Contraception 68 (2003) 409 – 420
`Original research article
`Pharmacological properties of mifepristone: toxicology and safety in
`animal and human studies
`
`Regine Sitruk-Warea,*, Irving M. Spitza,b
`aCenter for Biomedical Research, Population Council Regine Sitruk-Ware Center for Biomedical Research, 1230 York Avenue, 6th Floor,
`New York, NY 10021, USA
`bInstitute of Hormone Research, Shaare Zedek Medical Center, Jerusalem, Israel
`
`Received 17 March 2003; received in revised form 11 June 2003; accepted 17 June 2003
`
`Abstract
`
`Roussel Uclaf in partnership with the INSERM unit of Prof. E.E. Baulieu first discovered mifepristone (RU486) as part of a large
`research program on steroidal compounds with antihormone properties. Exhibiting a strong affinity to the progesterone and the
`glucocorticoid receptors, mifepristone exerted competitive antagonism to these hormones both in in vitro and in animal experiments.
`Due to its antiprogesterone activity, it was proposed that mifepristone be used for the termination of early human pregnancy.
`Mifepristone, at a dose of 600 mg initially used alone, was then used with a subsequent low dose of prostaglandin that led to a success
`rate of 95% as a medical method for early termination of pregnancy (TOP). Its use was extended to other indications, such as cervical
`dilatation prior to surgical TOP in the first trimester, therapeutic TOP for medical reasons beyond the first trimester, and for labor
`induction in case of fetal death in utero. The efficacy and safety of this treatment has been confirmed based on its use for over a decade,
`with close adherence to the approved recommendations. This paper describes the safety studies conducted in animals as well as the
`safety follow-up and side effects reported with use of the compound in various indications either approved or unapproved. The
`rationale for warnings and contraindications for use of the product are also explained. At lower doses, the molecule has proven
`promising for contraceptive purposes with few reported side effects. However, development of the product for this indication would
`require long-term studies. Although political and philosophical obstacles have delayed research, the use of mifepristone for other
`potential indications in gynecology or oncology should be investigated. © 2003 Elsevier Inc. All rights reserved.
`
`Keywords: Mifepristone; Pharmacology; Toxicology; Safety
`
`1. Introduction
`
`Mifepristone is an orally active synthetic steroid with
`antiprogesterone and antiglucocorticoid activities. To
`date, mifepristone is approved in several countries for use
`in four
`indications: early termination of pregnancy
`(TOP), cervical dilatation prior to surgical TOP, prepa-
`ration for prostaglandin-induced TOP during the second-
`trimester, and expulsion of a dead fetus during the third
`trimester. Although the molecule has several possible
`indications due to its unique properties [1], its potential
`has not been fully realized; the controversy and philo-
`sophical debate involving mifepristone has resulted in
`
`* Corresponding author. Fax: ⫹1-212-327-7678.
`E-mail address: regine@popcbr.rockefeller.edu (R. Sitruk-Ware).
`
`0010-7824/03/$ – see front matter © 2003 Elsevier Inc. All rights reserved.
`doi:10.1016/S0010-7824(03)00171-9
`
`opposition to further research of this compound. In spite
`of the controversy, there is continued interest in investi-
`gating the properties of antiprogestins, including its con-
`traceptive properties.
`
`2. Structure and physical properties
`
`Mifepristone, 17␤-hydroxy-11␤-(4-dimethylaminophe-
`nyl)-17␣-(prop-1-ynyl)-estra-4,9-dien-3-one,
`is
`derived
`from the estrane progestins (Fig. 1). Its molecular weight is
`429.5. It is insoluble in water, but very rapidly dissolves in
`the gastric milieu when ingested orally. The product is
`available in the form of tablets containing 200 mg of active
`ingredient, and remains stable after 3 years at ambient
`temperature [2].
`
`1
`
`TEVA1015
`
`

`

`410
`
`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`Fig. 1. Chemical structure of mifepristone.
`
`3. Mechanism of action and interaction with steroid
`receptors
`
`Mifepristone acts at the receptor level, binding strongly
`to the progesterone and glucocorticoid receptors, and to a
`lesser extent to the androgen receptor. Mifipristone is a
`potent antiprogesterone and antiglucocorticoid and a weak
`antiandrogen. Its relative binding affinity for progesterone,
`glucocorticoid and androgenic receptors is approximately
`five times greater than progesterone, three times greater than
`dexamethasone and four times less than testosterone, re-
`spectively [3]. In contrast, it is without affinity for the
`mineralocorticoid and estrogen receptors. Mifepristone’s
`binding association constant for these receptors (Ka at
`steady state) is clearly greater than those of the natural
`steroids, progesterone and corticosterone, and of the glu-
`cocorticoid receptor agonist, dexamethasone. The metabo-
`lites of mifepristone also bind to the progesterone receptor,
`with binding of the monodemethylated, hydroxylated and
`didemethylated metabolites being 50, 36 and 21, respec-
`tively, relative to 100 for progesterone.
`Mifepristone, like progesterone, enters target cells and
`reaches its receptors; however, it interacts differently from
`progesterone and may produce different conformational
`changes in the receptor. By occupying the progesterone
`receptor in the nucleus, progesterone modifies the receptor’s
`shape, enabling it to bind to chromatin, and this binding
`leads to gene transcription and protein synthesis. Mifepris-
`tone antagonizes these effects by occupying the receptor
`without stimulating gene transcription [4,5].
`The physiological effects of progesterone are mediated
`by its interaction with intracellular progesterone receptors
`(PRs) that are expressed as two protein isoforms, PR-A and
`PR-B, from a single gene. Both PRs have domains for DNA
`binding, hormone binding and activation [6]. PR-A and
`PR-B respond differently to progesterone as well as proges-
`terone antagonists (PA) [7]. Activation of transcription al-
`ways occurs with PR-B but often not with PR-A. Indeed,
`
`PR-A may function as a transdominant repressor not only of
`PR-B-mediated transcription, but also of estrogen receptor-
`mediated transcription. This may explain the antiestrogenic
`activity of certain PAs.
`Antiprogestins may function as pure antagonists to the
`PR or as mixed agonist-antagonist molecules, also known as
`progesterone receptor modulators (PRMs). Mifepristone be-
`haves as a pure antagonist on the McPhail test [1] that
`measures steroidal effect on the rabbit endometrium.
`The complex of mifepristone with the PR inhibits tran-
`scription resulting in the down-regulation of progesterone-
`dependent genes [4]. As compared with other more recently
`synthesized antiprogestins, mifepristone is predominantly
`an antagonist with minimal agonist activity [8]. Several PAs
`including mifepristone, administered at low doses in the
`monkey, were shown to exert antiproliferative effects in the
`endometrium [9,10]. Whether this effect is due to a partial
`progesterone agonistic effect, or an overexpression of the
`androgen receptor is unclear [9].
`
`4. Toxicology in animals
`
`Roussel Uclaf conducted a comprehensive toxicology
`program in the mid-1980s demonstrating the safety of the
`molecule and allowing mifepristone to be used in humans.
`Most of the program focused on the development of indi-
`cations using single-dose administration of the compound.
`Therefore, toxicology studies were conducted with dura-
`tions of animal exposure not exceeding 6 months [11]. The
`compound was shown to have no mutagenic potential and
`no toxic effect up to 1000 mg/kg in acute administration
`performed in several species.
`In subchronic toxicity studies conducted in rodents for
`30 days and 26 weeks, daily doses up to 200 mg/kg or 125
`mg/kg, respectively, displayed no toxicity but induced ef-
`fects related to the antihormonal effects of the compound.
`The antiprogesterone effects resulted in: frequent estrus, a
`
`2
`
`

`

`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`411
`
`decrease in uterine weight and in mammary development,
`and suppression of menstruation and a decrease in serum
`progesterone in monkeys. Antiglucocorticoid effects were
`observed with an increase in kidney and adrenal weights in
`rats and in monkeys and increases in serum adrenocortico-
`tropic hormone (ACTH) and cortisol concentrations. An
`antiandrogenic effect was observed in male rats with a
`decrease in prostate and seminal vesicle weights. Monkeys
`were more sensitive to the antiglucocorticoid effect of the
`molecule. Although doses of 4 mg/kg did not have any
`effect, doses of 15 or 20 mg/kg induced increases in serum
`cortisol and in ACTH levels.
`In summary, 1-month and 6-month treatments in rats and
`monkeys revealed no true toxicity that could not be attrib-
`uted to the antiglucocorticoid, antiprogesterone and antian-
`drogenic activities of mifepristone. No long-term toxicity
`and carcinogenicity studies were performed because the
`treatment was developed for a single-dose use for the indi-
`cations approved to date.
`The molecule has been shown to induce fetal loss at 0.5
`mg/kg in mice, 1 mg/kg in rats and 2 mg/kg in rabbits.
`Doses below these levels must be used to prevent the oc-
`currence of fetal loss in the animals and to study the devel-
`opment of exposed embryos.
`Embryotoxicity studies were conducted in rodents at
`these subabortive doses, and the surviving rat and mice
`fetuses showed no anomaly [11]. In rabbits, isolated anom-
`alies were observed but were not dose-dependent and, there-
`fore, could not be directly correlated to the drug. Rare
`malformations involving the encephalon were observed in
`another study [12] in which pregnant female rabbits were
`treated with a low dose of mifepristone (0.08 – 0.33 mg/kg/
`day). These abnormalities were attributed by the authors to
`a uterine retraction effect related to the antiprogesterone
`activity of mifepristone before or during the formation of
`the chondrocranium, rather than to a direct effect of the
`product on the embryo. Indeed, supplementary treatment
`with progesterone (100 mg/kg) totally suppressed the abor-
`tifacient effect of mifepristone, and under these circum-
`stances no malformations were observed.
`In a neonatal exposure study in rats, the administration of
`a subcutaneous dose of mifepristone up to 100 mg/kg on the
`first day after birth had no adverse effect on future repro-
`ductive function in males or females. The onset of puberty
`was observed to be slightly premature in female rats neo-
`natally exposed to mifepristone [13].
`
`5. Pharmacodynamic effects
`
`5.1. Antiprogestational effects
`
`In animal experiments conducted in vivo, at doses rang-
`ing from 3 to 10 mg/kg depending on the test, mifepristone
`totally inhibited all standard biological responses of exog-
`enous progesterone, such as preparation of the endometrium
`
`for nidation in the rabbit, the process of decidualization
`which occurs at the time of implantation, and maintenance
`of pregnancy in the female rat. By antagonizing endogenous
`progesterone, it displayed antinidatory and abortifacient ac-
`tivity in the rat at doses of 10 mg/kg, whatever the treatment
`period, except for days 1, 2 and 15 when the compound only
`proved effective at about 10 times the dose. Mifepristone
`also exerted abortifacient activity in the mouse. In the mon-
`key, it caused menstruation when administered during the
`luteal phase, regardless of the treatment period [3,4].
`Thirty normal female volunteers received a single dose
`of mifepristone (5, 10, 25, 50, 75, 100, 150, 200, 250, 300
`or 400 mg) between days 2 and 6 after the luteinizing
`hormone (LH) surge. An endometrial biopsy was performed
`3 days after mifepristone intake. Mifepristone induced in-
`hibition of glandular
`secretory activity, degenerative
`changes in glandular cells, vascular changes (reduction of
`stromal edema and stromal extravasation) and an increase in
`glandular mitotic activity. These endometrial responses
`were significantly related to the dose of mifepristone ad-
`ministered [14].
`
`5.1.1. Effects of mifepristone on ovulation
`Mifepristone, 3 mg/kg/day, was administered for 4 days
`to 6 normal women as soon as a dominant follicle had
`emerged. Mifepristone treatment provoked a fall in estradiol
`concentrations with a regression in the dominant follicle.
`LH and follicle-stimulating hormone levels had a tendency
`to diminish but subsequently increased with reinitiation of
`folliculogenesis and occurrence of an LH ovulatory surge
`13 days later [15].
`Three women received mifepristone, 25 mg/day, on days
`1 to 14 of the cycle, and 5received 25 mg/day on days 1 to
`21 [16] During mifepristone treatment, concentrations of
`estradiol remained low, indicating inhibition of folliculo-
`genesis. When mifepristone treatment was discontinued,
`there was an increase in estradiol levels followed by a rise
`in progesterone levels, indicating the occurrence of ovula-
`tion. During the follicular phase, when administered imme-
`diately before the ovulatory peak of LH, mifepristone
`caused a major delay in this peak, resulting in increased
`duration of the follicular phase, without affecting the luteal
`phase [4] administration of mifepristone during the follicu-
`lar phase thus interrupts normal follicular development re-
`sulting in delayed ovulation. This effect is probably related
`to the antigonadotropic activity of mifepristone.
`Mifepristone at a dose of 1 mg/day interferes with en-
`dometrial development while allowing for the occurrence of
`biphasic ovarian cycles and regular bleeding [17]. However,
`it also prevents ovarian cyclicity in a high proportion of
`treatment months; this is associated with increased endome-
`trial growth in some subjects, which may be of concern.
`
`5.1.2. Effect of mifepristone on uterine contractility
`During early pregnancy, the uterus is inactive, probably
`due to the inhibitory effect of progesterone. Doses of ⱖ1
`
`3
`
`

`

`412
`
`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`mg/kg of mifepristone have been shown to antagonize the
`endometrial and myometrial effects of progesterone in
`women. During pregnancy, the compound sensitizes the
`myometrium to the contraction-inducing activity of prosta-
`glandins [2,18].
`The time interval between mifepristone and the appear-
`ance of uterine contractions is 24 –36 h. Simultaneously
`with increased contractility, sensitivity to prostaglandin in-
`creases about fivefold. These data provide the rationale for
`combining mifepristone and a low dose of prostaglandin for
`termination of early pregnancy. In an earlier study [18], the
`increased sensitivity of the uterus to prostaglandin appeared
`24 h after the start of mifepristone treatment and was max-
`imal at 36 and 48 h. Pretreatment with mifepristone did not
`increase uterine sensitivity to oxytocin.
`
`5.2. Antiglucocorticoid effects
`
`Mifepristone has antiglucocorticoid properties and has
`antagonized the effects of dexamethasone in a number of
`models. Mifepristone totally inhibited effects of dexameth-
`asone, such as inhibition of ACTH secretion, as well as its
`thymolytic action and diuretic effects [1].
`In humans, mifepristone also exerts antiglucocorticoid
`activity. The antiglucocorticoid effect of mifepristone is
`exerted both on the central actions of cortisol (inhibition of
`feedback control of cortisol over its own production evi-
`denced by an increase in ACTH and lipotropin hormone)
`and on the peripheral effects (suppression of cutaneous
`vasoconstriction, or decrease in circulating eosinophils in-
`duced by glucocorticoids) [19 –24]. Mifepristone binds to
`the glucocorticoid receptor in human mononuclear leuko-
`cytes with an affinity about threefold higher than that of
`dexamethasone [1].
`The antiglucocorticoid effects of mifepristone are dose-
`dependent and are apparent at single doses of mifepristone
`in the order of 4 – 6 mg/kg. The antiglucocorticoid effect
`lasts for at least 24 h after a single dose of mifepristone
`given at midnight [21]. Mifepristone at a dose of 400 mg (6
`mg/kg) antagonized the suppressive effect of 1 mg of dexa-
`methasone on the hypothalamo–pituitary–adrenal axis
`(HPA) [23]. Administration of dexamethasone at doses ⬎1
`mg counteracts the antiglucocorticoid effects of a 6 mg/kg
`dose of mifepristone [23].
`In conclusion, mifepristone displays antiglucocorticoid
`activity, which is expressed at doses of 400 mg and above
`(single administration). This antiglucocorticoid activity oc-
`curs centrally and peripherally [19 –24]. However, no clin-
`ical or laboratory signs of adrenal failure have been ob-
`served during chronic administration of mifepristone to
`patients with normal adrenal function. This is probably
`related to compensation arising from hypersecretion of
`ACTH and cortisol, resulting from the central action of
`cortisol and weak agonist activity of mifepristone. Also, it
`must be kept in mind that mifepristone does not bind to the
`
`mineralocorticoid receptors, and hence the mineralocorti-
`coid axis is unaffected by the product.
`
`5.3. Antiandrogenic properties
`
`The product also has modest antiandrogenic action.
`When the different activities of mifepristone are compared
`in the same species (rat), the ED50 for the antiprogesterone
`and antiglucocorticoid activities is about 3 mg/kg, whereas
`it is about 30 mg/kg for antiandrogenic activity [1].
`
`6. Pharmacokinetic studies
`
`6.1. Studies in animals
`
`At active oral doses, mifepristone is absorbed satisfac-
`torily in rats and monkeys. Its bioavailability is reduced by
`a moderate presystemic effect in rats (bioavailability: 39%
`of the dose), which is very much more pronounced in
`monkeys (bioavailability: 15% of the dose). In humans, the
`absolute bioavailability of a 20-mg dose is 69% [2,25–29].
`After ingestion of doses of 100 – 800 mg, there is an initial
`redistribution phase of 6 –10 h followed by a plateau for
`24 h or more [26]. The terminal t1/2 of mifepristone is 4 h in
`rats, 15 h in monkeys and 30 h in humans [3,25,29].
`The clearance of mifepristone is 2.7 L/h/kg of body
`weight in rats and 1.45 L/h/kg of body weight in monkeys.
`These values are very much higher than in women (0.04
`L/h/kg of body weight) principally because of a very re-
`duced volume of distribution in women as a result of satu-
`rable high-affinity binding to ␣1-acid glycoprotein, a prop-
`erty not shared by corresponding animal species. However,
`the influence of the human protein, studied in vitro in the
`rat, is restricted to this phenomenon and does not affect the
`elimination rate of mifepristone or its concentration in target
`tissues, the uterus and thymus, and consequently does not
`alter its activity.
`
`6.2. Metabolism and excretion
`
`The metabolism of mifepristone is initiated by rapid
`demethylation and hydroxylation in man, rat and monkey
`[3]. Metabolism of mifepristone occurs primarily via path-
`ways involving N-demethylation and terminal hydroxyla-
`tion of the 17-propynyl chain. In vitro studies conducted
`with human liver microsomes have shown that CYP450
`3A4 is largely responsible for the oxidative metabolism of
`mifepristone [30]. The three major metabolites identified in
`humans are: (a) RU 42 633, which is the metabolite most
`widely found in plasma and is the N-monodemethylated
`metabolite; (b) RU 42 848, which results from the loss of
`two methyl groups from the 4-dimethylaminophenyl in po-
`sition 11␤ and (c) RU 42 698, which results from terminal
`hydroxylation of the 17-propynyl chain [2,25] (Fig. 2).
`Following oral administration of 100 mg or more to
`
`4
`
`

`

`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`413
`
`Fig. 2. The mean (⫾SEM) concentration of mifepristone and three of its metabolites after oral administration of 100 mg and 600 mg in five subjects. The
`upper panels illustrate mifepristone (left) and the monodemethylated (RU 42 633) metabolite (right). Circulating levels of the didemethylated (RU 42 848)
`and alcoholic nondemethylated (RU 42 698) derivatives are shown in the left and right lower panels, respectively. The concentrations of mifepristone were
`similar after these doses, but the concentrations of metabolites were higher after the 600 mg dose of mifepristone. Concentrations of the monodemethylated
`metabolites are higher than those of the parent compound at all time points. From Robbins A, Spitz I. Mifepristone: clinical pharmacology. Clin Obstet
`Gynecol 1996;39:436 –50. Reprinted with permission from Lippincott-Raven Publishers.
`
`humans, constant serum concentrations of mifepristone, but
`increasing concentrations of the monodemethylated, didem-
`ethylated and hydroxylated metabolites, are found [28].
`Within the dose range of 100 – 600 mg, serum concentra-
`tions of the monodemethylated metabolite exceed those of
`the parent drug; in addition, following oral administration of
`doses beyond 400 mg, levels of didemethylated and hy-
`droxylated metabolites exceed those of mifepristone. Al-
`though their affinity for the receptors and their potency are
`less than those of mifepristone, these metabolites may con-
`tribute to the overall effects of the drug in view of their high
`concentrations in serum [31].
`Excretion is essentially fecal (about 92% of the total
`excreted). Mifepristone was excreted by rats and monkeys
`after undergoing almost complete biotransformation [25].
`The presence in human plasma of the N-mono- and N-
`didemethylated metabolites and of the 22-hydroxylated me-
`tabolite shows that the two primary routes are also active in
`women. The demethylated and hydroxylated metabolites
`are further metabolized and excreted in bile, but in humans
`only a very small fraction of mifepristone can be detected in
`urine. The major difference between the pharmacokinetics
`
`of mifepristone in women and in animals, therefore, lies in
`the binding to human ␣1-acid glycoprotein, a characteristic
`that is absent in all the other species tested, including those
`currently used in pharmacology and toxicology. The only
`consequences are much higher plasma concentrations of
`mifepristone in women at active oral doses and the nonlin-
`earity of the kinetics observed clinically.
`
`6.3. Studies in women
`
`Various assay methods have been employed in the mea-
`surement of serum mifepristone and its metabolites; these
`include
`radioimmunoassay
`(RIA),
`radioreceptor-assay
`(RRA), and assays based on high-performance liquid chro-
`matography (HPLC) [26 –29,31].
`Following single-dose administration of mifepristone
`(600 mg), to healthy female volunteers, mean maximum
`plasma concentrations were about 2.0 mg/L at 1.35 h (tmax).
`After oral ingestion, mifepristone is rapidly absorbed, and
`the time to peak serum concentration (tmax) is approximately
`1–2 h (Table 1). When analyzed by specific RIA or HPLC,
`tmax is similar within the dose range of 200 – 600 mg. Peak
`
`5
`
`

`

`414
`
`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`Table 1
`Dose-related pharmacokinetic parameters of the RU486 equivalent [time
`to peak, peak level and AUC] after a single oral dose of 200, 400 and
`600 mg RU486
`
`RU486
`doses
`
`200 mg
`(n ⫽ 9)
`400 mg
`(n ⫽ 10)
`600 mg
`(n ⫽ 12)
`
`tmax (h)
`
`Cmax (␮mol/L)
`
`AUC (h ⫻ ␮mol/L)
`
`1.71 ⫾ 0.54
`
`9.30 ⫾ 2.22
`
`930.4 ⫾ 270.5
`
`2.05 ⫾ 0.98
`
`10.65 ⫾ 3.37
`
`978.1 ⫾ 227.0
`
`1.73 ⫾ 0.71
`
`12.30 ⫾ 3.52*
`
`1272.0 ⫾ 298.3**,***
`
`Values are mean ⫾ SD.
`From Foldesi et al. [26].
`Reprinted with permission from Elsevier Science, Inc.
`* p ⬍ 0.01, compared with 200 mg dose.
`**p ⬍ 0.01, compared with 200 mg dose.
`***p ⬍ 0.05, compared with 400 mg dose.
`
`drug plasma concentration (Cmax) rises according to the
`dose of mifepristone within the dose range of 2–25 mg.
`However, at doses of 100 – 800 mg, Cmax does not differ
`significantly; this is most likely due to the saturability of
`serum binding capacity of the ␣1-acid glycoprotein for
`mifepristone. The bioavailability of mifepristone has been
`estimated at 69% following oral intake of 20 mg as a result
`of presystemic metabolism. After administration of doses
`from 50 to 800 mg, mifepristone displays nonlinear kinetics.
`Following the absorption and distribution phase of approx-
`imately 4 – 6 h, the serum concentration of mifepristone
`remains in the micromolar range for the next 24 – 48 h (Fig.
`
`3). Within the dose range of 2–25 mg, serum concentrations
`of mifepristone, as well as the areas under the concentra-
`tion–time curves (AUC), increase according to dose [31].
`After administration of 400 or 600 mg to pregnant
`women, the pharmacokinetic parameters do not vary with
`the dose and do not differ from those obtained in nonpreg-
`nant women [26,32].
`At serum concentrations below 2.5 ␮mol/L, 94 –99% of
`mifepristone is protein-bound in human serum. As men-
`tioned, human serum, unlike rat serum, contains a protein
`with high-affinity binding for mifepristone, identified as
`␣1-acid glycoprotein (AAG) or orosomucoid. The highly
`significant correlations between serum concentrations of
`mifepristone and AAG suggest that AAG has a marked
`effect on the pharmacokinetics of mifepristone in humans,
`and studies using centrifugal ultrafiltration dialysis have
`shown that the 2.5 ␮mol/L plateau serum concentration of
`mifepristone represents saturation of AAG binding capacity
`[25]. Its volume of distribution and its clearance are dose-
`and time-dependent, and are inversely correlated with AAG
`concentration.
`In humans, the initial volume of distribution and the
`clearance rate of mifepristone are, respectively, 10% of
`body weight and 0.55 L/kg/day [⬇30 L/day, which is con-
`siderably slower than cortisol (200 L/day) and estrone sul-
`fate (160 L/day), two natural steroids that are considered to
`be cleared at slow rates]. Serum AAG seems to limit tissue
`availability of mifepristone. However, mifepristone in ex-
`cess of the binding capacity for AAG may be more suscep-
`
`Fig. 3. Half disappearance time of mifepristone after administration of 50 – 800 mg in the mid-luteal phase. Results show mean⫾SEM in 4 –5 subjects. From
`Robbins A, Spitz I. Mifepristone: clinical pharmacology. Clin Obstet Gynecol 1996;39:436 –50. Reprinted with permission from Lippincott-Raven Publishers.
`
`6
`
`

`

`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`415
`
`tible to excretion, or possibly to diffusion into peripheral
`tissues.
`After a distribution phase, elimination is at first slow,
`with concentration decreasing by only half between 12 and
`72 h, and subsequently becoming rapid with a half-life (t1/2)
`of 18 h. Following a single dose, the parent drug and its
`metabolites remain detectable for 6 –7 days by HPLC and
`for 10 days by RIA [26].
`Mifepristone crosses the placental barrier. Blood concen-
`trations are about one third of maternal blood concentrations
`while maternal:fetal ratios in plasma for mifepristone and its
`monodemethylated metabolite are 9:1 and 17:1, respectively.
`Fetal aldosterone levels are elevated 4 and 24 h after mater-
`nal ingestion of mifepristone, 600 mg, but the drug does not
`affect fetal progesterone, estradiol or cortisol levels [33].
`Although specific drug or food interactions with mife-
`pristone have not been studied, on the basis of this drug’s
`metabolism by CYP 3A4, it is possible that ketoconazole,
`itraconazole, erythromycin and grapefruit juice may inhibit
`its metabolism and increase serum levels of mifepristone.
`Furthermore, rifampicin, dexamethasone and certain anti-
`convulsants (phenytoin, phenobarbital, carbamazepine) may
`induce mifepristone metabolism and lower serum levels of
`mifepristone.
`Based on in vitro inhibition information, coadministra-
`tion of mifepristone may lead to an increase in serum levels
`of drugs that are CYP 3A4 substrates. Due to the slow
`elimination of mifepristone, such interaction may be ob-
`served for a prolonged period after its administration.
`Therefore, caution should be exercised when mifepristone is
`administered with drugs that are CYP 3A4 substrates and
`have narrow therapeutic range, including some agents used
`during general anesthesia [2].
`
`7. Safety in human trials
`
`7.1. Safety in long-term use of mifepristone
`
`The adverse events reported with the use of mifepristone
`vary according to the indication for which mifepristone was
`used and the mode of administration (single or repeated
`doses).
`
`7.1.1. Studies in male volunteers with repeated doses
`The safety of mifepristone was initially studied in
`healthy volunteers with high doses of mifepristone. Laue et
`al. [34] studied 11 healthy men who received either mife-
`pristone 10 mg/kg/day for 8 days or placebo. Eight cases of
`diffuse skin eruption were observed after 9 days of mife-
`pristone treatment. This high incidence of rashes was not
`observed in other studies involving daily administration of
`mifepristone at lower doses [35]. In clinical trials conducted
`in women with single-dose administration of 600 mg, the
`incidence of rashes was below 1% [36].
`
`7.1.2. Long-term use in pilot studies
`Information on safety of mifepristone after repeated ad-
`ministration is provided by pilot studies in breast cancer,
`meningioma, endometriosis, uterine leiomyomata, in which
`mifepristone was used in doses ranging from 50 to 200 mg
`daily for several months [35,37–39]. In clinical studies
`conducted in patients with endometriosis receiving daily
`mifepristone doses of 50 mg for 6 months, there was im-
`provement in pelvic pain and a decrease in the extent of
`disease as determined by laparoscopy. In the study from
`Kettel et al. [38], a few patients experienced a mild increase
`in transaminases.
`Mifepristone has also been used in patients with leiomy-
`omata. In a 3-month study of daily treatment with mifepris-
`tone in doses of 25 and 50 mg, there was a significant
`decrease in leiomyoma volume. All women were amenor-
`rheic, and regular cycles reappeared after treatment discon-
`tinuation. Four patients experienced mild atypical hot
`flushes, which resolved during the second month of therapy.
`After 3 months of therapy, one patient had an increase in
`serum transaminases, which normalized 1 month after dis-
`continuation of treatment [39].
`Studies in animals have suggested that antiprogestins
`could be used in other tumors including meningiomas, gli-
`omas and ovarian, prostate and endometrial cancer [4].
`Grunberg et al. [35] followed 53 patients with inoperable
`meningiomas who received daily doses of 200 mg mifepris-
`tone for several consecutive years. [Although no long-term
`carcinogenicity studies were conducted with mifepristone,
`chronic use of the product was allowed under the individual
`investigators’ Investigational New Drugs for cancer indica-
`tions.]
`Asthenia was the most common side effect and generally
`remained moderate. However, in two cases, severe fatigue
`was noted. One of these patients was a woman with pan-
`hypopituitarism who responded rapidly to an increase in
`exogenous glucocorticoid replacement.
`Maculopapular rash was noted in five patients. The rash
`generally responded to a short interruption of mifepristone
`and to symptomatic therapy. Rash did not recur with further
`mifepristone therapy.
`Two premenopausal women had endometrial hyperpla-
`sia. Laboratory investigations of the first 14 patients in-
`cluded in the study showed an elevation of mean cortisol
`levels (mean ⫾ SEM) from 13.12 ⫾ 1.69 to 26.48 ⫾ 3.83
`␮g/dL and an increase in mean thyroid-stimulating hormone
`(TSH) levels (mean ⫾ SEM) from 1.93 ⫾ 0.40 to 4.37 ⫾
`0.72 mIU/ml. Total T4 levels decreased (from 5.77 ⫾ 0.43
`to 4.47 ⫾ 0.44 ␮g/dL) by 2 to 3 months and then returned
`to normal. TSH, by contrast, remained persistently elevated.
`This represents biochemical hypothyroidism. Provided T4
`levels remain normal, no treatment with thyroxin is re-
`quired. However, it is advisable to monitor thyroid function
`with prolonged administration of mifepristone at high doses
`[40].
`The doses of mifepristone that showed efficacy for con-
`
`7
`
`

`

`416
`
`R. Sitruk-Ware, I.M. Spitz / Contraception 68 (2003) 409 – 420
`
`traceptive use, 2 or 5 mg/day, are 50 –100 times lower than
`the doses used in meningioma patients. Therefore, it is
`highly unlikely that changes in thyroid function would be
`observed with long-term use of mifepristone as an oral
`contraceptive.
`Mifepristone also binds to the glucocorticoid receptor
`and displays potent antiglucocorticoid properties [1,4].
`Higher doses of mifepristone are required to produce an
`antiglucocorticoid as compared to an antiprogestin effect
`[4]. High-dose continuous mifepristone administration
`(5–22 mg/kg/day) has been used to treat Cushing’s syn-
`drome due to ectopic ACTH secretion and adrenal carci-
`noma [41]. Mifepristone has been shown to normalize the
`Cushingoid phenotype, ameliorate depression, decrease hy-
`pertension, eliminate abnormal carbohydrate metabolism
`and correct glucocorticoid-induced gonadal and thyroid hor-
`mone suppression [41]. However, this drug cannot be used
`in Cushing