by HPLC ϩ radioimmunoassay analysis of the milk samples. The behaviour of NES in milk (r) was identical to that of a NES- standard (u), thus parent NES was passed to and measured in the Figure 3. Serum concentrations of prolactin (mean Ϯ SE) in the milk. nursing female monkeys during the use of NES implant.

(Ϯ 90) pmol/l during the use of the implants. The NES concentrations (mean Ϯ SD) in milk were 586 Ϯ 301 pmol/l. The ratio of serum/milk NES was 1.68 Ϯ 0.12 (mean Ϯ SE), and the concentrations of serum and milk NES were signi- ficantly correlated (r ϭ 0.75, P Ͻ 0.001). NES was not detectable (Ͻ13 pmol/l) in any of the infant samples. Following insertion high serum concentrations of NES were measured at 1 h; at 2 h the mean concentration of NES was 957 pmol/l, thus NES was rapidly distributed from the site of implantation. Following removal of the implants, serum concentrations of NES rapidly decreased; the calculated half- life (Ϯ SD) was 1.4 (Ϯ 0.4) h (Figure 1, insert). The identity of the measurable material by radioimmuno- assay in monkey milk was confirmed by HPLC fractionation of ether-extracted milk samples prior to radioimmunoassay. Practically all material measurable by radioimmunoassay in Figure 4. Prolactin concentrations during frequent blood sampling the milk samples eluted in a single peak with an identical in one of the monkeys before (u), during (r) and after (s) the use retention time as the NES standard (Figure 2), thus confirming of NES implant. Samples were collected via indwelling jugular cannule between 08:00 and 14:00 h at 2 weeks before insertion that the radioimmunoassay measured the parent NES in the (pre-implant), 2 weeks after insertion (implant) and 2 weeks milk samples. following removal of the implant (post-implant). Serum concentrations of prolactin (mean Ϯ SE) are depicted in Figure 3. The mean (ϮSD) concentrations of prolactin in the daily blood sample were 43 (Ϯ33), 26 (Ϯ17) and 29 (Ϯ26) Measurable concentrations of oestradiol were detected in µ g/l in the samples collected prior to, during and after the use some of the primates prior to implant use. However, during of the NES implants respectively. Comparison of the pooled the use of the NES implants, serum oestradiol remained below prolactin values using Student’s t-test revealed statistically the detection concentration of the currently used oestradiol significant differences when the pre-implant concentrations radioimmunoassay (Figure 5). Similarly, progesterone was were compared with those measured during (P Ͻ 0.001) and not detectable in any of the maternal serum samples (data ® after (P Ͻ 0.05) the use of the Nestorone implants. One way not shown). ANOVA, performed with prolactin concentrations at weekly The infant weight gain throughout the study period is shown intervals over the study period, indicated no statistically in Figure 6. The mean weight of 620 g at the beginning of significant differences at different time points of the study. the study increased to 734 g, measured at 2 weeks after implant Figure 4 shows the individual concentrations of prolactin removal. A notch in the mean weight at 4 weeks was due to concentrations during frequent blood sampling in one of the weight loss in one of the five infant primates. monkeys before, during and after the use of Nestorone® implant. The mean (ϮSD) concentrations of prolactin over the frequent sampling period in this individual monkey in the pre- Discussion implant, implant and post-implant samples were 61.1 (Ϯ37.2), High contraceptive efficacy and rapid first-pass metabolism, 70.0 (Ϯ26.7) and 66.8 (Ϯ6.0) µg/l respectively. and thus lack of effect following oral administration, make 1995 O.Heikinheimo et al.

such as medroxyprogesterone acetate (MPA), are transferred to milk at higher ratios (milk/serum ~0.88) than estrane or gonane progestins; milk/serum ratios of 0.34 and 0.15 have been reported for norethisterone and levonorgestrel (Nilsson et al., 1977; Heikkila¨ et al., 1982; Koetsawang et al., 1982). NES is a pregnane, thus higher milk/serum ratios are to be expected. Previously (La¨hteenma¨ki et al., 1990) a milk/serum ratio of 0.60 in lactating women using s.c. NES implant was reported. The NES concentrations in primate milk exceeded those measured in the serum, thus the milk/serum ratio was 1.68. Factors thought to increase the progestin content of human milk include the absence of high-affinity binding protein in maternal serum, and an increase in the milk fat content (Nilsson et al., 1977). In human serum NES is loosely bound by albumin (La¨hteenma¨ki et al., 1983), which might partly explain the high passage of NES to milk. Moreover, the Figure 5. Maternal serum concentrations of oestradiol during the fat content of the primate milk was not determined in the use of Nestorone® implants. The oestradiol concentrations appeared present study. to be suppressed during the first 2 weeks of the implant use. Despite the unusually high concentrations of NES in However, some follicular activity appeared towards the end of the implant use, and specially thereafter. Progesterone was not primate milk, no NES could be detected in the infant serum. detectable in any of the samples. Similarly in vitro, human fetal liver has been reported rapidly to metabolize 3H-NES (La¨hteenma¨ki, 1986). Also rat liver completely degraded the NES arriving via the portal blood (Heikinheimo et al., 1994). The fact that infant liver effectively metabolizes the NES in the milk endorses the safety of NES as a lactational contraceptive. The patterns of prolactin release have been previously evaluated during the use of depot MPA contraception in lactating women (Chaudhury et al., 1977). Increased mean concentrations as well as enhanced prolactin release to suckling stimulus was noted (Chaudhury et al., 1977). In the present study, a decrease in the pooled concentrations of prolactin was noted between the pre-treatment versus treatment and post-treatment values; however, when analysed at weekly intervals throughout the study, no statistically significant differ- ences were found in the circulating prolactin concentrations (Figure 3). In addition, the patterns of prolactin release were similar before, during and after the NES implant when analysed in during frequent sampling in one of the primates (Figure 4). Progestin contraception during lactation does not affect Figure 6. Infant weight (mean Ϯ SD) during the use of NES infant growth or development (WHO, 1994). In the present implants. study, the mean weight increased during the NES treatment. However, a small drop in the mean infant weight was noted ® Nestorone progestin a strong candidate for lactational contra- at 4 weeks. This was due to weight loss in one of the infants. ception. We studied the endocrine effects and distribution of The reason for this is unknown; however, the effect of repeated parenterally administered NES in nursing cynomolgus monkeys blood sampling on the infant growth cannot be ruled out. and their infants. In conclusion, use of Nestorone® implants in nursing In the present study, the use of 1 cm NES implants resulted cynomolgus monkeys results in high concentrations of NES in mean NES concentrations of above 270 pmol/l in all of the in maternal circulation and in milk. However, NES was not primates. Previous studies in women have shown that serum detectable in any of the infants. The weekly maternal serum NES concentrations of ~100 pmol/l are sufficient for inhibition concentrations of prolactin were unaffected during the study. of ovulation in women (La¨hteenma¨ki et al., 1982; Haukkamaa Furthermore, the infants’ weight gain continued during the et al., 1992). Thus the concentrations of NES seen in this study. Therefore it appears that NES approaches the ideal study are well within the clinically-proven effective range. In lactational contraception an effective contraception in the accordance with the rapid metabolism of NES in women, the mother and devoid of effects in the nursing infants. serum concentrations of NES decreased rapidly following removal of the implants, the half-life being 1.4 h in the monkey. Acknowledgements Progestins of different chemical structure are passed to We wish to thank Mrs Eeva Harju of the Steroid Research Laboratory human milk differently. The progestins of the prenane class, for her expert editorial help. This work was supported in part by the 1996 Nestorone® progestin during lactation contraceptive development programme sponsored by the International WHO (1998) World Health Organization Task Force on Methods for the Natural Committee for Contraception Research of The Population Council Regulation of Fertility; The World Health Organization Multinational Study Inc., New York, NY, USA. Also grants from The Jalmari and of Breast-feeding and Lactational Amenorrhea. I. Description of infant Rauha Ahokas and the Sigrid Juse´lius Foundation, Helsinki, Finland feeding patterns and of the return of menses. Fertil. Steril., 70, 448–460. (O.Heikinheimo) as well as The Andrew Mellon Foundation, New Received on January 27, 1999; accepted on April 21, 1999 York, NY, USA (K.Gordon) are gratefully acknowledged.

References Chaudhury, R.R., Chompootaweep, S., Dusitsin, N. et al. (1977) The release of prolactin by medroxy-progesterone acetate in human subjects. Br. J. Pharmacol., 59, 433–434. Coutinho, E., DaSilva, A.R., Carreira, C. and Sivin, I. (1981) Long-term contraception with a single implant of the progestin ST-1435. Fertil. Steril., 36, 737–740. Dı´az, S. and Croxatto, H.B. (1993) Contraception in lactating women. Curr. Opin. Obstet. Gynecol., 5, 815–822. Dı´az, S., Rodrı´guez, G., Peralta, O. et al. (1988) Lactational amenorrhea and the recovery of ovulation and fertility in fully nursing Chilean women. Contraception, 38, 53–67. Gordon, K., Hodgen, G. and Richardson, D. (1992) Postpartum lactational anovulation in a nonhuman primate (Macaca fascicularis): endogenous opiate mediation of suckling-induced hyperprolactinemia. J. Clin. Endocrinol. Metab., 75, 59–67. Haukkamaa, M., Laurikka-Routti, M. and Heikinheimo, O. (1991) Transdermal absorption of the progestin ST-1435: therapeutic serum steroid concentrations and high excretion of the steroid in saliva. Contraception, 44, 269–276. Haukkamaa, M., Laurikka-Routti, M., Heikinheimo, O. and Moo-Young, A. (1992) Contraception with subdermal implants releasing the progestin ST- 1435: a dose-finding study. Contraception, 45, 49–55. Heikinheimo, O., Noe´, G., Haukkamaa, M. and La¨hteenma¨ki, P. (1994) The progestin ST-1435 – rapid metabolism in man. Contraception, 50, 275–289. Heikkila¨, M., Haukkamaa, M. and Luukkainen, T. (1982) Levonorgestrel in milk and plasma of breast-feeding women with a levonorgestrel-releasing IUD. Contraception, 25, 41–49. Koetsawang, S., Nukulkarn, P., Fotherby, K. et al. (1982) Transfer of contraceptive steroids in milk of women using long-acting gestagens. Contraception, 25, 321–331. Kurunma¨ki, H., Toivonen, J., La¨hteenma¨ki, P.L.A. and Luukkainen, T. (1984) Intracervical release of ST-1435 for contraception. Contraception, 29, 411–421. Laurikka-Routti, M., Haukkamaa, M. and Heikinheimo, O. (1990) A contraceptive vaginal ring releasing ethinyl estradiol and the progestin ST- 1435: bleeding control, serum steroid concentrations, serum lipids and serum chemistry. Contraception, 41, 111–120. La¨hteenma¨ki, P.L.A. (1986) Contraceptive action of a synthetic progestin, ST- 1435 (Thesis). University of Helsinki, Helsinki, Finland. La¨hteenma¨ki, P., Weiner, E., La¨hteenma¨ki, P.L.A. et al. (1981) Contraception with subcutaneous capsules containing ST-1435. Pituitary and ovarian function and plasma concentrations of ST-1435. Contraception, 23, 63–75. La¨hteenma¨ki, P.L.A., Weiner, E., La¨hteenma¨ki, P. et al. (1982) Pituitary and ovarian function during contraception with one subcutaneous implant releasing the progestin ST-1435. Contraception, 25, 299–306. La¨hteenma¨ki, P.L.A., Hammond, G. and Luukkainen, T. (1983) Serum non- protein bound percentage and distribution of the progestin ST-1435: no effect of ST-1435 treatment on plasma SHBG and CBG binding capacities. Acta Endocrinol., 102, 307–313. La¨hteenma¨ki, P.L.A., Dı´az, S., Miranda, P. et al. (1990) Milk and plasma concentrations of the progestin ST-1435 in women treated parenterally with ST-1435. Contraception, 42, 555–562. Nilsson, S., Nygren, K-G. and Johansson, E.D.B. (1977) d-Norgestrel concentrations in maternal plasma, milk, and child plasma during administration of oral contraceptives to nursing women. Am. J. Obstet. Gynecol., 129, 178–184. Thapa, S., Short, R. and Potts, M. (1988) Breast feeding, birth spacing and their effects on child survival. Nature, 335, 679–682. WHO (1994) World Health Organization Task Force for Epidemiological Research on Reproductive Health; Special Programme of Research, Development and Research Training in Human Reproduction Progestogen- only contraceptives during lactation. I. Infant growth. Contraception, 50, 35–53. 1997