Advertisement
Research ArticleArticle

Utility of Hepatocytes in Predicting Drug Metabolism: Comparison of Hepatic Intrinsic Clearance in Rats and Humans in Vivo and in Vitro

Yoichi Naritomi, Shigeyuki Terashita, Akira Kagayama and Yuichi Sugiyama
Drug Metabolism and Disposition May 2003, 31 (5) 580-588; DOI: https://doi.org/10.1124/dmd.31.5.580

Abstract

We investigated hepatic in vitro intrinsic clearance (CLint,in vitro) in freshly isolated or cryopreserved hepatocytes and compared with CLint,in vivo by using nine model compounds, FK1052, FK480, diazepam, diltiazem, troglitazone, quinotolast, FK079, zidovudine, and acetaminophen, in rats and humans. The compounds showed a broad range of in vivo hepatic extraction ratios (rat, 0.05–0.93; humans, 0.03–0.76) and were metabolized by hepatic P450, UDP-glucuronosyltransferase, sulfotransferase, and/or esterase. CLint,in vitro was determined from substrate disappearance rate at 1 μM in hepatocytes. CLint,in vivowas calculated from in vivo pharmacokinetic data using two frequently used mathematical models (the well stirred and dispersion models). When estimating rat CLint,in vitro in freshly isolated hepatocytes, the rat scaling factor values (CLint,in vivo/CLint,in vitro) showed marked difference among the model compounds (0.2–73.1-fold). The rat CLint,in vitro values in freshly isolated hepatocytes were in good agreement with these in cryopreserved hepatocytes. Human CLint,in vitro were determined by use of cryopreserved hepatocytes. When human CLint,in vitro was regarded as the predicted CLint,in vivo, the observed and predicted CLint,in vivo for FK1052, FK480, troglitazone, and FK079 differed markedly (12.4–199.0-fold). In contrast, using human CLint,in vitro corrected with the rat scaling factors yielded better predictions of CLint,in vivo that were mostly within 5-fold of the actual values. These results make the evaluation using hepatocytes more useful and provide a basis for predicting hepatic clearance using hepatocytes.

Recently, pharmacokinetic investigation has played an increasingly important role in drug discovery. In particular, it is very important to predict human hepatic metabolic clearance because most drugs are eliminated from the body predominantly by hepatic metabolism. For predicting hepatic clearance, theoretical aspects of in vitro/in vivo scaling based on a physiological model and clearance concepts have been developed (Roberts and Rowland, 1986). Application of this method has been successful in predicting in vivo hepatic clearance in rats for many drugs metabolized by P4501 from in vitro metabolism data using rat liver microsomes and isolated hepatocytes (Sugiyama et al., 1988; Houston and Carlile, 1997). Because human liver samples have become more readily available, it would also be very useful to predict in vivo outcomes from in vitro data in humans. However, there has been relatively limited application of this approach (Hoener, 1994), and there have been some failed attempts at the prediction of human hepatic clearance. For example, Iwatsubo et al. (1997) and Houston and Carlile (1997)reported that CLint,in vitro generally exhibited a positive correlation with CLint,in vivo, but in some cases animal or human clearance values were not well predicted from in vitro studies. To improve the predictions of human hepatic clearance, a few investigators have described new methods and approaches. For example, Lave et al. (1997) have proposed allometric scaling techniques combined with in vitro data. Obach (1999) has reported that inclusion of microsome binding values in the prediction of clearance from in vitro data seems to be a more broadly applicable approach. Recently, we also investigated the quantitative prediction of human hepatic metabolic clearance from in vitro experiments using liver microsomes focusing on P450 metabolism with eight model compounds (Naritomi et al., 2001). The values of CLint,in vitro and CLint,in vivo for each model compound were compared in rats and/or dogs, and humans. As a result, 1) scaling factor values, which are the ratios of CLint,in vivo to CLint,in vitro, were similar in the different animal species; 2) scaling factor values were different for each compound; 3) successful predictions of human CLint,in vivo were obtained by considering animal scaling factor; and 4) use of human CLint,in vitro corrected with animal scaling factor produced good predictions of CLoral andEH in humans. Namely, the empirical prediction method is a simple one of incorporating additional information (which is compound-specific) from animal studies and is based on the assumption that any in vitro-in vivo difference seen in humans is also apparent in animals to approximately the same degree.

It is helpful to use hepatic microsomes for evaluating compounds metabolized by P450. In contrast to microsomes, intact hepatocytes contain phase I (P450), phase II (conjugative) enzymes, and their cofactors. In addition, it seems that the use of hepatocytes can evaluate the involvement of active transporters. Accordingly, isolated hepatocytes provide an in vitro system for studying the integrated metabolism and distribution of compounds. Some studies have reported the prediction of hepatic clearance from in vitro experiments using freshly isolated hepatocytes, especially in rats (Ashforth et al., 1995; Carlile et al., 1998). However, there has been limited application of the prediction using human freshly isolated hepatocytes (Bayliss et al., 1999). In general, there is limited availability of fresh human livers for research. Cryopreservation would greatly enhance the utility of human hepatocytes because cryopreserved hepatocytes could be used at any time for experiments. Successful cryopreservation of human hepatocytes has been reported by several researchers (Chesne et al., 1993; Coundouris et al., 1993). However, there have been few reports on the prediction of hepatic clearance using human cryopreserved hepatocytes.

In the present study, we examined CLint,in vitroobtained from in vitro experiments using hepatocytes and CLint,in vivo calculated from in vivo pharmacokinetic data with nine model compounds (FK1052, FK480, diazepam, diltiazem, troglitazone, quinotolast, FK079, zidovudine, and acetaminophen) in rats and humans. In particular, we evaluated CLint,in vitro using cryopreserved hepatocytes in humans. At the same time, the measurement method of CLint,in vitro, which is determined from substrate disappearance rate at 1 μM in hepatocytes, was used because it is a simple and useful method.

Materials and Methods

Chemicals.

FK1052, FK480, troglitazone, quinotolast, and FK079 were synthesized by Fujisawa Pharmaceutical Co., Ltd. (Osaka, Japan). Diltiazem hydrochloride, zidovudine, and acetaminophen were purchased from Sigma-Aldrich (St. Louis, MO). Diazepam was purchased from Wako Pure Chemicals (Osaka, Japan). The other regents and solvents used were of analytical and HPLC grade.

Selection of Model Compounds.

FK1052, FK480, diazepam, diltiazem, troglitazone, quinotolast, FK079, zidovudine, and acetaminophen (Fig. 1) were selected as the model compounds based on the following criteria: clearance of model compounds is determined by hepatic P450, UDP-glucuronosyltransferase, sulfotransferase, and/or esterase metabolism; extrahepatic clearances are assumed to be negligible; in vivo pharmacokinetic parameters in rats and humans are reported; and extent of absorption is good with no species difference.

Figure 1
Figure 1

Chemical structures of model compounds.

Preparation of Freshly Isolated Rat Hepatocytes.

Adult male Sprague-Dawley rats and Wistar-Imamichi rats (200–250 g;n = 3) were obtained from Charles River Japan Inc. (Yokohama, Japan) and Imamichi Institute for Animal Reproduction (Ibaraki, Japan), respectively. Rat hepatocytes were freshly prepared by the in situ collagenase perfusion method as described previously (Tanaka et al., 1978). The cells were suspended in Williams' medium E (WME) containing l-glutamine (2 mM) purged with 95% O2, 5% CO2. Cell viability was assessed by the trypan blue exclusion test and preparations that were more than 80% viable were used. The viability of freshly isolated rat hepatocytes used was 84.7 ± 2.7%.

Cryopreserved Rat and Human Hepatocytes.

Cryopreserved rat and human hepatocytes were obtained from In Vitro Technologies Inc. (Baltimore, MD). Rat hepatocytes that were the pooled preparations of 24 adult male Sprague-Dawley rat livers were used. Eleven cryopreserved human hepatocytes (race, Caucasian; age, 33–69 years; male, six donors and female, five donors) were used for the evaluation. Of the human hepatocytes, human hepatocytes from five to seven individual donors were used for each model compound. These hepatocytes were stored in liquid N2 until use. Thawing was achieved by gently shaking the vials of cryopreserved hepatocytes in a 37°C water bath. As soon as all contents had been thawed, the vials were placed immediately on ice and diluted with WME at 4°C. After dilution, the hepatocyte suspension was washed by centrifugation at 50g for 5 min and resuspended in WME (Ruegg et al., 1997). The viability of cryopreserved rat and human hepatocytes used was 51.3 ± 1.7 and 73.9 ± 7.5%, respectively.

In Vitro Metabolism in Hepatocytes.

In vitro experiments

Time courses of the unchanged model compounds in hepatocytes were obtained. Each compound was incubated with a reaction mixture (500 μl) of hepatocyte suspension. After preincubation at 37°C for 5 min, the reactions were initiated by adding 5 μl of model compound solution in methanol. The final concentration of each model compound used was 1 μM. After incubation at 37°C for various time periods, the reactions of FK1052, FK480, diltiazem, diazepam, quinotolast, and troglitazone were terminated by the addition of acetonitrile. The reaction of FK079 was terminated by the addition of chloroform. The reactions of zidovudine and acetaminophen were terminated by the addition of ethyl acetate. After stopping the metabolic reactions, the reaction mixtures for FK1052, FK480, diltiazem, diazepam, quinotolast, and troglitazone were centrifuged at 10,000g for 5 min, and an aliquot of the supernatant was injected into an HPLC for measuring the unchanged compound concentration. The reactions of FK079, zidovudine, and acetaminophen were processed by extraction. After adding each internal standard [FR117460 (FK079 derivative), 4-nitrophenol, and 3-acetamidophenol, respectively], the organic fractions were evaporated under N2 and the residues were reconstituted in the mobile phase (see below) for HPLC analysis.

Determination of unchanged model compound concentrations.

An LC module I plus (Millipore Corporation, Bedford, MA) was used. The column for analysis was an Inertsil ODS-3 (5 μm, 150 × 4.6 mm) (GL Science Inc., Tokyo, Japan). The flow rate was 1.0 ml/min. The mobile phase and detection wavelength for the analysis of each model compound were as follows: FK1052, mobile phase: buffer*/CH3CN (40:60), detection UV 242 nm; FK480, mobile phase: buffer*/CH3CN (40:60), detection UV 295 nm; diltiazem, mobile phase: buffer*/CH3CN (40:60), detection UV 240 nm; diazepam, mobile phase: buffer*/CH3CN (40:60), detection UV 254 nm; quinotolast, mobile phase: buffer*/CH3CN (70:30), detection, fluorescence, excitation 265 nm, emission 480 nm; troglitazone, mobile phase: buffer*/CH3CN (50:50), detection UV 230 nm; FK079, mobile phase: buffer*/CH3CN (70:30), detection UV 280 nm; zidovudine, mobile phase: buffer*/CH3OH (80:20), detection UV 267 nm; acetaminophen, mobile phase: 0.5% acetic acid/CH3CN (90:10), detection UV 254 nm; and buffer*: 5 mM phosphate buffer (pH 7.2).

All assay methods were in the concentration range of 0.1 to 2 μM. Reproducibility was evaluated by performing five replicate analyses of hepatocyte samples containing 0.1, 0.5, and 1 μM compound, respectively. The coefficient of variation was less than 10%, and the actual concentration of the compounds ranged from 97.5 to 106%. All assay methods thus provide good accuracy and precision.

Calculation of CLint,in vitro.

CLint,in vitro values were calculated from substrate disappearance rate in hepatocytes as follows. If substrate disappearance can be assumed to follow first-order reaction, the unchanged drug profile as a function of time [C(t)] is described as follows.C(t)=C0·exp(−ke·t) Equation 1where C0 is initial concentration of the compound, ke is the disappearance rate constant of unchanged drug (min−1).

Furthermore, initial metabolic rate (V0) (micromoles per minute per cell) is described by eq. 2:V0=ke·C0/Dcell Equation 2where Dcell is the cell density (cells per milliliter).

On the other hand, from the Michaelis-Menten equation,V0 is described by the eq. 3:V0=Vmax·C0(Km+C0) Equation 3If the substrate concentration used in the experiments (1 μM) is below the Km for the drug-metabolizing enzyme reactions, the drug concentration may be assumed to be much smaller than Km(Km ≫C0). Thus,V0 can be expressed by eq. 4:Embedded Image Equation 4Consequently,CLint,in vitro=Vmax/Km=V0/C0 Equation 5CLint,in vitro was thus calculated by eq. 5 based on the time course of unchanged drug concentrations by a least-square linear regression. The CLint,in vitro values expressed per cell calculated from the in vitro metabolism experiments were expressed per kilogram of body weight by taking the number of hepatocytes per gram liver and the liver weight per kilogram of body weight shown in Table1 into consideration.

View this table:
Table 1

Physiological parameters for calculation of intrinsic clearance in rats and humans

In Vivo Data.

Sources of pharmacokinetic data

In vivo clearance under linear conditions,fu andRB data, were obtained from in-house and published literature. The in vivo pharmacokinetic data were considered to be trustworthy because the in vivo pharmacokinetic experiments were performed based on good accuracy methods and the appropriate protocols. In vivo clearance value was calculated by dividing the dose by area under the curve. In the case of high-clearance compounds, CLoral was used, because CLint,in vivo calculated from CLtot is affected by changes ofQH (Iwatsubo et al., 1997). For FK480 and diazepam, which are intermediate-clearance drugs in rats, CLoral was also used in the same way as the in vitro-in vivo scaling using hepatic microsomes (Naritomi et al., 2001). When the in vivo clearance value was not expressed per kilogram of body weight, this value was converted so that it was expressed per kilogram of body weight by taking the mean value of body weight in published literature or body weight of 250 g, 10 kg, and 70 kg for rats, dogs, and humans, respectively.Fa of FK1052, FK480, quinotolast, and FK079 were estimated by summing the recoveries of radioactivity in bile and urine after oral administration of 14C model compounds to rats from our own data.Fa for diazepam, diltiazem, troglitazone, zidovudine, and acetaminophen were estimated from published data. FK1052, FK480, diazepam, and diltiazem were evaluated as the compounds metabolized mainly by P450. Troglitazone, quinotolast, FK079, zidovudine, and acetaminophen were evaluated as the compounds metabolized by different drug-metabolizing enzymes (Table2).

View this table:
Table 2

Estimation of CLint,in vivo from in vivo pharmacokinetic data for model compounds in rats and humans

Calculation of CLint,in vivo.

CLH values were determined from eqs. 6 and 7 by use of the CLtot and CLoralvalues, respectively. For quinotolast, CLH in rats were calculated from pharmacokinetic parameters after administration into femoral vein and hepatic portal vein (Katashima et al., 1993a). CLR was considered to be negligible for FK1052, FK480, diazepam, diltiazem, quinotolast, troglitazone, and FK079.CLH=(CLtot/RB)−CLR Equation 6CLH=(CLoral/RB)·(FH·Fa)−CLR Equation 7CLint,in vivo was calculated from the following equations using the well stirred (Pang and Rowland, 1977) and dispersion models (Roberts and Rowland, 1986):CLH=QH·(1−FH) Equation 8Well­stirred model:FH=QH/{QH+(fp/RB)·CLint} Equation 9Disperison model: Equation 10FH=4a(1+a)2exp{(a−1)/2DN}−(1−a)2exp{−(a+1)/2DN} a=(1+4RN·DN)1/2 Equation 11RN=(fp/RB)·CLint/QH Equation 12Embedded Image Equation 13EH was calculated from eq. 14:EH=1−FH=CLH/QH Equation 14

Estimation of Scaling Factor and Prediction of Human CLint,in vivo.

Scaling factor was estimated from the following equation:Scaling factor=CLint,in vivo/CLint,in vitro Equation 15Human CLint,in vivo were predicted based on human CLint,in vitro using the following two methods: 1) disregarding rat scaling factor,Predicted human CLint,in vivo=human CLint,in vitro Equation 16and 2) including rat scaling factor, Predicted human CLint,in vivo=human CLint,in vitro·rat scaling factor Equation 17

Binding of Model Compounds to Isolated Rat Hepatocytes.

Determination of fu,hepatocytes .

Binding to isolated rat hepatocytes was determined using the hepatocyte reaction mixture described above. After adding each model compound (final concentration, 1 μM) to the hepatocyte suspension, the mixture was placed into a centrifuge tube containing a mixture of silicone and mineral oil (density, 1.017) and centrifuged (10,000g for 10 s) at room temperature. The cells were separated through the oil layer, and the supernatant fraction was removed with a pipette (Yamazaki et al., 1993). All procedures were completed within 1 min, so that metabolism of the model compound in the hepatocytes could be neglected, and the concentration of the model compound in the supernatant fraction could be determined (Lin et al., 1980).

Calculation of CLuint,in vitro and the scaling factor.

By incorporating the correction with the unbound fraction in the hepatocyte incubation mixture, CLuint,in vitro is defined as (Obach, 1999):CLuint,in vitro=CLint,in vitro/fu,hepatocytes Equation 18The value of scaling factor, which is the ratio of CLint,in vivo to CLuint,in vitro, was calculated from eq. 15.

Results

In Vivo Pharmacokinetic Data for Model Compounds.

In vivo pharmacokinetic data for the model compounds are summarized in Table 2. The fractions absorbed from the intestinal tract (Fa) were high, in the range of 0.7 to 1.0. The values of unbound fraction in plasma (or serum) (fp) ranged widely from the highest values for zidovudine (fp: rat, 0.786; human, 0.8) and acetaminophen (fp: rat, 0.82; human, 0.79) to the lowest value for troglitazone (fp: rat, 0.000921; human, 0.000941). In vivo clearance, CLint,in vivo and EH values differed markedly between rats and humans for each compound. For example, FK480 and diazepam are characterized by intermediate to highEH (0.62 and 0.64) in rats. On the contrary, low EH values (0.12 and 0.03) were observed in humans. In vivo rat and humanEH ranged widely among the model compounds (rat: 0.05 for FK079 and ∼0.93 for FK1052; human: 0.03 for diazepam and ∼0.76 for diltiazem). The model compounds represented a variety of metabolic pathways. FK1052, FK480, diazepam, and diltiazem are metabolized mainly by P450. In contrast, quinotolast, FK079, and zidovudine were metabolized by UDP-glucuronosyltransferase or esterase besides P450. Troglitazone and acetaminophen were metabolized mainly by UDP-glucuronosyltransferase and sulfotransferase. Of the compounds metabolized by different drug-metabolizing enzymes, some compounds showed species differences in the elimination routes and metabolite profiles. For example, FK079 is metabolized mainly by P450 in rats and by esterase in humans, respectively (Tokuda et al., 1997). For zidovudine, approximately 75 and 10% of the dose were excreted as its unchanged and glucuronide in rats, respectively (Mays et al., 1991). On the other hand, approximately 14 and 75% of the dose were excreted as its unchanged and glucuronide in humans, respectively (Blum et al., 1996). For acetaminophen, the primary route of metabolism is glucuronidation in rats and sulfation in humans, respectively (Hjelle and Klaassen, 1984; Sonne et al., 1988).

In Vitro Metabolism in Rat Hepatocytes and Estimation of Scaling Factor.

Fig. 2 illustrates the time courses of the unchanged model compounds in freshly isolated rat hepatocytes. The time courses were estimated by use of Sprague-Dawley rat hepatocytes, except for troglitazone. For troglitazone, the time course was estimated by use of Wistar-Imamichi rat hepatocytes, because the in vivo pharmacokinetic data in Wistar-Imamichi rats were used for evaluation. The unchanged drug profiles at 1 μM showed linear log concentration declines so that the metabolism follows a first-order reaction under these conditions. Table 3shows the CLint,in vitro calculated from the time courses of the model compounds in freshly isolated rat hepatocytes and the values for rat scaling factor (the ratios of CLint,in vivo obtained from in vivo pharmacokinetic data to CLint,in vitro). The rat scaling factor values calculated using the well stirred model were about 73.1-fold for troglitazone, 43.6-fold for FK1052, 25.1-fold for FK480, 10.5-fold for quinotolast, 6.1-fold for FK079, and 0.5- to 2.2-fold for diazepam, diltiazem, zidovudine, and acetaminophen, showing a marked difference among the model compounds. In the same way, the rat scaling factor values calculated using the dispersion models were 0.2- to 47.1-fold, showing marked differences among the model compounds.

Figure 2
Figure 2

Time courses of unchanged compounds in freshly isolated rat hepatocytes.

Each compound (at a concentration 1 μM) was incubated for various time periods at 37°C in freshly isolated rat hepatocytes. The values for cell density in the reaction mixtures were 0.5 × 106 cells/ml (FK1052, FK480, diazepam, diltiazem, quinotolast, and troglitazone), 1 × 106 cells/ml (acetaminophen), 2 × 106 cells/ml (zidovudine), and 4 × 106 cells/ml (FK079). Each point and bar represents the mean ± S.D. of three experiments. The solid lines represent the linear regression lines by the least-squares method.

View this table:
Table 3

Estimation of rat CLint,in vitro in freshly isolated hepatocytes and their scaling factor values for the model compounds

Table 4 shows the unbound fraction in freshly isolated rat hepatocytes, rat CLuint,in vitro corrected withfu,hepatocytes and the scaling factor values. Rat fu,hepatocytes values were determined by use of Sprague-Dawley rat hepatocytes, except for thefu,hepatocytes value for troglitazone, which was determined by use of Wistar-Imamichi rat hepatocytes. Ratfu,hepatocytes values were dependent on the model compounds. For example, although FK1052, FK480, and troglitazone were highly bound to hepatocytes with free-fraction values ranging from 0.202 to 0.244, diazepam, quinotolast, FK079, zidovudine, and acetaminophen showed low binding with free-fraction values ranging from 0.705 to 1.000. For FK1052, FK480, and troglitazone, the rat scaling factor values obtained by correcting rat CLint,in vitro with fu,hepatocytesbecame smaller. However, the scaling factor values were still ranged from 3.1- to 15.8-fold. For diazepam, quinotolast, FK079, zidovudine, and acetaminophen, corrections of CLint,in vitro with fu,hepatocytesdid not change the rat scaling factor values significantly. Consequently, the scaling factor for quinotolast and FK079 remained at 8.4- to 9.8-fold and 4.2- to 4.3-fold, respectively.

View this table:
Table 4

Binding of the model compounds to rat freshly isolated hepatocytes and their effects on the values of rat scaling factor

Figure 3 shows the correlation between CLint,in vitro for the model compounds in freshly isolated and cryopreserved rat hepatocytes. CLint,in vitro values in freshly isolated hepatocytes were determined using Sprague-Dawley rat hepatocytes. For most of the model compounds, the CLint,in vitro in freshly isolated hepatocytes were in good agreement with those in cryopreserved hepatocytes.

Figure 3
Figure 3

Correlation of CLint,in vitrobetween freshly isolated and cryopreserved rat hepatocytes.

The dotted lines represent the lines of unity.

In Vitro Metabolism in Cryopreserved Human Hepatocytes and Prediction of Human CLint,in vivo.

Fig. 4 illustrates the values of CLint,in vitro in cryopreserved human hepatocytes for the model compounds. The CLint,in vitrovalues were expressed per kilogram of body weight by taking physiological parameters in Table 1 into consideration. There were significant interindividual differences in CLint,in vitro estimated in cryopreserved human hepatocytes from five to seven donors for each compound. The mean values of CLint,in vitro among the model compounds also ranged widely from undetectable for quinotolast to 72.4 ml/min/kg for diltiazem. For quinotolast, the CLint,in vitrovalues could not be detected because evident concentration declines of the unchanged compound were not observed. For this reason, prediction of human CLint,in vivo mentioned below was evaluated except for quinotolast.

Figure 4
Figure 4

CLint,in vitro in cryopreserved human hepatocytes for the model compounds.

CLint,in vitro values were estimated in cryopreserved human hepatocytes using five to seven donors for each compound and expressed per kilograms of body weight by taking physiological parameters in Table 1 into consideration. CLint,in vitro for quinotolast could not be detected as described under Results. The dotted lines represent the mean values of CLint,in vitro. Each value represents the mean ± S.D.

Table 5 shows predicted CLint,in vivo and the differences between observed and predicted CLint,in vivo in humans. The predicted CLint,in vivo values were estimated from the human CLint,in vitro corrected value with or without rat scaling factor. Without consideration of rat scaling factor, the differences showed various values (2.2–199.0-fold) among the model compounds. For FK1052, FK480, troglitazone, and FK079, the differences were especially large (FK1052, 135.6–199.0-fold; FK480, 32.2–33.7-fold; troglitazone, 117.4–148.4-fold; and FK079, 12.4–15.8-fold), resulting in discrepancies between CLint,in vitro and CLint,in vivo. In contrast, with rat scaling factor (Table 3), most of the values corrected were within 5-fold, except for FK1052 [10.9-fold (dispersion model)] and diltiazem [8.1-fold (well stirred model) and 11.0-fold (dispersion model)], significantly improving the predictability of human CLint,in vivo.

View this table:
Table 5

Predicted CLint,in vivo and the differences between observed and predicted CLint,in vivo

Discussion

In the present study, we investigated CLint,in vitro obtained from in vitro experiments using hepatocytes and CLint,in vivo calculated from in vivo pharmacokinetic data with nine model compounds (FK1052, FK480, diazepam, diltiazem, troglitazone, quinotolast, FK079, zidovudine, and acetaminophen) in rats and humans. As a result, 1) rat scaling factor values (CLint,in vivo/CLint,in vitro) were different among the compounds; 2) when human CLint,in vitro was regarded as the predicted CLint,in vivo, the observed and predicted CLint,in vivo differed markedly for some compounds; and 3) use of human CLint,in vitrocorrected with rat scaling factor improved the predictability of human CLint,in vivo.

In this study, CLint,in vitro values were estimated using hepatocytes. In drug discovery, enzymes involved in the metabolism of compounds are not sufficiently characterized. Compared with subcellular fractions such as hepatic microsomes, hepatocytes seem to be suitable for measuring CLint,in vitro, because they possess a larger complement of drug-metabolizing enzymes. In addition, the hepatocyte system is a useful tool to evaluate compounds that are metabolized by multiple drug-metabolizing enzymes. Therefore, we selected model compounds metabolized by various enzymes, such as P450, esterase, UDP-glucuronosyltransferase, and sulfotransferase, to confirm the advantage of hepatocytes (Table 2). Of the model compounds, troglitazone, quinotolast, FK079, zidovudine, and acetaminophen are metabolized by different drug-metabolizing enzymes.

When rat CLint,in vitro were evaluated using freshly isolated rat hepatocytes, rat scaling factor values for a few drugs (diazepam, diltiazem, zidovudine, and acetaminophen) were close to unity (Table 3). However, the values for some drugs (FK1052, FK480, quinotolast, troglitazone, and FK079) were from 6.0 to 73.1, resulting in CLint,in vivo values larger than the CLint,in vitro values (Table 3). These findings suggest that CLint,in vivo values are not always in agreement with the CLint,in vitrovalues. In the same way, when using cryopreserved human hepatocytes, the differences between observed and predicted human CLint,in vivo for FK1052, FK480, troglitazone, and FK079 were especially large (12.4–199.0-fold) (Table 5), resulting in discrepancies between CLint,in vitro and CLint,in vivo.

Several studies have reported the use of freshly isolated rat hepatocytes to predict quantitative hepatic clearance (Ashforth et al., 1995; Carlile et al., 1998). However, there have been few studies of prediction using freshly isolated human hepatocytes (Bayliss et al., 1999). One reason is that there is a limited availability of fresh human livers for research. Thus, human hepatocytes are not yet widely used as an experimental system. In this respect, cryopreservation is especially important for hepatocytes from species such as humans, where tissue supply is limited. There are several reports of the cryopreservation of human hepatocytes (Chesne et al., 1993; Coundouris et al., 1993). Furthermore, a few groups have reported the evaluation of metabolic pathways, enzyme induction, and inhibition using cryopreserved human hepatocytes applied to drug metabolism research (Ruegg et al., 1997; Li et al., 1999; Hewitt et al., 2001).

For predicting hepatic metabolic clearance using cryopreserved hepatocytes, there is a requirement that CLint,in vitro (or drug-metabolizing enzyme activity) in cryopreserved hepatocytes is in agreement with that in the freshly isolated hepatocytes. In this study, we compared CLint,in vitro in freshly isolated and cryopreserved rat hepatocytes for the model compounds. As a result, for most of the compounds, the CLint,in vitro in the freshly isolated hepatocytes were in good agreement with those in the cryopreserved hepatocytes (Fig. 3). In addition, Li et al. (1999) have reported that cryopreserved human hepatocytes had equivalent enzyme activities for P450 (CYP1A2, 2A6, 2C9, 2C19, 2D6, and 3A4), UDP-glucuronosyltransferase, and sulfotransferase to their corresponding freshly isolated human hepatocytes. Judging from these results, it is reasonable to use cryopreserved hepatocytes for predicting hepatic metabolic clearance.

CLint,in vitro was determined from substrate disappearance rate at a single drug concentration (1 μM) in hepatocytes. The measurement method can be thought of as a simple and useful method with advantages of the method as follows: 1) simple to conduct, 2) can be done for many compounds, 3) metabolites do not need to be known, 4) can be easily done without radiolabeling, and 5) can yield enzyme kinetic data based on the disappearance of parent compounds. On the other hand, disadvantages of the method are as follows: 1) it is difficult to measure very low CLint,in vitro values; 2) do not get individual metabolite information; and 3) do not obtain Kmand Vmax parameters (Naritomi et al., 2001). We could not detect CLint,in vitro for quinotolast in human hepatocytes because the CLint,in vitro values were very low (Fig. 4). In this case, it is difficult to predict CLint,in vivoquantitatively. However, it seems reasonable to assume that the predicted CLint,in vivo in humans was very low in comparison with that in rats, from a qualitative level.

Previously, we reported the quantitative prediction of human hepatic clearance using hepatic microsomes for compounds metabolized by P450 (Naritomi et al., 2001). As a result, we found that use of human CLint,in vitro corrected with rat and/or dog scaling factors yielded better predictions of human CLint,in vivo. In this study, we have examined the effect of rat scaling factor in the case of using hepatocytes. Without consideration of rat scaling factor, some compounds showed marked differences between observed and predicted human CLint,in vivo, as mentioned above. In contrast, with rat scaling factor, most of the differences were within 5-fold (Table 5). These results indicate that it is likely that the prediction method using human hepatic materials, which includes animal scaling factors, improves predictability. However, the differences were more than 2-fold except for FK480 (1.3–1.9-fold) (Table 5). Increasing the accuracy of the predictability requires the improvement of the prediction method. In addition, the results were obtained by use of the limited compounds. In the future, we would improve the prediction method for many more compounds and confirm its validity.

When using hepatic microsomes, inclusion of dog scaling factor also improved the predictability of human CLint,in vivo (Naritomi et al., 2001). In the present study, we evaluated the prediction by use of hepatocytes, focusing on the application of rat scaling factor. To give a better human prediction, future studies should also examine the prediction applying dog scaling factor. A major problem facing researchers using hepatocytes from the livers of larger species such as dogs and humans is the scarcity of tissue available. Recently, cryopreservation of dog hepatocytes has been reported (Swales and Utesch, 1998). Use of cryopreserved dog hepatocytes would be useful in prediction studies.

At the present time, it is not clear why each compound has an intrinsic scaling factor. However, there are a few possible reasons. To begin with, fu,hepatocytes in the reaction mixture may have an influence on the CLint,in vitro values. For in vitro-in vivo scaling, correction withfu in the reaction mixture has been reported to be important (Lin et al., 1980; Obach, 1999). Therefore, we have evaluated fu,hepatocytes of the model compounds in rats and the changes in scaling factor values when correcting with fu,hepatocytes. However, CLint,in vitro corrected withfu,hepatocytes were not in agreement with the CLint,in vivo for some compounds. Namely, the scaling factor values for FK1052, FK480, troglitazone, quinotolast, and FK079 still showed high values, 3.1- to 15.8-fold (Table 4). This suggests that a scaling factor different from unity might not be due only to the drug binding to hepatocytes. The following assumptions in the mathematical models may relate to the result: 1) a rapid equilibrium between blood and hepatocytes, 2) availability of only unbound drug for elimination and transport across membranes, and 3) homogeneous distribution of metabolic enzymes and transport carriers along the blood flow pathways in the liver (Iwatsubo et al., 1996). If any of these assumptions are incorrect, discrepancies between CLint,in vivo and CLint,in vitro will be observed.

In conclusion, we investigated CLint,in vitroobtained from in vitro experiments using freshly isolated or cryopreserved hepatocytes and compared with CLint,in vivo calculated from in vivo pharmacokinetic data with nine model compounds in rats and humans. When using freshly isolated rat hepatocytes, rat scaling factor values (CLint,in vivo/CLint,in vitro) showed marked difference among the model compounds. Human CLint,in vitro were determined by use of cryopreserved hepatocytes. When human CLint,in vitro was regarded as the predicted CLint,in vivo, the observed and predicted CLint,in vivo for some drugs differed markedly. In contrast, use of human CLint,in vitro corrected with rat scaling factors improved the predictability of human CLint,in vivo. These results make the evaluation using hepatocytes more useful and provide a basis for predicting hepatic clearance using hepatocytes.

Footnotes

  • Abbreviations used are::
    P450
    cytochrome P450
    HPLC
    high performance liquid chromatography
    CL
    plasma clearance
    WME
    Williams' medium E
    CLH
    hepatic clearance
    CLint
    intrinsic metabolic or hepatic clearance
    CLR
    renal clearance
    DN
    dispersion number
    EH
    hepatic extraction ratio
    Fa
    the fraction absorbed from the intestinal tract
    FH
    hepatic availability
    fp
    unbound fraction in plasma (or serum)
    fu,hepatocytes
    unbound fraction in hepatocytes
    QH
    hepatic blood flow rate
    RB
    blood-to-plasma concentration ratio
    • Received October 10, 2002.
    • Accepted February 4, 2003.

References

  1. ↵
    1. Ashforth EIL,
    2. Carlile DJ,
    3. Chenery R,
    4. Houston JB
    (1995) Prediction of in vivo disposition from in vitro systems: clearance of phenytoin and tolbutamide using rat hepatic microsomal and hepatocyte data. J Pharmcol Exp Ther 274:761–766.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Bayliss MK,
    2. Bell JA,
    3. Jenner WN,
    4. Park GR,
    5. Wilson K
    (1999) Utility of hepatocytes to model species differences in the metabolism of loxtidine and to predict pharmacokinetic parameters in rat, dog and man. Xenobiotica 29:253–268.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Blum MR,
    2. Liao SH,
    3. Good SS,
    4. de Miranda P
    (1996) Pharmacokinetics and bioavailability of zidovudine in humans. Am J Med 85 (Suppl 2A) 189–194.
    OpenUrl
  4. ↵
    1. Carlile DJ,
    2. Stevens AJ,
    3. Ashforth EIL,
    4. Waghela D,
    5. Houston JB
    (1998) In vivo clearance of ethoxycoumarin and its prediction from in vitro systems: use of drug depletion and metabolite formation methods in hepatic microsomes and isolated hepatocytes. Drug Metab Dispos 26:216–221.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Chesne C,
    2. Guyomard C,
    3. Fautrel A,
    4. Poullain MG,
    5. Fremond B,
    6. De Jong H,
    7. Guillouzo A
    (1993) Viability and function in primary culture of adult hepatocytes from various animal species and human beings after cryopreservation. Hepatology 18:406–414.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Chiou WL,
    2. Barve A
    (1998) Linear correlation of the fraction of oral dose absorbed of 64 drugs between humans and rats. Pharm Res (NY) 15:1792–1795.
    OpenUrl
  7. ↵
    1. Collins JM,
    2. Unadkat JD
    (1989) Clinical pharmacokinetics of zidovudine. An overview of current data. Clin Pharmacokinet 17:1–9.
    OpenUrlPubMed
  8. ↵
    1. Coundouris JA,
    2. Grant MH,
    3. Engeset J,
    4. Petrie JC,
    5. Hawksworth GM
    (1993) Cryopreservation of human adult hepatocytes for use in drug metabolism and toxicity studies. Xenobiotica 23:1399–1409.
    OpenUrlPubMed
  9. ↵
    1. Davies B,
    2. Morris T
    (1993) Physiological parameters in laboratory animals and humans. Pharm Res (NY) 10:1093–1095.
    OpenUrl
  10. ↵
    1. de Miranda P,
    2. Burnette TC,
    3. Good SS
    (1990) Tissue distribution and metabolic disposition of zidovudine in rats. Drug Metab Dispos 18:315–320.
    OpenUrlAbstract
  11. ↵
    1. Diaz-Garcia JM,
    2. Oliver-Botana J,
    3. Fos-Galve D
    (1992) Pharmacokinetics of diazepam in the rat: influence of a carbon tetrachloride-induced hepatic injury. J Pharm Sci 81:768–772.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Gazzard BG,
    2. Ford-Hutchinson AW,
    3. Smith MJ,
    4. Williams R
    (1973) The binding of paracetamol to plasma proteins of man and pig. J Pharm Pharmacol 25:964–967.
    OpenUrlPubMed
  13. ↵
    1. Hedaya MA,
    2. Elmquist WF,
    3. Sawchuk RJ
    (1990) Probenecid inhibits the metabolic and renal clearances of zidovudine (AZT) in human volunteers. Pharm Res (NY) 7:411–417.
    OpenUrl
  14. ↵
    1. Hewitt NJ,
    2. Bühring KU,
    3. Dasenbrock J,
    4. Haunschild J,
    5. Ladstetter B,
    6. Utesch D
    (2001) Studies comparing in vivo: in vitro metabolism of three pharmaceutical compounds in rat, dog, monkey and human using cryopreserved hepatocytes, microsomes and collagen gel immobilized hepatocyte cultures. Drug Metab Dispos 29:1042–1050.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Hjelle JJ,
    2. Klaassen CD
    (1984) Glucuronidation and biliary excretion of acetaminophen in the rats. J Pharmcol Exp Ther 228:407–413.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Hoener BA
    (1994) Predicting the hepatic clearance of xenobiotics in humans from in vitro data. Biopharm Drug Dispos 15:295–304.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Houston JB,
    2. Carlile DJ
    (1997) Prediction of hepatic clearance from microsomes, hepatocytes and liver slices. Drug Metab Rev 29:891–922.
    OpenUrlPubMed
  18. ↵
    1. Iwatsubo T,
    2. Hirota N,
    3. Ooie T,
    4. Suzuki H,
    5. Shimada N,
    6. Chiba K,
    7. Ishizaki T,
    8. Green CE,
    9. Tyson CA,
    10. Sugiyama Y
    (1997) Prediction of in vivo drug metabolism in the human liver from in vitro metabolism data. Pharmacol Ther 73:141–171.
  19. ↵
    1. Iwatsubo T,
    2. Hirota N,
    3. Ooie T,
    4. Suzuki H,
    5. Sugiyama Y
    (1996) Prediction of in vivo drug disposition from in vitro data based on physiological pharmacokinetics. Biopharm Drug Dispos 17:273–310.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Izumi T,
    2. Enomoto S,
    3. Hosiyama K,
    4. Sasahara K,
    5. Shibukawa A,
    6. Nakagawa T,
    7. Sugiyama Y
    (1996) Prediction of the human pharmacokinetics of troglitazone, a new and extensively metabolized antidiabetic agent, after oral administration, with an animal scale-up approach. J Pharmacol Exp Ther 277:1630–1641.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Katashima M,
    2. Fujimoto R,
    3. Ishibashi K,
    4. Tokuma Y,
    5. Noda K
    (1993a) Pharmacokinetics of quinotolast sodium in rats, guinea pigs and dogs. Xenobiol Metab Dispos 8:505–514.
    OpenUrl
  22. ↵
    1. Katashima M,
    2. Fujimoto R,
    3. Ishibashi K,
    4. Tokuma Y,
    5. Noda K
    (1993b) The protein binding of quinotolast to the serum and plasma and its distribution in the blood cells. Xenobiol Metab Dispos 8:515–520.
    OpenUrl
  23. ↵
    1. Klotz U,
    2. Antonin KH,
    3. Bieck PR
    (1976) Pharmacokinetics and plasma binding of diazepam in man, dog, rabbit, guinea pig and rat. J Pharmacol Exp Ther 199:67–73.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Kölle EU,
    2. Ochs HR,
    3. Vollner KO
    (1983) Pharmacokinetic model of diltiazem. Arzneim-Forsch 33:972–977.
    OpenUrlPubMed
  25. ↵
    1. Koss FW,
    2. Mayer D,
    3. Hanidl H,
    4. Kabbara T
    (1970) Pharmacokinetics of N-acetyl-para-aminophenol under the effect of phenylbutazone. Arzneim-Forsch 20:1218–1222.
    OpenUrlPubMed
  26. ↵
    1. Lave TH,
    2. Dupin S,
    3. Schmitt C,
    4. Chou RC,
    5. Jaeck D,
    6. Coassolo PH
    (1997) Integration of in vitro data into allometric scaling to predict hepatic metabolic clearance in man: application to 10 extensively metabolized drugs. J Pharm Sci 86:584–590.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Lee YH,
    2. Lee MH,
    3. Shim CK
    (1992) Decreased systemic clearance of diltiazem with increased hepatic metabolism in rats with uranyl nitrate-induced acute renal failure. Pharm Res (NY) 9:1599–1606.
    OpenUrl
  28. ↵
    1. Levet-Trafit B,
    2. Gruyer MS,
    3. Marjanovic M,
    4. Chou RC
    (1996) Estimation of oral drug absorption in man based on intestine permeability in rats. Life Sci 58:PL359–PL363.
  29. ↵
    1. Li AP,
    2. Lu C,
    3. Brent JA,
    4. Pham C,
    5. Fackett A,
    6. Ruegg CE,
    7. Silber PM
    (1999) Cryopreserved human hepatocytes: characterization of drug-metabolizing enzyme activities and applications in higher throughput screening assays for hepatotoxicity, metabolic stability and drug-drug interaction potential. Chem Biol Interact 121:7–16.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Lin JH,
    2. Levy G
    (1986) Effect of prevention of inorganic sulfate depletion on the pharmacokinetics of acetaminophen in rats. J Pharmacol Exp Ther 239:94–98.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Lin JH,
    2. Sugiyama Y,
    3. Awazu S,
    4. Hanano M
    (1980) Kinetic studies on the deethylation of ethoxybenzamide: a comparative study with isolated hepatocytes and liver microsomes of rat. Biochem Pharmacol 29:2825–2830.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Luurila H,
    2. Olkkola KT,
    3. Neuvonen PJ
    (1996) Interaction between erythromycin and the benzodiazepines diazepam and flunitrazepam. Pharmacol Toxicol 78:117–122.
    OpenUrlPubMed
  33. ↵
    1. Mays DC,
    2. Dixon KF,
    3. Balboa A,
    4. Pawluk LJ,
    5. Bauer MR,
    6. Nawoot S,
    7. Gerber N
    (1991) A nonprimate animal model applicable to zodovudine pharmacokinetics in humans: inhibition of glucuronidation and renal excretion of zidovudine by probenecid in rats. J Pharmacol Exp Ther 259:1261–1270.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Nakashima M,
    2. Kanamaru M,
    3. Terakawa M,
    4. Fujiwara S
    (1993) Phase I study of quinotolast sodium – single-dose and multiple-dose studies. Jpn Pharmacol Ther 21:451–472.
    OpenUrl
  35. ↵
    1. Naritomi Y,
    2. Terashita S,
    3. Kimura S,
    4. Suzuki A,
    5. Kagayama A,
    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
  36. ↵
    1. Navia MA,
    2. Chaturvedi PR
    (1996) Design principles for orally bioavailable drugs. Drug Discov Today 1:179–189.
  37. ↵
    1. Obach RS
    (1999) Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: an examination of in vitro half-life approach and nonspecific binding to microsomes. Drug Metab Dispos 27:1350–1359.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Pang KS,
    2. Rowland M
    (1977) Hepatic clearance of drugs. I. Theoretical considerations of a “well-stirred” model and a “parallel tube” model. Influence of hepatic blood flow, plasma and blood cell binding, and the hepatocellular enzymatic activity on hepatic drug clearance. J Pharmacokinet Biopharm 5:625–653.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Piepho RW,
    2. Bloedow DC,
    3. Lacz JP,
    4. Runser DJ,
    5. Dimmit DC,
    6. Browne RK
    (1982) Pharmacokinetics of diltiazem in selected animal species and human beings. Am J Cardiol 49:525–528.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Roberts MS,
    2. Rowland M
    (1986) Correlation between in-vitro microsomal enzyme activity and whole organ hepatic elimination kinetics; analysis with a dispersion model. J Pharm Pharmacol 38:177–181.
    OpenUrlPubMed
  41. ↵
    1. Ruegg CE,
    2. Silber PM,
    3. Mughal RA,
    4. Ismail J,
    5. Lu C,
    6. Bode DC,
    7. Li AP
    (1997) Cytochrome-P450 induction and conjugated metabolism in primary human hepatocytes after cryopreservation. In Vitro Toxicol 10:217–222.
    OpenUrl
  42. ↵
    1. Sawada Y,
    2. Hanano M,
    3. Sugiyama Y,
    4. Iga T
    (1985) Prediction of the disposition of nine weakly acidic and six weakly basic drugs in humans from pharmacokinetic parameters in rats. J Pharmacokinet Biopharm 13:477–492.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Sonne J,
    2. Poulsen HE,
    3. Loft S,
    4. Dossing M,
    5. Vollmer-Larsen A,
    6. Simonsen K,
    7. Thyssen H,
    8. Lundstrom K
    (1988) Therapeutic doses of codeine have no effect on acetaminophen clearance or metabolism. Eur J Clin Pharmacol 35:109–111.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Sugiyama Y,
    2. Sawada Y,
    3. Iga T,
    4. Hanano M
    (1988) Xenobiotic metabolism and disposition. Proceeding of the 2nd International ISSX Meeting (Kato R, Estabrook RW, and Cayen MN eds) pp 225–235, Taylor & Francis, London.
  45. ↵
    1. Swales NJ,
    2. Utesch D
    (1998) Metabolic activity of fresh and cryopreserved dog hepatocyte. Xenobiotica 28:937–948.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Tanaka K,
    2. Sato M,
    3. Tomita Y,
    4. Ichihara A
    (1978) Biochemical studies on liver functions in primary cultured hepatocytes of adult rats. I. Hormonal effects on cell viability and protein synthesis. J Biochem (Tokyo) 84:937–946.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Tokuda H,
    2. Miyao Y,
    3. Hata T
    (1997) In vitro-in vivo scaling for FK079 pharmacokinetics. Xenobiol Metab Dispos 19 (Suppl) S273.
    OpenUrl
  48. ↵
    1. Tsui BHC,
    2. Feng JDZ,
    3. Buckley SJ,
    4. Yeung PKF
    (1994) Pharmacokinetics and metabolism of diltiazem in rats following a single intra-arterial or single oral dose. Eur J Drug Metab Pharmacokinet 19:369–373.
    OpenUrlPubMed
  49. ↵
    1. Yamazaki M,
    2. Suzuki H,
    3. Hanano M,
    4. Tokui T,
    5. Komai T,
    6. Sugiyama Y
    (1993) Na+-independent multispecific anion transporter mediates active transport of pravastatin into rat liver. Am J Physiol 264:G36–G44.
  50. ↵
    1. Yazdanian M,
    2. Glynn SL,
    3. Wright JL,
    4. Hawi A
    (1998) Correlating partitioning and caco-2 cell permeability of structurally diverse small molecular weight compounds. Pharm Res (NY) 15:1490–1494.
    OpenUrl
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleArticle

Utility of Hepatocytes in Predicting Drug Metabolism: Comparison of Hepatic Intrinsic Clearance in Rats and Humans in Vivo and in Vitro

Yoichi Naritomi, Shigeyuki Terashita, Akira Kagayama and Yuichi Sugiyama
Drug Metabolism and Disposition May 1, 2003, 31 (5) 580-588; DOI: https://doi.org/10.1124/dmd.31.5.580
Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement