Biliary secretion of rosuvastatin and bile acids in humans during the absorption phase

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/ejps

Biliary secretion of rosuvastatin and bile acids in humans during the absorption phase Ebba Bergman a , Patrik Forsell b , Annica Tevell c , Eva M. Persson a , Mikael Hedeland c , ¨ a,∗ Ulf Bondesson c,d , Lars Knutson b , Hans Lennernas a

Department of Pharmacy, Uppsala University, P.O. Box 580, SE-751 23 Uppsala, Sweden Department of Surgery, University Hospital, 751 85 Uppsala, Sweden c Department of Chemistry, National Veterinary Institute, Uppsala, Sweden d Division of Analytical Pharmaceutical Chemistry, Uppsala University, P.O. Box 574, 751 23 Uppsala, Sweden b

a r t i c l e

i n f o

a b s t r a c t

Article history:

Aim: The aim of this study was to investigate the biliary secretion of rosuvastatin in healthy

Received 11 April 2006

volunteers using an intestinal perfusion method after administration of 10 mg rosuvastatin

Accepted 25 April 2006

dispersion in the intestine.

Published on line 12 May 2006

Methods: The Loc-I-Gut tube was positioned in the distal duodenum/proximal jejunum and a semi-open segment was created by inflating the proximal balloon in ten volunteers. A

Keywords:

dispersion of 10 mg rosuvastatin was administered below the inflated balloon and bile was

Rosuvastatin

collected proximally of the inflated balloon. Bile and plasma samples were withdrawn every

Intestinal intubation

20 min during a 4 h period (absorption phase) and additional plasma samples were collected

Biliary secretion

24 and 48 h post-dose.

Absorption

Results: The study showed that there is a substantial and immediate transport of rosuvastatin into the human bile, with the maximum concentration appearing 42 min after dosing, 39,000 ± 31,000 ng/ml. Approximately 11% of the administered intestinal dose was recovered in the bile after 240 min. At all time points the biliary concentration exceeded the plasma concentration, and the average bile to plasma ratio was 5200 ± 9200 (range 89–33,900, median 2000). We were unable to identify any bile-specific metabolites of rosuvastatin in the present study. Conclusion: Rosuvastatin is excreted via the biliary route in humans, and the transport and accumulation of rosuvastatin in bile compared to that in plasma is rapid and extensive. This intestinal perfusion technique offers a successful way to estimate the biliary secretion for drugs, metabolites and endogenous substances during the absorption phase in healthy volunteers. © 2006 Elsevier B.V. All rights reserved.

1.

Introduction

The impact of carrier-mediated membrane transport of drugs on their absorption, first-pass liver extraction, and disposition is not well established in vivo owing to the complexity asso-

∗

Corresponding author. Tel.: +46 18 471 43 17; fax: +46 18 471 42 23. ¨ E-mail address: [email protected] (H. Lennernas).

0928-0987/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2006.04.015

ciated with it and relevant in vivo based data are therefore often lacking, as a result of which one is generally restricted to in vitro based data (Meijer and Lennernas, 2005). Furthermore, an increasing awareness of drug induced hepatotoxicity has resulted in a focus on investigating local liver exposure of

206

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

drugs and metabolites as an important part of drug development (Lee, 2003). For instance, a population-based study performed in France indicated an underestimation of the reports of hepatic adverse drug events (hepatic ADR) by the regulatory authorities of 16 times, and the incidence rate of hepatic ADR was 13.9 in 100,000 habitants per year (Sgro et al., 2002). Consequently, a model enabling direct determinations to be made of the in vivo kinetics of intestinal and hepatobiliary transport and the metabolism of drugs would be a valuable tool. The use of such a model could increase our understanding of the mechanisms underlying the processes determining the bioavailability, local liver exposure and drug–drug interactions, and therefore have an impact on pharmacokinetics and the safety evaluation of drugs. In the present paper we have studied an in vivo approach to the direct collection of human bile in the intestine during the intestinal absorption phase in healthy subjects. Rosuvastatin (bis[(E)-7-[4-(4-fluorophenyl)-6-isopropyl-2[methyl(methylsulfonyl)amino] pyrimidin-5-yl](3R,5S)-3,5dihydroxyhept-6-enoic acid [calcium salt]], molecular weight 1001.14 g/mol (AstraZeneca, 2003)) is a statin and reduces blood cholesterol by inhibition of 3-hydroxy-3-methylglutaryl coenzyme A (HMG Co A) reductase in the hepatocyte. To reduce the systemic effects of statins, a high liver extraction, enterohepatic circulation and a low degree of metabolism are considered to be preferable drug properties and rat studies have shown that rosuvastatin is selectively distributed into the liver (Nezasa et al., 2002). The absolute oral bioavailability and hepatic extraction is estimated to be 20.1% and 63%, respectively, despite a low degree of metabolism both in vitro and in vivo (Fujino et al., 2004; Martin et al., 2003b,d). After oral administration of [14 C]rosuvastatin, 93% of the dose was recovered in the faeces, mainly in the form of the parent compound (76.8%), and the remaining 10% of the dose was recovered in the urine (Martin et al., 2003d). Previously published studies showing multiple peaks in the plasma profiles have suggested that rosuvastatin undergoes enterohepatic circulation, but the in vivo biliary secretion of rosuvastatin has not been previously investigated in humans (Martin et al., 2003b,c). Drug–drug interaction studies have shown a significant increase in the bioavailability, but not in the plasma half-life of rosuvastatin after co-administration with gemfibrozil and cyclosporine (Schneck et al., 2004; Simonson et al., 2004). However, modest to no interaction have been shown in vivo after co-administration with fenofibric acid (CYP3A4) (Martin et al., 2003a), erythromycin (CYP3A4) (Cooper et al., 2003a), digoxin (ABCB1) (Martin et al., 2002b), itraconazole (CYP3A4, ABCB1) (Cooper et al., 2003c), ketoconazole (CYP3A4, ABCB1) (Cooper et al., 2003b) and fluconazole (CYP2C9, CYP2C19) (Cooper et al., 2002). It has been suggested that the interactions with gemfibrozil and cyclosporine might be mediated by inhibiting OATP1B1 (OATP-C) that is considered to be the main uptake mechanism for rosuvastatin into the liver (Brown et al., 2001; Schneck et al., 2004; Simonson et al., 2004). Further studies have also indicated that rosuvastatin is a substrate for ABCG2 (BCRP, Km 14 ␮M) (AAPS meeting, 7–9 March, 2005). The more extensive effect of cyclosporine on the pharmacokinetics of rosuvastatin than that of gemfibrozil might be explained by dual inhibition by OATP1B1 mediated liver uptake and ABCG2

mediated biliary secretion by cyclosporine (Qadir et al., 2005; Schneck et al., 2004; Simonson et al., 2004). Taken together, the results emphasize the need to investigate biliary secretion of rosuvastatin during the absorption phase. The main aim of this study was to investigate the feasibility of an intensive bile sampling method in humans and to characterize the bile secretion of rosuvastatin and bile acids during the absorption phase. In addition, we performed a qualitative investigation of the presence of any bile specific rosuvastatin metabolites.

2.

Material and methods

2.1.

Subjects and study design

The study included five male (subjects 2, 4, 6, 8 and 9) and five female (subjects 1, 3, 5, 7 and 10) volunteers, all of whom were deemed to be healthy by a physician; they were of normal weight and were between 21 and 38 years old (Martin et al., 2002a). The subjects had fasted overnight (at least 10 h) before the start of the experiment, with the exception of subject eight, who had breakfast on the morning of the study day. Subject six received intravenously administered glucose solution (glucose 50 mg/ml, Baxter Medical AB) from the time of the drug administration to 240 min post-dose, at an infusion rate of 200 ml/h. Subject six was administered glucose intravenously since he started the experiment hours later than the other subjects and had therefore a longer time of fasting. Addition of glucose solution is not considered to affect the experiment. All subjects gave their written informed consent to their participation in the investigation. The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Uppsala University. The study was performed at the Clinical Research Department, University Hospital, Uppsala, Sweden.

2.2.

Experimental procedures

The Loc-I-Gut tube (Synectics Medical, Stockholm, Sweden) is a multichannel polyvinyl tube with an external diameter of 5.3 mm, it is 175 cm long with two inflatable balloons attached 10 cm apart and a tungsten weight at the end and is described in details elsewhere (Knutson et al., 1989; Lennernas et al., 1992). The tube was introduced orally and positioned in the distal duodenum/proximal jejunum below the papilla of Vater, using a Teflon-coated guide wire. The insertion of the tube was facilitated by using local anaesthetic to the upper throat with ¨ ¨ lidocain spray (Xylocain® ; AstraZeneca, Sodert alje, Sweden). The position in the jejunum was verified using fluoroscopy (Philips BV 21-S) and, once the tube was in position, the proximal balloon was inflated with 20–25 ml of air to create a semi-open segment. The distal balloon of the Loc-I-Gut tube was not inflated during the experiment. Gastric drainage was performed during the experiment to prevent discomfort for the subjects and to avoid contamination of the collected bile by gastric fluid (Salem sump nasogastric drainage tube: Sherwood Medical, Unilever PLC, London, UK). A 10 mg tablet of rosuvastatin (Crestor® , AstraZeneca) was dissolved in 100 ml

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

saline solution at 37 ◦ C for subjects 1–4 and in 50 ml saline solution at 37 ◦ C for subjects 5–10 (Natriumklorid, 9 mg/ml, Fresenius Kabi, Norway AS). For subjects 5–10 the dispersion vial was rinsed with an additional 100 ml saline solution, 37 ◦ C. The drug dispersion was administered as a bolus dose through the inlet located below the inflated proximal balloon of the tube and bile was collected above the tube’s proximal balloon. The balloon has previously been shown to completely separate the drug administration site and the bile collection site (Bonlokke et al., 1997, 1999, 2001; Ryde and Gustavsson, 1987). To facilitate bile collection, a vacuum pump was connected to the proximal drainage of the Loc-I-Gut tube (Ameda suction pump type 23; Ameda AG, Zug, Switzerland). Bile was collected quantitatively and weighed prior to administering the dose and at 20, 40, 60, 80, 100, 120, 140, 160, 180, 200, 220 and 240 min after the drug administration; it was immediately stored at −20 ◦ C until analysis. Blood samples were collected in tubes containing lithium heparin anticoagulant prior to and at 10, 30, 50, 70, 90, 110, 130, 150, 170, 190, 210, 230 min and 24 and 48 h after drug administration and centrifuged within 30 min of collection at 1000 × g for 10 min. The plasma samples were stored at −80 ◦ C pending analysis.

2.3.

Analytical methods

Calcium salts of rosuvastatin and atorvastatin were purchased from APIN Chemicals Ltd., Abingdon, Oxfordshire, UK. The acetonitrile, methanol, ethyl acetate, dichloromethane, ammonium formate, acetic acid, and sodium acetate were of analytical grade or better, and were used without further purification. The water was purified using a Milli-Q water purification system (Millipore, Bedford, MA, USA). High-performance liquid chromatography (HPLC) was performed with an HP series 1100 HPLC system (Hewlett-Packard, Waldbronn, Germany) with a binary pump, degasser and autosampler. The guard column was an ODS-octadecyl C18 (length 4 mm, i.d. 2.00 mm) and the chromatographic column was a Phenomenex Luna 5 ␮ C18 (length 50 mm, i.d. 2.00 mm, particle size 5 ␮m), both supplied from Scandina¨ ¨ vian GeneTech, Vastra Frolunda, Sweden. The mobile phase consisted of (A) ammonium formate buffer (2 mM ammonium, pH 3.4) and (B) acetonitrile and was adapted as a gradient with 30% B for 2 min, 30–90% B for 3 min and then kept constant at 90% B for 5 min, 90–30% B for 10 s and kept at 30% B for 5 min. The total run time was 15 min, the flow rate was 200 ␮l/min and the injection volume was 20 ␮l. The HPLC column outlet was connected to a Quattro LC quadrupole–hexapole–quadrupole mass spectrometer (Micromass, Manchester, UK) equipped with an electrospray interface (ESI) operated in the positive ion mode. The instruments were controlled using the MassLynx software, version 3.3. The MS parameters were manually optimised for sensitivity during the direct infusion of a rosuvastatin solution. The ESI parameters for the analysis were: capillary voltage 3.7 kV, cone 45 V, extractor 2 V, and RF lens 0.40 V. The desolvation and source block temperatures were 350 and 120 ◦ C, and the nebulizer and desolvation gas flows were 100 and 500 l/h, respectively. When running selected reaction monitoring (SRM) the

207

hexapole collision cell was filled with argon gas at a pressure of 1.0 × 10−3 mbar. The mass transitions used in SRM were m/z 482.1 → 258.1 for rosuvastatin and m/z 559.2 → 440.0 for the internal standard atorvastatin. The dwell time was 0.8 s and the collision energies were 37 and 22 eV for rosuvastatin and atorvastatin, respectively. Stock solutions of rosuvastatin and atorvastatin were made up in methanol at approximately 1 mg/ml (concentration of the free acid of the drug). The solutions were diluted 50 times in methanol and working solutions were then prepared in various concentrations by dilution in sodium acetate buffer (0.1 M, pH 4.0)/methanol (1:1). All stock solutions were stored at 2–8 ◦ C. For plasma, the standard curve concentrations were 0.1; 0.3; 0.5; 1.0; 5.0; 10.0; 15.0 and 20.0 ng/ml plasma prepared by adding 100 ␮l of the standard solution to 1000 ␮l of blank human plasma. For bile, the standard curve concentrations were 2.0; 20.0; 100; 1000; 5000; 10,000; 20,000 ng/ml bile prepared by adding 100 ␮l of the standard solution to 500 ␮l of blank human bile. The quality control (QC) samples were prepared from a separate stock solution of rosuvastatin in the concentrations, 0.1; 1.0 and 15 ng/ml plasma, and 40 and 1500 ng/ml bile. The QC samples were prepared in bulk of 7 and 13 ml at the time, respectively, and were stored at −20 ◦ C until analysis. One blank sample (blank plasma or bile) and one zero sample (blank plasma or bile with the addition of internal standard) was also prepared. To 500 ␮l of the plasma (sample, calibration or QC), 500 ␮l of sodium acetate buffer (0.1 M, pH 4.0) and 50 ␮l of internal standard solution (atorvastatin 11 ng/ml) was added. Extraction was performed with 3 ml of ethyl acetate/dichloromethane (1:1) for 20 min. After centrifugation for 10 min at 3500 rpm, the aqueous phase was removed and the organic phase was evaporated under nitrogen at 50 ◦ C. The residue was dissolved in 50 ␮l of 0.1% acetic acid/acetonitrile (1:1) before analysis. To 250 ␮l of the bile (sample, calibration or QC), 500 ␮l of sodium acetate buffer (0.1 M, pH 4.0) and 50 ␮l of internal standard solution (atorvastatin 6 ␮g/ml) were added. Extraction was performed in the same manner as for the plasma samples. For each run in the analysis, a verification of the method was performed using duplicates of the standard curve samples, with six of each QC sample, as well as one blank and one zero sample were extracted. The limit of detection (LOD), defined as three times the noise level, was 0.3 ng/ml for plasma and 2 ng/ml for bile. The limit of quantification (LOQ), defined as the lowest concentration repeatable with a precision of 20% and an accuracy of 80–120%, was 1.0 ng/ml for plasma and 20 ng/ml for bile. The standard curve was fitted to linearity using the MassLynx software. Residual plots were used to exclude points that diverged by more than 15% (20% for LOQ). The search for bile specific rosuvastatin metabolites was performed manually by interpretation of mass spectra and chromatograms. Bile was mixed with an equal volume of acetic acid buffer (0.1 M, pH 4.1) and centrifuged for 10 min. The supernatant was transferred to an LC vial. The mass spectral information for the samples, blank samples and synthetic standard solution of rosuvastatin was compared by looking at peaks in the total ion chromatogram (TIC) and known mass shifts, for example −14 for demethylation, +16 for hydroxylation, and +176 for glucuronidation. Neutral loss scan of 80

208

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

and 176 were used to recognize conjugations by sulfate and glucuronic acid, respectively.

The apparent biliary clearance, CLapp,bile during the first 4 h after dosing, was estimated using Eq. (3):

2.4.

CLapp,bile =

Bile acid analysis

The bile acid content of the collected bile was analysed by solid phase extraction followed by HPLC, using an evaporation light scattering (ELS) detector (Persson et al., unpublished data). Briefly, purification of bile acids in the bile extract was accomplished using pre-packed C18 columns (Isolute International Sorbent Technology, UK). The bile acids were separated on a Zorbax C18 Extend Column (length 150 mm, i.d. 4.6 mm, particle size 3.5 ␮m, Agilent Technologies, US) using a gradient with two mobile phases containing methanol:buffer (ammoniumacetate 15 mM, 0.2% triethylamine, 0.5% formic acid, pH 3.15). The following bile acids were quantified: cholic acid (CA), lithocholic-3-sulfate (LCA3S), glycocholic acid (GCA), taurocholic acid (TCA), chenodeoxycholic acid (CDCA), lithocholic acid (LCA), glycochenodeoxycholic acid (GCDCA), taurochenodeoxycholic acid (TCDCA), deoxycholic acid (DCA), glycodeoxycholic acid (GDCA) and taurodeoxycholic acid (TDCA). The limit of quantification was 0.1 mM.

2.5.

Adsorption and stability

Studies were conducted to measure the adsorption of rosuvastatin to the Loc-I-Gut tube and to various materials used in the study by conducting a maximal adsorption test at 37 ◦ C and at room temperature. The stability of rosuvastatin in human bile was measured at 37 ◦ C over a period of 90 min.

2.6.

Ae0−240 AUC0−230

(3)

where the accumulated amount of rosuvastatin in bile (Ae0–240 ) was divided by the plasma exposure of rosuvastatin over a period of 230 min (AUC0–230 ). Throughout this paper all data are presented as the arithmetic mean ± standard deviation (S.D.), unless stated otherwise.

3.

Results

The Loc-I-Gut tube was successfully positioned in all ten subjects within approximately 1 h (range 20 min–2 h) (Fig. 1). However, subject nine terminated the experiment directly after drug administration and subject three terminated the experiment after 150 min owing to discomfort caused by the tube. The remaining eight subjects were able to successfully complete the study. The average bile flow was 0.52 ± 0.22 ml/min, but there was high intra-individual variability during the 4 h long sampling period (Fig. 2). The pharmacokinetic variables of rosuvastatin in bile and plasma are presented in Table 1. The fraction of the rosuvastatin dose excreted as unchanged drug in the bile was 10.6 ± 5.1% (range 2.2–16.0%) and the accumulated amount rosuvastatin excreted in bile is presented in Fig. 3. The dose recovered in the bile was 11.8 ± 5.4% when corrected for the extrapolated area (Table 1).

Pharmacokinetic data

The plasma and bile concentration–time data were analysed using non-compartmental methods WinNonlin Professional Version 4.0; Pharsight Corp., Mountain View, CA, USA. The areas under the plasma and bile concentration–time curves (AUC0–230 and AUC0–240 ) were calculated by the linear/logarithmic trapezoidal rule. The AUC0–240 , was then extrapolated to infinity using z to obtain the area under the bile concentration–time curve from zero to infinity, AUC0–∞ . The accumulated amount of rosuvastatin excreted in the bile collected during a 240 min period (Ae0–240 ) is calculated according to Eq. (1):

Ae0−240 =

240 

Cbile × Vbile

(1)

0

where Cbile is the concentration of rosuvastatin in the bile sample collected at time t and Vbile is the volume of the fluid collected at time t. The fraction of administrated rosuvastatin that is excreted in the bile, febile , was calculated according to Eq. (2): febile [%] =

Ae0−240 × 100 Dose

(2)

where the dose is the 10 mg rosuvastatin given as a dispersion in the intestine and has not been corrected for bioavailability.

Fig. 1 – A schematic figure of the Loc-I-Gut tube positioned in the distal duodenum/proximal jejunum and below papilla of Vater. The proximal balloon of the Loc-I-Gut tube is inflated, separating the drug administration site and the bile sampling site and the distal balloon remains un-inflated during the experiment. Gastric drainage is performed throughout the experiment to avoid discomfort for the subject as well as to avoid contamination of the collected bile with gastric fluid.

209

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

Fig. 4 – The correlation between the accumulated amount rosuvastatin secreted in bile and the biliary excretion rate [mg/min] for rosuvastatin in eight healthy subjects.

Fig. 2 – The individual biliary flow [ml/min] in eight healthy volunteers during a period of 240 min of intensive bile sampling in the distal duodenum/proximal jejunum using the Loc-I-Gut tube.

rate of rosuvastatin (Fig. 4). There was significant and rapid secretion of rosuvastatin into bile, with the time taken to reach the maximum bile concentration (Tmax ) being approximately 40 min from the time at which the dose was administered (Fig. 5). The concentration of rosuvastatin in the bile was substantially higher than the corresponding concentration for the plasma throughout the entire 240 min period of bile sampling as shown in Figs. 5 and 6 and Table 1, respectively. For instance, the mean concentration of rosuvastatin in bile collected 0–20 min post-dose was 42,700 ± 42,100 ng/ml and the corresponding mean plasma concentration at 10 min was 1.80 ± 0.60 ng/ml. The apparent biliary clearance (CLapp,bile ) was 82.3 ± 69.1 l/h (range 8.1–195.2). Plasma samples were intensively collected during a 4 h period, and then 24 and 48 h after dosing. The mean plasma concentration–time profiles and the pharmacokinetics of rosuvastatin in plasma, from 0 to 230 min post-dose, are presented in Fig. 5 and in Table 1. With the sole exception of subject eight, the plasma concentrations at 24 and 48 h were all below detection level, i.e., <0.3 ng/ml. The kinetic variables of each bile acid collected in the aspirate are, respectively, presented in Table 2. The concentration–time profile of total bile acids in the human bile during 4 h is illustrated in Fig. 7. The concentration of CA, LCA3S, LCA, CDCA and DCA were below LOQ.

Fig. 3 – The accumulated amount rosuvastatin secreted in bile over time following a 10 mg intestinal dose of rosuvastatin and sampling of bile during 240 min post-dose in eight healthy subjects.

There was no correlation between the individual excretion rate of rosuvastatin and the individual biliary flow. However, there was a correlation (r2 = 0.67) between the accumulated amount rosuvastatin excreted in the bile and the excretion

Table 1 – Pharmacokinetic variables for rosuvastatin in plasma and bile presented as the arithmetic mean ± S.D. (range) Pharmacokinetic variables Tmax [h] Cmax [ng/ml] t1/2 [h] AUC0−t [h ng/ml] AUC0−∞ [h ng/ml] Area extrapolated [%] Biliary flow [ml/min] febile [%] Corrected febile [%] a

The variables are presented as median value (range).

Plasma a

3.54 (1.87–3.92) 11.93 ± 9.56 (2.90–28.00) – 18.98 ± 11.35 (5.73–36.30) – – – – –

Bile a

0.70 (0.32–2.00) 39000 ± 31000 (5200–99000) 1.05 ± 0.74 (0.33–2.68) 39000 ± 22000 (10500–74000) 44000 ± 23000 (17000–80000) 13.04 ± 12.91 (0.01–37.97) 0.52 ± 0.22 (0.20–1.06) 10.6 ± 5.1 11.8 ± 5.4

210

Kinetic variables TCA (N = 6) GCA (N = 6) TCDCA (N = 6) TDCA (N = 6) GCDCA (N = 7) GDCA (N = 6)

Cmax [mM]

z [h−1 ]

AUC0−t [h mM]

AUC0−∞ [h mM]

Area extrapolated [%]

2.07 ± 1.66 (0.29–4.98) 3.77 ± 3.04 (1.92–9.85) 0.96 ± 0.57 (0.21–1.57) 0.63 ± 0.66 (0.06–1.82) 2.74 ± 3.93 (0.41–11.6) 1.08 ± 1.01 (0.32–2.73)

0.94 ± 0.44 (0.51–1.46) 0.63 ± 0.19 (0.47–0.96) 0.94 ± 0.52 (0.46–1.62) 1.21 ± 0.52b (0.56–2.00) 1.70 ± 1.04 (0.32–3.04) 1.74 ± 0.91 (0.27–2.70)

3.26 ± 1.70 (0.56–5.16) 5.88 ± 1.82 (4.25–8.23) 1.87 ± 0.99 (0.53–3.27) 0.87 ± 0.71 (0.15–2.13) 2.28 ± 1.98 (0.60–6.31) 1.10 ± 0.49 (0.66–1.76)

3.83 ± 2.02 (0.67–6.42) 7.52 ± 2.41 (4.88–11.8) 2.13 ± 1.07 (0.88–3.83) 1.10 ± 0.80b (0.30–2.42) 2.44 ± 1.92 (0.60–6.31) 1.25 ± 0.51 (0.68–1.76)

14.8 ± 7.88 (3.42–26.4) 21.2 ± 10.0 (7.08–33.7) 15.1 ± 13.2 (2.39–39.4) 6.79 ± 3.26b (3.60–11.9) 8.56 ± 15.1 (0.13–42.2) 9.45 ± 18.8 (0.39–47.7)

Tmax [h] 1.83a 1.83a 2.17a 1.00a 2.33a 1.00a

(0.67–3.33) (0.00–3.33) (1.00–3.33) (0.00–3.33) (0.00–3.33) (0.00–2.67)

The bile collected was analysed for the content of the following bile acids: taurocholic acid (TCA), glycocholic acid (GCA), taurochenodeoxycholic acid (TCDCA), taurodeoxycholic acid (TDCA), glycochenodeoxycholic acid (GCDCA), glycodeoxycholic acid (GDCA), cholic acid (CA), lithocholic acid 3-sulfate (LCA3S), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA) and lithocholic acid (LCA). For the following bile acids was the concentration of the collected bile samples below detection level; CA, LCA3S, CDCA, DCA and LCA, respectively. a b

The variables are presented as median value (range). N = 5.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

Fig. 5 – The concentration–time profile of rosuvastatin in plasma and bile presented as the arithmetic mean ± S.D. in eight healthy subjects following a single dose of 10 mg rosuvastatin in the intestine.

Fig. 6 – The ratio of the concentration of rosuvastatin in bile and plasma during the absorption phase following a single dose of 10 mg rosuvastatin in the intestine (mean ± S.D.).

Fig. 7 – The total bile acid concentration [mM] depicted as a function of time in the bile samples collected from the distal duodenum/proximal jejunum during 20 min intervals over a period of 240 min.

Table 2 – The individual kinetic variables for six bile acids presented as the arithmetic mean ± S.D. (range)

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

4.

Discussion

This technique has proven to be a feasible method for the investigation of the human biliary secretion of drugs, metabolites and endogenous compounds following intestinal administration of drugs. The method enables intensive sampling to be made of the bile in the distal duodenum/proximal jejunum, distal of papilla of Vater, during the absorption phase, i.e., the first 4 h of drug absorption and disposition. Unlike previously existing biliary secretion methods, this modified LocI-Gut technique allows for intestinal drug administration and assessment of liver transport/metabolism during the absorption phase, which is the kinetic phase during which the liver is exposed to the highest possible concentration of drugs and metabolites. The present method is relatively easy to use and is associated with a low degree of discomfort for healthy volunteers, as a result of which it has a high success rate. It has been reported that the liver extraction of rosuvastatin in humans was 63% following a 4 h intravenous infusion of 8 mg of rosuvastatin (Martin et al., 2003b). In the present study, approximately 11% of a 10 mg intestinal dose was excreted in the bile in the unchanged form. Based on a previous report by Martin et al. (2003b), we assume that approximately 50% of the administered intestinal dose was absorbed (i.e., transported into the vena porta from the lumen as unchanged rosuvastatin) and, consequently, approximately 20% (ranging from 11 to 31%) of the absorbed dose was excreted via the biliary route in our study. There are several plausible explanations for the discrepancy between hepatic extraction assessed after intravenous dosing by Martin et al. (2003b) and the amount excreted in bile in our present study. For instance, it is possible that a large part of rosuvastatin is subjected to a hepatic metabolism into not yet identified metabolites, even if we were not able to find any metabolites in the human bile. Secondly, the relatively short sampling time in relation to the elimination half-life of rosuvastatin will result in an underestimation of the dose excreted in the bile. Other plausible contributory factors are incomplete emptying of the gallbladder and/or intra-hepatic storage of rosuvastatin. The biliary AUC0–∞ was 44,000 ± 23,000 ng h/ml and the extrapolated area ranged from 0.01 to 38%, indicating that rosuvastatin is maintained in the gallbladder and/or stored intra-hepatically for some subjects. The data indicates that a high percentage of extrapolated area results in a lower dose recovery, which further suggests the incomplete gallbladder emptying as one of the possible explanations for the differences in dose recovery. However, out of consideration for the subjects, the bile sampling time is limited to a maximum of 4 h. Fourth, it is possible that the drug directly sampled from the intestine using the present clinical method was incompletely recovered. As similar applications of the technique for investigation of in vivo dissolution have reported recovery values above 80%, which makes this a less likely explanation (Bonlokke et al., 1997, 1999, 2001). Fifth, the fraction dose absorbed and the hepatic extraction estimated by Martin et al. (2003b) are approximated values based on a number of assumptions and may therefore be overestimated. Accordingly, the lower recovery of the parent compound in the bile than expected is probably due to a combination of the short sampling time, poor knowledge of the fraction of dose

211

absorbed, formation of unidentifiable metabolites and incomplete gallbladder emptying. In the present study it was shown that there is a substantial and immediate transport of rosuvastatin into the bile. This is evident from the difference between the biliary and plasma concentrations of rosuvastatin; the average biliary to plasma ratio is 5200 ± 9200 and the variation of this ratio is between 89 and 33,900 (median value 2000), as presented in Fig. 6. A further important finding is that the transport into bile is very rapid, with the time taken to attain the maximum concentration (Tmax ) being 42 min (median) after the dose (Table 1). This has not been shown in humans before, but was possible in the present study using this novel intensive bile sampling method during the absorption and first-pass extraction processes. The accumulation of rosuvastatin in human bile compared to plasma is very high. As a comparison, it has been reported that the accumulation of digoxin in the bile was 20 times that of the plasma concentration in healthy subjects (Caldwell and Cline, 1976). It is important to understand that the high concentration of rosuvastatin in the collected intestinal samples is not due to any back leakage of the dose. Administered drug would have to move against the intestinal motility pattern to contaminate the sampling site. This is because the dose is given distally to the inflated balloon and the intestinal motility pattern pushes the fluid forward, making the likelihood of backflow across the Loc-I-Gut balloon insignificant. In several studies, we have shown that the semi-open segment produces a tightly fitting balloon thereby preventing fluid from moving backwards (Persson et al., unpublished data) (Bonlokke et al., 1997, 1999, 2001). Furthermore, an earlier report using a similar bile sampling technique in humans that included methylene blue as a marker substance showed no contamination of the sampling site by the administration site (Ryde and Gustavsson, 1987). The biliary flow displayed a high intra-individual variability and one of the major factors contributing to this high variability is the fact that the bile is emptied into the intestine as pulse doses rather than as constant secretion in the fasted state (Fig. 2) (Ghibellini et al., 2004). During the fasted state the contraction of the gallbladder and the activity of sphincter of Oddi are the major factors determining the rate at which bile enters the duodenum and occurs in a cyclic manner as a part of the migratory motor complex (MMC) (Lanzini et al., 1987, 1989; Qvist et al., 1991; Shaffer, 2000). Ghibellini and co-workers made biliary secretion and gallbladder emptying visual by using technetium and gamma camera showing the individual underlying mechanisms for biliary secretion and gallbladder emptying in four subjects. One of the subjects deposited all the bile produced into the gallbladder and only emptied the gallbladder after intravenous administration of cholecystokinin-8 at 120 min post-dosing, while another subject immediately secreted the bile produced into the jejunum without any intermediate storage in the gallbladder (Ghibellini et al., 2004). The high inter-individual variability in the biliary emptying pattern into the intestine is most likely one reason for the wide variation in the dose recovered in the present study (Fig. 3). A report by Agoston et al. (1973) using bile drainage for 30 h showed a high inter-individual variation in accumulated dose recovery (Agoston et al., 1973). Thus, if we were to increase the bile sampling time and include intravenous administra-

212

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

tion of cholecystokinin-8, the recovered amount of drug in the bile might increase and the inter-individual variability would be reduced. However, intravenous administration of cholecystokinin-8 and stimulation of gallbladder contraction is associated with safety and tolerability issues and discomfort for the subjects, and it may also disrupt the normal biliary flow and was therefore not applied in this study. In the past, biliary secretion studies in man were often performed with cholestasis patients, using tightly-fitting enddrainage in the common bile duct, enabling primary bile to be collected for up to 48 h (Loria et al., 1989; Meijer et al., 1971, 1983, 1988; Westphal et al., 1997; Westra et al., 1981; Wierda et al., 1991). The biliary flow for the initial 4 h after intravenous dosing in patients having a common bile duct tube has been reported to be approximately 16 ml/h, compared to our data in healthy volunteers where the biliary secretion rate is 31 ml/h (Meijer et al., 1983, 1988). It is likely that the effect of the disease state and surgical procedure affect the biliary flow, as well as the composition and concentration of bile acids and, therefore, it is unlikely that these earlier studies reflect the biliary secretion in the normal state (Hofmann, 1976; Roberts et al., 2002; Strasberg et al., 1990). The bile collected when using an intubation technique situated in the distal duodenum/proximal jejunum might be contaminated with intestinal and pancreatic secretion compared to primary bile. This may lead to an overestimation of the biliary secretion rate using intubation techniques, however, the two-fold differences in the biliary secretion rate observed for healthy volunteers and patients can probably be attributed to the disease state and post-surgical processes rather than intestinal and pancreatic secretion (Roberts et al., 2002; Strasberg et al., 1990). Several types of intubation techniques have been used in previous studies to investigate biliary secretion after intravenous dosing, yet in none of these studies was the jejunal sampling segment blocked, enabling biliary secretion to be quantitatively estimated (Arvidsson et al., 1982; Caldwell and Cline, 1976; Ghibellini et al., 2004; Gundert-Remy et al., 1982; Lanzini et al., 1987; Sutfin et al., 1990). The inclusion of an intestinal perfusion solution composed of essential amino acids and glucose to stimulate and preserve tonic gallbladder contraction, thereby obtaining the maximum biliary secretion rate is common, nonetheless the effect of this on the normal biliary secretion rate needs further investigation (Angelin et al., 1987; Arvidsson et al., 1982; Gundert-Remy et al., 1982). Ryde and Gustavsson (1987) performed a bile collection study using a luminal tube and intestinal administration of a olsalazine sodium infusion (50 ml/h) at a constant rate over a period of 1 h in the jejunum, although this method was not well tolerated and four out of six subjects experienced side-effects, probably related to irritation of the intestinal mucosa, compared to our study where a rapid bolus dose of a drug dispersion and no adverse effects where reported and a high rate of successful experiments (Ryde and Gustavsson, 1987). Few studies have been conducted where the individual bile acid concentrations have been determined for the fasted state, although a previous investigation by our group agrees well with the data obtained in this study. The mean total concentration of bile acids were 4.45 ± 2.05 mM during a 240 min

period at fasted state and are increased after a high fat liquid meal to approximately 9 mM (Fig. 7) (Hernell et al., 1990). All of the subjects who participated in this study had fasted for at least 10 h, except one, subject eight, who had breakfast on the morning of the study. The reason why this subject has bile acid concentrations similar to the other subjects is likely to be a consequence of the food intake occurred several hours before the start of the study, and therefore the bile acid concentrations being returned to the fasted state. According to a study by Yao et al. (2002), bile acid concentrations return to the fasted state approximately 1 h after a liquid high fat meal (Yao et al., 2002). Two metabolites of rosuvastatin have been previously identified in human plasma, urine and faeces, rosuvastatin lactone and N-desmethyl rosuvastatin (Martin et al., 2003d; Schneck et al., 2004). Hence, reports based on human microsomes have shown that the metabolic clearance of rosuvastatin is low and studies show discrepancies as to which CYP 450 enzymes are involved (Fujino et al., 2004; Prueksaritanont et al., 2002b). Research with human hepatocytes has shown that rosuvastatin and other statins are metabolised by UDP glucuronosyltransferases (UGTs) UGT1A1 and UGT1A3 to form ␤-1-O-acyl glucuronide (Prueksaritanont et al., 2002a,b). In the present study we were unable to detect any glucuronide conjugates of rosuvastatin in bile possibly due their instability in the alkaline bile and intestinal fluid, as well as enzymatic hydrolysis. It is well known that acyl glucuronides are difficult to detect owing to their instability and the fact that they are able to undergo hydrolysis, acyl migration and covalent binding to macromolecules in vitro as well as in vivo at physiological pH and temperature (Shipkova et al., 2003). However, it is of great importance to further elucidate the formation of acyl glucuronide metabolite of rosuvastatin since they have been associated with hepatoxicity because of their high reactive capacity. Using the modified Loc-I-Gut technique to sample human bile from an intestinal segment in healthy volunteers has proven to be an important tool with which to assess drug excretion kinetics in bile and to estimate the liver exposure of drugs and metabolites during the absorption and first-pass extraction processes. The biliary flow displayed high interindividual variability and occurred in a pulsed manner, rather than as a constant flow in the fasted state. Rosuvastatin was immediately and significantly accumulated in bile compared to plasma, and approximately 11% of an intestinally administered dose was excreted in the bile within the first 4 h after drug administration.

references

Agoston, S., Vermeer, G.A., Kertsten, U.W., Meijer, D.K., 1973. The fate of pancuronium bromide in man. Acta Anaesthesiol. Scand. 17, 267. Angelin, B., Arvidsson, A., Dahlqvist, R., Hedman, A., Schenck-Gustafsson, K., 1987. Quinidine reduces biliary clearance of digoxin in man. Eur. J. Clin. Invest. 17, 262. Arvidsson, A., Alvan, G., Angelin, B., Borga, O., Nord, C.E., 1982. Ceftriaxone: renal and biliary excretion and effect on the colon microflora. J. Antimicrob. Chemother. 10, 207.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

AstraZeneca, 2003. Crestor (Rosuvastatin) Product Monograph, 1. Bonlokke, L., Christensen, F.N., Knutson, L., Kristensen, H.G., Lennernas, H., 1997. A new approach for direct in vivo dissolution studies of poorly soluble drugs. Pharm. Res. 14, 1490. Bonlokke, L., Hovgaard, L., Kristensen, H.G., Knutson, L., Lennernas, H., 2001. Direct estimation of the in vivo dissolution of spironolactone, in two particle size ranges, using the single-pass perfusion technique (Loc-I-Gut) in humans. Eur. J. Pharm. Sci. 12, 239. Bonlokke, L., Hovgaard, L., Kristensen, H.G., Knutson, L., Lindahl, A., Lennernas, H., 1999. A comparison between direct determination of in vivo dissolution and the deconvolution technique in humans. Eur. J. Pharm. Sci. 8, 19. Brown, C.D.A., Windass, A., Bleasby, K., Lauffart, B., 2001. Rosuvastatin is a high affinity substrate of hepatic organic anion transporter OATP-C. Atherosclerosis (Suppl. 2), 90. Caldwell, J.H., Cline, C.T., 1976. Biliary excretion of digoxin in man. Clin. Pharmacol. Ther. 19, 410. Cooper, K.J., Martin, P.D., Dane, A.L., Warwick, M.J., Raza, A., Schneck, D.W., 2003a. The effect of erythromycin on the pharmacokinetics of rosuvastatin. Eur. J. Clin. Pharmacol. 59, 51. Cooper, K.J., Martin, P.D., Dane, A.L., Warwick, M.J., Raza, A., Schneck, D.W., 2003b. Lack of effect of ketoconazole on the pharmacokinetics of rosuvastatin in healthy subjects. Br. J. Clin. Pharmacol. 55, 94. Cooper, K.J., Martin, P.D., Dane, A.L., Warwick, M.J., Schneck, D.W., Cantarini, M.V., 2002. The effect of fluconazole on the pharmacokinetics of rosuvastatin. Eur. J. Clin. Pharmacol. 58, 527. Cooper, K.J., Martin, P.D., Dane, A.L., Warwick, M.J., Schneck, D.W., Cantarini, M.V., 2003c. Effect of itraconazole on the pharmacokinetics of rosuvastatin. Clin. Pharmacol. Ther. 73, 322. Fujino, H., Saito, T., Tsunenari, Y., Kojima, J., Sakaeda, T., 2004. Metabolic properties of the acid and lactone forms of HMG-CoA reductase inhibitors. Xenobiotica 34, 961. Ghibellini, G., Johnson, B.M., Kowalsky, R.J., Heizer, W.D., Brouwer, K.L.R., 2004. A novel method for the determination of biliary clearance in humans. AAPS J. 6, 1. Gundert-Remy, U., Frohnapfel, F., Jourdan, W., Weber, E., Stiehl, A., 1982. Estimation of biliary excretion of ureidopenicillins in healthy volunteers using marker dilution technique. Br. J. Clin. Pharmacol. 13, 795. Hernell, O., Staggers, J.E., Carey, M.C., 1990. Physical–chemical behavior of dietary and biliary lipids during intestinal digestion and absorption. 2. Phase analysis and aggregation states of luminal lipids during duodenal fat digestion in healthy adult human beings. Biochemistry 29, 2041. Hofmann, A.F., 1976. The enterohepatic circulation of bile acids in man. Adv. Intern. Med. 21, 501. Knutson, L., Odlind, B., Hallgren, R., 1989. A new technique for segmental jejunal perfusion in man. Am. J. Gastroenterol. 84, 1278. Lanzini, A., Jazrawi, R.P., Northfield, T.C., 1987. Simultaneous quantitative measurements of absolute gallbladder storage and emptying during fasting and eating in humans. Gastroenterology 92, 852. Lanzini, A., Pigozzi, M.G., Wuhrer, A., Facchinetti, D., Castellano, M., Bettini, L., Guerra, U.P., Beschi, M., Muiesan, G., 1989. Quantitative measurement of biliary excretion and of gall bladder concentration of drugs under physiological conditions in man. Gut 30, 104. Lee, W.M., 2003. Drug-induced hepatotoxicity. N. Engl. J. Med. 349, 474.

213

Lennernas, H., Ahrenstedt, O., Hallgren, R., Knutson, L., Ryde, M., Paalzow, L.K., 1992. Regional jejunal perfusion, a new in vivo approach to study oral drug absorption in man. Pharm. Res. 9, 1243. Loria, P., Carulli, N., Medici, G., Tripodi, A., Iori, R., Rovesti, S., Bergomi, M., Rosi, A., Romani, M., 1989. Determinants of bile secretion: effect of bile salt structure on bile flow and biliary cation secretion. Gastroenterology 96, 1142. Martin, P.D., Dane, A.L., Nwose, O.M., Schneck, D.W., Warwick, M.J., 2002a. No effect of age or gender on the pharmacokinetics of rosuvastatin: a new HMG-CoA reductase inhibitor. J. Clin. Pharmacol. 42, 1116. Martin, P.D., Dane, A.L., Schneck, D.W., Warwick, M.J., 2003a. An open-label, randomized, three-way crossover trial of the effects of coadministration of rosuvastatin and fenofibrate on the pharmacokinetic properties of rosuvastatin and fenofibric acid in healthy male volunteers. Clin. Ther. 25, 459. Martin, P.D., Kemp, J., Dane, A.L., Warwick, M.J., Schneck, D.W., 2002b. No effect of rosuvastatin on the pharmacokinetics of digoxin in healthy volunteers. J. Clin. Pharmacol. 42, 1352. Martin, P.D., Warwick, M.J., Dane, A.L., Brindley, C., Short, T., 2003b. Absolute oral bioavailability of rosuvastatin in healthy white adult male volunteers. Clin. Ther. 25, 2553. Martin, P.D., Warwick, M.J., Dane, A.L., Cantarini, M.V., 2003c. A double-blind, randomized, incomplete crossover trial to assess the dose proportionality of rosuvastatin in healthy volunteers. Clin. Ther. 25, 2215. Martin, P.D., Warwick, M.J., Dane, A.L., Hill, S.J., Giles, P.B., Phillips, P.J., Lenz, E., 2003d. Metabolism, excretion, and pharmacokinetics of rosuvastatin in healthy adult male volunteers. Clin. Ther. 25, 2822. Meijer, D.K., Lennernas, H., 2005. EUFEPS conference on drug transporters at Copenhagen: integrative approaches in ADME research. Eur. J. Pharm. Sci. 26, 130. Meijer, D.K., Weert, B., Vermeer, G.A., 1988. Pharmacokinetics of biliary excretion in man. VI. Indocyanine green. Eur. J. Clin. Pharmacol. 35, 295. Meijer, D.K., Weitering, J.G., Vermeer, G.A., 1983. Pharmacokinetics of biliary excretion in man. V. Dibromosulfophthalein. Eur. J. Clin. Pharmacol. 24, 549. Meijer, D.K., Vermeer, G.A., Kwant, G., 1971. The excretion of hexafluorenium in man and rat. Eur. J. Pharmacol. 14, 280. Nezasa, K., Higaki, K., Matsumura, T., Inazawa, K., Hasegawa, H., Nakano, M., Koike, M., 2002. Liver-specific distribution of rosuvastatin in rats: comparison with pravastatin and simvastatin. Drug Metab. Dispos. 30, 1158. Prueksaritanont, T., Subramanian, R., Fang, X., Ma, B., Qiu, Y., Lin, J.H., Pearson, P.G., Baillie, T.A., 2002a. Glucuronidation of statins in animals and humans: a novel mechanism of statin lactonization. Drug Metab. Dispos. 30, 505. Prueksaritanont, T., Tang, C., Qiu, Y., Mu, L., Subramanian, R., Lin, J.H., 2002b. Effects of fibrates on metabolism of statins in human hepatocytes. Drug Metab. Dispos. 30, 1280. Qadir, M., O’Loughlin, K.L., Fricke, S.M., Williamson, N.A., Greco, W.R., Minderman, H., Baer, M.R., 2005. Cyclosporin A is a broad-spectrum multidrug resistance modulator. Clin. Cancer Res. 11, 2320. Qvist, N., Rafaelsen, S., Oster-Jorgensen, E., Rasmussen, L., Hovendal, C., Pedersen, S.A., 1991. The simultaneous use of 99m-Tc-HIDA scintigraphy and ultrasound in the determination of gallbadder storage and emptying in the fasting state. Eur. J. Gastroenterol. Hepatol. 3, 657. Roberts, M.S., Magnusson, B.M., Burczynski, F.J., Weiss, M., 2002. Enterohepatic circulation: physiological, pharmacokinetic and clinical implications. Clin. Pharmacokinet. 41, 751.

214

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 205–214

Ryde, M., Gustavsson, S., 1987. Biliary excretion of olsalazine sodium in humans. Eur. J. Drug Metab. Pharmacokinet. 12, 17. Schneck, D.W., Birmingham, B.K., Zalikowski, J.A., Mitchell, P.D., Wang, Y., Martin, P.D., Lasseter, K.C., Brown, C.D., Windass, A.S., Raza, A., 2004. The effect of gemfibrozil on the pharmacokinetics of rosuvastatin. Clin. Pharmacol. Ther. 75, 455. Sgro, C., Clinard, F., Ouazir, K., Chanay, H., Allard, C., Guilleminet, C., Lenoir, C., Lemoine, A., Hillon, P., 2002. Incidence of drug-induced hepatic injuries: a French population-based study. Hepatology 36, 451. Shaffer, E.A., 2000. Review article: control of gall-bladder motor function. Aliment Pharmacol. Ther. 14 (Suppl. 2), 2. Shipkova, M., Armstrong, V.W., Oellerich, M., Wieland, E., 2003. Acyl glucuronide drug metabolites: toxicological and analytical implications. Ther. Drug Monit. 25, 1. Simonson, S.G., Raza, A., Martin, P.D., Mitchell, P.D., Jarcho, J.A., Brown, C.D., Windass, A.S., Schneck, D.W., 2004. Rosuvastatin pharmacokinetics in heart transplant recipients administered an antirejection regimen including cyclosporine. Clin. Pharmacol. Ther. 76, 167.

Strasberg, S.M., Harvey, P.R., Hofmann, A.F., 1990. Bile sampling, processing and analysis in clinical studies. Hepatology 12, 176S. Sutfin, T.A., Lind, T., Gabrielsson, M., Regardh, C.G., 1990. Biliary secretion of felodipine metabolites in man after intravenous [14C]felodipine. Eur. J. Clin. Pharmacol. 38, 421. Westphal, J.F., Brogard, J.M., Caro-Sampara, F., Adloff, M., Blickle, J.F., Monteil, H., Jehl, F., 1997. Assessment of biliary excretion of piperacillin-tazobactam in humans. Antimicrob. Agents Chemother. 41, 1636. Westra, P., Vermeer, G.A., de Lange, A.R., Scaf, A.H., Meijer, D.K., Wesseling, H., 1981. Hepatic and renal disposition of pancuronium and gallamine in patients with extrahepatic cholestasis. Br. J. Anaesth. 53, 331. Wierda, J.M., Szenohradszky, J., De Wit, A.P., Zentai, G., Agoston, S., Kakas, M., Kleef, U.W., Meijer, D.K., 1991. The pharmacokinetics, urinary and biliary excretion of pipecuronium bromide. Eur. J. Anaesthesiol. 8, 451. Yao, L., Heubi, J.E., Buckley, D.D., Fierra, H., Setchell, K.D., Granholm, N.A., Tso, P., Hui, D.Y., Woollett, L.A., 2002. Separation of micelles and vesicles within lumenal aspirates from healthy humans: solubilization of cholesterol after a meal. J. Lipid Res. 43, 654.