Prediction of pharmacokinetic drug-drug interactions using human hepatocyte suspension in plasma and cytochrome P450 phenotypic data. II. In vitro-in vivo correlation with ketoconazole

Predictive In Vitro-In Vivo Extrapolation for Time Dependent Inhibition of CYP1A2, CYP2C8, CYP2C9, CYP2C19 and CYP2D6 Using Pooled Human Hepatocytes, Human Liver Microsomes, and a Simple Mechanistic Static Model

Inactivation of Cytochrome P450 (CYP450) enzymes can lead to significant increases in exposure of co-medicants. The majority of reported in vitro to in vivo extrapolation (IVIVE) data have historically focused on CYP3A4 leaving the assessment of other CYP isoforms insubstantial. To this end, the utility of human hepatocytes (HHEP) and microsome (HLM) to predict clinically relevant DDIs was investigated with a focus on CYP1A2, CYP2C8, CYP2C9, CYP2C19 and CYP2D6. Evaluation of IVIVE for CYP2B6 was limited to only weak inhibition. A search of the University of Washington Drug-Drug Interaction Database was conducted to identify a clinically relevant weak, moderate and strong inhibitor for selective substrates of CYP1A2, CYP2C8, CYP2C9, CYP2C19 and CYP2D6, resulting in 18 inhibitors for in vitro characterization against 119 clinical interaction studies. Pooled human hepatocytes and HLM were pre-incubated with increasing concentrations of inhibitors for designated timepoints. Time dependent inhibition (TDI) was detected in HLM for four moderate/strong inhibitors suggesting that some optimization of incubation conditions (i.e. lower protein concentrations) is needed to capture weak inhibition. Clinical risk assessment was conducted by incorporating the in vitro derived kinetic parameters k_inact and K_I into static equations recommended by regulatory authorities. Significant overprediction was observed when applying the basic models recommended by regulatory agencies. Mechanistic static models (MSM), which consider the fraction of metabolism through the impacted enzyme, using the unbound hepatic inlet concentration lead to the best overall prediction accuracy with 92% and 85% of data from HHEPs and HLM, respectively, within 2-fold of the observed value. Significance Statement Collectively, the data demonstrate that coupling time-dependent inactivation parameters derived from pooled human hepatocytes and HLM with a mechanistic static model provides an easy and quantitatively accurate means to determine clinical DDI risk from in vitro data. Weak and moderate inhibitors did not show TDI under standard incubation conditions using HLM and optimization of incubation conditions is warranted. Recommendations are made with respect to input parameters for IVIVE of TDI with non-CYP3A enzymes using available data from HLM and HHEPs.

Human hepatocytes and cytochrome P450-selective inhibitors predict variability in human drug exposure more accurately than human recombinant P450s

BACKGROUND AND PURPOSE

Drugs metabolically eliminated by several enzymes are less vulnerable to variable compound exposure in patients due to drug-drug interactions (DDI) or if a polymorphic enzyme is involved in their elimination. Therefore, it is vital in drug discovery to accurately and efficiently estimate and optimize the metabolic elimination profile.

EXPERIMENTAL APPROACH

CYP3A and/or CYP2D6 substrates with well described variability in vivo in humans due to CYP3A DDI and CYP2D6 polymorphism were selected for assessment of fraction metabolized by each enzyme (fmCYP ) in two in vitro systems: (i) human recombinant P450s (hrP450s) and (ii) human hepatocytes combined with selective P450 inhibitors. Increases in compound exposure in poor versus extensive CYP2D6 metabolizers and by the strong CYP3A inhibitor ketoconazole were mathematically modelled and predicted changes in exposure were compared with in vivo data.

KEY RESULTS

Predicted changes in exposure were within twofold of reported in vivo values using fmCYP estimated in human hepatocytes and there was a strong linear correlation between predicted and observed changes in exposure (r2  = 0.83 for CYP3A, r2  = 0.82 for CYP2D6). Predictions using fmCYP in hrP450s were not as accurate (r2  = 0.55 for CYP3A, r2  = 0.20 for CYP2D6).

CONCLUSIONS AND IMPLICATIONS

The results suggest that variability in human drug exposure due to DDI and enzyme polymorphism can be accurately predicted using fmCYP from human hepatocytes and CYP-selective inhibitors. This approach can be efficiently applied in drug discovery to aid optimization of candidate drugs with a favourable metabolic elimination profile and limited variability in patients.

Quantitative prediction of the extent of drug-drug interaction using a physiologically based pharmacokinetic model that includes inhibition of drug metabolism determined in cryopreserved hepatocytes

1. A physiologically based pharmacokinetic (PBPK) model that includes inhibition constant evaluated in cryopreserved hepatocytes was used to predict drug-drug interactions (DDIs) between orally administered nifedipine, a CYP substrate, and fluconazole or ketoconazole, CYP inhibitors, in rats. 2. The Kp,uu, ratio of unbound inhibitor concentration in liver ([I]liver,u) to that in plasma ([I]sys,u), of fluconazole and ketoconazole was 1.0 and 13.0, indicating that ketoconazole accumulates in liver. The ratios of inhibition constants in rat liver microsomes (Ki,mic,u) to that in rat cryopreserved hepatocytes (Ki,hep,u) for fluconazole and ketoconazole were 1.5 and 25.5, which were similar to the Kp,uu and suggested that cryopreserved hepatocytes could mimic the hepatic accumulation of inhibitors. 3. The increases in AUC of nifedipine predicted by the minimal PBPK model using [I]liver,u/Ki,mic,u and [I]sys,u/Ki,hep,u were within 1.5-fold of the observed values for both inhibitors, whereas the model using [I]sys,u/Ki,mic,u underestimated the AUC increase caused by ketoconazole 21-fold. 4. These results indicated that hepatic accumulation factor of an inhibitor is required for a precise DDI projection and that cryopreserved hepatocytes would be useful to obtain the Ki including hepatic accumulation factor. It was demonstrated that PBPK model using Ki,hep,u could be a valuable approach for quantitative DDI projection.

Evaluation of a New Molecular Entity as a Victim of Metabolic Drug-Drug Interactions-an Industry Perspective

Under the guidance of the International Consortium for Innovation and Quality in Pharmaceutical Development (IQ), scientists from 20 pharmaceutical companies formed a Victim Drug-Drug Interactions Working Group. This working group has conducted a review of the literature and the practices of each company on the approaches to clearance pathway identification (fCL), estimation of fractional contribution of metabolizing enzyme toward metabolism (fm), along with modeling and simulation-aided strategy in predicting the victim drug-drug interaction (DDI) liability due to modulation of drug metabolizing enzymes. Presented in this perspective are the recommendations from this working group on: 1) strategic and experimental approaches to identify fCL and fm, 2) whether those assessments may be quantitative for certain enzymes (e.g., cytochrome P450, P450, and limited uridine diphosphoglucuronosyltransferase, UGT enzymes) or qualitative (for most of other drug metabolism enzymes), and the impact due to the lack of quantitative information on the latter. Multiple decision trees are presented with stepwise approaches to identify specific enzymes that are involved in the metabolism of a given drug and to aid the prediction and risk assessment of drug as a victim in DDI. Modeling and simulation approaches are also discussed to better predict DDI risk in humans. Variability and parameter sensitivity analysis were emphasized when applying modeling and simulation to capture the differences within the population used and to characterize the parameters that have the most influence on the prediction outcome.

Human hepatocyte assessment of imatinib drug-drug interactions - complexities in clinical translation

AIM

Inducers and inhibitors of CYP3A, such as ritonavir and efavirenz, may be used as part of the highly active antiretroviral therapy (HAART) to treat HIV patients. HIV patients with chronic myeloid leukemia or gastrointestinal stromal tumour may need imatinib, a CYP3A4 substrate with known exposure response-relationships. Administration of imatinib to patients on ritonavir or efavirenz may result in altered imatinib exposure leading to increased toxicity or failure of therapy, respectively. We used primary human hepatocyte cultures to evaluate the magnitude of interaction between imatinib and ritonavir/efavirenz.

METHODS

Hepatocytes were pre-treated with vehicle, ritonavir, ketoconazole, efavirenz or rifampicin, and the metabolism of imatinib was characterized over time. Concentrations of imatinib and metabolite were quantitated in combined lysate and medium, using LC-MS.

RESULTS

The predicted changes in imatinib CLoral (95% CI) with ketoconazole, ritonavir, rifampicin and efavirenz were 4.0-fold (0, 9.2) lower, 2.8-fold (0.04, 5.5) lower, 2.9-fold (2.2, 3.5) higher and 2.0-fold (0.42, 3.5) higher, respectively. These predictions were in good agreement with clinical single dose drug-drug interaction studies, but not with reports of imatinib interactions at steady-state. Alterations in metabolism were similar after acute or chronic imatinib exposure.

CONCLUSIONS

In vitro human hepatocytes predicted increased clearance of imatinib with inducers and decreased clearance with inhibitors of CYP enzymes. The impact of HAART on imatinib may depend on whether it is being initiated or has already been dosed chronically in patients. Therapeutic drug monitoring may have a role in optimizing imatinib therapy in this patient population.

In vitro-in vivo correlation of the effect of supersaturation on the intestinal absorption of BCS Class 2 drugs

The aim of this study was to establish an in vitro method for evaluating the effect of supersaturation on oral absorption of poorly water-soluble drugs in vivo. Albendazole, dipyridamole, gefitinib, and ketoconazole were used as model drugs. Supersaturation of each drug was induced by diluting its stock solution by fasted state simulated intestinal fluid (FaSSIF) (solvent-shift method), then dissolution and precipitation profile of the drug was observed in vitro. The crystalline form of the precipitate was checked by differential scanning calorimetry (DSC). For comparison, control suspension was prepared by suspending a drug powder directly into FaSSIF (powder-suspending method). In vivo intestinal absorption of the drug was observed in rats by determined the plasma concentration after intraduodenal administration of drug suspensions. For all drugs, suspensions prepared by solvent-shift method showed significantly higher dissolved concentration in vitro than that prepared by powder-suspending method, clearly indicated the induction of supersaturation. DSC analysis revealed that crystalline form of the precipitate profoundly affects the extent and the duration of supersaturation. A rat in vivo study confirmed that the supersaturation of these drugs increased the fraction absorbed from the intestine, which corresponded well to the in vitro dissolution and precipitation profile of drugs except for ketoconazole. For ketoconazole, an in vivo absorption study was performed in rats pretreated with 1-aminobenzotriazole, a potent inhibitor of CYP mediated metabolism. CYP inhibition study suggested that the high luminal concentration of ketoconazole caused by supersaturation saturated the metabolic enzymes and further increased the systemic exposure of the absorbed drug. The additional effects of supersaturation on the absorption of ketoconazole are consistent with previous studies in humans under differing gastric pH conditions. In conclusion, effects of supersaturation on the intestinal absorption of poorly water-soluble drugs could be predicted from in vitro dissolution and a precipitation study. However if supersaturation affects the pharmacokinetic profiles of drugs, such as a first-pass metabolism, a combination with in vivo study should be required to evaluate its impact on oral bioavailability.

Irreversible enzyme inhibition kinetics and drug-drug interactions

This chapter describes the types of irreversible inhibition of drug-metabolizing enzymes and the methods commonly employed to quantify the irreversible inhibition and subsequently predict the extent and time course of clinically important drug-drug interactions.

Text Mining Driven Drug-Drug Interaction Detection

Identifying drug-drug interactions is an important and challenging problem in computational biology and healthcare research. There are accurate, structured but limited domain knowledge and noisy, unstructured but abundant textual information available for building predictive models. The difficulty lies in mining the true patterns embedded in text data and developing efficient and effective ways to combine heterogenous types of information. We demonstrate a novel approach of leveraging augmented text-mining features to build a logistic regression model with improved prediction performance (in terms of discrimination and calibration). Our model based on synthesized features significantly outperforms the model trained with only structured features (AUC: 96% vs. 91%, Sensitivity: 90% vs. 82% and Specificity: 88% vs. 81%). Along with the quantitative results, we also show learned "latent topics", an intermediary result of our text mining module, and discuss their implications.

Pharmacointeraction network models predict unknown drug-drug interactions

Drug-drug interactions (DDIs) can lead to serious and potentially lethal adverse events. In recent years, several drugs have been withdrawn from the market due to interaction-related adverse events (AEs). Current methods for detecting DDIs rely on the accumulation of sufficient clinical evidence in the post-market stage – a lengthy process that often takes years, during which time numerous patients may suffer from the adverse effects of the DDI. Detection methods are further hindered by the extremely large combinatoric space of possible drug-drug-AE combinations. There is therefore a practical need for predictive tools that can identify potential DDIs years in advance, enabling drug safety professionals to better prioritize their limited investigative resources and take appropriate regulatory action. To meet this need, we describe Predictive Pharmacointeraction Networks (PPINs) – a novel approach that predicts unknown DDIs by exploiting the network structure of all known DDIs, together with other intrinsic and taxonomic properties of drugs and AEs. We constructed an 856-drug DDI network from a 2009 snapshot of a widely-used drug safety database, and used it to develop PPIN models for predicting future DDIs. We compared the DDIs predicted based solely on these 2009 data, with newly reported DDIs that appeared in a 2012 snapshot of the same database. Using a standard multivariate approach to combine predictors, the PPIN model achieved an AUROC (area under the receiver operating characteristic curve) of 0.81 with a sensitivity of 48% given a specificity of 90%. An analysis of DDIs by severity level revealed that the model was most effective for predicting “contraindicated” DDIs (AUROC = 0.92) and less effective for “minor” DDIs (AUROC = 0.63). These results indicate that network based methods can be useful for predicting unknown drug-drug interactions.

Time-dependent inhibition and estimation of CYP3A clinical pharmacokinetic drug-drug interactions using plated human cell systems

The current studies assessed the utility of freshly plated hepatocytes, cryopreserved plated hepatocytes, and cryopreserved plated HepaRG cells for the estimation of inactivation parameters k(inact) and K(I) for CYP3A. This was achieved using a subset of CYP3A time-dependent inhibitors (fluoxetine, verapamil, clarithromycin, troleandomycin, and mibefradil) representing a range of potencies. The estimated k(inact) and K(I) values for each time-dependent inhibitor were compared with those obtained using human liver microsomes and used to estimate the magnitude of clinical pharmacokinetic drug-drug interaction (DDI). The inactivation kinetic parameter, k(inact), was most consistent across systems tested for clarithromycin, verapamil, and troleandomycin, with a high k(inact) of 0.91 min(-1) observed for mibefradil in HepaRG cells. The apparent K(I) estimates derived from the various systems displayed a range of variability from 3-fold for clarithromycin (5.4-17.7 μM) to 6-fold for verapamil (1.9-12.6 μM). In general, the inactivation kinetic parameters derived from the cell systems tested fairly replicated what was observed in time-dependent inhibition studies using human liver microsomes. Despite some of the observed differences in inactivation kinetic parameters, the estimated DDIs derived from each of the tested systems generally agreed with the clinically reported DDI within approximately 2-fold. In addition, a plated cell approach offered the ability to conduct longer primary incubations (greater than 30 min), which afforded improved ability to identify the weak time-dependent inhibitor fluoxetine. Overall, results from these studies suggest that in vitro inactivation parameters generated from plated cell systems may be a practical approach for identifying time-dependent inhibitors and for estimating the magnitude of clinical DDIs.

A quantitative framework and strategies for management and evaluation of metabolic drug-drug interactions in oncology drug development: new molecular entities as object drugs

This article outlines general strategies for the management and evaluation of pharmacokinetic drug-drug interactions (DDIs) resulting from perturbation of clearance of investigational anticancer drug candidates by concomitantly administered agents in a drug development setting, with a focus on drug candidates that cannot be evaluated in first-in-human studies in healthy subjects. A risk level classification is proposed, based on quantitative integration of knowledge derived from preclinical drug-metabolism studies evaluating the projected percentage contribution [f(i)(%)] of individual molecular determinants (e.g. cytochrome P450 isoenzymes) to the overall human clearance of the investigational agent. The following classification is proposed with respect to susceptibility to DDIs with metabolic inhibitors: a projected maximum DDI expected to result in a ≤1.33-fold increase in exposure, representing a low level of risk; a projected maximum DDI expected to result in a >1.33-fold but <2-fold increase in exposure, representing a moderate level of risk; and a projected maximum DDI expected to result in a ≥2-fold increase in exposure, representing a potentially high level of risk. For DDIs with metabolic inducers, the following operational classification is proposed, based on the sum of the percentage contributions of enzymes that are inducible via a common mechanism to the overall clearance of the investigational drug: <<25%, representing a low level of risk; <50%, representing a moderate level of risk; and ≥50%, representing a potentially high level of risk. To ensure patient safety and to minimize bias in determination of the recommended phase II dose (RP2D), it is recommended that strong and moderate inhibitors and inducers of the major contributing enzyme are excluded in phase I dose-escalation studies of high-risk compounds, whereas exclusion of strong inhibitors and inducers of the contributing enzyme(s) is recommended as being sufficient for moderate-risk compounds. For drugs that will be investigated in diseases such as glioblastoma, where there may be relatively frequent use of enzyme-inducing antiepileptic agents (EIAEDs), a separate dose-escalation study in this subpopulation is recommended to define the RP2D. For compounds in the high-risk category, if genetic deficiencies in the activity of the major drug-metabolizing enzyme are known, it is recommended that poor metabolizers be studied separately to define the RP2D for this subpopulation. Whereas concomitant medication exclusion criteria that are utilized in the phase I dose-escalation studies will probably also need to be maintained for high-risk compounds in phase II studies unless the results of a clinical DDI study indicate the absence of a clinically relevant interaction, these exclusion criteria can potentially be relaxed beyond phase I for moderate-risk compounds, if supported by the nature of clinical toxicities and the understanding of the therapeutic index in phase I. Adequately designed clinical DDI studies will not only inform potential relaxation of concomitant medication exclusion criteria in later-phase studies but, importantly, will also inform the development of pharmacokinetically derived dose-modification guidelines for use in clinical practice when coupled with adequate safety monitoring, as illustrated in the prescribing guidance for many recently approved oncology therapeutics.

SuperCYP: a comprehensive database on Cytochrome P450 enzymes including a tool for analysis of CYP-drug interactions

Much of the information on the Cytochrome P450 enzymes (CYPs) is spread across literature and the internet. Aggregating knowledge about CYPs into one database makes the search more efficient. Text mining on 57 CYPs and drugs led to a mass of papers, which were screened manually for facts about metabolism, SNPs and their effects on drug degradation. Information was put into a database, which enables the user not only to look up a particular CYP and all metabolized drugs, but also to check tolerability of drug-cocktails and to find alternative combinations, to use metabolic pathways more efficiently. The SuperCYP database contains 1170 drugs with more than 3800 interactions including references. Approximately 2000 SNPs and mutations are listed and ordered according to their effect on expression and/or activity. SuperCYP (http://bioinformatics.charite.de/supercyp) is a comprehensive resource focused on CYPs and drug metabolism. Homology-modeled structures of the CYPs can be downloaded in PDB format and related drugs are available as MOL-files. Within the resource, CYPs can be aligned with each other, drug-cocktails can be 'mixed', SNPs, protein point mutations, and their effects can be viewed and corresponding PubMed IDs are given. SuperCYP is meant to be a platform and a starting point for scientists and health professionals for furthering their research.

Nonlinear absorption of methylprednisolone by absorptive and secretory transporters

The intestinal absorption rate constant of methylprednisolone (MP) evaluated by the loop method increased significantly with increasingly higher concentrations of the drug up to 500 microM in a nonlinear fashion but did not increase further at higher concentrations. Mucosal-to-serosal directed permeation of MP across rat jejunal sheets also increased in a nonlinear fashion in a low concentration range (100-150 microM), followed by a decrease as the concentration increased further, whereas serosal-to-mucosal directed permeation decreased in a concentration-dependent manner. Vectorial transport of MP across Caco-2 cell monolayers was observed, with greater transport in the basolateral-to-apical direction at 37 degrees C. These observations suggest that MP is taken up in the intestinal epithelial cells by a carrier-mediated transport mechanism. The absorptive and secretory clearance of MP increased and decreased with P-glycoprotein (P-gp) inhibitors, respectively. These results strongly suggest that MP is secreted into the intestinal lumen predominantly by P-gp. We conclude that intestinal transport of MP involves P-gp or some other transporters in both the absorptive and secretory directions, and complex nonlinear intestinal absorption characteristics can be ascribed to the existence of multiple transport mechanisms.