- Email: [email protected]
Transplacental passage of a progesterone antagonist in monkeys
Transplacental passage of a progesterone antagonist in monkeys Jean Philippe Wolf, MD," Claudio F. Chillik, MD," Joseph ltskovitz, MD," Donna Weyman, MD," Ted L. Anderson, PhD," Andre Ulmann, MD," Etienne E. Baulieu, MD; and Gary D. Hodgen, PhD• Norfolk, Virginia, and Paris, France The progesterone antagonist RU 486 dramatically increases myometrial contractility of the pregnant uterus, making it a potential adjunctive therapy for labor induction or therapeutic pregnancy termination. Sixteen female cynomolgus monkeys were studied during the second or third trimester of pregnancy. Hysterotomies were performed with the animals under anesthesia, providing access to the intact placental vasculature. RU 486 (25 mg) was injected intravenously into the mothers. Serial blood samples were drawn from the maternal and fetal-placental compartments for a period of 2 hours. RU 486 achieved a gradient equilibrium between the maternal and fetal-placental circulation within 5 minutes, suggesting free passage by simple diffusion. The clearance kinetics of immunoreactive RU 486 are consistent with an open three-compartment system in mother and fetus. The fetal-placental index decreased from 31.2% to 17.8% between the second and the third trimester of pregnancy. There was no acute toxicity of the RU 486 noticed during the experimental course. (AM J OssrET GvNECOL 1988;159:238-42.)
Key words: Progesterone antagonist, pharmacokinetics, transplacental passage
RU 486 is a steroid with potent progesterone antagonistic effects that has been studied extensively for abortifacient, 1. 2 menstrual induction," and antiovulatory' properties. This novel agent may also have other useful applications. Cabrol et al. 5 have made preliminary reports on human fetal expulsion through antiprogesterone treatment in cases of intrauterine death. These investigators found expulsion of the conceptus within 72 hours after administration of the progesterone antagonist RU 486 (mifepristone) at a dosage of 400 mg for 2 days in 9 of I I cases. Frydman et al." reported pregnancy termination for medical indications after 16 weeks of amenorrhea; RU 486 induced a "ripening" of the cervix along with which dilation was calibrated. This procedure dramatically increased the Bishop score and eased fetal-placental explusion. These earlier studies offered some evidence that the antiprogesterone may serve as a useful adjunct in the management of labor induction. Accordingly, preclinical pharmacokinetic and toxicologic studies are needed for
From The Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology, Eastern Virginia Medical School, Norfolk," and Roussel Uclaf and Unite 33 INSERM, Hopital de Bice/re,' Paris. Supported in part by an academic I corporate agreement between Eastern Virginia Medical School and Roussel Uclaf ReceivedforpublicationMarch31, 1987; revisedJanuary 13, 1988; accepted February 16, 1988. Reprint requests: Gary D. Hodgen, PhD, Professor and Scientific Director, The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Lewis Hall, Room 2011, 700 Olney Road, Norfolk, VA 23507.
238
assessment of the safety and efficacy of RU 486 in adjunctive obstetric management. Whereas previous studies have addressed the effects of RU 486 on monkeys in advanced pregnancy 7 as well as initial pharmacokinetic measurements of RU 486 in nonpregnant monkeys,8 no reports on the pharmacokinetics of this progesterone antagonist in pregnant monkeys and their concepti have been made. The present study was undertaken to assess the passage and clearance of RU 486 and its measurable metabolites between maternal and fetal-placental compartments in both the second and third trimesters of pregnancy in laboratory primates.
Material and methods Primates. Sixteen female cynomolgus monkeys (Macaca f ascicularis) with a history of regular menstrual cycles were studied. The primate husbandry practices used have been described elsewhere." Matings took place on days I 2 and I 4 of the menstrual cycle; day I 4 was counted as the first day of pregnancy. Diagnosis of pregnancy was made by detection of macaque chorionic gonadotropin. 10 Pregnant monkeys (body weight range 3.9 to 5.4 ± 0.450 kg) were randomly divided into two groups: Group A subjects were between 100 and 130 days of gestation (mean ± SE = I I 0 ± I 0.3 days), and those in group B were between 130 and 160 days of pregnancy (I 39 ± 8.5 days). Because the full gestational span of these monkeys is approximately 167 days, 11 these primate models were physiologically comparable to second and third trimester human pregnancies.
Transplacental passage of antiprogesterone
Volume 159 Number I
239
Table I. Serial plasma levels of immunodetectable RU 486 in mothers and fetuses after an intravenous bolus in ethanol Time (min) RU486
plasma leveL1
0
I
Group A (I 00-130 days' gestation) Mother (ng/ml) 0 Fetus and placenta 0 (ng/ml) Group B (130-160 days' gestation) Mother (ng/ml) 0 Fetus and placenta (ng/ml)
0
5
I
15
I
30
I
60
I
90
I
120
7034
3153
1857
974
652
529
:t 4352
:t 2786
:t 1272
:t 501
:t 345
:t 272
1644 :t 807
848 :t 350
589 :t 273
323 :t 115
336 :t 163
142 :t 51
10576
3511
1775
1115
663
357
:t 5455
:t 2320
:t 981
:t 372
:t 530
:t 57
982 :t 499
343 :t 163
369 :t 191
299 :t 183
182 :t 63
319 :t 144
r*
0.99
0.94
Data are mean :t SEM. *Derived from temporal comparisons of RU 486 levels in maternal and fetal-placental circulation.
The husbandry practices used were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Methods. Laparotomy was performed with the monkeys under ketamine (Parke-Davis, Morris Plains, N .J .) anesthesia (100 mg per dose). The placental disks were located by fiberoptic transillumination, so that conservative hysterotomy (incision of 3 to 4 cm) was performed in a relatively a vascular zone of the uterus. The primary disk of the (usually) bidiscoid placenta was located and access to its amniotic cavity exposed vascular surface was provided. About 35 ml of amniotic fluid was removed before cutting of the amniotic membrane. After collection of pretreatment blood samples from both the maternal and fetal-placental compartments, the progesterone antagonist RU 486 (25 mg in 1.0 ml of 100% ethanol) was injected as an intravenous bolus into the maternal saphenous vein. The dose of RU 486 is not adjusted for weight when administered clinically to women ranging from 50 to 70 kg (40% variation); similarly, no adjustment of dose for weight was incorporated for these monkeys (<40% weight variation). Venous blood samples were then collected serially from mother and conceptus at 5, 15, 30, 60, and 120 minutes after dosing. Each blood sample from mother and conceptus was 2.0 and 1.0 ml, respectively; a total of 12 ml was taken from the mother and 6.0 ml from the conceptus. Red blood cells were harvested by gentle centrifugation and returned in sterile saline solution (0.9%) to maintain consistency of blood volume in each compartment and adequate oxygenation. Plasma samples were kept frozen at - 20° C until measurement of RU 486 and its cross-reacting metabolites." Radioimmunoassay. Anti-RU 486 serum (rabbit) was provided by Roussel Uclaf (Paris, France). 12 Radioimmunoassay procedures included a single diethyl ether extraction that recovered 89.3% ± 1.3% of labeled RU 486 added to plasma. Assay sensitivity was approximately 51.6 ± 4.6 pg per tube, with a 9% inter-
assay coefficient of variation at 70% bound; 100 pg of RU 486 reduced maximum binding by about half. In addition to binding to RU 486 itself, the antiserum cross-reacts mainly with three metabolites of RU 486: N-Didemethyl RU 486 (84%), N-monodemethyl RU 486 (60%), and propargyl alcohol RU 486 (0.80%). The cross-reactivity with other steroids is <0.01%. 12 Statistics. Plasma concentrations of RU 486 are reported as mean ± SEM. The fetal index, defined as the ratio of the area under the fetal-placental clearance curve versus that of the mother, gives an indication of the efficiency of the drug's passage to (and from) the fetal-placental compartment. Fetal index was calculated for each pair (mother and conceptus); results from second- versus third-trimester groups were compared by the Student t test. Results
Plasma values of RU 486 showed some indi\·idual variability (Table I), but levels between groups A and B did not differ statistically in maternal plasma at any time interval. The apparent initial volume of distribution, which corresponds to the dose of the drug divided by the mean initial concentration (here, 25 mg/8124 ng/ml = 3077 ml with weight about :~ kg) was less than the body weight for pregnant mothers, and the half-life of distribution was < 15 minutes. The correlation coefficients between the data obtained for mothers and concepti were equal to or exceeding r = 0.9 in 13 of 16 cases. In three instances blood samples from the fetal-placental circulation were technically problematic and subsequently were associated with poor correlations of maternal to fetal-placental levels of RU 486 (r < 0.5). Because of these procedural dif~ ficulties, the data from these three subjects were excluded from the overall calculation. RU 486 achieved immediate passage into the fetal-placental compartment, reaching a gradient equilibrium within 5 min-
240
Wolf et al.
July 1988 Am J Obstet Gynecol
A.
8000
co
co
Mother ----Fetus
6000
111:1'
:::>
a: E
n:9 Ratio of area under the curves: 31.2 %
ca -m 4000 E c ca
• Q.
2000
0
•' '...... 5 15 30
60
90
120
Minutes
B.
12000 - - Mother ----Fetus
9600
co
co
111:1'
:::>
a:
~
ca c:n E c :
a:
n=4 Ratio of area under the curves: 17.8%
7200
48
2400
5 15 30
60 Minutes
Fig. I. Mean (±SE) acute pharmacokinetics of RU 486 in maternal and fetal-placental compartments. A, Second-trimester pregnancies at I 00 to 130 days of gestation. B, Third-trimester pregnancies at 130 to 160 days of gestation (term of 167 days). Fetal index is the ratio of area under the curve for the fetus to that of mother. Error measurements are presented in Table I.
utes. Concentrations of RU 486 in mothers and concepti manifested similar distribution and elimination curves (Fig. 1): an initial rapid decline followed by a much slower rate of elimination. It should be noted that in the first hour a permanent gradient was established between mother and fetus. Moreover, although a relative steady state was achieved rapidly, the RU 486 plasma levels in maternal versus fetal-placental circulation were not equal. Indeed, the gradient remains significantly different (p < 0.05) through 60 minutes, disappearing only when the clearance of RU 486 or its metabolites approached zero.
Considered separately, the disappearance curves of RU 486 in groups A and B (Fig. l) have very similar profiles; however, the ratio (fetal index) of areas beneath the curves (fetal-placental: maternal)" shifted significantly from 31.2% for second-trimeste r pregnancies to 17.8% for third-trimester pregnancies (p < 0.01). Comment
Our pharmacokinetic data on RU 486 in pregnant monkeys reported here agrees well with data of Deraedt et al., 8 who showed that the plasma levels of this progesterone antagonist correlated well with distribu-
Volume 159 Number 1
tion into an open three-compartment system. After intravenous injection of RU 486, 3 mg/kg in polyethyleneglycol 300, there was an initial rapid decrease in the plasma level followed by a much slower descent. The half-life of distribution was about l hour, whereas the half-life of elimination was about 15 hours. The plasma levels of RU 486 and its principal metabolites were 1.14 mg/ml at 5 minutes, 43 ng/ml at 6 hours, and 4 ng/ml at 72 hours. The apparent initial volume of distribution was approximately twice the body weight.' We administered RU 486 dissolved in 100% ethanol. This vehicle may account in part for the apparent increase in the diffusion rate of RU 486 and its principal metabolites compared with the polyethyleneglycol 300 vehicle. The half-life of distribution (<15 minutes) is much smaller, probably because of the combined effects of increased permeability of the blood vessel in pregnant monkeys and the use of ethanol as the vehicle. Surprisingly, the weight of the apparent initial volume of distribution is below the mean body weight of the monkeys, whereas in non pregnant monkeys it was twice the body weight." In both the maternal and fetal-placental compartments, the data in our pharmacokinetic curves are consistent with an open three-compartment system with half-lives of distribution of 15 and 25 minutes, respectively. Indeed, the results illustrate an almost immediate relative steady-state equilibrium between maternal and fetal-placental compartments. The patterns observed are consistent with rapid transplacental passage of RU 486, perhaps via simple diffusion. However, the evidence that a gradient develops between the mother and her conceptus suggests some limit to this "free" passage. The relative equilibria were achieved after 2 hours, suggesting that RU 486 passage through the placenta has a limited Aux. However, if the drug had been administered chronically, we would anticipate that higher levels could be reached in the fetal-placental compartment. It is possible that the metabolites measured in the radioimmunoassay are more restricted in their placental passage than the parent molecule, RU 486 itself. It is noteworthy that the antibody used in this assay could not distinguish RU 486 from its principal metabolites; no antibody of greater discrimination was available at the time of this study. Accordingly, calculations to prove whether an open three-compartment system exists cannot be undertaken appropriately when four molecular species are recognized by a single antibody. Each metabolite has an individual rate of formation and clearance from the parent RU 486 compound, and investigati?n of these kinetics would require administration of each of the four molecular species to be studied independently; such an objective is beyond the principle raised in the present study.
Transplacental passage of antiprogesterone
241
Of interest is that the fetal index decreased to 18% in the third trimester relative to the second (31 %). This may be because of the increased volume of the fetalplacental compartment in the third trimester; alternatively, perhaps the mature or senescent placenta is less facilitatory to RU 486 passage from the maternal compartment. We must emphasize that the data presented are limited to initial metabolic kinetics within the first 2 hours after intravenous administration of the antiprogesterone agent. Such acute observations may have a limited bearing on the pharmacokinetics of RU 486 metabolism in more chronic situations. The three problematic cases in which very little of the drug reached the fetuses, despite appropriate elevations in the mothers, were related to technical complications in which the vascular system of the fetalplacental unit was compromised at surgery by compression or myometrial spasm early on in the surgical procedure. In summary, the progesterone antagonist RU 486 quickly achieved passage through the placenta to the fetus in a relatively unrestricted fashion. If we are correct in interpreting that the maternal fetal-placental compartment diffusion patterns are consistent with an open three-compartment system, the gradient that developed between them during the first 2 hours of the pharmacokinetic study illustrates only minimal resistance to "free" passage of RU 486. Interestingly, the efficiency of RU 486 passage through the placenta decreased (p < 0.01) as pregnancy advanced to the latter part of the third trimester. If our data accurately reflect a progressive decrease in the fetal index as gestation progresses, then the bioavailability of the drug reaching fetal-placental tissues may diminish significantly, thereby reducing theoretical risks of RU 486 side effects on fetuses and newborns when RU 486 is used as an adjunct medication in the management of labo.r induction. Whereas these findings encourage consideration of a serious clinical study and confirm previous indications of free placental passage of RU 486 in women," recommendations for cervical "ripening" should await further evidence of safety in maternal and newborn responses from the primate model. We thank Ms. Lynn Sharpe for efficient technical assistance during surgery and Ms. Rose-Marie Bradley Jones for editing. REFERENCES I. Hermann W, Wyss R, Riondel A, et al. The effects of an antiprogesterone steroid on women: Interruption of the menstrual cycle and of early pregnancy. Compres Rendu Acad Med 1982;294:933. 2. Couzinet B, Le Strat N, Ulmann A, Baulieu EE, Schaison G. Termination of early pregnancy by the progesterone antagonist RU 486 (mifepristone). N Engl J Med 1986; 315:1565. 3. Healy DL, Baulieu EE, Hodgen GD. Induction of men-
Wolf et al.
4.
5. 6.
7.
8.
struation by an antiprogesterone steroid (RU 486) in primates: Site of action, dose response relationships, and hormonal effects. Fertil Steril 1983:40:253. Collins RL, Hodgen GD. Blockade of the spontaneous midcycle gonadotropin surge in monkeys by RU 486: A progesterone antagonist or agonist? J Clin Endocrinol Metab 1986;63: 1270. Cabral D, Bouvier D'Yvoire M, et al. Induction of labor with mifepristone after intrauterine fetal death [Letter]. Lancet 1985;2:1019. Frydman R, Taylor S, Frenandez H, Pons JC, Forman RG, Ulmann A. Obstetrical indication of mifepristone (RU 486) [Abstract). Presented at the sixth annual meeting of the Society for Advances in Contraception. Chicago: Society for Advances in Contraception, 1986. Germain G, Philibert D, Pottier J, Mouren M, Baulieu EE, Sureau C. Effects of the antiprogesterone agent RU 486 on the natural cycle and gestation in intact cynomolgus monkeys. In: Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum, 1985:155-67. Deraedt R, Bonnat C, Busigny M, et al. Pharmacokinetics of RU 486. In: Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum, 1985:103-22.
July 1988 Am J Obstet Gynecol
9. Goodman AL, Hodgen GD. Postpartum patterns of circulating FSH, LH, prolactin, estradiol and progesterone in nonsuckling cynomolgus monkeys. Steroids 1978;3 I: 731. IO. Hodgen GD, Ross GT. Pregnancy diagnosis by a hemagglutination inhibition test for urinary macaque chorionic gonadotropin (mCG). J Clin Endocrinol Metab I 974;38:927. 11. Hodgen .GD. Primate models for pregnancy hormone secretion in man: Fetal, maternal and placental factors. In: Alexander NJ, ed. Animal models for research on contraception and fertility. Hagerstown, Maryland: Harper & Row, 1979:425-36. 12. Salmon J, Mourne R. Radioimmunoassay of RU 486. In Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum Press, 1985 :99-10 I. 13. Levy G, Hayton WL. Pharmacokinetic aspects of placental drug transfer. In: Boreus LO, ed. Fetal pharmacology. New York: Raven, 1973:29-39. 14. Frydman R, Taylor S, Ulmann A. Transplacental passage of mifepristone [Letter]. Lancet 1983;2: 1252.
Immunoreactive substance P in the human ovary David H. Barad, MD, Kenneth J. Ryan, MD, Karen Elkind-Hirsch, PhD, and Anatasia Makris, PhD Boston, Massachusetts Although substance P has been demonstrated in the nonhuman ovary, it has not previously been identified in normal human ovarian tissue. We examined ovarian surgical specimens from 30 women for the presence of immunoreactive substance P. With the use of a specific radioimmunoassay, immunoreactive substance P was identified in these ovarian extracts. The geometric mean of the immunoreactive substance P concentration was 0.55 pg/mg wet weight (95% confidence limits 0.04 and 6.15). Thecal and stromal concentrations of immunoreactive substance P were greater than those in the corpora lutea and tunica albuginea. Large follicles had a lower concentration of immunoreactive substance P than did small follicles. We conclude that the neuropeptide substance P is present in some normal human ovaries and may be a modulator of ovarian function. (AM J OssTET GvNECOL 1988;159:242-6.)
Key words: Substance P, ovary, neuropeptides Substance P is a neuropeptide first described in the digestive system' and later found to be present in neuroendocrine cells and in the central and peripheral nervous systems! The amino acid sequence of substance
From the Laboratory of Human Reproduction and Reproductive Biology, Haroard Medical School. Supported by National Institute of Child Health and Human Development Grant No. ROI HD 07923-15. Received for publication June 18, 1987; revised January 5, 1988; accepted March 5, 1988. Reprint requests: David H. Barad, MD, Albert Einstein College of Medicine, Ullmann Bldg. Room 123, l300 Morris Park Ave., Bronx, NY 10461.
242
P was first identified in 1971.' Nerves containing substance P have been observed in extracts of the human fallopian tube, cervix, and uterus and throughout the female reproductive systems of other species.1 Extracts of rat, guinea pig, and hamster ovaries have all been demonstrated to contain substance P.2· 5 •7 However, Kannisto et al." were not able to demonstrate substance P in the human ovary by immunohistochemistry. Although substance P has not been previously isolated from normal human ovarian tissue, it was noted in patients with primary ovarian carcinoids. 9 · '° Substance P has diverse effects in vivo and in vitro, acting as a smooth muscle contractant, a hypotensive agent, and
Key words: Progesterone antagonist, pharmacokinetics, transplacental passage
RU 486 is a steroid with potent progesterone antagonistic effects that has been studied extensively for abortifacient, 1. 2 menstrual induction," and antiovulatory' properties. This novel agent may also have other useful applications. Cabrol et al. 5 have made preliminary reports on human fetal expulsion through antiprogesterone treatment in cases of intrauterine death. These investigators found expulsion of the conceptus within 72 hours after administration of the progesterone antagonist RU 486 (mifepristone) at a dosage of 400 mg for 2 days in 9 of I I cases. Frydman et al." reported pregnancy termination for medical indications after 16 weeks of amenorrhea; RU 486 induced a "ripening" of the cervix along with which dilation was calibrated. This procedure dramatically increased the Bishop score and eased fetal-placental explusion. These earlier studies offered some evidence that the antiprogesterone may serve as a useful adjunct in the management of labor induction. Accordingly, preclinical pharmacokinetic and toxicologic studies are needed for
From The Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology, Eastern Virginia Medical School, Norfolk," and Roussel Uclaf and Unite 33 INSERM, Hopital de Bice/re,' Paris. Supported in part by an academic I corporate agreement between Eastern Virginia Medical School and Roussel Uclaf ReceivedforpublicationMarch31, 1987; revisedJanuary 13, 1988; accepted February 16, 1988. Reprint requests: Gary D. Hodgen, PhD, Professor and Scientific Director, The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Lewis Hall, Room 2011, 700 Olney Road, Norfolk, VA 23507.
238
assessment of the safety and efficacy of RU 486 in adjunctive obstetric management. Whereas previous studies have addressed the effects of RU 486 on monkeys in advanced pregnancy 7 as well as initial pharmacokinetic measurements of RU 486 in nonpregnant monkeys,8 no reports on the pharmacokinetics of this progesterone antagonist in pregnant monkeys and their concepti have been made. The present study was undertaken to assess the passage and clearance of RU 486 and its measurable metabolites between maternal and fetal-placental compartments in both the second and third trimesters of pregnancy in laboratory primates.
Material and methods Primates. Sixteen female cynomolgus monkeys (Macaca f ascicularis) with a history of regular menstrual cycles were studied. The primate husbandry practices used have been described elsewhere." Matings took place on days I 2 and I 4 of the menstrual cycle; day I 4 was counted as the first day of pregnancy. Diagnosis of pregnancy was made by detection of macaque chorionic gonadotropin. 10 Pregnant monkeys (body weight range 3.9 to 5.4 ± 0.450 kg) were randomly divided into two groups: Group A subjects were between 100 and 130 days of gestation (mean ± SE = I I 0 ± I 0.3 days), and those in group B were between 130 and 160 days of pregnancy (I 39 ± 8.5 days). Because the full gestational span of these monkeys is approximately 167 days, 11 these primate models were physiologically comparable to second and third trimester human pregnancies.
Transplacental passage of antiprogesterone
Volume 159 Number I
239
Table I. Serial plasma levels of immunodetectable RU 486 in mothers and fetuses after an intravenous bolus in ethanol Time (min) RU486
plasma leveL1
0
I
Group A (I 00-130 days' gestation) Mother (ng/ml) 0 Fetus and placenta 0 (ng/ml) Group B (130-160 days' gestation) Mother (ng/ml) 0 Fetus and placenta (ng/ml)
0
5
I
15
I
30
I
60
I
90
I
120
7034
3153
1857
974
652
529
:t 4352
:t 2786
:t 1272
:t 501
:t 345
:t 272
1644 :t 807
848 :t 350
589 :t 273
323 :t 115
336 :t 163
142 :t 51
10576
3511
1775
1115
663
357
:t 5455
:t 2320
:t 981
:t 372
:t 530
:t 57
982 :t 499
343 :t 163
369 :t 191
299 :t 183
182 :t 63
319 :t 144
r*
0.99
0.94
Data are mean :t SEM. *Derived from temporal comparisons of RU 486 levels in maternal and fetal-placental circulation.
The husbandry practices used were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Methods. Laparotomy was performed with the monkeys under ketamine (Parke-Davis, Morris Plains, N .J .) anesthesia (100 mg per dose). The placental disks were located by fiberoptic transillumination, so that conservative hysterotomy (incision of 3 to 4 cm) was performed in a relatively a vascular zone of the uterus. The primary disk of the (usually) bidiscoid placenta was located and access to its amniotic cavity exposed vascular surface was provided. About 35 ml of amniotic fluid was removed before cutting of the amniotic membrane. After collection of pretreatment blood samples from both the maternal and fetal-placental compartments, the progesterone antagonist RU 486 (25 mg in 1.0 ml of 100% ethanol) was injected as an intravenous bolus into the maternal saphenous vein. The dose of RU 486 is not adjusted for weight when administered clinically to women ranging from 50 to 70 kg (40% variation); similarly, no adjustment of dose for weight was incorporated for these monkeys (<40% weight variation). Venous blood samples were then collected serially from mother and conceptus at 5, 15, 30, 60, and 120 minutes after dosing. Each blood sample from mother and conceptus was 2.0 and 1.0 ml, respectively; a total of 12 ml was taken from the mother and 6.0 ml from the conceptus. Red blood cells were harvested by gentle centrifugation and returned in sterile saline solution (0.9%) to maintain consistency of blood volume in each compartment and adequate oxygenation. Plasma samples were kept frozen at - 20° C until measurement of RU 486 and its cross-reacting metabolites." Radioimmunoassay. Anti-RU 486 serum (rabbit) was provided by Roussel Uclaf (Paris, France). 12 Radioimmunoassay procedures included a single diethyl ether extraction that recovered 89.3% ± 1.3% of labeled RU 486 added to plasma. Assay sensitivity was approximately 51.6 ± 4.6 pg per tube, with a 9% inter-
assay coefficient of variation at 70% bound; 100 pg of RU 486 reduced maximum binding by about half. In addition to binding to RU 486 itself, the antiserum cross-reacts mainly with three metabolites of RU 486: N-Didemethyl RU 486 (84%), N-monodemethyl RU 486 (60%), and propargyl alcohol RU 486 (0.80%). The cross-reactivity with other steroids is <0.01%. 12 Statistics. Plasma concentrations of RU 486 are reported as mean ± SEM. The fetal index, defined as the ratio of the area under the fetal-placental clearance curve versus that of the mother, gives an indication of the efficiency of the drug's passage to (and from) the fetal-placental compartment. Fetal index was calculated for each pair (mother and conceptus); results from second- versus third-trimester groups were compared by the Student t test. Results
Plasma values of RU 486 showed some indi\·idual variability (Table I), but levels between groups A and B did not differ statistically in maternal plasma at any time interval. The apparent initial volume of distribution, which corresponds to the dose of the drug divided by the mean initial concentration (here, 25 mg/8124 ng/ml = 3077 ml with weight about :~ kg) was less than the body weight for pregnant mothers, and the half-life of distribution was < 15 minutes. The correlation coefficients between the data obtained for mothers and concepti were equal to or exceeding r = 0.9 in 13 of 16 cases. In three instances blood samples from the fetal-placental circulation were technically problematic and subsequently were associated with poor correlations of maternal to fetal-placental levels of RU 486 (r < 0.5). Because of these procedural dif~ ficulties, the data from these three subjects were excluded from the overall calculation. RU 486 achieved immediate passage into the fetal-placental compartment, reaching a gradient equilibrium within 5 min-
240
Wolf et al.
July 1988 Am J Obstet Gynecol
A.
8000
co
co
Mother ----Fetus
6000
111:1'
:::>
a: E
n:9 Ratio of area under the curves: 31.2 %
ca -m 4000 E c ca
• Q.
2000
0
•' '...... 5 15 30
60
90
120
Minutes
B.
12000 - - Mother ----Fetus
9600
co
co
111:1'
:::>
a:
~
ca c:n E c :
a:
n=4 Ratio of area under the curves: 17.8%
7200
48
2400
5 15 30
60 Minutes
Fig. I. Mean (±SE) acute pharmacokinetics of RU 486 in maternal and fetal-placental compartments. A, Second-trimester pregnancies at I 00 to 130 days of gestation. B, Third-trimester pregnancies at 130 to 160 days of gestation (term of 167 days). Fetal index is the ratio of area under the curve for the fetus to that of mother. Error measurements are presented in Table I.
utes. Concentrations of RU 486 in mothers and concepti manifested similar distribution and elimination curves (Fig. 1): an initial rapid decline followed by a much slower rate of elimination. It should be noted that in the first hour a permanent gradient was established between mother and fetus. Moreover, although a relative steady state was achieved rapidly, the RU 486 plasma levels in maternal versus fetal-placental circulation were not equal. Indeed, the gradient remains significantly different (p < 0.05) through 60 minutes, disappearing only when the clearance of RU 486 or its metabolites approached zero.
Considered separately, the disappearance curves of RU 486 in groups A and B (Fig. l) have very similar profiles; however, the ratio (fetal index) of areas beneath the curves (fetal-placental: maternal)" shifted significantly from 31.2% for second-trimeste r pregnancies to 17.8% for third-trimester pregnancies (p < 0.01). Comment
Our pharmacokinetic data on RU 486 in pregnant monkeys reported here agrees well with data of Deraedt et al., 8 who showed that the plasma levels of this progesterone antagonist correlated well with distribu-
Volume 159 Number 1
tion into an open three-compartment system. After intravenous injection of RU 486, 3 mg/kg in polyethyleneglycol 300, there was an initial rapid decrease in the plasma level followed by a much slower descent. The half-life of distribution was about l hour, whereas the half-life of elimination was about 15 hours. The plasma levels of RU 486 and its principal metabolites were 1.14 mg/ml at 5 minutes, 43 ng/ml at 6 hours, and 4 ng/ml at 72 hours. The apparent initial volume of distribution was approximately twice the body weight.' We administered RU 486 dissolved in 100% ethanol. This vehicle may account in part for the apparent increase in the diffusion rate of RU 486 and its principal metabolites compared with the polyethyleneglycol 300 vehicle. The half-life of distribution (<15 minutes) is much smaller, probably because of the combined effects of increased permeability of the blood vessel in pregnant monkeys and the use of ethanol as the vehicle. Surprisingly, the weight of the apparent initial volume of distribution is below the mean body weight of the monkeys, whereas in non pregnant monkeys it was twice the body weight." In both the maternal and fetal-placental compartments, the data in our pharmacokinetic curves are consistent with an open three-compartment system with half-lives of distribution of 15 and 25 minutes, respectively. Indeed, the results illustrate an almost immediate relative steady-state equilibrium between maternal and fetal-placental compartments. The patterns observed are consistent with rapid transplacental passage of RU 486, perhaps via simple diffusion. However, the evidence that a gradient develops between the mother and her conceptus suggests some limit to this "free" passage. The relative equilibria were achieved after 2 hours, suggesting that RU 486 passage through the placenta has a limited Aux. However, if the drug had been administered chronically, we would anticipate that higher levels could be reached in the fetal-placental compartment. It is possible that the metabolites measured in the radioimmunoassay are more restricted in their placental passage than the parent molecule, RU 486 itself. It is noteworthy that the antibody used in this assay could not distinguish RU 486 from its principal metabolites; no antibody of greater discrimination was available at the time of this study. Accordingly, calculations to prove whether an open three-compartment system exists cannot be undertaken appropriately when four molecular species are recognized by a single antibody. Each metabolite has an individual rate of formation and clearance from the parent RU 486 compound, and investigati?n of these kinetics would require administration of each of the four molecular species to be studied independently; such an objective is beyond the principle raised in the present study.
Transplacental passage of antiprogesterone
241
Of interest is that the fetal index decreased to 18% in the third trimester relative to the second (31 %). This may be because of the increased volume of the fetalplacental compartment in the third trimester; alternatively, perhaps the mature or senescent placenta is less facilitatory to RU 486 passage from the maternal compartment. We must emphasize that the data presented are limited to initial metabolic kinetics within the first 2 hours after intravenous administration of the antiprogesterone agent. Such acute observations may have a limited bearing on the pharmacokinetics of RU 486 metabolism in more chronic situations. The three problematic cases in which very little of the drug reached the fetuses, despite appropriate elevations in the mothers, were related to technical complications in which the vascular system of the fetalplacental unit was compromised at surgery by compression or myometrial spasm early on in the surgical procedure. In summary, the progesterone antagonist RU 486 quickly achieved passage through the placenta to the fetus in a relatively unrestricted fashion. If we are correct in interpreting that the maternal fetal-placental compartment diffusion patterns are consistent with an open three-compartment system, the gradient that developed between them during the first 2 hours of the pharmacokinetic study illustrates only minimal resistance to "free" passage of RU 486. Interestingly, the efficiency of RU 486 passage through the placenta decreased (p < 0.01) as pregnancy advanced to the latter part of the third trimester. If our data accurately reflect a progressive decrease in the fetal index as gestation progresses, then the bioavailability of the drug reaching fetal-placental tissues may diminish significantly, thereby reducing theoretical risks of RU 486 side effects on fetuses and newborns when RU 486 is used as an adjunct medication in the management of labo.r induction. Whereas these findings encourage consideration of a serious clinical study and confirm previous indications of free placental passage of RU 486 in women," recommendations for cervical "ripening" should await further evidence of safety in maternal and newborn responses from the primate model. We thank Ms. Lynn Sharpe for efficient technical assistance during surgery and Ms. Rose-Marie Bradley Jones for editing. REFERENCES I. Hermann W, Wyss R, Riondel A, et al. The effects of an antiprogesterone steroid on women: Interruption of the menstrual cycle and of early pregnancy. Compres Rendu Acad Med 1982;294:933. 2. Couzinet B, Le Strat N, Ulmann A, Baulieu EE, Schaison G. Termination of early pregnancy by the progesterone antagonist RU 486 (mifepristone). N Engl J Med 1986; 315:1565. 3. Healy DL, Baulieu EE, Hodgen GD. Induction of men-
Wolf et al.
4.
5. 6.
7.
8.
struation by an antiprogesterone steroid (RU 486) in primates: Site of action, dose response relationships, and hormonal effects. Fertil Steril 1983:40:253. Collins RL, Hodgen GD. Blockade of the spontaneous midcycle gonadotropin surge in monkeys by RU 486: A progesterone antagonist or agonist? J Clin Endocrinol Metab 1986;63: 1270. Cabral D, Bouvier D'Yvoire M, et al. Induction of labor with mifepristone after intrauterine fetal death [Letter]. Lancet 1985;2:1019. Frydman R, Taylor S, Frenandez H, Pons JC, Forman RG, Ulmann A. Obstetrical indication of mifepristone (RU 486) [Abstract). Presented at the sixth annual meeting of the Society for Advances in Contraception. Chicago: Society for Advances in Contraception, 1986. Germain G, Philibert D, Pottier J, Mouren M, Baulieu EE, Sureau C. Effects of the antiprogesterone agent RU 486 on the natural cycle and gestation in intact cynomolgus monkeys. In: Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum, 1985:155-67. Deraedt R, Bonnat C, Busigny M, et al. Pharmacokinetics of RU 486. In: Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum, 1985:103-22.
July 1988 Am J Obstet Gynecol
9. Goodman AL, Hodgen GD. Postpartum patterns of circulating FSH, LH, prolactin, estradiol and progesterone in nonsuckling cynomolgus monkeys. Steroids 1978;3 I: 731. IO. Hodgen GD, Ross GT. Pregnancy diagnosis by a hemagglutination inhibition test for urinary macaque chorionic gonadotropin (mCG). J Clin Endocrinol Metab I 974;38:927. 11. Hodgen .GD. Primate models for pregnancy hormone secretion in man: Fetal, maternal and placental factors. In: Alexander NJ, ed. Animal models for research on contraception and fertility. Hagerstown, Maryland: Harper & Row, 1979:425-36. 12. Salmon J, Mourne R. Radioimmunoassay of RU 486. In Baulieu EE, Segal SJ, eds. The antiprogestin steroid RU 486 and human fertility control. New York: Plenum Press, 1985 :99-10 I. 13. Levy G, Hayton WL. Pharmacokinetic aspects of placental drug transfer. In: Boreus LO, ed. Fetal pharmacology. New York: Raven, 1973:29-39. 14. Frydman R, Taylor S, Ulmann A. Transplacental passage of mifepristone [Letter]. Lancet 1983;2: 1252.
Immunoreactive substance P in the human ovary David H. Barad, MD, Kenneth J. Ryan, MD, Karen Elkind-Hirsch, PhD, and Anatasia Makris, PhD Boston, Massachusetts Although substance P has been demonstrated in the nonhuman ovary, it has not previously been identified in normal human ovarian tissue. We examined ovarian surgical specimens from 30 women for the presence of immunoreactive substance P. With the use of a specific radioimmunoassay, immunoreactive substance P was identified in these ovarian extracts. The geometric mean of the immunoreactive substance P concentration was 0.55 pg/mg wet weight (95% confidence limits 0.04 and 6.15). Thecal and stromal concentrations of immunoreactive substance P were greater than those in the corpora lutea and tunica albuginea. Large follicles had a lower concentration of immunoreactive substance P than did small follicles. We conclude that the neuropeptide substance P is present in some normal human ovaries and may be a modulator of ovarian function. (AM J OssTET GvNECOL 1988;159:242-6.)
Key words: Substance P, ovary, neuropeptides Substance P is a neuropeptide first described in the digestive system' and later found to be present in neuroendocrine cells and in the central and peripheral nervous systems! The amino acid sequence of substance
From the Laboratory of Human Reproduction and Reproductive Biology, Haroard Medical School. Supported by National Institute of Child Health and Human Development Grant No. ROI HD 07923-15. Received for publication June 18, 1987; revised January 5, 1988; accepted March 5, 1988. Reprint requests: David H. Barad, MD, Albert Einstein College of Medicine, Ullmann Bldg. Room 123, l300 Morris Park Ave., Bronx, NY 10461.
242
P was first identified in 1971.' Nerves containing substance P have been observed in extracts of the human fallopian tube, cervix, and uterus and throughout the female reproductive systems of other species.1 Extracts of rat, guinea pig, and hamster ovaries have all been demonstrated to contain substance P.2· 5 •7 However, Kannisto et al." were not able to demonstrate substance P in the human ovary by immunohistochemistry. Although substance P has not been previously isolated from normal human ovarian tissue, it was noted in patients with primary ovarian carcinoids. 9 · '° Substance P has diverse effects in vivo and in vitro, acting as a smooth muscle contractant, a hypotensive agent, and