Skip to main content
Advertisement

Main menu

  • Home
  • Articles
    • Current Issue
    • Fast Forward
    • Latest Articles
    • Special Sections
    • Archive
  • Information
    • Instructions to Authors
    • Submit a Manuscript
    • FAQs
    • For Subscribers
    • Terms & Conditions of Use
    • Permissions
  • Editorial Board
  • Alerts
    • Alerts
    • RSS Feeds
  • Virtual Issues
  • Feedback
  • Submit
  • Other Publications
    • Drug Metabolism and Disposition
    • Journal of Pharmacology and Experimental Therapeutics
    • Molecular Pharmacology
    • Pharmacological Reviews
    • Pharmacology Research & Perspectives
    • ASPET

User menu

  • My alerts
  • Log in
  • Log out
  • My Cart

Search

  • Advanced search
Drug Metabolism & Disposition
  • Other Publications
    • Drug Metabolism and Disposition
    • Journal of Pharmacology and Experimental Therapeutics
    • Molecular Pharmacology
    • Pharmacological Reviews
    • Pharmacology Research & Perspectives
    • ASPET
  • My alerts
  • Log in
  • Log out
  • My Cart
Drug Metabolism & Disposition

Advanced Search

  • Home
  • Articles
    • Current Issue
    • Fast Forward
    • Latest Articles
    • Special Sections
    • Archive
  • Information
    • Instructions to Authors
    • Submit a Manuscript
    • FAQs
    • For Subscribers
    • Terms & Conditions of Use
    • Permissions
  • Editorial Board
  • Alerts
    • Alerts
    • RSS Feeds
  • Virtual Issues
  • Feedback
  • Submit
  • Visit dmd on Facebook
  • Follow dmd on Twitter
  • Follow ASPET on LinkedIn
Research ArticleArticle

Confidence Assessment of the Simcyp Time-Based Approach and a Static Mathematical Model in Predicting Clinical Drug-Drug Interactions for Mechanism-Based CYP3A Inhibitors

Ying-Hong Wang
Drug Metabolism and Disposition July 2010, 38 (7) 1094-1104; DOI: https://doi.org/10.1124/dmd.110.032177
Ying-Hong Wang
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • Article
  • Figures & Data
  • Info & Metrics
  • eLetters
  • PDF + SI
  • PDF
Loading

Abstract

Accurate prediction of the extent of mechanism-based CYP3A inhibition is critical in determining the timing of clinical drug interaction studies in drug development. To evaluate the prediction accuracy of the static and Simcyp time-based approaches, 54 clinical drug interactions involving mechanism-based CYP3A inhibitors were predicted using both methods. The Simcyp time-based approach generated better prediction when 0.03 h−1 was used as the hepatic CYP3A enzyme degradation rate constant (kdeg) value. Of the predictions 87 and 55% had an error less than 2 and 0.5, respectively, relative to the observed values, compared with 57 and 20%, respectively, when the Simcyp default kdeg value of 0.0077 h−1 was used. Accuracy improvement using the kdeg value of 0.03 over 0.0077 h−1 was most evident for trials with observed magnitude of interaction greater than 2-fold; predictions with an error less than 0.5 relative to clinical observations increased from 8 to 48%. For the static approach, 76 and 35% of the predictions had an error less than 2 and 0.5, respectively. Both methods generated good predictions for weak and moderate inhibitors. The prediction accuracy could be affected by our knowledge of disposition of a substrate compound, in vitro inactivation parameter estimates, and the ability of Simcyp to accurately simulate the pharmacokinetics of inhibitors. Nonetheless, both the Simcyp and static approaches are useful tools for assessing the drug-drug interaction potential of a mechanism-based CYP3A inhibitor, especially when human pharmacokinetics of the inhibitor is known and 0.03 h−1 is used as the hepatic CYP3A kdeg value.

Drug-drug interactions remain an area of focus in drug discovery and development. Mechanism-based inhibition of CYP3A is one of the major causes of clinical drug-drug interactions and generally leads to greater concern, as highlighted by the list of moderate and strong CYP3A inhibitors in the Food and Drug Administration (2006) Drug Interaction Guidance. The inhibitory effects of a mechanism-based CYP3A inhibitor persist long after the compound is eliminated from the body because the recovery of CYP3A enzyme activity requires de novo protein synthesis or slow release of the enzyme from the enzyme-inhibitor complex. Because of the primary role of CYP3A in drug disposition, prediction of the clinical drug interaction potential of a mechanism-based CYP3A inhibitor would be helpful in guiding the timing and design of clinical studies.

Several approaches have been developed to predict clinical outcomes of mechanism-based inhibition. The static mathematical model developed by Mayhew et al. (2000) is a commonly used approach to predict mechanism-based drug interactions from in vitro estimated inactivation parameters. Several modifications to the original model have been made to incorporate the effects of intestinal wall metabolism (Wang et al., 2004b), competitive inhibition, and induction (Fahmi et al., 2009). Nonetheless, these static models are only capable of predicting the average magnitude of drug interactions across a population, assuming that the steady state of enzyme inhibition has been reached. Temporal changes in inhibitor concentrations and CYP3A enzyme activities as well as interindividual variability in CYP3A enzyme levels and rate constants of enzyme degradation are not considered. In addition, it is difficult to assess the effects of dosing regimens (e.g., irregular dosing) on the extent of drug interactions using a static model.

Several physiologically based pharmacokinetic models (PBPK) were developed to address some of the aforementioned limitations with static models (Kanamitsu et al., 2000; Zhang et al., 2009; Fenneteau et al., 2010). These PBPK models take into account temporal changes in inhibitor and substrate concentrations and enzyme activities as well as the enzyme inhibition concept in the static models. They can be used to simulate drug concentration-time profiles and to explore the effects of various dosing regimens. Simcyp (Simcyp Limited, Sheffield, UK) is a commercially available absorption, distribution, metabolism, and elimination simulator (Jamei et al., 2009). The Simcyp drug interaction prediction model is a population-based PBPK model. It considers the interplay among demographic, genetic, anatomical, physiological, and drug-specific factors. The objectives of the present study were to evaluate the accuracy of the static and Simcyp time-based models in predicting mechanism-based drug interactions of CYP3A, explore factors that could affect the accuracy of drug interaction predictions, and assess the confidence in drug interaction predictions when human pharmacokinetic data are available.

Materials and Methods

Clinical Drug Interaction Data.

Clinical drug interaction data were collected from PubMed and the University of Washington Drug Interaction Database (http://www.druginteractioninfo.org/). Clinical drug interaction studies involving seven mechanism-based CYP3A inhibitors and eight substrates were chosen based on available inhibitor and substrate profiles in Simcyp.

Simcyp Simulations.

Time-based simulations were performed using the Simcyp Population-based ADME Simulator (version 8.2, SP2, version 9.0, or version 9.10) according to the trial design described in the publication including the same number of subjects, age range, gender ratio, and dosing regimen. For studies conducted in Japan, a Sim-Japanese population was used. For studies conducted in Egypt, the United States, or Europe, a Sim-Healthy Volunteer population was used. Ten trials were simulated for each inhibitor-substrate pair.

Input parameters were default values supplied by the program unless otherwise stated. The clarithromycin profile was built using the parameters listed in Table 1. To simulate plasma concentration-time profiles for sustained-release (SR) formulations of verapamil or diltiazem, each dose, as indicated in Table 2, was divided into 10 equal minidoses administered at fixed intervals on 10 consecutive occasions. For each divided minidose, the simulation was run using the first-order absorption model and one-compartment distribution model. In addition, for the diltiazem SR formulation, the values of systemic clearance and volume of distribution of diltiazem were changed to 48.3 l/h with a coefficient of variation (CV) value of 27% and 5.2 l/kg with a CV value of 27.2%, respectively (Ochs and Knüchel, 1984) to simulate the observed diltiazem plasma profile (Robbins-Weilert et al., 1999). In vitro mechanism-based inactivation parameters were added to the interaction profiles of saquinavir, ritonavir, and fluoxetine (Table 3) when simulations were performed using 0.03 h−1 as the kdeg value for hepatic CYP3As. Microsomal protein bindings were predicted using the quantitative structure-activity relationship model (Turner et al., 2006).

View this table:
  • View inline
  • View popup
TABLE 1

Model parameters used in clarithromycin simulation

View this table:
  • View inline
  • View popup
TABLE 2

Summary of the 54 clinical studies and predictions based on Simcyp time-based model and static model

Values highlighted in bold had an absolute difference between the predicted and observed AUC changes >2.

View this table:
  • View inline
  • View popup
TABLE 3

Key inhibitor parameters provided in the Simcyp program

Static Prediction Model.

The extent of drug interactions was predicted using the eq. 1 that incorporates the effects of competitive inhibition and mechanism-based inhibition in both the intestine and liver (Fahmi et al., 2008): Embedded Image where AUCp.o., i and AUCp.o. are the areas under the curve of an affected substrate in the presence and absence of an inhibitor, respectively, FG is the intestinal bioavailability of the substrate, and fm is the fraction of the substrate metabolized by CYP3A in the liver. A and Y are the mechanism-based inhibition components in the liver and intestine, respectively; B and Z are the competitive inhibition components in the liver and intestine, respectively. They can be expressed as (eqs. 2 and 3): Embedded Image and Embedded Image where kinact represents the maximal inactivation rate constant, KI represents the inhibitor concentration when the rate of inactivation reaches half of the maximal value, and kdeg represents the rate constant of enzyme degradation. The mean kdeg value of 0.03 h−1 was used for both hepatic and intestinal CYP3As in the static model; 0.03 h−1 was the default kdeg value for intestinal CYP3As in Simcyp. Iu represents the unbound inhibitor concentration in the liver or intestine. The inhibitor concentration in the liver was assumed to be equal to the average systemic plasma concentration of the inhibitor calculated from observed AUCτ divided by τ, where τ represents the dosing interval (Table 4). When the observed AUCτ was not available, Simcyp-simulated AUCτ was used to calculate the average plasma concentration. The inhibitor concentration in the intestine was calculated from eq. 4 (Fahmi et al., 2008), Embedded Image where Dose is total daily dose of an inhibitor given orally, ka represents first-order absorption rate constant, fa represents the fraction of the dose absorbed, QG is the intestinal (villous) blood flow (17.9 l/h) (Yang et al., 2007), and Freq is the frequency of daily dose. Unbound inhibitor concentration in the intestine was assumed to be equal to IG.

View this table:
  • View inline
  • View popup
TABLE 4

Comparison of Simcyp-simulated inhibitor concentrations with clinical observations

Absorption rate constants, unbound fractions in plasma (fu, p) and microsomes (fu, mic), mechanism-based inactivation parameters, and competitive inhibition constants of the inhibitors used in the static model were the same as those in the Simcyp program (Table 3). Fraction absorbed (fa) was assumed to be 1 for inhibitors other than clarithromycin (Table 1). Intestinal bioavailabilities (FG) of the substrates, which were the same as those in the Simcyp, were 0.57, 0.75, 0.66, 0.78, 0.99, 0.56, and 0.96 for midazolam, triazolam, simvastatin, nifedipine, alprazolam, sildenafil, and zolpidem, respectively. The fraction of the dose metabolized by hepatic CYP3As (fm) was assumed to be 0.9 for midazolam, triazolam, simvastatin, nifedipine, alprazolam, and sildenafil and 0.48 for zolpidem (Farkas et al., 2009). When competitive inhibition was not considered in the prediction, B and Z in eq. 1 were equal to 1.

Data Analysis.

The simulated mean AUC ratios or geometric mean AUC ratios of a substrate in the presence and absence of an inhibitor were compared with the corresponding AUC ratios observed from clinical studies. Prediction error was calculated from the absolute difference between the predicted and observed AUC ratios. In addition, the predicted mean or geometric mean values of AUC and Cmax values of a substrate or an inhibitor were compared with those published values.

The average deviation (AD) of the predicted AUC ratio from the observed AUC ratio (eq. 5) was used to measure the prediction bias of different methods. A smaller absolute value of average deviation indicates a smaller prediction bias. The geometric mean-fold error (GMFE) (eq. 6) was calculated for readers' convenience. The root mean square error (RMSE) (eq. 7) was calculated to measure the precision of these methods. A smaller root mean square error indicates greater precision. Embedded Image Embedded Image Embedded Image where Predi represents the predicted AUC ratio, Obsi represents the observed AUC ratio of the ith drug interaction trial, and n is the total trial number in each category.

Results

Thirty-two clinical studies involving 54 inhibitor-substrate pairs were included in this analysis (Table 2). These clinical studies included trials investigating the effects of doses, duration of pretreatment, and the CYP3A recovery process on the extent of mechanism-based inhibition. The inhibitors included strong inhibitors such as clarithromycin, ritonavir, and saquinavir; moderate inhibitors such as erythromycin, diltiazem, and verapamil; and weak inhibitors such as fluoxetine. The affected substrates included sensitive CYP3A substrates such as midazolam, simvastatin, triazolam, sildenafil, and nifedipine; a substrate having low extraction ratio (i.e., alprazolam); and a substrate only partially metabolized by CYP3A (i.e., zolpidem). In addition, data from both intravenous and oral administration of midazolam were included in this analysis. For verapamil and diltiazem, both immediate-release (IR) and SR formulations were included.

To evaluate the accuracy of Simcyp in predicting drug interactions for mechanism-based CYP3A inhibitors, drug interaction trials were first simulated using the default values provided by Simcyp. Figure 1A and Table 2 show that the majority of drug interactions were overpredicted, except that the extent of drug interactions of saquinavir and ritonavir was underpredicted, which probably occurred because the mechanism-based inactivation for these two compounds was not considered by the program. Overall, 57% of the predictions had an error (i.e., absolute difference between the predicted and observed values) less than 2 relative to clinical observations, and 20% of the predictions had an error less than 0.5. Predictions with an error less than 0.5 were significantly reduced when only the trials with observed AUC ratios greater than 2-fold were considered (Table 6).

Fig. 1.
  • Download figure
  • Open in new tab
  • Download powerpoint
Fig. 1.

Comparisons of the Simcyp time-based approach (A and B) and the static model (C) to predict mean AUC ratios of clinical drug interactions from 54 inhibitor-substrate pairs. The points represent the observed mean (or geometric mean) AUC change versus predicted AUC change for each inhibitor-substrate pair. The solid line represents the unity, and the dotted lines represent a prediction error (= |predicted AUC ratio − observed AUC ratio|) of 2 relative to observed values. Insets are the enlarged areas between 0 and 10. CLAR, clarithromycin; DTZ, diltiazem; ERM, erythromycin; RTV, ritonavir; SQV, saquinavir; VER, verapamil; FLU, fluoxetine.

Hepatic CYP3A Degradation Rate Constant.

The default kdeg value for both hepatic CYP3A4 and CYP3A5 in the Simcyp model is 0.0077 h−1, corresponding to a half-life of 90 h. A recent study investigated the effects of age and sex on the time course of CYP3A recovery after multiple-dose clarithromycin administration using intravenous and oral midazolam as probe substrates. The results showed that the mean kdeg value for hepatic CYP3A in the population studied was approximately 0.03 h−1 (Wang et al., 2004a), corresponding to a half-life of 23 h. To examine whether Simcyp could reproduce the results from this study when 0.03 h−1 was used as the kdeg value, the clinical study was simulated using the same study design in the Sim-Healthy Volunteer population. The predicted magnitudes of clarithromycin-midazolam interactions for both intravenous and oral midazolam were in good agreement with the observed values when a kdeg value of 0.03 h−1 were used (Table 2). In contrast, significant overpredictions were observed when a kdeg value of 0.0077 h−1 were used. Figure 2 and Table 4 show that the simulated plasma concentration profile and pharmacokinetic parameters of clarithromycin were comparable with those observed in the clinical study (Wang et al., 2004a).

Fig. 2.
  • Download figure
  • Open in new tab
  • Download powerpoint
Fig. 2.

Comparison between simulated and observed plasma concentration profiles of inhibitors. Symbols represent the mean observed plasma concentrations. Solid lines represent the mean simulated plasma concentrations, dashed lines represent the 5th percentile, and dotted lines represent the 95th percentile. Error bars represent S.D. The dosages and formulations of the inhibitors are provided in the graph titles, and figure legends provide the source of the observed data points. MDZ, midazolam; ZLP, zolpidem; TZM, triazolam; MD, multiple-dose; SD, single-dose; NT, nighttime; DT, daytime.

To examine whether using a kdeg value of 0.03 h−1 could improve the prediction accuracy of the Simcyp time-based approach, default kdeg values for both hepatic CYP3A4 and CYP3A5 in the program were replaced with 0.03 h−1 while the CVs were maintained at 68%. In addition, mechanism-based inhibition parameters were added to the interaction profiles of saquinavir, ritonavir, and fluoxetine (Table 3). Compared with the default kdeg values, using kdeg values of 0.03 h−1 significantly improved the prediction accuracy (Fig. 1B; Table 2). Approximately 90% of the predictions had an error less than 2 relative to the observed values, and the predictions with an error less than 0.5 were approximately 54%, which represented more than a 2-fold improvement over those generated using the default kdeg values. These improvements to the prediction accuracy were most evident when only the trials with observed AUC ratio greater than 2-fold were considered (Table 6).

The Static Prediction Model.

The Simcyp time-based approach used in the aforementioned studies considered the effects of both competitive inhibition and mechanism-based inhibition. To make a valid comparison of the Simcyp time-based approach and the static approach, the static model (eq. 1) combined with the competitive inhibition components and gut first-pass effect was used. Parameters used in this prediction are detailed under Materials and Methods. Approximately 76% of the predictions had an error less than 2 compared with the observed values, and approximately 35% of the predictions had an error less than 0.5 (Fig. 1C; Table 2). Similar prediction accuracy was obtained when only the trials with observed AUC ratio greater than 2-fold were considered (Table 6). Significant overpredictions of the magnitude of interaction were observed with strong CYP3A inhibitors such as ritonavir, saquinavir, and clarithromycin (Table 2). Prediction results were the same as those shown in Table 2 when the static model in eq. 1 without the competitive inhibition components was used.

Inhibitor and Substrate Concentration Profiles.

To explore factors that could affect the prediction accuracy of the Simcyp approach, simulated inhibitors and substrate concentration profiles were compared with those observed in clinical studies. Plasma concentration-time profiles for inhibitors were simulated using a first-order absorption model and a one-compartment distribution model with the default parameters provided by the program unless otherwise specified. Figure 2 shows the mean plasma concentration-time profiles of the inhibitors. The AUC and Cmax values of these inhibitors are summarized in Table 4. The simulated and observed profiles were generally comparable with the exception of those for verapamil IR and diltiazem IR (Fig. 2). Simcyp overpredicted the AUC value of verapamil in the verapamil IR-simvastatin study and underpredicted the AUC and Cmax values of fluoxetine and saquinavir (Table 4).

The observed and predicted mean plasma concentration profiles of substrates at the baseline are shown in Fig. 3 and Supplemental Fig. 1. The observed and predicted AUC and Cmax values of these substrates were generally comparable (Fig. 4); 92% of the predicted Cmax and 87% of the predicted AUC values fell within 2-fold of the observed values, similar to a previous report (Einolf, 2007). Several significant deviations in the observed and predicted pharmacokinetic values were observed. Mean midazolam AUC in the verapamil SR-midazolam IV study was overpredicted by 5-fold compared with the clinical observation, but its predicted clearance was similar to the observed clearance (data not shown). The predicted Cmax values of sildenafil were approximately 3- to 7-fold lower than the observed values. In addition, the predicted AUC values of alprazolam were approximately 2-fold lower than the observed values, and the predicted plasma profiles of alprazolam deviated from those observed when the intrinsic clearance of alprazolam was estimated using recombinant CYP3As (version 9.0) instead of human liver microsomes (version 8.2) (Fig. 3). The predicted profiles of triazolam and nifedipine generally agreed with the observed profiles, but most of the predicted plasma profiles of sildenafil and simvastatin deviated from the observed profiles (Fig. 3; Supplemental Fig. 1).

Fig. 3.
  • Download figure
  • Open in new tab
  • Download powerpoint
Fig. 3.

Comparison between predicted and observed plasma concentration profiles for substrates at the baseline. Symbols represent the mean observed plasma concentrations of a substrate, manually adapted from the publication noted in the graph titles. The inhibitor-substrate pair is shown in the symbol legend. Unless otherwise noted in the symbol legend, solid lines represent the mean predicted plasma concentrations of a substrate, dashed lines represent the 5th percentile, and dotted lines represent the 95th percentile. RTV, ritonavir; ALP, alprazolam; CLAR, clarithromycin; MDZ, midazolam; DTZ, diltiazem; SIM, simvastatin; SDF, sildenafil; NIF, nifedipine; VER, verapamil; TZM, triazolam; ZLP, zolpidem.

Fig. 4.
  • Download figure
  • Open in new tab
  • Download powerpoint
Fig. 4.

Predicted and observed pharmacokinetic parameters Cmax (A) and AUC (B) for substrates obtained from the control groups of the clinical trials (n = 42). The solid line represents the unity, and the dotted line represents a 2-fold deviation from the unity. ALP, alprazolam; MDZ, midazolam; NIF, nifedipine; SDF, sildenafil; SIM, simvastatin; TZM, triazolam; ZLP, zolpidem.

In Vitro Inactivation Parameters.

The inactivation parameters kinact and KI represent two of the six important factors in the prediction models that may affect the prediction accuracy because they are sensitive to the in vitro study design (Yang et al., 2005). The effects of inactivation parameters generated from different laboratories on the prediction accuracy were examined (Table 5). For verapamil and diltiazem, the predicted magnitude of AUC changes was comparable with the observed data despite the different parameter estimates. For clarithromycin, there were large differences in KI estimates. However, incorporating fu, mic values minimized the differences in the KI estimates and resulted in similar predictions, which underscores the importance of considering the nonspecific microsomal binding in drug interaction predictions. For erythromycin, large differences in the kinact and KI estimates resulted in a significant difference in the predicted extent of interactions. In most cases, clinical drug interactions were predicted using these in vitro kinact and KI estimates.

View this table:
  • View inline
  • View popup
TABLE 5

Summary of drug interaction predictions using inactivation parameter estimates from various sources

Predictions were performed using Simcyp version 9.10.

Summary of Predictions Using Different Models.

To assess the confidence in the different prediction approaches, the prediction accuracy in various categories is summarized in Table 6. The static model generated more accurate predictions for weak and moderate CYP3A inhibitors. The Simcyp time-based approach yielded the most accurate predictions when the kdeg value of 0.03 h−1 was used. Greater than 90% of the predictions had an error less than 2 relative to the observed AUC ratios for trials using midazolam as a probe substrate and trials with moderate CYP3A inhibitors. For trials with an observed AUC ratio less than 2-fold including trials with weak inhibitors and trials with the substrate only partially metabolized by CYP3A (i.e., zolpidem), approximately 90% of the predictions had an error less than 1, despite the fact that some of the predicted substrate or inhibitor plasma profiles deviated from those observed (Figs. 2 and 3; Supplemental Fig. 1). For trials with strong inhibitors, the prediction accuracy mostly depended on how well Simcyp predicted the substrate profiles whereas overprediction of inhibitor concentrations such as ritonavir and saquinavir seemed less critical probably because maximal CYP3A inhibition was achieved at lower concentrations (Tables 2 and 4; Figs. 2 and 3). For trials with moderate inhibitors, the prediction accuracy primarily depended on adequate simulation of substrate and inhibitor profiles, e.g., similar AUC and Cmax, and overlapping profiles of observed and simulated plasma inhibitor concentrations (Tables 2 and 4; Fig. 2).

View this table:
  • View inline
  • View popup
TABLE 6

Summary of prediction accuracy of different models

Discussion

Several retrospective analyses were conducted to compare the prediction accuracy of the Simcyp models with that of static models (Einolf, 2007; Fahmi et al., 2009). However, only a small portion of the entire dataset in these studies involved mechanism-based inhibitors of CYP3A. In some cases, different kdeg values were used for hepatic CYP3A when the accuracy among prediction models was compared. As a result, it is difficult to evaluate the accuracy of Simcyp in predicting mechanism-based inhibitions from these studies. The present study examined the prediction accuracy of Simcyp as an out-of-box application, compared with those of the static and Simcyp time-based models using 0.03 h−1 as the kdeg value for hepatic CYP3A. In addition, factors that affect prediction accuracy of these two models were examined, which could help to determine the confidence in drug interaction predictions and enable informed decision-making.

The kdeg value of hepatic CYP3A is one of the important factors that affect the accuracy of mechanism-based drug interaction predictions. However, this parameter cannot be directly measured in human subjects. Various in vitro and in vivo approaches have been used to estimate the kdeg value of hepatic CYP3A (Yang et al., 2008). The mean values estimated from these studies range from 0.007 to 0.02 h−1, and no consensus has been reached on the appropriate kdeg value for hepatic CYP3A. A recent clinical study was conducted to examine the time course of CYP3A activity recovery after multiple doses of clarithromycin using intravenous and oral midazolam as probe substrates (Wang et al., 2004a). By fitting the ratios of midazolam hepatic intrinsic clearance to a mechanism-based enzyme inactivation model, the mean kdeg value estimated from this study was approximately 0.03 h−1. The present study showed that predictions conducted using both the static and Simcyp models with the kdeg value of 0.03 h−1 were in excellent agreement with clinical observations from the studies that investigated the time course of inhibitory effects of CYP3A inhibitors such as clarithromycin, erythromycin, and diltiazem (Table 2). Predictions generated with the kdeg value of 0.03 h−1 were more accurate and less biased than those generated with 0.0077 h−1. In fact, when studies with observed AUC changes less than 2-fold were excluded, predictions using the kdeg value of 0.03 h−1 were approximately 2 to 6 times better than those generated with 0.0077 h−1 (Table 6). In addition, the kdeg value of 0.03 h−1 also generated better prediction accuracy compared with the kdeg values used by Fahmi et al. (2009) and Xu et al. (2009) when the common data from these studies were compared. Taken together, these findings suggest that using 0.03 h−1 as the mean kdeg value for hepatic CYP3A generates better predictions for drug interactions involving mechanism-based CYP3A inhibitors.

Simcyp-simulated results were comparable to those generated with other PBPK models (Zhang et al., 2009; Fenneteau et al., 2010). For midazolam drug interaction studies, greater than 90% of the predictions had an error less than 2; more accurate predictions were achieved when the observed and simulated profiles were superimposed (e.g., diltiazem SR and verapamil SR). These data suggest that the Simcyp approach can be useful in predicting the extent of midazolam interactions for an investigational new drug that has a mechanism-based CYP3A inhibition potential. In addition, interindividual variability of drug interactions could also be assessed; the predicted ranges of AUC changes were comparable with the actual ranges observed in clinical studies (Table 5).

Several drug interactions were overpredicted by Simcyp when the kdeg value of 0.03 h−1 was used, which were probably due to the deviation of the predicted substrate profiles from those observed. For example, the design of the clarithromycin-midazolam study (Gurley et al., 2008) was similar to those of two other clarithromycin-midazolam studies (Table 2), but the observed extent of interactions among these studies differed by nearly 3-fold. Predicted mean baseline midazolam concentrations were lower compared with those observed (Fig. 3), indicating that additional information is needed for Simcyp to capture the interindividual variability in this study. In the case of the ritonavir-alprazolam interaction, when the alprazolam plasma profile was predicted using the intrinsic clearance estimated from human liver microsomes, the predicted magnitude of interactions was reduced to 4.4-fold. Consequently, the prediction error was less than 2. The above examples show the importance of accurately predicting substrate profiles, especially for moderate and strong inhibitors. Reasons for the poor predictions of simvastatin or sildenafil interaction studies need further investigation.

In the current version of Simcyp, the first-order absorption model is the only absorption model available for inhibitors. To simulate plasma concentration profiles of diltiazem SR and verapamil SR, the daily doses need to be given in 10 equally divided minidoses consecutively at 10 fixed dosing intervals. This approach was proved to be valid because the simulated profiles and pharmacokinetic parameter estimates were comparable to those observed (Fig. 2; Table 4). In the present study, inhibitor concentration profiles were simulated using pharmacokinetic parameters obtained from clinical studies, assuming first-order absorption and one-compartment distribution; nonlinear pharmacokinetic properties such as autoinhibition and metabolite interaction were not considered. Nonetheless, the simulated profiles of the inhibitors reported in this study were generally comparable with those observed. This finding may be attributed to several factors. For example, although metabolites of several inhibitors are potent CYP3A inhibitors (Mayhew et al., 2000; Wang et al., 2004b; Zhao et al., 2007), the parent compounds have to be metabolized first to inactivate the enzyme; therefore, it is difficult to distinguish whether the parent compounds or their metabolites are responsible for CYP3A inactivation. In addition, for most inhibitors used in this study, the effect of autoinhibition may not be significant at the doses studied and could sometimes be dealt with by applying a large coefficient of variation of the clearance parameter (e.g., ritonavir). As with most other methods, prediction accuracy of Simcyp for a potential victim of mechanism-based inhibition may be limited by our knowledge of the disposition of the victim compound. In addition, Simcyp has limited ability to simulate the plasma concentration profiles of potential inhibitors with target-mediated saturable distribution kinetics (data not shown).

For a potential mechanism-based inhibitor, if the plasma concentration profile of the inhibitor or its victim cannot be confidently simulated or predicted by Simcyp, the static model may be used as an alternative approach to predict the extent of interaction. For an inhibitor whose human plasma profiles can be simulated using Simcyp, prediction using the static model may be used to categorize the inhibitor and assess confidence in the Simcyp prediction. Approximately 90% of predictions generated using the static model had an error less than 2 for weak and moderate inhibitors, which were comparable with the predictions generated using the Simcyp time-based approach with a kdeg value of 0.03 h−1. The Simcyp approach was more accurate than the static model when predictions with an error less than 0.5 and 1 were predicted, which may be more relevant for weak and moderate inhibitors. For the static approach, the majority of the inaccuracies occurred when drug interactions involving strong CYP3A inhibitors were predicted, probably because inhibitor concentrations lower than the input values were sufficient to exhibit maximal inactivation of CYP3A, highlighting the importance of considering the temporal changes in active enzyme levels in the body. It is worth noting that unbound average inhibitor concentrations at steady state were used as the input values in the static model. When unbound maximal inhibitor concentrations were used, the overall prediction accuracy was reduced to 61 from 76% (data not shown). In addition, prediction results were the same for this dataset irrespective of whether the competitive inhibition components in eq. 1 were considered or not.

The cutoffs for prediction errors were arbitrary, and the prediction accuracy summarized from this study varied slightly when simulations were performed in different in silico populations or with different parameters. Sometimes, the predicted AUC ratios were comparable with those observed even though the simulated substrate plasma profiles deviated from clinical observations (e.g., sildenafil); therefore, the prediction results should be interpreted with caution. No matter which method is used, there are uncertainties about factors that determine the prediction accuracy such as the in vitro estimated inactivation parameters. For example, there is large variability in reported kinact and KI estimates for erythromycin. The observed inconsistencies in the predicted AUC changes may be due to variability in enzyme sources, differences in experimentally measured versus model-predicted nonspecific protein binding, and/or the physiochemical properties of erythromycin.

In conclusion, the results from this study suggest that Simcyp provides good prediction of the magnitude of drug interactions with mechanism-based CYP3A inhibitors when human pharmacokinetics of a substrate can be adequately predicted, the inhibitor profile can be simulated from human pharmacokinetic data, and a hepatic kdeg value of 0.03 h−1 is used. The static model is also a useful tool for weak and moderate inhibitors with known human pharmacokinetics. The prediction accuracy summarized in this study may be applied to situations for which human pharmacokinetics of a potential inhibitor or substrate is unknown, but there is adequate confidence in the predicted concentrations.

Acknowledgments.

I thank Amy O. Johnson-Levonas and Kathleen Newcomb (Merck Research Laboratories) for help in preparing this article for publication, Regina Wang for contributing inactivation parameters of the mechanism-based inhibitors, and Drs. Charlie Thompson, Christopher Gibson, and Deborah Nicoll-Griffith for reading this article and for helpful suggestions.

Footnotes

  • Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

    doi:10.1124/dmd.110.032177.

  • ↵Embedded Image The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material.

  • ABBREVIATIONS:

    PBPK
    physiologically based pharmacokinetics
    SR
    sustained-release formulation
    CV
    coefficient of variation
    AUC
    area under the concentration-time curve
    IR
    immediate-release formulation
    GMFE
    geometric mean-fold error
    RMSE
    root mean square error.

  • Received January 12, 2010.
  • Accepted April 2, 2010.
  • Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics

References

  1. ↵
    1. Allard S,
    2. Sainati S,
    3. Roth-Schechter B,
    4. MacIntyre J
    (1998) Minimal interaction between fluoxetine and multiple-dose zolpidem in healthy women. Drug Metab Dispos 26:617–622.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Backman JT,
    2. Olkkola KT,
    3. Aranko K,
    4. Himberg JJ,
    5. Neuvonen PJ
    (1994) Dose of midazolam should be reduced during diltiazem and verapamil treatments. Br J Clin Pharmacol 37:221–225.
    OpenUrlPubMed
  3. ↵
    1. Brannan MD,
    2. Reidenberg P,
    3. Radwanski E,
    4. Shneyer L,
    5. Lin CC,
    6. Cayen MN,
    7. Affrime MB
    (1995) Loratadine administered concomitantly with erythromycin: pharmacokinetic and electrocardiographic evaluations. Clin Pharmacol Ther 58:269–278.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Chu S,
    2. Wilson DS,
    3. Deaton RL,
    4. Mackenthun AV,
    5. Eason CN,
    6. Cavanaugh JH
    (1993) Single- and multiple-dose pharmacokinetics of clarithromycin, a new macrolide antimicrobial. J Clin Pharmacol 33:719–726.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Chu SY,
    2. Deaton R,
    3. Cavanaugh J
    (1992) Absolute bioavailability of clarithromycin after oral administration in humans. Antimicrob Agents Chemother 36:1147–1150.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Dilger K,
    2. Eckhardt K,
    3. Hofmann U,
    4. Kucher K,
    5. Mikus G,
    6. Eichelbaum M
    (1999) Chronopharmacology of intravenous and oral modified release verapamil. Br J Clin Pharmacol 47:413–419.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Einolf HJ
    (2007) Comparison of different approaches to predict metabolic drug-drug interactions. Xenobiotica 37:1257–1294.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Ernest CS 2nd,
    2. Hall SD,
    3. Jones DR
    (2005) Mechanism-based inactivation of CYP3A by HIV protease inhibitors. J Pharmacol Exp Ther 312:583–591.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Fahmi OA,
    2. Hurst S,
    3. Plowchalk D,
    4. Cook J,
    5. Guo F,
    6. Youdim K,
    7. Dickins M,
    8. Phipps A,
    9. Darekar A,
    10. Hyland R,
    11. et al
    . (2009) Comparison of different algorithms for predicting clinical drug-drug interactions, based on the use of CYP3A4 in vitro data: predictions of compounds as precipitants of interaction. Drug Metab Dispos 37:1658–1666.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Fahmi OA,
    2. Maurer TS,
    3. Kish M,
    4. Cardenas E,
    5. Boldt S,
    6. Nettleton D
    (2008) A combined model for predicting CYP3A4 clinical net drug-drug interaction based on CYP3A4 inhibition, inactivation, and induction determined in vitro. Drug Metab Dispos 36:1698–1708.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Farkas D,
    2. Volak LP,
    3. Harmatz JS,
    4. von Moltke LL,
    5. Court MH,
    6. Greenblatt DJ
    (2009) Short-term clarithromycin administration impairs clearance and enhances pharmacodynamic effects of trazodone but not of zolpidem. Clin Pharmacol Ther 85:644–650.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Fenneteau F,
    2. Poulin P,
    3. Nekka F
    (2010) Physiologically based predictions of the impact of inhibition of intestinal and hepatic metabolism on human pharmacokinetics of CYP3A substrates. J Pharm Sci 99:486–514.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Ferrero JL,
    2. Bopp BA,
    3. Marsh KC,
    4. Quigley SC,
    5. Johnson MJ,
    6. Anderson DJ,
    7. Lamm JE,
    8. Tolman KG,
    9. Sanders SW,
    10. Cavanaugh JH
    (1990) Metabolism and disposition of clarithromycin in man. Drug Metab Dispos 18:441–446.
    OpenUrlAbstract
  14. ↵
    Food and Drug Administration (2006) FDA Guidance for Industry on Drug Interaction Studies- Study Design, Data Analysis, and Implication for Dosing and Labeling, US Department of Health and Human Services; Rockville, MD.
  15. ↵
    1. Gascon MP,
    2. Dayer P
    (1991) In vitro forecasting of drugs which may interfere with the biotransformation of midazolam. Eur J Clin Pharmacol 41:573–578.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Greenblatt DJ,
    2. Preskorn SH,
    3. Cotreau MM,
    4. Horst WD,
    5. Harmatz JS
    (1992) Fluoxetine impairs clearance of alprazolam but not of clonazepam. Clin Pharmacol Ther 52:479–486.
    OpenUrlPubMed
  17. ↵
    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,
    10. Shader RI
    (1998) Inhibition of triazolam clearance by macrolide antimicrobial agents: in vitro correlates and dynamic consequences. Clin Pharmacol Ther 64:278–285.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Greenblatt DJ,
    2. von Moltke LL,
    3. Harmatz JS,
    4. Durol AL,
    5. Daily JP,
    6. Graf JA,
    7. Mertzanis P,
    8. Hoffman JL,
    9. Shader RI
    (2000a) Alprazolam-ritonavir interaction: implications for product labeling. Clin Pharmacol Ther 67:335–341.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Greenblatt DJ,
    2. von Moltke LL,
    3. Harmatz JS,
    4. Durol AL,
    5. Daily JP,
    6. Graf JA,
    7. Mertzanis P,
    8. Hoffman JL,
    9. Shader RI
    (2000b) Differential impairment of triazolam and zolpidem clearance by ritonavir. J Acquir Immune Defic Syndr 24:129–136.
    OpenUrlPubMed
  20. ↵
    1. Gurley B,
    2. Hubbard MA,
    3. Williams DK,
    4. Thaden J,
    5. Tong Y,
    6. Gentry WB,
    7. Breen P,
    8. Carrier DJ,
    9. Cheboyina S
    (2006) Assessing the clinical significance of botanical supplementation on human cytochrome P450 3A activity: comparison of a milk thistle and black cohosh product to rifampin and clarithromycin. J Clin Pharmacol 46:201–213.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Gurley BJ,
    2. Swain A,
    3. Hubbard MA,
    4. Hartsfield F,
    5. Thaden J,
    6. Williams DK,
    7. Gentry WB,
    8. Tong Y
    (2008) Supplementation with goldenseal (Hydrastis canadensis), but not kava kava (Piper methysticum), inhibits human CYP3A activity in vivo. Clin Pharmacol Ther 83:61–69.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Hedaya MA,
    2. El-Afify DR,
    3. El-Maghraby GM
    (2006) The effect of ciprofloxacin and clarithromycin on sildenafil oral bioavailability in human volunteers. Biopharm Drug Dispos 27:103–110.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Jacobson TA
    (2004) Comparative pharmacokinetic interaction profiles of pravastatin, simvastatin, and atorvastatin when coadministered with cytochrome P450 inhibitors. Am J Cardiol 94:1140–1146.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Jamei M,
    2. Marciniak S,
    3. Feng K,
    4. Barnett A,
    5. Tucker G,
    6. Rostami-Hodjegan A
    (2009) The Simcyp population-based ADME simulator. Expert Opin Drug Metab Toxicol 5:211–223.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Kanamitsu S,
    2. Ito K,
    3. Green CE,
    4. Tyson CA,
    5. Shimada N,
    6. Sugiyama Y
    (2000) Prediction of in vivo interaction between triazolam and erythromycin based on in vitro studies using human liver microsomes and recombinant human CYP3A4. Pharm Res 17:419–426.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Kantola T,
    2. Kivistö KT,
    3. Neuvonen PJ
    (1998) Erythromycin and verapamil considerably increase serum simvastatin and simvastatin acid concentrations. Clin Pharmacol Ther 64:177–182.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Kosuge K,
    2. Nishimoto M,
    3. Kimura M,
    4. Umemura K,
    5. Nakashima M,
    6. Ohashi K
    (1997) Enhanced effect of triazolam with diltiazem. Br J Clin Pharmacol 43:367–372.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Lam YW,
    2. Alfaro CL,
    3. Ereshefsky L,
    4. Miller M
    (2003) Pharmacokinetic and pharmacodynamic interactions of oral midazolam with ketoconazole, fluoxetine, fluvoxamine, and nefazodone. J Clin Pharmacol 43:1274–1282.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Lefebvre M,
    2. Lacasse Y,
    3. Spénard J,
    4. Geadah D,
    5. Moisan R,
    6. Gossard D,
    7. Landriault H,
    8. Du Souich P,
    9. Caillé G
    (1994) Pharmacokinetics and pharmacodynamics of a slow-release formulation of diltiazem after the administration of a single and repeated doses to healthy volunteers. Biopharm Drug Dispos 15:227–242.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Mayhew BS,
    2. Jones DR,
    3. Hall SD
    (2000) An in vitro model for predicting in vivo inhibition of cytochrome P450 3A4 by metabolic intermediate complex formation. Drug Metab Dispos 28:1031–1037.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Mousa O,
    2. Brater DC,
    3. Sunblad KJ,
    4. Hall SD
    (2000) The interaction of diltiazem with simvastatin. Clin Pharmacol Ther 67:267–274.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Muirhead GJ,
    2. Faulkner S,
    3. Harness JA,
    4. Taubel J
    (2002) The effects of steady-state erythromycin and azithromycin on the pharmacokinetics of sildenafil in healthy volunteers. Br J Clin Pharmacol 53 (Suppl 1):37S–43S.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Muirhead GJ,
    2. Wulff MB,
    3. Fielding A,
    4. Kleinermans D,
    5. Buss N
    (2000) Pharmacokinetic interactions between sildenafil and saquinavir/ritonavir. Br J Clin Pharmacol 50:99–107.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Ochs HR,
    2. Knüchel M
    (1984) Pharmacokinetics and absolute bioavailability of diltiazem in humans. Klin Wochenschr 62:303–306.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Ohashi K,
    2. Sudo T,
    3. Sakamoto K,
    4. Tateishi T,
    5. Fujimura A,
    6. Kumagai Y,
    7. Ebihara A
    (1993) The influence of pretreatment periods with diltiazem on nifedipine kinetics. J Clin Pharmacol 33:222–225.
    OpenUrlPubMed
  36. ↵
    1. Okudaira T,
    2. Kotegawa T,
    3. Imai H,
    4. Tsutsumi K,
    5. Nakano S,
    6. Ohashi K
    (2007) Effect of the treatment period with erythromycin on cytochrome P450 3A activity in humans. J Clin Pharmacol 47:871–876.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Olkkola KT,
    2. Aranko K,
    3. Luurila H,
    4. Hiller A,
    5. Saarnivaara L,
    6. Himberg JJ,
    7. Neuvonen PJ
    (1993) A potentially hazardous interaction between erythromycin and midazolam. Clin Pharmacol Ther 53:298–305.
    OpenUrlPubMed
  38. ↵
    1. Palkama VJ,
    2. Ahonen J,
    3. Neuvonen PJ,
    4. Olkkola KT
    (1999) Effect of saquinavir on the pharmacokinetics and pharmacodynamics of oral and intravenous midazolam. Clin Pharmacol Ther 66:33–39.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Piergies AA,
    2. Sweet J,
    3. Johnson M,
    4. Roth-Schechter BF,
    5. Allard S
    (1996) The effect of co-administration of zolpidem with fluoxetine: pharmacokinetics and pharmacodynamics. Int J Clin Pharmacol Ther 34:178–183.
    OpenUrlPubMed
  40. ↵
    1. Robbins-Weilert DK,
    2. Giesing DH,
    3. Weir SJ
    (1999) Steady-state pharmacokinetics of high-dose diltiazem hydrochloride (Cardizem CD) administered once daily in healthy volunteers. Am J Ther 6:211–216.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Suwa T,
    2. Yoshida H,
    3. Kohno Y,
    4. Yoshitumi S,
    5. Ohta K
    (1988) Metabolic fate of TE-031 (A-56268). III. Absorption, distribution and excretion of 14C-TE-0341 in rats, mice and dogs. Chemotherapy 36:213–226.
    OpenUrl
  42. ↵
    1. Tateishi T,
    2. Ohashi K,
    3. Sudo T,
    4. Sakamoto K,
    5. Toyosaki N,
    6. Hosoda S,
    7. Toyo-oka T,
    8. Kumagai Y,
    9. Sugimoto K,
    10. Fujimura A
    (1989) Dose dependent effect of diltiazem on the pharmacokinetics of nifedipine. J Clin Pharmacol 29:994–997.
    OpenUrlPubMed
  43. ↵
    1. Turner DB,
    2. Rostami-Hodjegan A,
    3. Tucker GT,
    4. Rowland-Yeo K
    (2006) Prediction of nonspecific hepatic microsomal binding from readily available physiochemical properties. Proceedings of the 9th European ISSX Meeting; 2006 Jun 4–7; Manchester, UK. International Society for the Study of Xenobiotics, Washington, DC.
  44. ↵
    1. Varhe A,
    2. Olkkola KT,
    3. Neuvonen PJ
    (1996) Diltiazem enhances the effects of triazolam by inhibiting its metabolism. Clin Pharmacol Ther 59:369–375.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Wang YH,
    2. Gorski JC,
    3. Hamman MA,
    4. Hall SD
    (2004a) Age and sex effects on the recovery of CYP3A inhibition by clarithromycin. Drug Metab Rev 36 (Suppl 1):287.
    OpenUrl
  46. ↵
    1. Wang YH,
    2. Jin Y,
    3. Ho H,
    4. Hilligoss JK,
    5. Hu Z,
    6. Gorski JC,
    7. Hall SD
    (2005) Effect of CYP3A5 genotype on the extent of CYP3A inhibition by verapamil (Abstract). Clin Pharmacol Ther 77:P3.
    OpenUrl
  47. ↵
    1. Wang YH,
    2. Jones DR,
    3. Hall SD
    (2004b) Prediction of cytochrome P450 3A inhibition by verapamil enantiomers and their metabolites. Drug Metab Dispos 32:259–266.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Wright CE,
    2. Lasher-Sisson TA,
    3. Steenwyk RC,
    4. Swanson CN
    (1992) A pharmacokinetic evaluation of the combined administration of triazolam and fluoxetine. Pharmacotherapy 12:103–106.
    OpenUrlPubMed
  49. ↵
    1. Xu L,
    2. Chen Y,
    3. Pan Y,
    4. Skiles GL,
    5. Shou M
    (2009) Prediction of human drug-drug interactions from time-dependent inactivation of CYP3A4 in primary hepatocytes using a population-based simulator. Drug Metab Dispos 37:2330–2339.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Yamano K,
    2. Yamamoto K,
    3. Katashima M,
    4. Kotaki H,
    5. Takedomi S,
    6. Matsuo H,
    7. Ohtani H,
    8. Sawada Y,
    9. Iga T
    (2001) Prediction of midazolam-CYP3A inhibitors interaction in the human liver from in vivo/in vitro absorption, distribution, and metabolism data. Drug Metab Dispos 29:443–452.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Yang J,
    2. Jamei M,
    3. Yeo KR,
    4. Tucker GT,
    5. Rostami-Hodjegan A
    (2005) Kinetic values for mechanism-based enzyme inhibition: assessing the bias introduced by the conventional experimental protocol. Eur J Pharm Sci 26:334–340.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Yang J,
    2. Jamei M,
    3. Yeo KR,
    4. Tucker GT,
    5. Rostami-Hodjegan A
    (2007) Prediction of intestinal first-pass drug metabolism. Curr Drug Metab 8:676–684.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Yang J,
    2. Liao M,
    3. Shou M,
    4. Jamei M,
    5. Yeo KR,
    6. Tucker GT,
    7. Rostami-Hodjegan A
    (2008) Cytochrome p450 turnover: regulation of synthesis and degradation, methods for determining rates, and implications for the prediction of drug interactions. Curr Drug Metab 9:384–394.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Yasui N,
    2. Otani K,
    3. Kaneko S,
    4. Ohkubo T,
    5. Osanai T,
    6. Sugawara K,
    7. Chiba K,
    8. Ishizaki T
    (1996) A kinetic and dynamic study of oral alprazolam with and without erythromycin in humans: in vivo evidence for the involvement of CYP3A4 in alprazolam metabolism. Clin Pharmacol Ther 59:514–519.
    OpenUrlCrossRefPubMed
  55. ↵
    1. Yeates RA,
    2. Laufen H,
    3. Zimmermann T
    (1996) Interaction between midazolam and clarithromycin: comparison with azithromycin. Int J Clin Pharmacol Ther 34:400–405.
    OpenUrlPubMed
  56. ↵
    1. Yeo KR,
    2. Yeo WW
    (2001) Inhibitory effects of verapamil and diltiazem on simvastatin metabolism in human liver microsomes. Br J Clin Pharmacol 51:461–470.
    OpenUrlCrossRefPubMed
  57. ↵
    1. Zhang X,
    2. Quinney SK,
    3. Gorski JC,
    4. Jones DR,
    5. Hall SD
    (2009) Semiphysiologically based pharmacokinetic models for the inhibition of midazolam clearance by diltiazem and its major metabolite. Drug Metab Dispos 37:1587–1597.
    OpenUrlAbstract/FREE Full Text
  58. ↵
    1. Zhao P,
    2. Kunze KL,
    3. Lee CA
    (2005) Evaluation of time-dependent inactivation of CYP3A in cryopreserved human hepatocytes. Drug Metab Dispos 33:853–861.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    1. Zhao P,
    2. Lee CA,
    3. Kunze KL
    (2007) Sequential metabolism is responsible for diltiazem-induced time-dependent loss of CYP3A. Drug Metab Dispos 35:704–712.
    OpenUrlAbstract/FREE Full Text
PreviousNext
Back to top

In this issue

Drug Metabolism and Disposition: 38 (7)
Drug Metabolism and Disposition
Vol. 38, Issue 7
1 Jul 2010
  • Table of Contents
  • Table of Contents (PDF)
  • About the Cover
  • Index by author
  • Back Matter (PDF)
  • Editorial Board (PDF)
  • Front Matter (PDF)
Download PDF
Article Alerts
Sign In to Email Alerts with your Email Address
Email Article

Thank you for sharing this Drug Metabolism & Disposition article.

NOTE: We request your email address only to inform the recipient that it was you who recommended this article, and that it is not junk mail. We do not retain these email addresses.

Enter multiple addresses on separate lines or separate them with commas.
Confidence Assessment of the Simcyp Time-Based Approach and a Static Mathematical Model in Predicting Clinical Drug-Drug Interactions for Mechanism-Based CYP3A Inhibitors
(Your Name) has forwarded a page to you from Drug Metabolism & Disposition
(Your Name) thought you would be interested in this article in Drug Metabolism & Disposition.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Citation Tools
Research ArticleArticle

Confidence Assessment of the Simcyp Time-Based Approach and a Static Mathematical Model in Predicting Clinical Drug-Drug Interactions for Mechanism-Based CYP3A Inhibitors

Ying-Hong Wang
Drug Metabolism and Disposition July 1, 2010, 38 (7) 1094-1104; DOI: https://doi.org/10.1124/dmd.110.032177

Citation Manager Formats

  • BibTeX
  • Bookends
  • EasyBib
  • EndNote (tagged)
  • EndNote 8 (xml)
  • Medlars
  • Mendeley
  • Papers
  • RefWorks Tagged
  • Ref Manager
  • RIS
  • Zotero

Share
Research ArticleArticle

Confidence Assessment of the Simcyp Time-Based Approach and a Static Mathematical Model in Predicting Clinical Drug-Drug Interactions for Mechanism-Based CYP3A Inhibitors

Ying-Hong Wang
Drug Metabolism and Disposition July 1, 2010, 38 (7) 1094-1104; DOI: https://doi.org/10.1124/dmd.110.032177
Reddit logo Twitter logo Facebook logo Mendeley logo
  • Tweet Widget
  • Facebook Like
  • Google Plus One

Jump to section

  • Article
    • Abstract
    • Materials and Methods
    • Results
    • Discussion
    • Acknowledgments.
    • Footnotes
    • References
  • Figures & Data
  • Info & Metrics
  • eLetters
  • PDF + SI
  • PDF

Related Articles

Cited By...

More in this TOC Section

  • Metabolic enzymes in nintedanib metabolism
  • Mechanism of AO Inactivation by Hydralazine
  • Warfarin PBPK modeling with target binding
Show more Articles

Similar Articles

Advertisement
  • Home
  • Alerts
Facebook   Twitter   LinkedIn   RSS

Navigate

  • Current Issue
  • Fast Forward by date
  • Fast Forward by section
  • Latest Articles
  • Archive
  • Search for Articles
  • Feedback
  • ASPET

More Information

  • About DMD
  • Editorial Board
  • Instructions to Authors
  • Submit a Manuscript
  • Customized Alerts
  • RSS Feeds
  • Subscriptions
  • Permissions
  • Terms & Conditions of Use

ASPET's Other Journals

  • Journal of Pharmacology and Experimental Therapeutics
  • Molecular Pharmacology
  • Pharmacological Reviews
  • Pharmacology Research & Perspectives
ISSN 1521-009X (Online)

Copyright © 2023 by the American Society for Pharmacology and Experimental Therapeutics