Advertisement
Research ArticleArticle

The Prediction of the Relative Importance of CYP3A/P-glycoprotein to the Nonlinear Intestinal Absorption of Drugs by Advanced Compartmental Absorption and Transit Model

Junichi Takano, Kazuya Maeda, Michael B. Bolger and Yuichi Sugiyama
Drug Metabolism and Disposition November 2016, 44 (11) 1808-1818; DOI: https://doi.org/10.1124/dmd.116.070011

Abstract

Intestinal CYP3A and P-glycoprotein (P-gp) decrease the intestinal absorption of substrate drugs. Since substrate specificity of CYP3A often overlaps that of P-gp, and estimation of their saturability in the intestine is difficult, dose-dependent FaFg (fraction of the administered drugs that reach the portal blood) of substrate drugs and the relative importance of CYP3A and P-gp have not been clarified in many cases. Thus, we tried to establish the universal methodology for predicting the in vivo absorption of several CYP3A and/or P-gp substrates from in vitro assays. One of the key points is to set up the scaling factor (SF), correcting the difference between the observed in vivo clearance and the predicted clearance from in vitro data. The SFs of Vmax for CYP3A (SFCYP3A) and P-gp (SFP-gp) were simultaneously optimized to explain the FaFg of CYP3A and/or P-gp substrate drugs. The best predictability of FaFg was achieved when considering both SFCYP3A and SFP-gp. The simulation also clarified the relative importance of CYP3A and P-gp in determining FaFg. In particular, the nonlinear intestinal absorption of verapamil was caused by the saturation of intestinal CYP3A, whereas that of quinidine was governed by the saturation of both CYP3A and P-gp. In addition, the dose-dependent FaFg of selective and dual CYP3A and/or P-gp substrates was well predicted. We therefore propose a methodology for predicting the FaFg of drugs using a mathematical model with optimized SFCYP3A and SFP-gp. Our methodology is applicable to in vitro–in vivo extrapolation of intestinal absorption, even if absolute in vivo functions of enzymes/transporters are unclear.

Introduction

It is well known that CYP3A and P-glycoprotein (P-gp) in the enterocytes contribute to the suppression of intestinal absorption of substrate drugs. CYP3A involves the decrease in the fraction of absorbed drugs that reach the portal vein. Whereas CYP3A4 is the dominant CYP3A isoform in the liver and small intestine, CYP3A5 is also found in these tissues, and its expression is polymorphic (Wrighton et al., 1990; Guengerich, 1995; Paine et al., 1997). In addition, P-gp plays a role in the drug efflux from enterocytes to the intestinal lumen (Thiebaut et al., 1987; Saitoh and Aungst, 1995; Wacher et al., 1998). Thus, the saturation of CYP3A-mediated metabolism and/or P-gp–mediated efflux may lead to the increase in the intestinal absorption as the dose increases. Indeed, it has been reported that felodipine, midazolam, and sildenafil (substrates of CYP3A); celiprolol and talinolol (substrates of P-gp); and indinavir and 3-ethyl-5-{5-[(4-ethylpiperazin-1-yl)sulfonyl]-2-propoxyphenyl}-2-(pyridin-2-ylmethyl)-2,6-dihydro-7H-pyrazolo(4,3-d)pyrimidin-7-one (UK-393,664) (substrates of both CYP3A and P-gp) show nonlinear pharmacokinetics at therapeutically relevant doses, possibly caused by the saturation of intestinal CYP3A and/or P-gp (Bornemann et al., 1985; Nakashima et al., 1988, 1992; Wetterich et al., 1996; Yeh et al., 1998; Nichols et al., 2002; Abuasal et al., 2012).

Microdose (MD) studies are useful to obtain information on human clinical pharmacokinetics of new drug candidates in the early stage of drug development. Some researchers are concerned with whether the dose-normalized drug concentration at therapeutic dose (ThD) is identical to that at MD (Maeda and Sugiyama, 2011). In theory, if all of the molecules involved in metabolism and transport are not saturated at ThD, drug concentration normalized by dose should be equal, regardless of the dose. We and other groups have already demonstrated the linear pharmacokinetics of drugs at MD and ThD for several MD studies (Lappin et al., 2006, 2011; Yamane et al., 2007). However, it has been pointed out that the nonlinearity between MD and ThD renders it difficult to predict the pharmacokinetics. Based on the pharmacokinetic theory, although low bioavailability of drugs is confirmed in an MD study, it is still possible that drugs show high bioavailability in the clinical situation if the saturation of CYP3A-mediated metabolism and/or P-gp–mediated efflux occurs at a high dose, but not at MD. This may lead to the incorrect conclusion of poor bioavailability, which is not appropriate for a new chemical entity during exploratory research. Thus, it is important to quantitatively clarify the role of the saturation of intestinal CYP3A and P-gp in predicting the pharmacokinetics at ThD from the results of MD study.

Yu and Amidon (1999) proposed a compartmental absorption and transit model to simulate the rate and extent of drug absorption. The compartmental absorption and transit model was subsequently enhanced by Simulations Plus, Inc. (Lancaster, CA) to reflect more physiologic gastrointestinal characteristics along with an enterocyte compartment to account for gut metabolism, and was called the advanced compartmental absorption and transit (ACAT) model. The ACAT model was proposed to mechanistically predict drug dissolution and intestinal absorption including gut metabolism and active transport processes after oral administration (Agoram and et al., 2001). Bolger et al. (2009) demonstrated the ability of the ACAT model to simulate the nonlinear pharmacokinetics for substrates of influx and efflux transporters in the intestine. Theoretically, in vivo functions of enzymes/transporters should be predicted from in vitro metabolic/transport activities simply by scaling up in vitro parameters by using the ratio of expression level of molecules in whole tissue to that in the expression systems. However, in the actual situations, various reports indicated that some empirical scaling factors (SFs) further improved the predictability of cytochrome P450 (P450)–mediated metabolism and transporter-mediated membrane permeation (e.g., Hirota et al., 2001; Naritomi et al., 2001; Galetin et al., 2004; Emoto and Iwasaki, 2007). Differences in activity per unit enzyme amount between cDNA-expressed P450s and liver microsomes are sometimes caused by the levels of accessory proteins and cofactors and buffer composition (Venkatakrishnan et al., 2000; Taguchi et al., 2001; Kudo et al., 2016). Materials, experimental conditions, and analytical methods also affect interlaboratory differences in in vitro parameters. Moreover, information on an absolute expression level of P-gp in each region of human intestine is still lacking. Thus, the prediction of intestinal absorption of drugs by “pure bottom-up approach” is not always successful for various reasons. Therefore, we tried to establish an appropriate method for the quantitative prediction of the dose-dependent intestinal availability (FaFg) of drugs based on the in vitro saturation kinetics and relative importance of CYP3A and P-gp in their nonlinear pharmacokinetics by the ACAT model. In addition, we aimed to discriminate whether or not pharmacokinetics of CYP3A/P-gp substrates maintain linearity even at the therapeutic dose.

Materials and Methods

Materials.

Felodipine, midazolam, triazolam, and verapamil hydrochloride were purchased from Wako Pure Chemicals (Kyoto, Japan). Digoxin and quinidine sulfate dehydrate salt were purchased from Sigma Chemical Co., Ltd. (St. Louis, MO). Sildenafil citrate, talinolol, and fexofenadine hydrochloride were purchased from Toronto Research Chemicals Inc. (North York, Canada). Celiprolol hydrochloride was purchased from AvaChem Scientific (San Antonio, TX). Indinavir and saquinavir mesylate were purchased from the U.S. Pharmacopeial Convention (Rockville, MD). Recombinant human CYP3A4 and CYP3A5 supersomes were purchased from BD Gentest Co. (Woburn, MA). Human intestine homogenate was purchased from Biopredic International (Rennes, France). Dulbecco’s modified Eagle’s medium with 4500 mg/l glucose, nonessential amino acids, and penicillin-streptomycin solution were purchased from Gibco BRL (Gaithersburg, MD). All other reagents used were of analytical grade.

In Vitro Assay for CYP3A4-Mediated Metabolism.

Kinetic parameters for CYP3A4-mediated metabolism were estimated using a substrate depletion assay (Obach and Reed-Hagen, 2002). Recombinant human CYP3A4 microsomes (10 pmol of P450/ml) and different concentrations of felodipine (0.03–10 μM), midazolam (0.05–5 μM), sildenafil (0.03–30 μM), triazolam (0.5–100 μM), indinavir (0.03–10 μM), quinidine (2–100 μM), saquinavir (0.03–10 μM), or verapamil (0.5–50 μM) were incubated in 100 mM KH2PO4 (pH 7.4) containing 3.3 mM MgCl2 for 5 minutes at 37°C. The reaction was initiated by the addition of 1.3 mM NADPH in a total volume of 500 μl. Reaction mixture (0.05 ml) was removed at 0, 0.5, 1, 2, 5, 10, and 20 minutes. The reaction was stopped by the addition of 50 μl of acetonitrile containing an internal standard, triazolam (for felodipine and midazolam), diazepam (for sildenafil), saquinavir (for indinavir), dextromethorphan (for verapamil and quinidine), and indinavir (for saquinavir) on ice. The mixture was centrifuged for 5 minutes at 20,400 × g at 4°C, and the supernatants were collected. Terminated reaction mixtures were analyzed by liquid chromatography–tandem mass spectrometry (LC-MS/MS). In addition, nonspecific binding to microsomes was estimated by ultracentrifugation. All substrates were incubated with recombinant human CYP3A4 microsomes in 100 mM KH2PO4 (pH 7.4) containing 3.3 mM MgCl2 for 5 minutes at 37°C. After incubation, the mixtures were centrifuged for 60 minutes at 105,000 × g at 4°C, and the supernatants were collected and analyzed by LC-MS/MS. The affinity of the CYP3A4-mediated metabolism (Km value) was determined by fitting one of the following equations (eq. 1 or 2) to the CLint values at the designated unbound substrate concentrations corrected with the unbound fraction in microsomes:Embedded Image(1)Embedded Image(2)where S represents the initial unbound substrate concentration in the medium corrected with the unbound fraction in microsomes, and Vmax is the theoretical maximum depletion rate. The fitting was performed with the nonlinear least-squares method using the MULTI program (http://www.pharm.kyoto-u.ac.jp/byoyaku/Kinetics/download.html) (Yamaoka et al., 1981). Verapamil was analyzed by eq. 2 because two saturable components were optically observed. All other substrates were analyzed by eq. 1. The experimental data are shown in Supplemental Fig. 1.

Cell Culture.

Caco-2 cells, provided by the Department of Molecular Pharmacokinetics, Graduate School of Pharmaceutical Sciences, The University of Tokyo (Tokyo, Japan), were cultured at 37°C in a 5% CO2 atmosphere using Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 1% nonessential amino acids, 100 U/ml penicillin, and 100 μg/ml streptomycin. Cells were passaged upon reaching approximately 80% confluence using trypsin-EDTA and plated at ratios of 1:5. For the preparation of Caco-2 cell monolayers, cells (passage numbers 4–16) were plated in a Corning 24-well Transwell plate (0.33 cm2 membrane surface area, 6.5 mm diameter; Corning Inc., Tewksbury, MA) at a density of 4.0 × 104 cells/insert. The culture medium was replaced every second day with 250 μl and 1 ml into the apical and basolateral side, respectively. After 21–25 days in culture, the Caco-2 cell monolayers were used for the transport experiments.

In Vitro Assay for P-gp–Mediated Transport.

In vitro transcellular transport assays were performed to determine the kinetic parameters for P-gp–mediated transport. The Caco-2 cell monolayer was preincubated in Hanks’ balanced salt solution (HBSS) with 10 mM HEPES (pH 7.4) for 30 minutes at 37°C. After preincubation, the transepithelial electrical resistance was estimated with the Millicell-ERS system (Millipore Co., Bedford, MA). The Caco-2 monolayers with transepithelial electrical resistance values over 300 Ωcm2 were used in the experiments. Thereafter, an HBSS buffer containing celiprolol (10–1000 μM), digoxin (0.3–30 μM), fexofenadine (1–300 μM), talinolol (3–1000 μM), indinavir (1–30 μM), quinidine (0.3–100 μM), saquinavir (1–30 μM), or verapamil (0.3–100 μM) was added to the apical side (0.25 ml, HBSS, pH 6.5). To determine the passive permeability of substrates, verapamil was coincubated at a concentration of 100 μM. After incubation for 30, 60, and 90 minutes at 37°C, 0.1 ml of the medium was taken from the opposite side. The substrate concentration was determined by LC-MS/MS. The apparent permeability coefficient (Papp, cm/s) was calculated using the following equation:Embedded Image(3)where dQ/dt is the permeation rate (mol/min), A is the surface area of the insert (0.33 cm2), and C0 is the initial concentration (μM). The kinetic parameter for P-gp–mediated efflux was estimated from the concentration-dependent Papp. The Km value with regard to the unbound concentration in the cells was calculated based on the model analysis method (Tachibana et al., 2010). The overall apical-to-basolateral permeability was shown as eqs. (4) and (5). The affinity of the P-gp–mediated efflux (Km value) was determined by fitting with the following eq. (6):Embedded Image(4)Embedded Image(5)Embedded Image(6)where PSAB is the permeation clearance in the apical to basolateral direction, PS1 is the passive permeation clearance from the apical side to cells, PS2 is the passive permeation clearance from cells to the apical side, and PS3 is the passive permeation clearance from cells to the basolateral side. PSP-gp is the active efflux clearance mediated by P-gp. Passive permeation clearance was assumed to be the same for each substrate (PS1 = PS2 = PS3). Then, a total of three parameters (Km, Vmax, and PS1) were fitted to the concentration-dependent permeation clearance. Obvious saturation was not observed on concentration-dependent permeation of celiprolol, digoxin, and saquinavir. Assuming PSP-gp was zero in eq. (4), PSAB can be simply described as follows:Embedded Image(7)Thus, we could assume that PSAB in the presence of 100 μM verapamil (P-gp inhibitor) was equal to half of the passive permeation clearance. Then, after fixing PS1, Km and Vmax for P-gp were optimized based on the concentration-dependent permeation clearance of substrate drugs. The fitting was performed with the numerical mode by SAAMII version 1.2 (The Epsilon Group, Charlottesville, VA). The experimental data are shown in Supplemental Fig. 2.

Apparent Permeability.

The apparent permeability was estimated by an in vitro transcellular transport assay using Caco-2 cells. Assay conditions are described in the In Vitro Assay for P-gp–Mediated Transport section. HBSS buffer containing 1 μM felodipine, 1 μM midazolam, 1 μM sildenafil, 1 μM triazolam, 100 μM atenolol, 100 μM cimetidine, 1 μM ketoprofen, 1 μM metoprolol, 1 μM propranolol, or 100 μM ranitidine was added to the apical side [0.25 ml of HBSS (pH 6.5)]. P-gp substrates, 10 μM celiprolol, 0.3 μM digoxin, 1 μM fexofenadine, 3 μM talinolol, 1 μM indinavir, 1 μM quinidine, or 1 μM saquinavir was added to the apical side with 100 μM verapamil as a P-gp inhibitor. After incubation for 30, 60, and 90 minutes at 37°C, 0.1 ml of the medium was taken from the opposite side and analyzed by LC-MS/MS (experimental data are shown in Supplemental Table 1).

Protein Binding in Enterocyte.

The unbound fraction in enterocytes was estimated by equilibrium dialysis using a Rapid Equilibrium Dialysis (RED) device (Thermo Fisher Scientific Inc., Waltham, MA) or a 96-well Equilibrium DIALYZER plate with a molecular weight cutoff of 10,000 Da (Harvard Apparatus, Holliston, MA). Twenty-nine percent human intestine homogenate was diluted to obtain 7, 14, and 21% homogenate containing 3 μM felodipine, 50 μM midazolam, 10 μM sildenafil, 50 μM triazolam, 100 μM celiprolol, 100 μM digoxin, 100 μM fexofenadine, 100 μM talinolol, 10 μM indinavir, 100 μM quinidine, 10 μM saquinavir, and 100 μM verapamil with phosphate-buffered saline at pH 7.4. Aliquots (75 μl; 96-well Equilibrium DIALYZER, 100 μl; RED device) of each sample and phosphate-buffered saline (75 μl; 96-well Equilibrium DIALYZER, 300 μl; RED device) were added to the equilibrium dialysis device. The plate was put into a 37°C incubator and gently stirred for 4 hours. After incubation, 30 μl of samples and buffers was removed and mixed with 50 μl of acetonitrile containing an internal standard. The mixture was centrifuged for 5 minutes at 20,400 × g at 4°C, and the supernatants were collected. Substrate concentration was determined by LC-MS/MS.

Solubility.

pH-dependent solubility of felodipine, midazolam, sildenafil citrate, digoxin, fexofenadine hydrochloride, indinavir, quinidine sulfate, saquinavir, and verapamil hydrochloride was determined. Acetate buffer (0.1 M, pH 4 and 5), 0.1 M phosphate buffer (pH 6, 7, and 8), and glycine-NaOH buffer (pH 9 and 10) were prepared. Excess amounts of substrate were added to each different pH buffer and incubated for 24 hours at 25°C. After incubation, the suspensions were centrifuged for 5 minutes at 20,400 × g at 25°C, and the supernatants were filtered through a Millipore Millex-HA filter 0.45 μm. The substrate concentration was determined by LC-MS/MS. The pH was determined using a pH meter (Shindengen, Tokyo, Japan). The pKa and solubility factor (defined as a ratio of the solubility of completely ionized drug to completely unionized drug) of each compound were derived by fitting the preinstalled model, named “Solubility vs. pH Model version 6.1” in GastroPlus (Simulations Plus, Inc., Lancaster, CA) to all of the solubility data at the different pH values tested (experimental data are shown in Supplemental Fig. 3 and Table 2). The pKa and solubility factor of triazolam, celiprolol, and talinolol were calculated by the reported solubility data (package insert of triazolam; Kobayashi Kako Co., Ltd., Fukui, Japan) (Gramatté et al., 1996; Bergström et al., 2004).

LC-MS/MS Analysis.

Three LC-MS/MS systems—a Quattro micro API tandem mass spectrometer equipped with an alliance 2695 separation module (Waters, Milford, MA), an AB SCIEX QTRAP 5500 mass spectrometer (AB SCIEX, Foster City, CA) equipped with a prominence liquid chromatograph (Shimadzu, Kyoto, Japan), and an AB SCIEX API-4000 mass spectrometer (AB SCIEX) equipped with a Agilent 1100 series liquid chromatograph (Agilent Technologies, Santa Clara, CA)—were used for the determination of compounds. The mass spectrometers were operated in positive-ion electrospray ionization and multiple-reaction monitoring mode for the detection of compounds. The details of the LC conditions and multiple-reaction monitoring transitions are shown in Supplemental Table 2. The collected supernatants were diluted with the mobile phase. The diluted samples (10 μl) were injected into the LC-MS/MS connected with an Atlantis T3 column (2.1 × 50 mm, 5 μm; Waters). The range of calibration curve of each compound was 0.01–10 μmol/l (felodipine), 0.1–10 μmol/l (sildenafil), 1–1000 nmol/l (celiprolol), 0.1–100 nmol/l (digoxin), 0.3–300 nmol/l (fexofenadine), 1–300 nmol/l (talinolol), 0.003–10 μmol/l (indinavir), 1–3000 nmol/l (quinidine, verapamil), and 1–100 nmol/l (midazolam, triazolam, atenolol, cimetidine, ranitidine, metoprolol, propranolol, ketoprofen). The correlation coefficient of calibration curve was over 0.990 in all analysis.

Intestinal Availability.

The total clearance (CLtot), renal clearance (CLr), and blood-to-plasma concentration ratio (Rb) were calculated from literature information. Bioavailability (F), hepatic availability (Fh), and FaFg values are shown in Table 1. FaFg was calculated using the following equation:

View this table:
TABLE 1

Bioavailability and intestinal availability for CYP3A and P-gp substrates

Embedded Image(8)Embedded Image(9)Embedded Image(10)Embedded Image(11)Embedded Image(12)

where AUCpo, AUCiv, Dosepo, Doseiv, CLh, Fh, and Qh are plasma area under the concentration-time curve (AUC) after oral administration, plasma AUC after intravenous administration, dose on oral administration, dose on intravenous administration, hepatic blood clearance, hepatic availability, and hepatic blood flow rate, respectively. The Qh was set at 25.5 ml/min/kg (Wynne et al., 1989) except for felodipine. According to a previous report, felodipine caused a 1.7-fold increase in the Qh after oral dosing (Bengtsson-Hasselgren et al., 1990). Thus, the Qh was set at 43.4 ml/min/kg for felodipine only. Pharmacokinetic parameters were obtained from previous reports shown in Table 1. FaFg estimated from the clinical pharmacokinetic study was defined as “observed FaFg,” whereas FaFg calculated by the simulation of the ACAT model with in vitro and in silico parameters was defined as “predicted FaFg.” In cases where the body weight of subjects was unknown in a clinical study report, it was assumed to be 70 kg. Dose-escalation studies of celiprolol and saquinavir, and clinical studies of triazolam shown in Table 6 were conducted under nonfasted conditions (Nakashima et al., 1988; Greenblatt et al., 1998a,b; Hsu et al., 1998). Conditions of food intake in clinical studies of celiprolol and fexofenadine could not be obtained from previous reports (Caruso et al., 1985; Lappin et al., 2010). All other clinical studies after oral administration, shown in Table 1 and Table 6, were conducted under fasted conditions.

ACAT Model.

GastroPlus version 9.0 was used for quantitative prediction of nonlinear intestinal absorption. The human physiologic fasted model and Opt logD model were used in the ACAT model. As shown in Supplemental Table 3, default values in version 6.1 were used for physiologic parameters (intestinal length, radius, pH, transit time, and regional distribution factors for CYP3A4 and P-gp) in GastroPlus. Oral administrations in all clinical studies shown in Table 1 and Table 6 were conducted using oral solution, capsule, or tablet, not including controlled-release formulations. Thus, we assumed immediate-release formulation at oral administration in the ACAT model. The time length of calculation for all substrates was set to 48 hours after dosing. CYP3A4-mediated gut metabolism and P-gp–mediated efflux were defined in the ACAT model. GastroPlus includes the physiologic expression level of CYP3A4 in each segment of the human intestine to adjust the Vmax for gut enzymes. In contrast, since the absolute expression level of P-gp was lacking, Vmax was calculated based on the relative expression level in each segment. The SFs are included in the equation for gut metabolism and the efflux and influx transport in GastroPlus. These are empirical SFs which can be optimized to fit the predicted FaFg to explain the observed FaFg. In the ACAT model, the FaFg value was defined as the fraction of administered drugs that reach the portal blood. Each process of intestinal absorption was calculated using following equations:Embedded Image(13)Embedded Image(14)Embedded Image(15)Embedded Image(16)Embedded Image(17)where ka, Peff, R, L, i, j, dMabsorbed/dt, V, Clumen, Cent, SFVmax, SFKm, ESFi,j, and fue are absorption rate constant, effective permeability, radius of the compartment, length of the compartment, the index for the ith enzyme or influx/efflux transporter, the index for the jth intestine compartment, absorption rate, volume of intestinal lumen, drug concentration in the lumen, drug concentration in the enterocytes, SF for Vmax (default = 1), SF for Km (default = 1), regional distribution factors for enzyme or transporter i in compartment j, and the unbound fraction in enterocytes, respectively. The maximum Cuent (unbound concentration in the enterocyte compartment) was calculated as the maximum concentration among all intestinal compartments in the ACAT model. Peff in humans is calculated by measuring the rate of disappearance of a drug from a section of the gastrointestinal tract at steady state. In this study, Peff was converted from Papp of Caco-2 cells by the following regression equation (Sun et al., 2002):Embedded Image(18)where Peff and Papp are expressed in (×10−4 cm/s) and (×10−6 cm/s). A and B are coefficients, which were calculated to be −0.162 and 0.6447 based on Peff and Papp of reference compounds. Peff of a test substrate was converted from its Papp with optimized coefficients A and B with eq. 18.

Consideration of CYP3A5-Mediated Metabolism.

The expression level of CYP3A5 in each segment of human intestine is not defined in the GastroPlus default settings. The mean expression level of CYP3A5 was calculated from the literature. Lin et al. (2002) reported that the CYP3A4 and CYP3A5 expression levels in jejunal homogenate were 17.2 and 14.0 pmol/mg protein in CYP3A5*1/*1 subjects, 2.4 and 3.6 pmol/mg protein in CYP3A5*1/*3 subjects, and 18.2 and 0.5 pmol/mg protein in CYP3A5*3/*3 subjects, respectively. The allelic frequency of CYP3A5 was calculated as the weighted average from Caucasian kidney transplant patients (Haufroid et al., 2004; Mai et al., 2004). The frequency of CYP3A5*1/*1, CYP3A5*1/*3, and CYP3A5*3/*3 were calculated to be 1, 14, and 85%, respectively. The mean expression levels of CYP3A4 and CYP3A5 were calculated using the following equation:Embedded Image(19)where i is the index of the ith genotype. The calculated mean expression levels of CYP3A4 and CYP3A5 were 16.0 and 1.07 pmol/mg protein, respectively. Assuming that the ratio of CYP3A4/CYP3A5 expression is constant in each segment of the human intestine, the CYP3A5 expression level was defined in GastroPlus. SFCYP3A was defined as a common parameter for both CYP3A4 and CYP3A5.

Optimization of SFCYP3A and SFP-gp.

Our analysis strategy is shown in Fig. 1. In the ACAT model in the GastroPlus software, predicted gut metabolism rate and efflux rate were estimated based on eqs. 15 and 17. We assumed that in vivo Km values of drugs are the same as those obtained from in vitro experiments for P-gp and CYP3A. The ESF value for CYP3A is set up based on the absolute expression level of CYP3A in the gut preset in GastroPlus. Since the absolute expression level of P-gp in each region of the intestine has not been clarified, the relative expression level of each compartment is set as the ESF value for P-gp. One of the critical points required to construct an accurate model is to determine appropriate SFCYP3A and SFP-gp. SFCYP3A and SFP-gp should be common parameters for all substrates of CYP3A and P-gp, respectively. We selected 12 test compounds, including four P-gp substrates which are not metabolized by CYP3A (celiprolol, digoxin, fexofenadine, and talinolol), four CYP3A substrates which are not transported by P-gp (felodipine, midazolam, sildenafil, and triazolam), and four dual substrates of CYP3A and P-gp (indinavir, quinidine, saquinavir, and verapamil), based on the literature (Tachibana et al., 2012) to collect abundant information on the relationship between kinetic parameters for P-gp and CYP3A and intestinal availability of various drugs to clearly discriminate the contribution of P-gp and CYP3A. Clinical data for quinidine at doses of 0.1 and 1 mg and verapamil at doses of 0.1 and 3 mg were used for our analyses. SFCYP3A and SFP-gp were optimized by simultaneous fitting of predicted FaFg values (using GastroPlus based on in vitro parameters of all test compounds) to clinically observed ones.

Fig. 1.
Fig. 1.

The outline of our strategy of the prediction of intestinal availability. FaFg was predicted from physiochemical and in vitro kinetic parameters of drugs with the ACAT model. SFCYP3A and SFP-gp were simultaneously estimated from the relationship between predicted and observed FaFg.

The optimization module was used to optimize the SF for Vmax. The objective function was constructed from the predicted and observed FaFg:Embedded Image(20)where W is weight. SFCYP3A and SFP-gp were simultaneously optimized to explain the FaFg of selective and dual CYP3A and/or P-gp substrate drugs. The weight was set at 1/FaFg Observed. If the observed FaFg of substrates under linear conditions was available, this was applied to the optimization of SFs. The bias and precision of prediction of FaFg were evaluated using average fold error (afe) and root mean squared error (rmse) (Sheiner and Beal, 1981; Obach et al., 1997), as follows:

Embedded Image(21)Embedded Image(22)

Results

In Vitro Experimental Data for Predicting the Intestinal Absorption of Drugs.

Physicochemical parameters for each drug to predict the intestinal absorption of CYP3A and P-gp substrates are shown in Table 2. The passive permeability and the unbound fraction in enterocytes are shown in Table 3. The estimated Peff for each substrate was calculated from the Papp estimated in Caco-2 cells. The kinetic parameters for CYP3A4 and P-gp are shown in Tables 4 and 5. All CYP3A substrates showed saturation kinetics, and their metabolic clearances showed 50-fold difference from 0.00423 ml/min/pmol P450 of quinidine to 0.211 ml/min/pmol P450 of felodipine. Tachibana et al. (2010) calculated the kinetic parameters for P-gp–mediated efflux with regard to the intracellular unbound concentration in Caco-2 cells. This method was used to estimate the Km and Vmax values for P-gp. The efflux clearance of 12 tested drugs showed 50-fold difference from 1.18 μl/min/106 cells of verapamil to 63.6 μl/min/106 cells of saquinavir. The selective and dual CYP3A and/or P-gp substrates selected in this study showed a wide range of metabolic and efflux clearances. The metabolic clearances of CYP3A4 and CYP3A5 are shown in Supplemental Table 4. Midazolam showed the highest metabolic clearance ratio of CYP3A5/CYP3A4 (1.86), whereas quinidine had the lowest value (0.299) among the CYP3A substrates. The calculated mean expression level of CYP3A4 in the intestine was 10.8-fold higher than that of CYP3A5, thus contribution of CYP3A5 to the intestinal metabolism was expected to be minor compared with CYP3A4.

View this table:
TABLE 2

Physicochemical and kinetic parameters for predicting the intestinal absorption of drugs with GastroPlus

View this table:
TABLE 3

Passive permeability and unbound fraction in enterocytes for predicting the intestinal absorption of drugs with GastroPlus

View this table:
TABLE 4

Kinetic parameters for CYP3A4-mediated metabolism of tested drugs

Kinetic parameter for CYP3A4 was estimated from the elimination of the parent compound using recombinant human CYP3A4. Each point represents the mean ± computer-calculated S.D. (n = 6).

View this table:
TABLE 5

Kinetic parameters for P-gp–mediated transport of tested drugs

Each parameter represents the mean ± computer-calculated S.D. (n = 3).

Optimization of SFCYP3A and SFP-gp to Determine the Ratio of Their Expression Levels between In Vitro Experimental Systems and In Vivo Intestine.

As shown in Fig. 2A, most of the substrates were well absorbed when both CYP3A-mediated metabolism and P-gp–mediated efflux were not considered. Then, SFCYP3A and SFP-gp were defined as the ratio of their functions between in vitro experimental systems and in vivo intestine and optimization was attempted to minimize the differences between observed and predicted FaFg values of 12 drugs. The optimized SFCYP3A and SFP-gp values were calculated to be 0.0740 and 0.00651, respectively. After optimization of SFCYP3A and SFP-gp, the predicted FaFg values of 8 of 12 substrates were within 1.5-fold deviations from the observed ones. As shown in Table 6, to estimate the relative contribution of CYP3A and P-gp in the intestinal absorption, the simulation was performed under four different conditions: both the optimized SFCYP3A and SFP-gp were used (SFCYP3A = 0.0740, SFP-gp = 0.00651, simulation I), only the optimized SFCYP3A was used and P-gp–mediated efflux was not considered (SFCYP3A = 0.0740, SFP-gp = 0, simulation II), only the optimized SFP-gp was used and CYP3A-mediated metabolism was not considered (SFCYP3A = 0, SFP-gp = 0.00651, simulation III), and both P-gp–mediated efflux and CYP3A-mediated metabolism were not considered (SFCYP3A and SFP-gp = 0, simulation IV). In simulations II and III, the FaFg values were overestimated for the P-gp and CYP3A substrates, respectively. The overall predictability of FaFg values of 12 CYP3A and/or P-gp substrates was best improved when considering both SFCYP3A and SFP-gp since simulation I showed the lowest afe and rmse values.

Fig. 2.
Fig. 2.

Comparison of the predicted (pred.) and observed (obs.) intestinal availabilities of CYP3A and P-gp substrates. (A) Before optimization, FaFg was predicted by only considering the membrane permeation by passive diffusion. CYP3A-mediated metabolism and P-gp–mediated efflux were not considered. (B) After optimization, FaFg was predicted with optimized SFCYP3A and SFP-gp. The broken line represents a 1.5-fold deviation from the unity. CYP3A substrates are felodipine (1), midazolam (2), sildenafil (3), and triazolam (4). P-gp substrates are celiprolol (5), digoxin (6), fexofenadine (7), and talinolol (8). Dual substrates are indinavir (9), quinidine 0.1 mg (10-1), quinidine 1 mg (10-2), saquinavir (11), verapamil 0.1 mg (12-1), and verapamil 3 mg (12-2).

View this table:
TABLE 6

Role of CYP3A and P-gp in the intestinal availability of CYP3A and P-gp substrates

Prediction of Dose-Dependent FaFg of CYP3A/P-gp Substrate Drugs with Predefined SFs.

To further verify the predictability of dose-dependent FaFg of several CYP3A/P-gp substrate drugs with our approach, we predicted the dose-dependent FaFg of substrates by using the ACAT model with our optimized SFs. The predicted FaFg values and their ratio to that at minimum dose, and the observed AUC/dose ratio to that at minimum dose obtained from the clinical reports are shown in Table 7. The predicted FaFg ratios of triazolam, celiprolol, digoxin, fexofenadine, and indinavir were constant in their clinical dose range. In contrast, the predicted FaFg ratios of felodipine, midazolam, sildenafil, talinolol, quinidine, and verapamil increased in a dose-dependent manner, whereas that of saquinavir decreased.

View this table:
TABLE 7

Prediction of the dose-dependent intestinal availability of CYP3A and P-gp substrates

Discussion

In this study, we tried to establish an appropriate method for the quantitative prediction of the dose-dependent FaFg of drugs based on the in vitro saturation kinetics of CYP3A and P-gp using a mathematical model for the small intestine. SFCYP3A and SFP-gp were simultaneously optimized to well explain the FaFg of selective and dual CYP3A and/or P-gp substrate drugs. The best overall predictability of FaFg values of these drugs was achieved when considering both SFCYP3A and SFP-gp.

To consider the relative importance of CYP3A4 and CYP3A5 in the intestine, when considering only CYP3A4-mediated metabolism in the intestine, the optimized SFCYP3A and SFP-gp were estimated to be 0.0811 and 0.00643, which are comparable to the default SFs (0.0740 and 0.00651). This result indicated that CYP3A5-mediated metabolism had a minor contribution to the intestinal absorption of tested substrates.

When the FaFg of substrates was calculated considering only the passive permeability, the predicted FaFg of most substrates was close to 100%, and was not consistent with the observed FaFg (Fig. 2A). This result indicated that all of the tested substrates had good transcellular permeability, and that P-gp and CYP3A were clearly involved in suppressing the intestinal absorption of those substrates. After optimization of SFCYP3A and SFP-gp, the predicted FaFg values of 8 of 12 substrates were within 1.5-fold deviations from the observed FaFg. Therefore, our method can quantitatively predict the FaFg of CYP3A/P-gp substrates. However, the predicted FaFg value of indinavir with optimized SFCYP3A and SFP-gp was 0.113, which was much lower than the observed FaFg (Fig. 2B). When comparing the observed FaFg of drugs with the predicted ones in simulation IV (assuming that both intestinal P-gp and CYP3A are not functional), only predicted FaFg of indinavir was lower than the observed one (Table 6). One of the explanations for this discrepancy is that solubility of indinavir in the intestinal lumen was underestimated in our model. Actually, the simulated dissolved fraction in the intestinal lumen estimated with our ACAT model (Supplemental Table 5) revealed that all of the tested substrates, except for indinavir and saquinavir, were well dissolved at 48 hours after dosing. Gertz et al. (2011) performed Fg prediction of CYP3A substrates with a physiologically-based pharmacokinetic model using in vitro clearance data and solubility in fasted state–simulated intestinal fluid (FaSSIF). It has been reported that the solubilities of indinavir in FaSSIF (pH 6.5) and fed state–simulated intestinal fluid (FeSSIF; pH 5.0) are 0.048 and 1.4 mg/ml, respectively (Holmstock et al., 2013), whereas those of indinavir in the buffer at pH 6.1 and 5.1 in our experiment were 0.035 and 0.18 mg/ml, respectively. The solubility in FaSSIF was comparable to that in buffer, whereas the solubility in FeSSIF was 8-fold higher than that in buffer. The clinical study of indinavir was conducted under fasted condition (Yeh et al., 1999). The underestimation of predicted FaFg of indinavir is not improved, even if the solubility in FaSSIF is used for the simulation. Assuming that solubility in buffer was 8-fold higher at each pH, the predicted FaFg of indinavir with optimized SFCYP3A and SFP-gp increased to 0.530. The solubility of saquinavir in FeSSIF at pH 6.5 was reported as 0.456 mg/ml (Kawai et al., 2011), which is 2-fold higher than that in the buffer (0.239 mg/ml at pH 6.9). The predicted FaFg of saquinavir was increased to 0.0631 considering the similar assumption of high solubility as indinavir. Thus, by considering the solubility in simulated intestinal fluid, the predicted FaFg of indinavir largely approached the real one, whereas that of saquinavir was still maintained as low. GastroPlus incorporates the effect of bile salts on the solubility of drug in the gastrointestinal tract by taking into account the solubilization by bile salts. For indinavir, predicted FaFg was estimated to be 0.331 under fed conditions considering the effect of bile acids. Thus, for low-solubility compounds, the predictability of FaFg may be improved by considering the effect of bile acid, and the estimation of solubility needs to be further optimized. When we tried to perform an optimization of SFs for P-gp and CYP3A without using the data for indinavir, SFCYP3A and SFP-gp were estimated to be 0.0781 and 0.00682, respectively, which were almost the same as the optimized SFs with all substrates (SFCYP3A = 0.0778 and SFP-gp = 0.00651). Thus, the inclusion of indinavir in the test set did not largely affect the optimized SFs in this study.

Using the ACAT model with our optimized SFs, the relative importance of CYP3A and P-gp to the intestinal absorption of drugs was also estimated. As for verapamil, its predicted FaFg was identical between simulations I and II, but the predicted FaFg was higher than the observed one in simulation III. This result suggested that CYP3A is the main contributor to the low FaFg of verapamil. Conversely, the predicted FaFg of quinidine was higher than the observed one in both simulations II and III, suggesting both CYP3A and P-gp contributed to the low FaFg of quinidine. We previously demonstrated that the dose-normalized plasma AUCs of verapamil and quinidine increased in a dose-dependent manner, possibly due to the saturation of P-gp and CYP3A in the gut (Maeda et al., 2011). As shown in Fig. 3, the dose-dependent FaFg values of verapamil and quinidine were well reproduced with the optimized SFCYP3A and SFP-gp. Moreover, nonlinear intestinal absorption of verapamil was dominated by the saturation of intestinal CYP3A, whereas that of quinidine was caused by the saturation of both CYP3A and P-gp. In the case of saquinavir, which is a dual substrate of CYP3A and P-gp, its predicted FaFg values in simulations I and III were identical. However, in simulation II, the predicted FaFg of saquinavir was higher than the observed one. This result suggested that P-gp mainly contributed to the low FaFg of saquinavir at clinical dose. Due to the low Km value (0.395 μM) of saquinavir for CYP3A4, intestinal CYP3A4 was thought to be completely saturated at clinical dose.

Fig. 3.
Fig. 3.

Role of CYP3A and P-gp in the nonlinear intestinal absorption of verapamil and quinidine. Intestinal availability after oral administration of verapamil at a dose range of 0.1–80 mg (A) and that of quinidine at a dose range of 0.1–100 mg (B) were simulated. Simulation was performed using optimized SFCYP3A and SFP-gp (− − −), optimized SFCYP3A and SFP-gp = 0 (− • −), SFCYP3A = 0 and optimized SFP-gp (• • •), and SFCYP3A and SFP-gp = 0 (—-). Square symbols represent the mean FaFg value observed in vivo. Each point represents the mean ± S.D. [n = 8 (verapamil) and 7 (quinidine)].

We also predicted the dose-dependent FaFg of substrates with the optimized SFs (Table 7). Felodipine, midazolam, sildenafil, talinolol, quinidine, and verapamil showed nonlinear pharmacokinetics in previous clinical studies (Bornemann et al., 1985; Nakashima et al., 1992; Wetterich et al., 1996; Nichols et al., 2002; Maeda et al., 2011). Their predicted FaFg ratios increased in a similar manner as the dose-dependent increase in the observed AUC/dose ratios. Cuin max at a high dose of quinidine (200 and 600 mg) was higher than the Km value for CYP3A, indicating hepatic CYP3A would be saturated. In contrast, the predicted FaFg of indinavir did not change as the dose increased. Intestinal CYP3A was saturated at 400 mg because predicted FaFg was not changed between simulations I and III in Table 6. Since the Cuin max of indinavir (17.5 μM at 400-mg dose) was much higher than its KmCYP3A4, its supraproportional AUC increase with dose was possibly due to the saturation of hepatic CYP3A, as previously reported (Ke et al., 2012). Although celiprolol showed nonlinear pharmacokinetics in the clinical dose range, the predicted FaFg of celiprolol was constant at different doses. Celiprolol is a substrate of efflux transporter (P-gp) and influx transporter (OATP2B1) on the apical membrane of enterocytes. In our in vitro assay with Caco-2 cells, the concentration-dependent nonlinear transcellular transport was not observed. It has been reported that OATP2B1 is expressed on Caco-2 cells (Englund et al., 2006). Both influx and efflux transporters may have been involved in transcellular transport of celiprolol. However, we could not find any clear reasons for such discrepancy. According to the clinical report, the AUC/dose ratio of saquinavir decreased at higher doses. The predicted FaFg ratio of saquinavir also decreased dose dependently, and was close to the clinically observed AUC/dose ratios. Cuin max and maximum Cuent of saquinavir at doses of 400 and 600 mg were higher than KCYP3A4 (Table 7), suggesting hepatic and intestinal CYP3A-mediated metabolism was expected to be saturated. In addition, due to the low solubility of saquinavir (simulated solubility; 0.44 mg/ml at pH 6.5), drug absorption would be limited by solubility. Therefore, nonlinear pharmacokinetics of saquinavir may be caused by complex mechanisms, such as saturation of both metabolism and dissolution. Previous reports demonstrated that triazolam, digoxin, and fexofenadine showed linear pharmacokinetics within the clinical dose range (Table 7). Their predicted FaFg values were maintained as constant at different doses. Overall, the dose-dependent FaFg was well predicted even if substrates showed nonlinear intestinal absorption.

In conclusion, using the optimized SFCYP3A and SFP-gp based on the data of 12 P-gp and/or CYP3A substrate drugs, FaFg can be well predicted by in vitro parameters for CYP3A and P-gp using the ACAT model. The linearity of the dose-dependent intestinal absorption of those drugs could be theoretically judged by incorporating kinetic parameters estimated by in vitro experiments into the ACAT model. We determined that one of the key points to successfully obtain universal SFs is to use the reliable kinetic parameters obtained from unified in vitro experiments, since their large interlaboratory differences were reported. Once in-house SFs are set up with a set of test compounds, FaFg of a new compound can surely be predicted at any dose from in vitro data.

Authorship Contributions

Participated in research design: Takano, Maeda, Bolger, Sugiyama.

Conducted experiments: Takano, Maeda, Sugiyama.

Contributed new reagents or analytic tools: Takano, Maeda.

Performed data analysis: Takano, Maeda, Sugiyama.

Wrote or contributed to the writing of the manuscript: Takano, Maeda, Bolger, Sugiyama.

Footnotes

    • Received February 10, 2016.
    • Accepted August 17, 2016.

  • J.T. and K.M. equally contributed to this work.

  • This study was supported in part by a Grant-in-Aid for Young Scientists (A) [Grant 24689009] from the Ministry of Education, Culture, Sports, Science and Technology (to K.M.).

  • dx.doi.org/10.1124/dmd.116.070011.

  • ↵Embedded ImageThis article has supplemental material available at dmd.aspetjournals.org.

Abbreviations

ACAT model
advanced compartmental absorption and transit model
afe
average fold error
AUC
area under the concentration-time curve
ESF
enzyme/transporter scaling factor
FaFg
intestinal availability
FaSSIF
fasted state–simulated intestinal fluid
FeSSIF
fed state–simulated intestinal fluid
HBSS
Hanks’ balanced salt solution
LC-MS/MS
liquid chromatography–tandem mass spectrometry
MD
microdose
P450
cytochrome P450
P-gp
P-glycoprotein
RED
Rapid Equilibrium Dialysis
rmse
root mean squared error
SF
scaling factor
ThD
therapeutic dose
UK-393,664
3-ethyl-5-{5-[(4-ethylpiperazin-1-yl)sulfonyl]-2-propoxyphenyl}-2-(pyridin-2-ylmethyl)-2,6-dihydro-7H-pyrazolo(4,3-d)pyrimidin-7-one

References

  1. ↵
    1. Abuasal BS,
    2. Bolger MB,
    3. Walker DK, and
    4. Kaddoumi A
    (2012) In silico modeling for the nonlinear absorption kinetics of UK-343,664: a P-gp and CYP3A4 substrate. Mol Pharm 9:492–504.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Agoram B,
    2. Woltosz WS, and
    3. Bolger MB
    (2001) Predicting the impact of physiological and biochemical processes on oral drug bioavailability. Adv Drug Deliv Rev 50 (Suppl 1):S41–S67.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Bengtsson-Hasselgren B,
    2. Rönn O,
    3. Blychert LO,
    4. Edgar B, and
    5. Raner S
    (1990) Acute effects of felodipine and nifedipine on hepatic and forearm blood flow in healthy men. Eur J Clin Pharmacol 38:529–533.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Bergström CAS,
    2. Luthman K, and
    3. Artursson P
    (2004) Accuracy of calculated pH-dependent aqueous drug solubility. Eur J Pharm Sci 22:387–398.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Bolger MB,
    2. Lukacova V, and
    3. Woltosz WS
    (2009) Simulations of the nonlinear dose dependence for substrates of influx and efflux transporters in the human intestine. AAPS J 11:353–363.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Bornemann LD,
    2. Min BH,
    3. Crews T,
    4. Rees MM,
    5. Blumenthal HP,
    6. Colburn WA, and
    7. Patel IH
    (1985) Dose dependent pharmacokinetics of midazolam. Eur J Clin Pharmacol 29:91–95.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Caruso FS,
    2. Doshan HD,
    3. Hernandez PH,
    4. Costello R,
    5. Applin W, and
    6. Neiss ES
    (1985) Celiprolol: pharmacokinetics and duration of pharmacodynamic activity. Br J Clin Pract Suppl 40:12–16.
    OpenUrlPubMed
    1. Damkier P,
    2. Hansen LL, and
    3. Brøsen K
    (1999) Effect of fluvoxamine on the pharmacokinetics of quinidine. Eur J Clin Pharmacol 55:451–456.
    OpenUrlCrossRefPubMed
    1. Darbar D,
    2. Dell’Orto S,
    3. Mörike K,
    4. Wilkinson GR, and
    5. Roden DM
    (1997) Dietary salt increases first-pass elimination of oral quinidine. Clin Pharmacol Ther 61:292–300.
    OpenUrlCrossRefPubMed
    1. Edgar B,
    2. Regårdh CG,
    3. Johnsson G,
    4. Johansson L,
    5. Lundborg P,
    6. Löfberg I, and
    7. Rönn O
    (1985) Felodipine kinetics in healthy men. Clin Pharmacol Ther 38:205–211.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Emoto C and
    2. Iwasaki K
    (2007) Approach to predict the contribution of cytochrome P450 enzymes to drug metabolism in the early drug-discovery stage: the effect of the expression of cytochrome b(5) with recombinant P450 enzymes. Xenobiotica 37:986–999.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Englund G,
    2. Rorsman F,
    3. Rönnblom A,
    4. Karlbom U,
    5. Lazorova L,
    6. Gråsjö J,
    7. Kindmark A, and
    8. Artursson P
    (2006) Regional levels of drug transporters along the human intestinal tract: co-expression of ABC and SLC transporters and comparison with Caco-2 cells. Eur J Pharm Sci 29:269–277.
    OpenUrlCrossRefPubMed
    1. Fromm MF,
    2. Busse D,
    3. Kroemer HK, and
    4. Eichelbaum M
    (1996) Differential induction of prehepatic and hepatic metabolism of verapamil by rifampin. Hepatology 24:796–801.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Galetin A,
    2. Brown C,
    3. Hallifax D,
    4. Ito K, and
    5. Houston JB
    (2004) Utility of recombinant enzyme kinetics in prediction of human clearance: impact of variability, CYP3A5, and CYP2C19 on CYP3A4 probe substrates. Drug Metab Dispos 32:1411–1420.
    OpenUrlAbstract/FREE Full Text
    1. Gerecke M
    (1983) Chemical structure and properties of midazolam compared with other benzodiazepines. Br J Clin Pharmacol 16 (Suppl 1):11S–16S.
    OpenUrlCrossRefPubMed
    1. Gertz M,
    2. Harrison A,
    3. Houston JB, and
    4. Galetin A
    (2010) Prediction of human intestinal first-pass metabolism of 25 CYP3A substrates from in vitro clearance and permeability data. Drug Metab Dispos 38:1147–1158.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Gertz M,
    2. Houston JB, and
    3. Galetin A
    (2011) Physiologically based pharmacokinetic modeling of intestinal first-pass metabolism of CYP3A substrates with high intestinal extraction. Drug Metab Dispos 39:1633–1642.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Gramatté T,
    2. Oertel R,
    3. Terhaag B, and
    4. Kirch W
    (1996) Direct demonstration of small intestinal secretion and site-dependent absorption of the beta-blocker talinolol in humans. Clin Pharmacol Ther 59:541–549.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Greenblatt DJ,
    2. von Moltke LL,
    3. Harmatz JS,
    4. Counihan M,
    5. Graf JA,
    6. Durol AL,
    7. Mertzanis P,
    8. Duan SX,
    9. Wright CE, and
    10. Shader RI
    (1998a) Inhibition of triazolam clearance by macrolide antimicrobial agents: in vitro correlates and dynamic consequences. Clin Pharmacol Ther 64:278–285.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Greenblatt DJ,
    2. Wright CE,
    3. von Moltke LL,
    4. Harmatz JS,
    5. Ehrenberg BL,
    6. Harrel LM,
    7. Corbett K,
    8. Counihan M,
    9. Tobias S, and
    10. Shader RI
    (1998b) Ketoconazole inhibition of triazolam and alprazolam clearance: differential kinetic and dynamic consequences. Clin Pharmacol Ther 64:237–247.
    OpenUrlCrossRefPubMed
    1. Grube S,
    2. Langguth P,
    3. Junginger HE,
    4. Kopp S,
    5. Midha KK,
    6. Shah VP,
    7. Stavchansky S,
    8. Dressman JB, and
    9. Barends DM
    (2009) Biowaiver monographs for immediate release solid oral dosage forms: quinidine sulfate. J Pharm Sci 98:2238–2251.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Oritiz de Montellano PR
    1. Guengerich FP
    (1995) Human cytochrome P450 enzymes., in Cytochrome P450 (Oritiz de Montellano PR ed) pp 473–535, Plenum Press, New York.
  23. ↵
    1. Haufroid V,
    2. Mourad M,
    3. Van Kerckhove V,
    4. Wawrzyniak J,
    5. De Meyer M,
    6. Eddour DC,
    7. Malaise J,
    8. Lison D,
    9. Squifflet J-P, and
    10. Wallemacq P
    (2004) The effect of CYP3A5 and MDR1 (ABCB1) polymorphisms on cyclosporine and tacrolimus dose requirements and trough blood levels in stable renal transplant patients. Pharmacogenetics 14:147–154.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Hirota N,
    2. Ito K,
    3. Iwatsubo T,
    4. Green CE,
    5. Tyson CA,
    6. Shimada N,
    7. Suzuki H, and
    8. Sugiyama Y
    (2001) In vitro/in vivo scaling of alprazolam metabolism by CYP3A4 and CYP3A5 in humans. Biopharm Drug Dispos 22:53–71.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Holmstock N,
    2. De Bruyn T,
    3. Bevernage J,
    4. Annaert P,
    5. Mols R,
    6. Tack J, and
    7. Augustijns P
    (2013) Exploring food effects on indinavir absorption with human intestinal fluids in the mouse intestine. Eur J Pharm Sci 49:27–32.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Hsu A,
    2. Granneman GR,
    3. Cao G,
    4. Carothers L,
    5. el-Shourbagy T,
    6. Baroldi P,
    7. Erdman K,
    8. Brown F,
    9. Sun E, and
    10. Leonard JM
    (1998) Pharmacokinetic interactions between two human immunodeficiency virus protease inhibitors, ritonavir and saquinavir. Clin Pharmacol Ther 63:453–464.
    OpenUrlCrossRefPubMed
    1. Hughes IE,
    2. Ilett KF, and
    3. Jellett LB
    (1975) The distribution of quinidine in human blood. Br J Clin Pharmacol 2:521–525.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Kawai Y,
    2. Fujii Y,
    3. Tabata F,
    4. Ito J,
    5. Metsugi Y,
    6. Kameda A,
    7. Akimoto K, and
    8. Takahashi M
    (2011) Profiling and trend analysis of food effects on oral drug absorption considering micelle interaction and solubilization by bile micelles. Drug Metab Pharmacokinet 26:180–191.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Ke AB,
    2. Nallani SC,
    3. Zhao P,
    4. Rostami-Hodjegan A, and
    5. Unadkat JD
    (2012) A PBPK Model to Predict Disposition of CYP3A-Metabolized Drugs in Pregnant Women: Verification and Discerning the Site of CYP3A Induction. CPT Pharmacometrics Syst Pharmacol 1:e3.
    OpenUrlCrossRef
    1. Kroboth PD,
    2. McAuley JW,
    3. Kroboth FJ,
    4. Bertz RJ, and
    5. Smith RB
    (1995) Triazolam pharmacokinetics after intravenous, oral, and sublingual administration. J Clin Psychopharmacol 15:259–262.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Kudo T,
    2. Ozaki Y,
    3. Kusano T,
    4. Hotta E,
    5. Oya Y,
    6. Komatsu S,
    7. Goda H, and
    8. Ito K
    (2016) Effect of buffer conditions on CYP2C8-mediated paclitaxel 6α-hydroxylation and CYP3A4-mediated triazolam α- and 4-hydroxylation by human liver microsomes. Xenobiotica 46:241–246.
    OpenUrlCrossRefPubMed
    1. Kupferschmidt HH,
    2. Fattinger KE,
    3. Ha HR,
    4. Follath F, and
    5. Krähenbühl S
    (1998) Grapefruit juice enhances the bioavailability of the HIV protease inhibitor saquinavir in man. Br J Clin Pharmacol 45:355–359.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Lappin G,
    2. Kuhnz W,
    3. Jochemsen R,
    4. Kneer J,
    5. Chaudhary A,
    6. Oosterhuis B,
    7. Drijfhout WJ,
    8. Rowland M, and
    9. Garner RC
    (2006) Use of microdosing to predict pharmacokinetics at the therapeutic dose: experience with 5 drugs. Clin Pharmacol Ther 80:203–215.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Lappin G,
    2. Shishikura Y,
    3. Jochemsen R,
    4. Weaver RJ,
    5. Gesson C,
    6. Brian Houston J,
    7. Oosterhuis B,
    8. Bjerrum OJ,
    9. Grynkiewicz G,
    10. Alder J,
    11. et al.
    (2011) Comparative pharmacokinetics between a microdose and therapeutic dose for clarithromycin, sumatriptan, propafenone, paracetamol (acetaminophen), and phenobarbital in human volunteers. Eur J Pharm Sci 43:141–150.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Lappin G,
    2. Shishikura Y,
    3. Jochemsen R,
    4. Weaver RJ,
    5. Gesson C,
    6. Houston B,
    7. Oosterhuis B,
    8. Bjerrum OJ,
    9. Rowland M, and
    10. Garner C
    (2010) Pharmacokinetics of fexofenadine: evaluation of a microdose and assessment of absolute oral bioavailability. Eur J Pharm Sci 40:125–131.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Lin YS,
    2. Dowling ALS,
    3. Quigley SD,
    4. Farin FM,
    5. Zhang J,
    6. Lamba J,
    7. Schuetz EG, and
    8. Thummel KE
    (2002) Co-regulation of CYP3A4 and CYP3A5 and contribution to hepatic and intestinal midazolam metabolism. Mol Pharmacol 62:162–172.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Maeda K and
    2. Sugiyama Y
    (2011) Novel strategies for microdose studies using non-radiolabeled compounds. Adv Drug Deliv Rev 63:532–538.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Maeda K,
    2. Takano J,
    3. Ikeda Y,
    4. Fujita T,
    5. Oyama Y,
    6. Nozawa K,
    7. Kumagai Y, and
    8. Sugiyama Y
    (2011) Nonlinear pharmacokinetics of oral quinidine and verapamil in healthy subjects: a clinical microdosing study. Clin Pharmacol Ther 90:263–270.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Mai I,
    2. Perloff ES,
    3. Bauer S,
    4. Goldammer M,
    5. Johne A,
    6. Filler G,
    7. Budde K, and
    8. Roots I
    (2004) MDR1 haplotypes derived from exons 21 and 26 do not affect the steady-state pharmacokinetics of tacrolimus in renal transplant patients. Br J Clin Pharmacol 58:548–553.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Nakashima M,
    2. Kanemaru M,
    3. Takiguchi Y, and
    4. Mizuno A
    (1988) Phase I Study of NBP-582 (Celiprolol). Rinsyoiyaku 4:1075–1090.
    OpenUrl
  41. ↵
    1. Nakashima M,
    2. Kanamaru M,
    3. Ujita S,
    4. Shimizu H,
    5. Sakamoto M, and
    6. Yokota N
    (1992) Study on Safety, Pharmacodynamics and Pharmacokinetics of Felodipine Following Single Oral Administration in Healthy Adult Male Volunteers. Rinsyoiyaku 8:1763–1780.
    OpenUrl
  42. ↵
    1. Naritomi Y,
    2. Terashita S,
    3. Kimura S,
    4. Suzuki A,
    5. Kagayama A, and
    6. Sugiyama Y
    (2001) Prediction of human hepatic clearance from in vivo animal experiments and in vitro metabolic studies with liver microsomes from animals and humans. Drug Metab Dispos 29:1316–1324.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Nichols DJ,
    2. Muirhead GJ, and
    3. Harness JA
    (2002) Pharmacokinetics of sildenafil after single oral doses in healthy male subjects: absolute bioavailability, food effects and dose proportionality. Br J Clin Pharmacol 53 (Suppl 1):5S–12S.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Obach RS,
    2. Baxter JG,
    3. Liston TE,
    4. Silber BM,
    5. Jones BC,
    6. MacIntyre F,
    7. Rance DJ, and
    8. Wastall P
    (1997) The prediction of human pharmacokinetic parameters from preclinical and in vitro metabolism data. J Pharmacol Exp Ther 283:46–58.
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Obach RS and
    2. Reed-Hagen AE
    (2002) Measurement of Michaelis constants for cytochrome P450-mediated biotransformation reactions using a substrate depletion approach. Drug Metab Dispos 30:831–837.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Paine MF,
    2. Khalighi M,
    3. Fisher JM,
    4. Shen DD,
    5. Kunze KL,
    6. Marsh CL,
    7. Perkins JD, and
    8. Thummel KE
    (1997) Characterization of interintestinal and intraintestinal variations in human CYP3A-dependent metabolism. J Pharmacol Exp Ther 283:1552–1562.
    OpenUrlAbstract/FREE Full Text
    1. Perdaems N,
    2. Blasco H,
    3. Vinson C,
    4. Chenel M,
    5. Whalley S,
    6. Cazade F, and
    7. Bouzom F
    (2010) Predictions of metabolic drug-drug interactions using physiologically based modelling: Two cytochrome P450 3A4 substrates coadministered with ketoconazole or verapamil. Clin Pharmacokinet 49:239–258.
    OpenUrlCrossRefPubMed
    1. Rengelshausen J,
    2. Göggelmann C,
    3. Burhenne J,
    4. Riedel K-D,
    5. Ludwig J,
    6. Weiss J,
    7. Mikus G,
    8. Walter-Sack I, and
    9. Haefeli WE
    (2003) Contribution of increased oral bioavailability and reduced nonglomerular renal clearance of digoxin to the digoxin-clarithromycin interaction. Br J Clin Pharmacol 56:32–38.
    OpenUrlCrossRefPubMed
    1. Rodgers T and
    2. Rowland M
    (2007) Mechanistic approaches to volume of distribution predictions: understanding the processes. Pharm Res 24:918–933.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Saitoh H and
    2. Aungst BJ
    (1995) Possible involvement of multiple P-glycoprotein-mediated efflux systems in the transport of verapamil and other organic cations across rat intestine. Pharm Res 12:1304–1310.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Sheiner LB and
    2. Beal SL
    (1981) Some suggestions for measuring predictive performance. J Pharmacokinet Biopharm 9:503–512.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Sun D,
    2. Lennernas H,
    3. Welage LS,
    4. Barnett JL,
    5. Landowski CP,
    6. Foster D,
    7. Fleisher D,
    8. Lee K-D, and
    9. Amidon GL
    (2002) Comparison of human duodenum and Caco-2 gene expression profiles for 12,000 gene sequences tags and correlation with permeability of 26 drugs. Pharm Res 19:1400–1416.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Tachibana T,
    2. Kato M, and
    3. Sugiyama Y
    (2012) Prediction of nonlinear intestinal absorption of CYP3A4 and P-glycoprotein substrates from their in vitro Km values. Pharm Res 29:651–668.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Tachibana T,
    2. Kitamura S,
    3. Kato M,
    4. Mitsui T,
    5. Shirasaka Y,
    6. Yamashita S, and
    7. Sugiyama Y
    (2010) Model analysis of the concentration-dependent permeability of P-gp substrates. Pharm Res 27:442–446.
    OpenUrlCrossRefPubMed
  55. ↵
    1. Taguchi M,
    2. Imaoka S,
    3. Yoshii K,
    4. Kobayashi K,
    5. Hosokawa M,
    6. Shimada N,
    7. Funae Y, and
    8. Chiba K
    (2001) Kinetics of testosterone 6beta-hydroxylation in the reconstituted system with similar ratios of purified CYP3A4, NADPH-cytochrome p450 oxidoreductase and cytochrome B5 to human liver microsomes. Res Commun Mol Pathol Pharmacol 109:53–63.
    OpenUrlPubMed
  56. ↵
    1. Thiebaut F,
    2. Tsuruo T,
    3. Hamada H,
    4. Gottesman MM,
    5. Pastan I, and
    6. Willingham MC
    (1987) Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 84:7735–7738.
    OpenUrlAbstract/FREE Full Text
    1. Tsunoda SM,
    2. Velez RL,
    3. von Moltke LL, and
    4. Greenblatt DJ
    (1999) Differentiation of intestinal and hepatic cytochrome P450 3A activity with use of midazolam as an in vivo probe: effect of ketoconazole. Clin Pharmacol Ther 66:461–471.
    OpenUrlCrossRefPubMed
    1. Uchimura T,
    2. Kato M,
    3. Saito T, and
    4. Kinoshita H
    (2010) Prediction of human blood-to-plasma drug concentration ratio. Biopharm Drug Dispos 31:286–297.
    OpenUrlPubMed
    1. Ueda CT,
    2. Williamson BJ, and
    3. Dzindzio BS
    (1976) Absolute quinidine bioavailability. Clin Pharmacol Ther 20:260–265.
    OpenUrlCrossRefPubMed
    1. Vanderveen RP,
    2. Jirak JL,
    3. Peters GR,
    4. Cox SR, and
    5. Bombardt PA
    (1991) Effect of ranitidine on the disposition of orally and intravenously administered triazolam. Clin Pharm 10:539–543.
    OpenUrlPubMed
  61. ↵
    1. Venkatakrishnan K,
    2. von Moltke LL,
    3. Court MH,
    4. Harmatz JS,
    5. Crespi CL, and
    6. Greenblatt DJ
    (2000) Comparison between cytochrome P450 (CYP) content and relative activity approaches to scaling from cDNA-expressed CYPs to human liver microsomes: ratios of accessory proteins as sources of discrepancies between the approaches. Drug Metab Dispos 28:1493–1504.
    OpenUrlPubMed
  62. ↵
    1. Wacher VJ,
    2. Silverman JA,
    3. Zhang Y, and
    4. Benet LZ
    (1998) Role of P-glycoprotein and cytochrome P450 3A in limiting oral absorption of peptides and peptidomimetics. J Pharm Sci 87:1322–1330.
    OpenUrlCrossRefPubMed
  63. ↵
    1. Wetterich U,
    2. Spahn-Langguth H,
    3. Mutschler E,
    4. Terhaag B,
    5. Rösch W, and
    6. Langguth P
    (1996) Evidence for intestinal secretion as an additional clearance pathway of talinolol enantiomers: concentration- and dose-dependent absorption in vitro and in vivo. Pharm Res 13:514–522.
    OpenUrlCrossRefPubMed
    1. Winiwarter S,
    2. Bonham NM,
    3. Ax F,
    4. Hallberg A,
    5. Lennernäs H, and
    6. Karlén A
    (1998) Correlation of human jejunal permeability (in vivo) of drugs with experimentally and theoretically derived parameters. A multivariate data analysis approach. J Med Chem 41:4939–4949.
    OpenUrlCrossRefPubMed
  65. ↵
    1. Wrighton SA,
    2. Brian WR,
    3. Sari MA,
    4. Iwasaki M,
    5. Guengerich FP,
    6. Raucy JL,
    7. Molowa DT, and
    8. Vandenbranden M
    (1990) Studies on the expression and metabolic capabilities of human liver cytochrome P450IIIA5 (HLp3). Mol Pharmacol 38:207–213.
    OpenUrlAbstract
  66. ↵
    1. Wynne HA,
    2. Cope LH,
    3. Mutch E,
    4. Rawlins MD,
    5. Woodhouse KW, and
    6. James OF
    (1989) The effect of age upon liver volume and apparent liver blood flow in healthy man. Hepatology 9:297–301.
    OpenUrlCrossRefPubMed
  67. ↵
    1. Yamane N,
    2. Tozuka Z,
    3. Sugiyama Y,
    4. Tanimoto T,
    5. Yamazaki A, and
    6. Kumagai Y
    (2007) Microdose clinical trial: quantitative determination of fexofenadine in human plasma using liquid chromatography/electrospray ionization tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 858:118–128.
    OpenUrlCrossRefPubMed
  68. ↵
    1. Yamaoka K,
    2. Tanigawara Y,
    3. Nakagawa T, and
    4. Uno T
    (1981) A pharmacokinetic analysis program (multi) for microcomputer. J Pharmacobiodyn 4:879–885.
    OpenUrlCrossRefPubMed
    1. Yamazaki A,
    2. Kumagai Y,
    3. Yamane N,
    4. Tozuka Z,
    5. Sugiyama Y,
    6. Fujita T,
    7. Yokota S, and
    8. Maeda M
    (2010) Microdose study of a P-glycoprotein substrate, fexofenadine, using a non-radioisotope-labelled drug and LC/MS/MS. J Clin Pharm Ther 35:169–175.
    OpenUrlCrossRefPubMed
  70. ↵
    1. Yeh KC,
    2. Deutsch PJ,
    3. Haddix H,
    4. Hesney M,
    5. Hoagland V,
    6. Ju WD,
    7. Justice SJ,
    8. Osborne B,
    9. Sterrett AT,
    10. Stone JA,
    11. et al.
    (1998) Single-dose pharmacokinetics of indinavir and the effect of food. Antimicrob Agents Chemother 42:332–338.
    OpenUrlAbstract/FREE Full Text
  71. ↵
    1. Yeh KC,
    2. Stone JA,
    3. Carides AD,
    4. Rolan P,
    5. Woolf E, and
    6. Ju WD
    (1999) Simultaneous investigation of indinavir nonlinear pharmacokinetics and bioavailability in healthy volunteers using stable isotope labeling technique: study design and model-independent data analysis. J Pharm Sci 88:568–573.
    OpenUrlCrossRefPubMed
  72. ↵
    1. Yu LX and
    2. Amidon GL
    (1999) A compartmental absorption and transit model for estimating oral drug absorption. Int J Pharm 186:119–125.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleArticle

Predictive Model for Nonlinear Intestinal Absorption

Junichi Takano, Kazuya Maeda, Michael B. Bolger and Yuichi Sugiyama
Drug Metabolism and Disposition November 1, 2016, 44 (11) 1808-1818; DOI: https://doi.org/10.1124/dmd.116.070011
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement