Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Tibolone [(7,17)-17-hydroxy-7-methyl-19-norpregn-5(10)-en-20-yn-3-one] is a tissue-specific compound with favorable effects on bone, vagina, climacteric symptoms, mood, and sexual well being in postmenopausal women. It has not been observed to stimulate the endometrium (Moore, 1999) or the breast, as demonstrated by the lower incidence of breast tenderness and lower mammographic density (Valdivia and Ortega, 2000). Therefore, in some tissues, tibolone has different effects than estrogens.

The metabolism of tibolone has been studied in female rats, rabbits, and dogs. A considerable number of metabolites were identified in this study using ¹H NMR and mass spectroscopy, and qualitative and quantitative differences between species were observed (Jacobs et al., 1992; Verhoeven et al., 2002). Major phase I metabolic routes were the reduction of 3-keto to 3- or 3-hydroxy moieties, and the major phase II metabolic route was sulfate conjugation of the hydroxy groups at C3 and C17. Profiling of the target organs showed a tissue-specific distribution of metabolites. The majority of these metabolites existed as sulfate conjugates. These data in animals indicate that tibolone exerts its tissue-specific activities, at least partly, due to its tissue-specific metabolism and distribution. In addition, the presence of local sulfatases may convert inactive sulfated metabolites to active forms.

The linearity of the pharmacokinetic profile of tibolone was studied in three groups of nine healthy female volunteers using 1.25, 2.5, and 5 mg of tibolone, respectively. The pharmacokinetic profile was mainly based on the primary phase I plasma metabolites (i.e., tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, and ⁴-tibolone). The steady state was attained by day 5 in all three dose groups. Since in most cases the plasma concentration of tibolone and ⁴-tibolone was below the detection limit, their elimination half-life could not reliably be determined. The geometric mean value of the elimination half-life of 3-hydroxy tibolone for the three dose levels ranged from 7.2 to 8.5 h. The very low plasma concentrations of the parent compound and the even lower concentrations of the ⁴-isomer, in combination with the considerably higher concentrations of the 3- and 3-hydroxy metabolites, indicated that tibolone is extensively metabolized, predominantly by hydroxylation at C3.

Human biotransformation pathways need to be identified, and the possibility of metabolic or pharmacokinetic interactions occurring with coadministered compounds need to be addressed during the development of a drug. In vitro approaches are usually used to study these issues and to evaluate their potential clinical relevance. These in vitro studies generally focus on cytochrome P450, which is a collective term for a group of enzymes that play a critical role in the oxidative metabolism (phase I metabolism) of the majority of drugs (Guengerich and Turvy, 1991; Shimada et al., 1994).

The present study investigated the in vivo human metabolism of tibolone in postmenopausal volunteers under steady-state conditions using [¹⁴C]tibolone. In addition, the interaction of tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, and ⁴-tibolone with the cytochrome P450 enzymes CYP1A2, CYP2C9, CYP2E1, and CYP3A4 was studied in human liver microsomes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reference Compounds. The following nonlabeled and labeled compounds were supplied by NV Organon (Oss, The Netherlands).

• Tibolone (Org OD14¹).
• ⁴-Tibolone (Org OM38).
• 3-Hydroxy tibolone (Org 4094).
• 3-Hydroxy tibolone (Org 30126).
• [¹⁴C]Tibolone, specific activity 6.2 MBq · mg¹, labeled at the 7-methyl substituent; its radiochemical purity was 98%.
• 3-[16-³H]Hydroxy tibolone, specific activity 1.5 TBq · mmol¹; its radiochemical purity was 80%.
• 3-[16-³H]Hydroxy tibolone, specific activity 1.5 TBq · mmol¹; its radiochemical purity was 80%.
• 3,17-Disulfate tibolone, 3-hydroxy-17-sulfate tibolone, 3-sulfate tibolone, isolated from animal studies and identified by mass spectometry and NMR (Jacobs et al., 1992; Verhoeven et al., 2002), and Org OD14, Org OM38, Org 4094, and Org 30126 were used as reference compounds for profiling purposes.

Clinical Investigator and Study Center. The clinical part of this study was performed under the supervision of the principal investigator, S. P. van Marle, in the clinical research center of Pharma Bio-Research International BV (Zuidlaren, The Netherlands). The Local Ethical Committee approved the study protocol, and each subject gave her written informed consent before participation. The study was conducted in compliance with the Declaration of Helsinki and with Good Clinical Practice.

Study Design, Dosing and Sample Collection. This was an open-label, multiple-dose study with no blinding procedures. Three healthy postmenopausal female subjects received a capsule of 2.5 mg of nonlabeled tibolone once daily for 5 consecutive days to attain steady-state plasma concentrations. On day 6, they were given one capsule of 2.5 mg  555 kBq of [¹⁴C]tibolone. Blood samples were collected predose and at 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, and 24 h after the radioactive dose and every 24 h thereafter, until discharged from the clinic (last sample taken at 264 h). Plasma was prepared by centrifugation.

1

Determination of Radioactivity Concentrations. The concentration of radioactivity in plasma and urine was determined by liquid scintillation counting using a type Tri-Carb 2500 TR/2 Canberra Packard liquid scintillation counter (Packard Instrument Co., Meriden, CT).

Pharmacokinetics. The area under the total plasma concentration versus time curves (AUC) were determined by the trapezoidal rule. The half-life of elimination of radioactivity was obtained as follows: from visual inspection of the individual log-concentration versus time plots, it was determined from which time point the plot was approximately linear. Using log-linear regression on these terminal data points of the concentration-time curve, the elimination half-life was calculated.

Pooling of Samples for Metabolite Profiling. For metabolite profiling, plasma samples containing sufficient radioactivity were pooled per sampling time point (1, 1.5, 2, 3, 4, 6, 8, 24, and 48 h after dosing). The plasma samples of later time points did not contain enough radioactivity for the purpose of metabolite profiling. For urine and feces samples, samples containing at least 1.7% of the administered dose were selected for metabolite profiling. This resulted in analyzed urine and feces samples representing 70 to 86% of the administered dose.

Sample Treatment for the Analysis of Metabolite Profiles. Selected biological fluids, containing an adequate amount of radioactivity, were concentrated and then profiled by direct injection on the HPLC column. These profiles were qualitatively and quantitatively compared with the corresponding extracted samples. In case of no significant differences (determined visually), the remaining samples were analyzed for practical reasons by pretreatment procedures. Plasma proteins were precipitated by the addition of ice-cold acetonitrile. After centrifugation, the supernatant was concentrated by vacuum centrifugation and subjected to HPLC analysis. The extraction recovery of radioactivity was 62 ± 9%.

HPLC Analysis of Metabolite Profiles in Plasma, Urine, and Feces. HPLC analysis of the plasma, urine, and feces samples was performed using a µBondapak C₁₈ column (internal diameter, 3.9 mm; internal length, 300 mm) and a gradient of ammonium acetate buffer (0.1 M, pH 4.2) (solvent A) and methanol (solvent B). Elution was performed with a linear gradient of 25 to 90% solvent B (v/v) for 20 min at 50°C. The flow rate was 1.7 ml/min. HPLC analysis was performed with a type HP1090 liquid chromatograph equipped with a type HP1040 diode array detector (Hewlett Packard, Waldbronn, Germany). Radioactivity in the HPLC effluent was determined on-line using a type A525 flow-through Flo-One beta radioactivity detector or by the collection of fractions followed by liquid scintillation counting. Samples were spiked with unlabeled ⁴-tibolone before HPLC analysis as an internal reference for the retention time (UV signal at 254 nm). Reference compounds (isolated or authentic synthesized) were analyzed for their retention times in "in-between" runs. Quantification of tibolone and ⁴-tibolone is not possible by the described HPLC method. Tibolone is not stable enough under applied conditions and will isomerize into ⁴-tibolone. No attempts have been made to correct for this phenomenon.

Inhibition study.

Microsomes Human liver microsomes from different organ donors were supplied by Human Biologics, Inc. (Phoenix, AZ). Human Biologics, Inc. fully complies with all applicable laws governing the sale of processed human biomaterials for commercial research, including the Uniform Anatomical Gift Act. The cytochrome P450 content of the pooled microsomal preparation was 0.48 nmol/mg of protein (data from Human Biologics, Inc.).

Test compounds and model inhibitors. Model inhibitors, fluvoxamine, and ketoconazole (selective inhibitors for CYP1A2 and CYP3A4, respectively) were obtained from Dispensing Services and Control, NV Organon. [¹⁴C]Testosterone, to be used as a substrate for CYP3A incubations, was obtained from Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). Model inhibitors, sulfaphenazole, and diethyldithiocarbamate (selective inhibitors for CYP2C9 and CYP2E1, respectively) were obtained from Ultrafine Chemicals (Manchester, UK) and Aldrich Chemie (Zwdrecht, The Netherlands), respectively.

Incubation Experiments. The study consisted of three types of incubations involving a pool of human liver microsomes and 1) the enzyme-selective substrates, 7-ethoxyresorufin (Burke and Mayer, 1974; Dutton and Parkinson, 1989), diclofenac (Leemann et al., 1992), chloroxazone (Peter et al., 1990) (Sigma Chemical Co., St. Louis, MO), and testosterone (Brian et al., 1990); 2) the enzyme-selective substrates in the presence of Org OD14, Org 4094, Org 30126, or Org OM38; or 3) the enzyme-selective substrates in the presence of model inhibitors known to be selective for each of the human cytochrome P450 enzymes.

Data Analysis. For Org OD14, Org 4094, Org 30126, Org OM38, and each of the model inhibitors, K_i values were determined with the curve-fitting program for the analysis of enzyme kinetic data called "EZ-FIT" (Perella, 1988). The type of inhibition was established by means of Hanes-Woolf plots.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Excretion of Radioactivity in Urine and Feces. Data on the excretion of radioactivity in urine and feces (0-192 h) are given in Table 1. Urine was sampled up to at least 192 h after the radioactive dose. The recovery of radioactivity in urine over the 0- to 192-h interval ranged from 19.7 to 40.4% and averaged 31.2 ± 10.5%.

Radioactivity in Plasma. Radioactivity appeared in plasma within 1 h after the radioactive dose. Peak concentrations were reached 3 h after the dose (Table 2). The terminal half-life of radioactivity was 129, 121, and 123 h for subjects 1, 2, and 3, respectively.

HPLC Metabolite Profiles in Plasma. The metabolite profiles in pooled plasma samples obtained from postmenopausal female volunteers after daily oral administration of 2.5 mg of tibolone for 5 days, followed by a single oral dose of 2.5 mg of [¹⁴C]tibolone, are included in Table 2. Tibolone and its metabolites were quantified as nanograms of equivalents per milliliter of plasma.

HPLC Metabolite Profiles in Urine and Feces. Metabolite profiles of the urine and feces samples obtained from the three postmenopausal volunteers after oral administration of [¹⁴C]tibolone are given in Tables 3 and 4 for urine and feces, respectively. Tibolone and its metabolites were quantified as a percentage of the radioactive dose.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Summary of the metabolic routes of tibolone in postmenopausal women. A, 2-hydroxylation; B, reduction of 3-keto moiety to a 3-hydroxy group, followed by sulfation or reduction to a 3-hydroxy group; C, isomerization; D, reduction of the 5(10)-double bond; E, 7-hydroxylation; F, sulfation of the 17-hydroxy group.

Inhibition Study. K_i values and the type of inhibition (competitive or noncompetitive) were determined for tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, ⁴-tibolone, and each of the model inhibitors (Table 6). The type of inhibition was determined on the basis of Hanes-Woolf plots.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The in vivo human results presented in this article confirm the metabolic routes previously shown in in vitro studies (Sandker et al., 1994) and in vivo in rat, rabbit, and dog (Verhoeven et al., 2002). The most important phase I metabolic reaction found for tibolone in postmenopausal women is the reduction of the 3-keto group to 3- and 3-hydroxy metabolites, a reaction catalyzed by 3-hydroxy-steroid-dehydrogenase/isomerase (HSD) and 3-HSD, respectively. Other phase I reactions are a shift of the 5(10) double bond to a 4(5) double bond catalyzed by ^4-5-isomerase, reduction of the 4(5) double bond to 5,10-dihydro or 5,10-dihydro metabolites catalyzed by 5-reductase, and hydroxylation at C2 and C7 catalyzed by cytochrome P450. The most important phase II metabolic reaction is sulfation of the C3 hydroxy groups formed during phase I metabolism and the C17 hydroxyl group of tibolone. The presence of sulfatases that are able to locally convert the inactive sulfated metabolites to active metabolites may contribute to the tissue-specific effects of tibolone. In addition, tibolone, in contrast to estrogens, inhibits sulfatase activity in the breast, resulting in reduced formation of estradiol from estrone sulfate in breast tissue (Pasqualini and Chetrite, 1999).

To obtain information on the effect of tibolone on the metabolism/pharmacokinetic profile of a coadministered compound, the inhibition of cytochrome P450 enzymes by tibolone and its major phase 1 metabolites was studied in vitro. Tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, and ⁴-tibolone appeared to be weak inhibitors of CYP1A2, CYP2C9, CYP2E1, and CYP3A4 in comparison to the enzyme-selective inhibitors used in this study.

The extent of in vivo inhibition by tibolone and its metabolites can be predicted as:
where [I] is the concentration of inhibitor, K_i is the inhibition constant, [S] is the concentration of substrate, and K_m is the Michaelis-Menten constant (Segel, 1975). Since the local concentrations of drug and inhibitor are difficult to determine, the in vivo plasma concentrations are often used to approximate the concentration at the site of potential metabolic interaction.

Since the peak plasma concentration of most drugs is far below their K_m, the contribution of [S]/K_m can generally be neglected, and the equation for the prediction of the extent of in vivo inhibition becomes:
which is identical with the equation for prediction of the in vivo inhibition for noncompetitive inhibitors (Segel, 1975).

In a pharmacokinetic multiple dose study using the maximum dose of tibolone for clinical use (2.5 mg/day), mean peak plasma concentrations at steady state were 1.7, 14.2, 3.8, and 0.4 ng/ml (5, 50, 12, and 1.3 nM) for tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, and ⁴-tibolone, respectively. The extent of in vivo inhibition of the metabolism of coadministered drugs would be expected to be highest for CYP3A and 3-hydroxy tibolone/3-hydroxy tibolone, for which the lowest K_i values were observed in combination with the highest mean peak plasma concentrations at steady state. The extent of in vivo inhibition of the metabolism of substrates of CYP3A is predicted to be 1.4 and 0.2% for 3-hydroxy tibolone and 3-hydroxy tibolone, respectively. Assuming a 10-fold higher concentration in the liver compared with plasma, the in vivo inhibition would be predicted to be less than 12 and 1.9% for 3-hydroxy tibolone and 3-hydroxy tibolone, respectively. From this study, it can be concluded that tibolone, 3-hydroxy tibolone, 3-hydroxy tibolone, and ⁴-isomer are not likely to display a clinically significant inhibition at the level of CYP1A2, CYP2C9, CYP2E1, and CYP3A4 with regard to the metabolism of coadministered drugs.

Generally, in vitro studies include the characterization of the cytochrome P450 enzymes involved in the primary metabolic routes of the drug. We concluded that these data for tibolone would be irrelevant for the clinical situation because tibolone is transformed in vivo very rapidly by the 3-HSD and 3-HSD enzymes. Oxidation by CYP450 enzymes presumably occurs thereafter and competes with sulfation by the sulfotransferases. The in vitro CYP450 microsomal studies with tibolone cannot mimic this in vivo situation. Incubations with 3- and 3-hydroxy tibolone should offer a more relevant alternative. However, because it is not clear which metabolite forms the predecessor for oxidation at C2 or C7 and because less than 12% of the metabolites are oxidized at position C2 or C7, we concluded that this study would not provide the relevant information.

C2 and C7 hydroxylation are the only phase I metabolic reactions catalyzed by cytochrome P450 and are only minor pathways in quantitative terms. Consequently, it is unlikely that inhibition or induction of cytochrome P450 by coadministered compounds will affect the metabolic elimination of tibolone. Furthermore, tibolone is metabolized by multiple pathways, indicating that other routes of biotransformation may compensate for interactions at one of these pathways. To our knowledge, interactions at the level of 3- or 3-HSD or ^4-5-isomerase have never been reported.

In conclusion, our data show that the metabolism of tibolone in humans is comparable to its metabolism in animal species. The main metabolic routes involve reduction of the 3-keto function, sulfation of the 3-hydroxy metabolites and the hydroxy function at C17. Since cytochrome P450 is not involved in these major metabolic routes and tibolone and its main phase I metabolites are only weak inhibitors of CYP450, clinically relevant interactions at the CYP450 level are not to be expected.