0090-9556/03/3111-1448-1455$20.00
DMD 31:1448-1455, 2003
PHARMACOKINETICS AND METABOLISM OF [14C]EPLERENONE AFTER ORAL ADMINISTRATION TO HUMANS
Chyung S. Cook,
Loren M. Berry,
Roy H. Bible,
Jeremy D. Hribar,
Elisabeth Hajdu, and
Norman W. Liu
Pfizer Corporation, Skokie, Illinois
(Received May 5, 2003;
Accepted July 23, 2003)
 |
Abstract
|
|---|
A pharmacokinetics and metabolism study was conducted in eight healthy human volunteers. After oral administration of [14C]eplerenone (EP) at a dose of 100 mg per person as an aqueous solution, blood, saliva, breath, urine, and fecal samples were collected at various time points. All matrices were analyzed for total radioactivity and/or for EP and its open-lactone-ring form (EPA). EP was well absorbed, and a mean EP Cmax of 1.72 µg/ml was achieved 1.2 h postdose. After the Cmax, plasma concentrations of EP declined with a half-life of 3.0 h. Plasma concentrations of EPA were much lower than EP concentrations, and the area under the plasma-concentration time curve (AUC) for EPA was only 4% of the EP AUC. Plasma protein binding was moderate (33-60%) but concentration-dependent over the therapeutic concentration range. EP and its metabolites did not preferentially partition into the red blood cells and blood concentrations of total radioactivity were lower than plasma concentrations. Approximately 66.6% and 32.0% of the radioactive dose were excreted in urine and feces, respectively. The majority of urinary and fecal radioactivity was due to metabolites, indicating extensive metabolism of EP. The major metabolic pathways were 6ß- and/or 21-hydroxylation and 3-keto reduction. There was no evidence for any alteration of the 9,11-epoxide ring or the methyl ester. As a percentage of dose, the primary metabolic products excreted in urine and feces included 6ß-hydroxy-EP (6ß-OHEP) (32.0%), 6ß,21-OHEP (20.5%), 21-OHEP (7.89%), and 2
,3ß,21-OHEP (5.96%). The amounts of the other metabolites excreted were less than 5% each.
Eplerenone (EP1; Fig. 1) is methyl hydrogen 9,11-epoxy-17-hydroxy-3-oxopregn-4-ene-7,21-dicarboxylate,
-lactone and a highly selective aldosterone blocker (McMahon, 2001
). EP has been approved for the treatment of hypertension in the United States, and a New Drug Application for treatment of heart failure has been filed in the United States in 2003. Blockade of aldosterone is known to be beneficial in the treatment of cardiovascular and renovascular diseases. Although angiotensin-converting enzyme inhibitors block the renin-angiotensin-aldosterone system, the reduction in aldosterone is only partial and transient, leaving the potentially deleterious effects of aldosterone unchecked. Spironolactone, a nonselective aldosterone blocker, is effective in suppressing the actions of aldosterone. In the Randomized Aldactone Evaluation Study (RALES), addition of spironolactone to standard therapy reduced mortality rates by 30% in patients with severe heart failure caused by systolic left ventricular dysfunction (Pitt et al., 1999). However, it is associated with progestational and antiandrogenic side effects due to its promiscuous binding to other steroid receptors (Sun et al., 1997; Pitt et al., 1999). To avoid these side effects, a new, highly selective aldosterone blocker, EP, has been developed. The presence of a 9,11-epoxide group in the EP molecule dramatically reduced progestational and antiandrogenic side effects compared with spironolactone, while maintaining aldosterone-blocking activity (McMahon, 2001). In a 12-month clinical trial of 499 patients with mild to moderate hypertension, systolic and diastolic blood pressures were reduced from baseline by 14.5 and 11.2 mm Hg for patients treated with EP, whereas the blood pressures were reduced by 12.7 and 11.3 mm Hg, respectively, for patients treated with enalapril (Liew et al., 2003). The Eplerenone Postacute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) of which primary end points were death from any cause and death or hospitalization from cardiovascular causes found a significant 15% relative reduction in all-cause mortality and 13% relative reduction in the combined endpoint of cardiovascular death and hospitalizations among the patients treated with EP, compared with patients on standard therapy (Pitt et al., 2003). Because of this proven efficacy and aldosterone selectivity, EP is expected to provide important clinical benefits not previously available with spironolactone. The present study was conducted to investigate absorption, distribution, metabolism and elimination of EP in humans after oral administration of [14C]EP.
View larger version (15K):
[in this window]
[in a new window]
|
FIG. 1. Chemical and enzymatic conversion between EP and EPA.
The asterisk indicates the position of the labeled carbon atom.
|
|
|
Materials and Methods
|
|---|
Materials. [14C]EP, with radiochemical purity greater than 98%, and EP, with chemical purity greater than 97%, were obtained from Pfizer (Kalamazoo, MI). The metabolite 6ß-OHEP, with chemical purity greater than 95%, was also obtained from Pfizer. All other chemicals and reagents were commercially available.
Pharmacokinetic Study. A pharmacokinetic study was performed to determine the absorption, distribution, metabolism, and elimination of orally administered [14C]EP. The study was designed as an open-label, single-dose study with eight subjects. Eligible subjects were chosen on the bases of a pretreatment screen of relevant parameters including medical history, physical examination, vital signs, 12-lead electrocardiogram, hepatitis B surface antigen test, hepatitis C antibody test, human immunodeficiency virus (HIV) test, drug/alcohol tests, and a number of clinical laboratory tests monitoring for hematology, biochemistry, and urinalysis. These parameters were determined during the subject selection process and then, again (excluding hepatitis and HIV tests), 24 h before administration of the study medication. The clinical and pharmacokinetic portions of the study were conducted in compliance with the institutional review board regulations in Title 21 of the Code of Federal Regulations (CFR), Part 56 and regulations in 21 CFR, Part 50, as well as in accordance with Food and Drug Administration Good Clinical Practice Guidelines and the Declaration of Helsinki.
The study consisted of one treatment: a solution of 100 mg of [14C]EP, containing 75 µCi per dose. The dose solution was prepared by dissolving the [14C]EP (preweighed in a hermetically sealed bottle; packaging lot #RCT10035) in 20 ml of 65°C, 15% hydroxypropyl-betacyclodextrin solution. This was combined at room temperature with 60 ml of commercially purchased apple juice. After the apple juice/hydroxypropyl-betacyclodextrin/[14C]EP solution was administered to the subject, the bottle was rinsed with an additional 60 ml of apple juice. The rinse was then administered to the subject.
Blood, saliva, breath, urine, and fecal samples were collected at predetermined intervals before and after the administration of [14C]EP. Samples from all matrices were analyzed for concentrations of total radioactivity using a validated liquid scintillation counting (LSC) procedure. For the LSC procedure, the shapes of all quench curves were good for all matrices and the quench range was sufficient for all samples. The 14C counting efficiency of the liquid scintillation analyzer varied between 95.8% and 96.8%, and the background varied between 19.6 and 22.3 dpm. Before analysis by LSC, fecal samples were combusted on a biological oxidizer. The mean (±S.D.) 14C combustion recovery of the oxidizer was 98.3 ± 0.9%, the mean 14C carryover was 0.19 ± 0.10%, and the mean correction factor was 1.05 ± 0.03. EP and EPA concentrations were also quantified using a liquid chromatography-tandem mass spectrometric (LC/MS-MS) assay validated for their determination in plasma (Zhang et al., 2003) and urine. The sensitivity for the LC/MS-MS assays in the plasma and urine was 10 ng/ml and 50 ng/ml, respectively.
Plasma Protein Binding. The in vitro plasma protein binding of [14C]EP was determined in whole blood obtained in heparinized BD Vacutainer tubes with plasticizer-free rubber stoppers (BD Biosciences-Discovery Labware, Bedford, MA) from four healthy male subjects 20 to 60 years old. The subjects were fasted overnight and had no drug intake for at least 6 days. The binding was determined by an ultrafiltration method as follows. [14C]EP was added to plasma or 0.1 M phosphate buffer (pH 7.4) to achieve the final concentrations of 0.02, 0.2, 1, 5, and 60 µg/ml. Plasma samples were filtered through a CentriFree filtration device (Millipore Corporation, Bedford, MA) at 1,000g for 1 h at 25°C. Triplicate aliquots (0.8 ml) of the buffer sample at each concentration also were centrifuged under the same conditions for 15 min. Approximately 0.2 to 0.4 ml of filtrate was collected from each device and total radioactivity in the filtrate was determined by LSC.
Plasma protein binding was also determined in the clinical study, following an oral dose of [14C]EP. Plasma samples from 1.5 h postdose were also filtered through the CentriFree filtration device using the same filtration method as described above. The filtrate was collected and total radioactivity was measured by LSC. In each case, the fraction of radioactivity bound in plasma was obtained by the following equation:
 | (1) |
where [D]F is the drug concentration in the filtrate and [D]T is the concentration in plasma or buffer.
Metabolic Profiles of Plasma, Urine, and Fecal Samples. The representative distribution profiles of radioactivity in pooled plasma, urine, and fecal samples were determined as follows using high performance liquid chromatography (HPLC) with radiochemical detection.
Plasma samples were prepared by precipitating plasma proteins with a volume of acetonitrile equal to that of the plasma aliquot from each subject (400 µl), vortexing briefly, and centrifuging. The supernatants were combined, evaporated to dryness under a stream of nitrogen, reconstituted in 10% aqueous acetonitrile, with or without acidifying with 20% (v/v) 1 N HCl, and injected onto the HPLC system described below.
Greater than 90% of the radioactivity excreted in urine was excreted within the first 48 h. Therefore, aliquots of samples collected up to 48 h were pooled proportionately by volume. An aliquot, 600 µl, of the pooled urine was evaporated to dryness under a stream of nitrogen, reconstituted in 10% aqueous acetonitrile, with or without acidifying with 20% (v/v) 1 N HCl, centrifuged, and injected onto the HPLC system.
Greater than 90% of the radioactivity in feces was excreted within 96 h. Portions of the fecal samples collected over the first 96 h were pooled proportionately by weight. Approximately 1.5 g of pooled fecal sample were extracted with 10 ml of acetonitrile by end-over-end rotation for 1 h in a polypropylene centrifuge tube using a rotating mixer. The extract was centrifuged at 1,900g for 10 min at 4°C. The supernatant was transferred to a new tube. The pellet was re-extracted as described above and the supernatants were combined. The extract was evaporated to dryness under a stream of nitrogen, reconstituted in 10% aqueous acetonitrile, with or without acidifying with 20% (v/v) 1 N HCl, centrifuged, and injected onto the HPLC system.
Processed plasma, urine, and fecal samples were analyzed for radioactivity distribution on an HP1050 HPLC system (Agilent Technologies, Inc., Palo Alto, CA) using 5-µm Alltima C18 guard (7.5 x 4.6 mm) and analytical (150 x 4.6 mm) columns (Alltech Associates, Deerfield, IL). A linear gradient of 10% aqueous acetonitrile to 50% aqueous acetonitrile over 40 min at a flow rate of 1 ml/min was used. Radioactivity eluting from the column was detected by a Flo-One Radiomatic detector (Flo-One/ß A-500; PerkinElmer Instruments, Norwalk, CT).
Percentage of dose excreted as each metabolite in urine and feces was calculated by multiplying the percentage of the eluting radiochromatographic peak by the fraction of radioactivity excreted in urine or feces, as appropriate.
Metabolite Identification. Metabolites corresponding to the radioactive peaks eluting in metabolic profiles of urine and feces were identified by LC/MS-MS and nuclear magnetic resonance spectroscopy (NMR). Metabolites in plasma were identified by comparison with urinary and fecal profiles. Metabolites were isolated by separation on an HPLC system similar to that described above. Urine and fecal samples from the [14C]EP clinical study were prepared as scaled-up versions of those used in the radiochromatographic profiling. Repetitive injections were made on the HPLC system while 30-s fractions were collected from the column eluate. When sufficient quantities of each metabolite peak were collected, each metabolite was profiled on the above HPLC system with radiochemical detection to check for purity. The metabolites were further purified on the HPLC system with fraction collection, as necessary, for analysis by LC/MS-MS and NMR.
The LC/MS-MS conditions were designed to determine the mol. wt. and the unique fragmentation pattern of each metabolite. The HP1050 LC system employed a 3.5-µm Zorbax SB-C18 analytical column (150 x 3.0 mm) (Agilent Technologies) at 50°C. A linear gradient was used, consisting of 5% aqueous methanol, 12.5 mM ammonium acetate to 95% aqueous methanol, 12.5 mM ammonium acetate over 30 min followed by a 10-min hold. LC/MS-MS data were obtained on a Finnigan TSQ700 MS system using the Finnigan APCI source and interface. The APCI source and interface were operated with the following conditions: the sprayer temperature was 500°C, the capillary temperature was 200°C, the sheath flow was 10 ml/min N2, the nebulizing gas pressure was 80 psi, and the source offset was -15 V. Positive-ion LC/MS-MS spectra were recorded using an argon collision gas pressure of 0.4 mTorr and a collision-energy of 30 V. Also, electron impact mass spectra for each metabolite were obtained by desorption probe induction on a Finnigan MAT8430 mass spectrometry system at 70 eV and a 250°C source temperature.
The NMR spectra of the fecal metabolites were determined using solutions in D3-acetonitrile, a 2.5-mm triple gradient indirect detection probe, and a Bruker DRX-600 MHz instrument (Bruker Biospin; Bruker Instruments Inc., Billerica, MA). The NMR spectra of the urinary metabolites were determined using solutions in D3-acetonitrile placed in 2.5-mm concentric tubes, a 5-mm single-gradient indirect detection probe, and a Bruker AMX-500 MHz instrument. The samples were placed under high vacuum before dissolving in D3-acetonitrile. The assignment of all the proton and carbon-13 chemical shifts for the parent drug, EP, in CD3CN solution were made by analyzing the one-dimensional proton and carbon-13 NMR spectra, and the two-dimensional COSY, heterocorrelation, transverse rotating frame nuclear Overhauser enhancement spectroscopy, heteronuclear multiple quantum coherence spectroscopy, and heteronuclear multiple-bond correlation spectroscopy data. The structures of the metabolites are based on the mass spectral data, on the changes in the proton NMR signals as determined by direct comparison with the corresponding signals observed for EP, and on the COSY data.
Pharmcokinetic Analysis. The observed peak plasma concentrations (Cmax) of total radioactivity, EP, and EPA, and time to reach peak plasma concentrations (Tmax) were obtained for each subject. The areas under the plasma concentration-time curve (AUCs) were calculated using the linear trapezoidal rule. Concentration-time curves of EP and EPA in plasma and amounts of EP and EPA excreted in urine after oral administration were simultaneously fit according to the pharmacokinetic model (Fig. 2) using the SAAM II computer program (SAAM Institute, Seattle, WA). The differential equations for the proposed pharmacokinetic model are given in eqs. 2 through 6.
 | (2) |
 | (3) |
 | (4) |
 | (5) |
 | (6) |
Upon integration of the above equations, plasma concentrations of EP (CB) and EPA (CC) can be described by eqs. 7 and 8, respectively, and the amounts of EP (D) and EPA (E) in the urine can be described by eqs. 9 and 10, respectively.
 | (7) |
 | (8) |
 | (9) |
 | (10) |
where k = k20 + k23 + k24.
|
Results
|
|---|
EP Concentrations and Pharmacokinetics. After oral administration of [14C]EP at a dose of 100 mg per person, mean Cmax and AUC values of total radioactivity in plasma were higher than those in whole blood (Table 1). Over time, the blood to plasma ratio of total radioactivity was constant, ranging from 0.70 to 0.78 (Fig. 3). These results suggest that binding of total drug-related materials to plasma is greater than red blood cell partitioning, and total radioactivity was not preferably partitioned into red blood cells. The Cmax values of total radioactivity in saliva were similar to those in plasma, whereas saliva Tmax was much shorter than that in plasma. However, Tmax and Cmax are predominantly influenced by residual radioactivity in the mouth after administration of the radioactive dose solution. Saliva to plasma and saliva to blood ratios of total radioactivity were 0.35 to 0.42 and 0.45 to 0.52, respectively, and were relatively constant 2 h or later after the oral dose. Low saliva concentrations of total radioactivity, relative to plasma or blood concentrations, were most likely due to the equilibrium of only unbound drug and metabolites between blood and saliva.
View this table:
[in this window]
[in a new window]
|
TABLE 1 Noncompartmental pharmacokinetic parameters for total radioactivity, EP, and EPA after oral administration of [14C] EP
|
|
View larger version (20K):
[in this window]
[in a new window]
|
FIG. 3. Concentrations of total radioactivity in plasma, whole blood, and saliva following a single 100-mg oral dose of [14C]EP.
|
|
Mean (±S.D.) Cmax of EP was 1.72 ± 0.28 µg/ml, and the mean Tmax was 1.3 ± 0.8 h (Table 1). The corresponding values for EPA were 0.083 ± 0.017 µg/ml and 1.1 ± 0.3 h, respectively. The mean AUC for EP and EPA were 9.54 ± 3.20 and 0.352 ± 0.115 µg · hr/ml, respectively. Plasma concentration and amounts of EP and EPA excreted in urine were simultaneously fitted according to the pharmacokinetic model presented in Fig. 2. The mean experimental and simulated values for EP and EPA are shown in Fig. 4. The mean t1/2 values for absorption (k12) phase and total elimination (k, sum of k20, k23, and k24) phase were 3.01 and 0.26 h, respectively (Table 2). Thus, the apparent elimination half-life of 3 h appears to be due to slow absorption of the drug. EPA was much more rapidly eliminated in urine compared with EP, and its elimination rate constant (k35) was approximately 180-fold greater than that of EP (k24).
View larger version (14K):
[in this window]
[in a new window]
|
FIG. 4. Plasma concentrations of EP and EPA (A) and cumulative percentage of dose excreted in urine as EP and EPA (B).
Circles indicate EP. Triangles indicate EPA. Data points are actual mean values ± standard error of the mean. Solid lines are the model-simulated curves.
|
|
Plasma Protein Binding. The plasma protein binding of EP was moderate but concentration-dependent over the concentration range of 0.02 µg/ml to 60 µg/ml. At concentrations of 0.02 µg/ml, 0.2 µg/ml, 1.0 µg/ml, 5.0 µg/ml, and 60 µg/ml, the percentages of [14C]EP bound were 60.6 ± 5.2%, 59.0 ± 6.2%, 38.2 ± 4.8%, 33.3 ± 4.2%, and 16.8 ± 2.4%, respectively. The percentage of protein binding in the concentration range at the therapeutic dose (0.02-5.0 µg/ml) was approximately 60 to 33%. However, in the clinical study, the percentage bound of total radioactivity at 1.5 h postdose (mean concentration of 2.36 µg Eq/ml) was 49.6 ± 12.7%. Since the majority (>90%) of total radioactivity in these plasma samples was due to the parent drug, this value is expected to reflect plasma protein binding of [14C]EP in these subjects. Therefore, the plasma protein binding data obtained in the clinical study appeared to be slightly higher than those obtained in the in vitro study. The exact reason for this discrepancy is not known. However, one possibility is that the extent of binding of EP was different among the individuals. The other possibility is that the protein binding of metabolites was higher than that of EP, although the metabolites were present in much smaller amounts.
Metabolic Profiles. The HPLC chromatograms of plasma (acetonitrile extracts), urine, and feces (acetonitrile extracts) with acidification are shown in Fig. 5. When plasma, urine, and fecal samples were acidified, the percentages associated with the parent drug increased slightly compared with those without acidification (data not shown). This was due to cyclization of the lactone ring open form, EPA, under mildly acidic conditions. The majority of radioactivity in 1.5-h plasma was due to the parent drug. In plasma, the 6ß-hydroxy metabolite was the major metabolite, and small amounts of 6ß,21-hydroxy, 3,6ß-hydroxy, and 21-hydroxy metabolites were present. In urine, 6ß-OHEP and 6ß,21-OHEP were major metabolites. Small amounts of the 3,6ß,21-hydroxy, 6ß,15-hydroxy, 3,6ß-hydroxy, and 21-hydroxy metabolites, and the parent drug were also present in the urine. In feces, more than 11 metabolite peaks were observed. The highest radioactive peak was associated with 6ß,21-OHEP. The other radioactive peaks present in feces were 3ß,6ß,21-OHEP, 2,3ß,6ß-OHEP, 3,6ß,21-OHEP, 6ß,15-OHEP, 3ß,6ß-OHEP, 3,6ß-OHEP, 6ß-OHEP, 21-OHEP, and EP. Details of identification of these metabolites are described below.
View larger version (15K):
[in this window]
[in a new window]
|
FIG. 5. Metabolic profiles of [14C]EP in 1.5-h plasma (A), 0- to 48-h urine (B), and 0- to 96-h feces (C) after acidification.
Peak 2: 3ß,6ß,21-OHEP. Peak 3: 2,3ß,6ß-OHEP. Peak 4: 3,6ß,21-OHEP. Peak 5: 6ß,15-OHEP. Peak 6: 6ß,21-OHEP. Peak 7: 3ß,6ß-OHEP. Peak 8: 3,6ß-OHEP. Peak 9: 6ß-OHEP. Peak 10: 21-OHEP. Peak 12: EP. Peaks 1 and 11: not identified.
|
|
Identification of Metabolites. The structures proposed for each of the metabolite peaks in urinary and fecal profiles are shown in Fig. 6. Table 3 shows the HPLC radiochromatographic retention time, mass spectral, and 1H NMR data of EP and its metabolites. Electron impact mass spectra (not shown) gave intense molecular ions for all metabolites except peak 6. LC/MS-MS using APCI gave either the ammonium adduct, MNH4+, the protonated molecular ion, MH+, or both. From these data the mol. wt. was obtained for each metabolite (Table 3 for APCI). In this series of metabolites, the mol. wt. permitted the immediate assessment of the extent and nature of the metabolism. The general approach that was used in establishing the structures of the metabolites shown in Fig. 6 is briefly described below. The proposed structures of all metabolites were fully consistent with the mass spectral data and all proton NMR data.
View this table:
[in this window]
[in a new window]
|
TABLE 3 HPLC radiochromatographic LC/MS-MS, and 600-MHz NMR characteristics for EP and its metabolites identified in human 0-48 hour urine and 0-96 hour feces
|
|
The NMR spectra of all metabolites showed the same pattern for the C-11 and C-12 protons, thus indicating that the epoxide group was unchanged. The NMR spectrum of all the metabolites had a singlet at about 3.6 ppm that was assigned to the methyl ester protons, indicating that all the metabolites retained the methyl ester.
Hydroxylation at the C-6ß position was indicated by the appearance of a doublet near 4.5 ppm (J = 2.1) and changes in the patterns due to the C-7 and C-8 protons. In those compounds in which the C-3 ketone was not reduced, hydroxylation at C-6ß also caused a significant downfield shift of the C-19 methyl proton signal. In this respect, peaks 2, 3, 4, 5, 6, 7, 8, and 9 were identified as having a C-6ß-hydroxyl group. Peak 9 (mol. wt. of 430) has simply a 6ß-hydroxyl group added to the structure of EP. Therefore, peak 9 was identified as 6ß-OHEP.
Hydroxylation at C-21 caused a decrease in the chemical shift difference between the two C-12 protons and the appearance of a single proton doublet of doublets near 4.4 ppm (J = 10, 8.8 Hz). Based on these criteria, peaks 2, 4, 6, and 10 were assigned as C-21 hydroxylated derivatives. Peak 10 (mol. wt. of 430) has simply a 21-hydroxyl group added to the structure of EP. Therefore, peak 10 was identified as 21-OHEP. Peak 6 (mol. wt. of 446) has both a 6ß-hydroxyl group and a 21-hydroxyl group added. Therefore, peak 6 was identified as 6ß,21-OHEP. However, the proton spectrum of peak 6 had two signals (0.928 and 0.902 ppm) in the region expected for a single C-18 methyl signal, and an apparent quartet rather than a triplet in the 4.4 ppm region. The apparent quartet in the 4.4 ppm region was, in fact, composed of two overlapping triplets. These observations were interpreted as showing that peak 6 consisted of a mixture of the two C-21 stereoisomers.
The C-4H signal of several C-6ß-hydroxyl metabolites was split either by a 2.0-Hz (peaks 2, 3, and 7) or a 4.5-Hz (peaks 4 and 8) coupling with C-3H. These couplings and the additional two hydrogen atoms demonstrate that the C-3 carbonyl in each of these metabolites has been reduced to the hydroxyl group. In the case of the 2.0-Hz coupling, the hydroxyl group was determined to be in the ß, or equatorial, configuration. In the case of the 4.5-Hz coupling, the hydroxyl group was determined to be in the , or axial, configuration. Peaks 7 and 8 (mol. wt. of 432) each have a 6ß-hydroxyl group as well as the reduced C-3 carbonyl group with isomeric configuration at the C-3 position. Thus, these peaks were identified as 3ß,6ß-OHEP and 3,6ß-OHEP, respectively. Peaks 2 and 4 (mol. wt. of 448) each had C-6ß-hydroxyl and C-21 hydroxyl groups as well as the reduced C-3 carbonyl group with isomeric configuration at the C-3 position. Thus, peaks 2 and 4 were identified as 3ß,6ß,21-OHEP and 3,6ß,21-OHEP, respectively.
Peak 3 had a C-6ß hydroxyl group and a C-3ß hydroxyl group, as well as an additional hydroxyl group as shown by an MNH4+ peak at m/z 466 and the overlapping doublet of doublets of doublets near 3.4 ppm in the NMR spectrum. The position of the additional hydroxyl group could be deduced by noting in the COSY that the proton with the signal near 3.4 ppm is coupled to the C-3 proton, and hence must be located at C-2. The ß configuration of the C-2 hydroxyl group was established by the large coupling (9.7 Hz) between the C-2H and C-1H. Therefore, peak 3 was identified as 2,3ß,6ß-OHEP.
Peak 5 has a C-6ß hydroxyl group and an additional hydroxyl group, resulting in a molecular weight of 446. The patterns of C-1H, C-6H, and C-11H and the chemical shift difference between the C-12Hs were unchanged from the patterns observed for EP. Therefore, the additional hydroxylation cannot be in rings A, B, D, or E. The remaining possibilities were for the hydroxyl group to be located at C-15 or C-16. The signal due to the proton attached to the carbon bearing the second hydroxyl group was split into an overlapping doublet of doublet of doublets with coupling constants of 9.5, 7.4, and 2.4 Hz. This means that there probably are at least three protons on the adjacent carbon atoms. The two large couplings require that the proton on the hydroxylated carbon is pseudoaxial; thus, peak 5 was identified as 6ß,15-OHEP.
Finally, the structure of peak 12 was shown by the HPLC retention time and mass spectral data to be the parent compound, EP. The quantities of peak 1 and 11 isolated from fecal samples were too small to permit determination of masses by LC-MS/MS or structures by NMR.
Excretion. The mean (±S.D.) percentages of the dose excreted as total radioactivity in urine and feces were 66.6 ± 3.11% and 32.0 ± 3.68%, respectively (Fig. 7). The majority of urinary radioactivity was excreted within the first 24 h, indicating rapid elimination of the drug and metabolites. Although plasma concentrations of EPA were lower than EP concentrations (Fig. 4), mean EPA excreted in urine (4.98 ± 1.50%) was greater than that of EP (1.65 ± 0.28%). Approximately 28% of the dose was excreted as 6ß-OHEP and 13% of the dose was excreted as 6ß,21-OHEP in urine (Table 4). The percentages of all the other metabolites excreted in urine were approximately 5% of the dose or less. The percentage of the dose excreted as total EP (EP plus EPA) in urine was 6.8% using radioactive assay, and this was consistent with the value obtained from the LC-MS/MS assay (6.6%). The percentages of EP and EPA in feces were 0.81 ± 0.44 and 1.65 ± 0.77%, respectively. The percentages of dose excreted as 3,6ß,21-OHEP, 6ß,21-OHEP, and 6ß-OHEP in feces were 3.42, 7.41, and 3.72%, respectively. The percentages of other metabolites were less than 3%. In the breath, there was no measurable amount of 14CO2 found, indicating that the radiolabeled carbon was metabolically stable and the carboxylic group was intact.
View larger version (17K):
[in this window]
[in a new window]
|
FIG. 7. Cumulative percentage of radioactive dose excreted in urine and feces following a single 100-mg oral dose of [14C]EP.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 4 Percentage of dose excreted as various metabolites identified in 0 to 48-h urine and 0- to 96-h feces following an oral dose of [14C]EP to humansab
|
|
|
Discussion
|
|---|
The present study demonstrates that EP was well absorbed after oral administration, with a mean Cmax value of 1.72 µg/ml achieved at 1.2 h. Good absorption of EP in humans is evidenced by the fact that approximately 67% of the administered dose was excreted in urine. EP is highly permeable with Caco-2 cell permeability of approximately 24 x 10-6 cm/s and has a reasonable aqueous solubility of approximately 0.5 mg/ml. EP was quantitatively absorbed in the dog after oral administration (Cook and Zhang, 2000). Furthermore, bioavailabilities of EP after duodenal, jejunal, colonic, and rectal administration in the dog were similar to that after oral administration. Therefore, as in the dog, good absorption of EP in humans appears to be the result of good absorption of the drug throughout the gastrointestinal tract. Following the Cmax, plasma concentrations of EP declined with a mean half-life of 3.0 h. However, detailed pharmacokinetic analysis revealed that EP was rapidly metabolized with a total elimination half-life (t1/2 - k) of approximately 0.26 h (Table 2). The apparent elimination phase, with a 3.0-h half-life, was due to the fact that absorption was slower than total elimination, demonstrating "flipflop" pharmacokinetic phenomena. Since the drug is absorbed in all parts of the gastrointestinal tract and flip-flop pharmacokinetic phenomena were evident even after oral administration of an aqueous dose solution, EP is expected to be a good candidate for an extended formulation to prolong clinically relevant blood levels.
Plasma protein binding of EP was moderate. However, the binding was concentration-dependent, ranging from 60% to 33% in the concentration range after a therapeutic dose (100 mg) administration. Because plasma protein binding was not extensive and the concentration dependence was not dramatic, any clinically significant dose-dependent pharmacokinetics in humans, due to concentration-dependent plasma protein binding, is not expected over the dose range used in the clinic.
EP is slowly converted to its open-form, EPA, in the body, and plasma concentrations of EPA were much lower than EP concentrations. The mean AUC of EPA was approximately 4% of the EP AUC. This is in contrast to the findings in the rat and dog. Plasma concentrations of EPA in the rat were similar to that of EP (Cook et al., 2003). In the dog, plasma concentrations of EPA were lower than that of EP; however, the mean AUC of EPA was approximately 55% of EP AUC (Cook and Zhang, 2000). These notable species differences in the ratios of EPA/EP suggest that hydrolysis of the lactone ring after EP administration was mediated by an enzyme or a group of enzymes rather than by a chemical process. This hypothesis is further supported by the species differences in the in vitro stability of the lactone ring in plasma (C. S. Cook, L. M. Berry, and J. Zhang, unpublished data). In these experiments, EP was more stable in human plasma than in the rat and dog plasma. It has been reported that lactone rings in statins (mevastatin, lovastatin, and simvastatin) and spironolactone are opened in the liver and plasma by enzymatic hydrolysis with esterases to form the hydroxy acid (Wittztum, 1996; Billecke et al., 2000; Teiber et al., 2003).
Nonhydrolytic metabolic pathways of EP include hydroxylation and 3-keto reduction. Hydroxylation occurred at C-2, C-6, C-15, and C-21. However, all the metabolites except one contained 6ß-hydroxyl, indicating that 6ß-hydroxylation is the major metabolic pathway for EP. In human liver microsomes, 6ß-hydroxylation and 21-hydroxylation were the major metabolic pathways, and the products were formed primarily by the CYP3A4 enzyme (Cook et al., 2002). Metabolic pathways such as 6ß-hydroxylation (Yamazaki and Shimada, 1997) and 3-keto reduction (Jacobi and Wilson, 1977; Chanoine et al., 1987; Kammerer et al., 1990) are very common for steroid molecules, and it is not surprising to find that EP is also metabolized via these routes. However, interestingly, there was no evidence for any alteration of the epoxide ring or the methyl ester function. Stability of the epoxide or methyl ester of EP may be due to steric hindrance of the steroid molecule at the 7-, 9-, and 11-positions. The stability of the epoxide ring is important for the selectivity of EP for aldosterone receptor as compared with spironolactone.
EP was extensively metabolized, and only 6.8% of the dose was excreted in urine and feces as the parent. The majority of the total radioactivity was excreted in urine, accounting for 66.6% of the dose. However, some metabolites (3ß,6ß,21-OHEP, 2,3ß,6ß-OHEP, and 3ß,6ß-OHEP) were excreted only in feces, and the number of metabolites excreted in feces was greater than that in urine.
In summary, EP was well absorbed but rapidly eliminated by extensive metabolism. Plasma protein binding was moderate but concentration-dependent over the therapeutic concentration range. The major metabolic pathways were 6ß- and/or 21-hydroxylation and 3-keto reduction. There was no evidence of any alteration of the epoxide ring or methyl ester by metabolism. The majority of the radioactive dose (approximately 67%), as either metabolites or the parent drug, was excreted in urine.
|
Acknowledgments
|
|---|
We thank Aziz Karim, Barbara Roniker, Thomas S. Burns, and Barbara W. L. Nelson for work on the clinical portion and Mary E. Lonien for contribution to the pharmacokinetic portion of this study.
|
Footnotes
|
|---|
1 Abbreviations used are: EP, eplerenone; OHEP, hydroxy-eplerenone; HIV, human immunodeficiency virus; LSC, liquid scintillation counting; EPA, open-lactone ring (hydrolyzed) form of eplerenone; LC/MS-MS, liquid chromatography/tandem mass spectrometry; HPLC, high performance liquid chromatography; NMR, nuclear magnetic resonance spectroscopy; APCI, atmospheric pressure chemical ionization; COSY, correlation spectroscopy; AUC, area under the plasma-concentration time curve. 
Address correspondence to: Dr. Chyung S. Cook, Baxter Healthcare, Life Sciences, Route 120 and Wilson Rd., Round Lake, IL 60073. E-mail: chyung_cook{at}baxter.com
|
References
|
|---|
Billecke S, Draganov D, Counsell R, Stetson P, Watson C, Hsu C, and La Du BN (2000) Human serum paraoxonase (PON1) isozymes Q and R hydrolyze lactones and cyclic carbonate esters. Drug Metab Dispos 28: 1335-1342.[Abstract/Free Full Text]
Chanoine F, Grenot C, Sellier N, Barrett WE, Thompson RM, Fentiman AF, Nixon JR, Goyer R, and Junien JL (1987) Isolation and identification of major metabolites of tixocortol pivalate in human urine. Drug Metab Dispos 15: 868-876.[Abstract]
Cook CS, Berry LM, Kim DH, Burton EG, Hribar J, and Zhang L (2002) Involvement of CYP3A in the metabolism of eplerenone in humans and dogs: differential metabolism by CYP3A4 and CYP3A5. Drug Metab Dispos 30: 1344-1351.[Abstract/Free Full Text]
Cook CS and Zhang L (2000) Absorption and disposition of a selective aldosterone receptor antagonist, eplerenone, in the dog. Pharm Res (NY) 17: 1426-1431.[CrossRef]
Cook CS, Zhang L, Ames GB, Fischer J, Zhang J, and Levin S (2003) Single- and repeated-dose pharmacokinetics of eplerenone, a selective aldosterone receptor blocker, in rats. Xenobiotica 33: 305-321.[CrossRef][Medline]
Jacobi GH and Wilson JD (1977) Formation of 5alpha-androstane-3alpha, 17beta-diol by normal and hypertrophic human prostate. J Clin Endocrinol Metab 44: 107-115.[Abstract/Free Full Text]
Kammerer RC, Merdink JL, Jagels M, Catlin DH, and Hui KK (1990) Testing for fluoxymesterone (Halotestin) administration to man: identification of urinary metabolites by gas chromatography-mass spectrometry. J Steroid Biochem 36: 659-666.[Medline]
Liew D, Martin J, and Krum H (2003) Eplerenone. Pharmacia. Cur Opin Investig Drugs 4: 316-322.
McMahon EG (2001) Recent studies with eplerenone, a novel selective aldosterone receptor antagonist. Curr Opin Pharmacol 1: 190-196.[CrossRef][Medline]
Pitt B, Remme WJ, Zannad F, Neaton J, Martinez F, Roniker B, Bittman R, Hurley S, Kleiman J, and Gatlin M; Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study Investigators (2003) Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 348: 1309-1321.[Abstract/Free Full Text]
Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky A, and Wittes J (1999) The effect of spironolactone on morbidity and mortality in patients with severe heart failure. N Engl J Med 341: 709-717.[Abstract/Free Full Text]
Sun Y, Ramirez FJA, and Weber KT (1997) Fibrosis of atria and great vessels in response to angiotensin II or aldosterone infusion. Cardiovasc Res 35: 138-147.[Abstract/Free Full Text]
Teiber JF, Draganov DI, and Du BN (2003) Lactonase and lactonizing activities of human serum paraoxygenase (PON1) and rabbit serum PON3. Biochem Pharmacol 66: 887-896.[CrossRef][Medline]
Wittztum JF, Draganov DI, and Du BN (2003) Lactonase and lactonizing activities of human serum para oxygenase (PON1) and rabbit serum PON3. Biochem Pharmacol 66: 887-896.
Wittztum JL (1996) Drugs used in the treatment of hyperlipoproteinemias, in The Pharmacological Basis of Therapeutics, 9th ed. (Hardman JG and Limbird LE, eds), pp 875-897, The McGraw-Hill Companies, Inc., New York.
Yamazaki H and Shimada T (1997) Progesterone and testosterone hydroxylation by cytochromes P450 2C19, 2C9 and 3A4 in human liver microsomes. Arch Biochem Biophys 346: 161-169.[CrossRef][Medline]
Zhang JY, Fast DM, and Breau AP (2003) Development and validation of a liquid chromatography-tandem mass spectrometric assay for eplerenone and its hydrolyzed metabolite in human plasma. J Chromatogr B 787: 333-344.
This article has been cited by other articles:
|
|
|
|
 
T. R. Marcy and T. L. Ripley
Aldosterone antagonists in the treatment of heart failure
Am. J. Health Syst. Pharm.,
January 1, 2006;
63(1):
49 - 58.
[Abstract]
[Full Text]
[PDF]
|
|
|
![[Feedback]](/icons/banner/feedbackACT.gif)
![[For Subscribers]](/icons/banner/subscriberACT.gif)
![[Archive]](/icons/banner/archiveACT.gif)
![[Search]](/icons/banner/searchACT.gif)
![[Contents]](/icons/banner/tocACT.gif)
Top
Abstract
Materials and Methods
