- Email: [email protected]
In Vitro–In Vivo Extrapolation of Metabolism- and Transporter-Mediated Drug–Drug Interactions—Overview of Basic Prediction Methods
Accepted Manuscript In Vitro-In Vivo Extrapolation of Metabolism- and Transporter-Mediated Drug-Drug Interactions – Overview of Basic Prediction Methods Kenta Yoshida, Ping Zhao, Lei Zhang, Darrell R. Abernethy, Dinko Rekić, Kellie S. Reynolds, Aleksandra Galetin, Shiew-Mei Huang PII:
S0022-3549(17)30281-2
DOI:
10.1016/j.xphs.2017.04.045
Reference:
XPHS 768
To appear in:
Journal of Pharmaceutical Sciences
Received Date: 7 February 2017 Revised Date:
10 April 2017
Accepted Date: 20 April 2017
Please cite this article as: Yoshida K, Zhao P, Zhang L, Abernethy DR, Rekić D, Reynolds KS, Galetin A, Huang SM, In Vitro-In Vivo Extrapolation of Metabolism- and Transporter-Mediated Drug-Drug Interactions – Overview of Basic Prediction Methods, Journal of Pharmaceutical Sciences (2017), doi: 10.1016/j.xphs.2017.04.045. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
In Vitro-In Vivo Extrapolation of Metabolism- and TransporterMediated Drug-Drug Interactions – Overview of Basic Prediction Methods Kenta Yoshida1,2,3, Ping Zhao1, Lei Zhang1, Darrell R. Abernethy1, Dinko Rekić1, Kellie S Reynolds1, Aleksandra Galetin1,2,4, Shiew-Mei Huang1 1
2
Oak Ridge Institution for Science and Education (ORISE) Fellow
3
RI PT
Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD, USA
Current affiliation: Clinical Pharmacology, Genentech Research and Early Development, 1 DNA Way, South San Francisco, CA, USA
4
SC
Centre for Applied Pharmaceutical Research, School of Health Sciences, University of Manchester, Manchester, UK
Shiew-Mei Huang, PhD
M AN U
Corresponding author:
Office of Clinical Pharmacology, Office of Translational Sciences, Center of Drug Evaluation and Research, U.S. Food and Drug Administration Silver Spring, MD 20993 Phone: 301-796-1541
Figures and Tables: 1 Words in Abstract: 119
TE D
Email: [email protected]
EP
Words in Manuscript: 4632 (including abstract, tables, figures and references)
AC C
Keywords: Drug interaction, Cytochrome P450s, Membrane transport/transporters, Regulatory, Guidance/Guideline
1
ACCEPTED MANUSCRIPT List of abbreviations: ADME: absorption, distribution, metabolism and excretion BCRP: breast cancer resistance protein CYP: cytochrome P-450 d: induction calibration factor DDI: drug-drug interaction
RI PT
EC50: concentration of inducer associated with half-maximum induction EMA: European Medicines Agency Emax: maximum induction effect FaFg: intestinal bioavailability fu,p: fraction unbound in plasma [I]: perpetrator (inhibitor or inducer) concentration
SC
FDA: Food and Drug Administration
Imax: maximum total (unbound plus bound) inhibitor concentration in plasma ITC: International Transporter Consortium
M AN U
Imax,u: maximum unbound inhibitor or inducer concentration
Iu,in,max: theoretical maximum unbound hepatic inlet concentration IVIVE: in vitro-in vivo extrapolation ka: absorption rate constant
kdeg: degradation rate constant of CYP enzyme
TE D
Ki: inhibition constant KI: apparent inactivation constant
kinact: maximal inactivation rate constant
kobs: the apparent inactivation rate constant
MATE: multidrug and toxin extrusion protein
EP
OAT: organic anion transporter
OATPs: organic anion transporting polypeptides
AC C
OCT: organic cation transporter
PET: positron emission tomography P-gp: P-glycoprotein
PK: pharmacokinetic
PMDA: Pharmaceuticals and Medical Devices Agency of Japan Qh: hepatic blood flow rate
RB: blood-to-plasma concentration ratio RIS: relative induction score UGT: UDP-glucuronosyltransferase
2
ACCEPTED MANUSCRIPT
Abstract
RI PT
Evaluation of drug-drug interaction (DDI) risk is vital to establish benefit-risk profiles of investigational new drugs during drug development. In vitro experiments are routinely conducted as an important first step to assess metabolism- and/or transporter-mediated DDI potential of investigational new drugs. Results from these experiments are interpreted, often with the aid of in vitro-in vivo extrapolation (IVIVE) methods, to determine whether and how DDI should be evaluated clinically to provide the basis for proper DDI management strategies, including dosing recommendations, alternative therapies, or contraindications under various DDI scenarios and in different patient population. This article provides an overview of currently available in vitro experimental systems and basic IVIVE methodologies for metabolism- and transporter-mediated DDIs.
SC
1 Background
TE D
M AN U
Many factors can affect pharmacokinetic (PK) profiles of drugs after administration to different patients, including concomitant medications, disease status, and genetic polymorphisms. For optimal use of medications, it is important to understand which factors are relevant to the pharmacokinetics of each drug. DDIs are one of the important factors that alter the pharmacokinetics of drugs, and they have in some cases resulted in fatal adverse events leading to withdrawal of drugs from the market1. Because of their importance, the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and Ministry of Health, Labour and Welfare of Japan (MHLW) have published a draft guidance or guideline (FDA and MHLW) or a final guideline (EMA) on methodologies to evaluate DDIs for investigational new drugs during drug development2-4. The evaluation of DDIs generally involves multiple processes, including 1) prediction of clinical DDIs with preclinical or in vitro systems, 2) evaluation/confirmation of the extent of DDIs with clinical studies, and/or 3) modelbased simulations of DDIs. This review article focuses on the first element, prediction of DDI using data from preclinical and in vitro systems, especially focusing on basic IVIVE methodologies.
EP
2 Clinically relevant metabolizing enzymes and transporters for DDI evaluation
AC C
The most well-known and important proteins involved in PK-based DDIs are enzymes of the cytochrome P-450 (CYP) family. The inhibition or induction of CYP enzymes can directly affect the exposures to substrate drugs and result in unexpected changes in efficacy and/or safety profiles without appropriate dose adjustment. Many of CYP enzymes have broad substrate specificity and they mediate the metabolism of the majority of marketed small molecule drugs5. Clinically important CYP enzymes relevant for PK-based DDIs include CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, and CYP3A. Many of these CYP enzymes are predominantly expressed in the liver. CYP3A is also expressed significantly in the intestine and it can contribute to the intestinal first-pass extraction of orally administered drugs. Other drug metabolizing enzymes, such as UDPglucuronosyltransferase (UGT), are less frequently evaluated for potential PK-based DDIs, mainly due to the lack of selective inhibitor and inducer drugs that can significantly influence PK of their substrates. In contrast to drug metabolizing enzymes, the alteration of transporter function can not only affect drug elimination, but also drug distribution. For example, inhibition of organic anion transporting polypeptides (OATPs) by rifampin significantly reduces radioactivity measured in the 3
ACCEPTED MANUSCRIPT
3 In vitro experimental systems
RI PT
liver with positron emission tomography (PET) after the administration of (15R)-11C-TIC-Me in humans6 or rosuvastatin in rats7. A list of drug transporters that may alter drug disposition has been recommended by the International Transporter Consortium (ITC)8-11. Drug transporters that may be involved in clinically significant DDIs include OATP1B1/OATP1B3, organic anion transporter 1/3 (OAT1/OAT3), organic cation transporter 2 (OCT2), breast cancer resistance protein (BCRP), Pglycoprotein (P-gp), and multidrug and toxin extrusion protein 1/2-K (MATE1/MATE2-K). The regulatory agencies from the U.S., Europe, and Japan adopted this recommended list in their recently published DDI guidance/guidelines2-4.
M AN U
SC
The gold standard for preclinical DDI evaluation has been the use of enzyme- or transporterexpressing systems or human cellular and subcessular fractions. Because of species differences in metabolizing enzymes and transporters between human and animals, results of DDI studies in animals are not frequently used to predict clinical DDIs. In general, the purpose of using human expression systems is to evaluate specific interaction of one particular human protein with the drug of interest. Use of holistic cellular systems (e.g., hepatocytes) allows characterization of the relative contribution of individual metabolizing enzymes or transporters to the drug clearance in the presence of all other proteins. In this section, we briefly explain available in vitro experimental systems for DDI evaluation, focusing predominantly on the inhibition type interactions. More detailed descriptions of individual experimental systems can be found in other review articles12,13.
TE D
Human liver microsomes and human hepatocytes are the most commonly utilized human in vitro systems to evaluate metabolism-mediated DDI. In many cases, in vivo metabolism pathways of a drug can be effectively evaluated with these systems. Probe substrates, inhibitors and inducers for individual CYP enzymes have been well established 14,15 allowing the evaluation of DDI potential for an investigational drug with these systems with the drug acting either as a substrate or as a perpetrator. Microsomes prepared from cells overexpressing a specific isoform of enzyme (e.g., recombinant CYP enzymes) are useful to determine the involvement of a specific enzyme to the metabolism of the investigational drug and can be applied for evaluation of DDIs (hepatic and intestinal).
AC C
EP
Compared to metabolism-mediated DDIs, cell lines transftected with individual transporters (e.g., HEK-OATP1B1) are used more frequently for the evaluation of transporter-mediated DDIs. The system of choice depends on questions of interest, function and directional characteristics of transporters13,16,17. Transfected cell lines expressing individual transporters of interest, coupled with probe substrates, are useful for studying a drug as a potential transporter inhibitor. When evaluating the drug as a potential substrate, these systems address a “yes/no” question to determine the involvement of a particular transporter. Human hepatocytes express multiple transporters and are useful tools to determine quantitative contributions of a transporter, investigate rate limiting processes, and complex metabolism-transporter interplay13. Certain methodological aspects of transporter in vitro inhibition studies are still challenging and may affect ability to predict transporter-mediated DDI risk. For example, selective probe substrates and inhibitors for each transporter have not been well established. Occurrence of substratedependent inhibition, associated with possible multiple binding sites, highlights the need to use relevant substrate co-medications/clinical probe in the in vitro inhibition studies16. Some OATP inhibitors (e.g., cyclosporine) have shown increased inhibitory potency following a pre-incubation step 18,19. In addition, quantitative estimation of inhibition parameters may require mathematical modeling of in vitro data. This is of particular relevance for the efflux transporters, as the interaction 4
ACCEPTED MANUSCRIPT of these transporters with the unbound intracellular drug concentration is considered in the model, rather than the nominal incubation concentration16. Furthermore, large inter-laboratory variations have been reported for in vitro inhibition parameters (IC50 values) for certain transporters (e.g, OATP1B1 and P-gp) 20,21. This may require additional effort for successful IVIVEs, such as standardization of in vitro methods or calibration of cellular systems using a standard set of model inhibitors, as discussed in digoxin case22 to generate “customized” decision criteria for DDI prediction.
4 In vitro-in vivo extrapolation of DDI
SC
RI PT
The next step of DDI evaluation is the extrapolation of in vitro parameters to in vivo interaction predictions. For a new drug as a substrate, knowing the contribution of a specific metabolizing enzyme or transporter to the overall absorption, distribution, metabolism and excretion (ADME) processes, based on IVIVE and clinical PK data, is required for an effective evaluation of clinical DDI potential. For example, in the case of metabolism-based DDI, it is recommended to conduct a clinical DDI study with strong index inhibitors and/or inducers if an enzyme of interest is responsible for ≥25% of the drug’s elimination based on in vitro phenotyping and human PK data.
Prediction of metabolism-mediated DDI
TE D
4.1
M AN U
For a new drug as a perpetrator, the first step of extrapolation is often a simple comparison of in vitro parameters with in vivo exposure to the drug (also known as a basic model). If a basic model shows the potential of clinical DDI according to a predefined cut-off criterion, the interaction potential may be usefully investigated via a clinical DDI study or with the use of mechanistic models such as physiologically-based pharmacokinetic (PBPK) models2-4. In the following sections, we provide an overview of the basic models for evaluating metabolism- and transporter-mediated DDIs involving a new drug as a perpetrator.
AC C
EP
The essential components of any metabolism- and transporter-mediated DDI predictions are the estimation of perpetrator concentration at the site of interaction and the in vitro potencies of the perpetrator (defined by inhibition constant, Ki). A basic model uses a constant value of perpetrator concentration to simplify prediction calculations. In many cases, the maximum steady-state concentration of the perpetrator at the highest clinical dose is used to minimize the likelihood of false negative prediction during this first step of evaluation. Basic models include other simplifications, such as assumption of 100% involvement of the pathway of interest, no consideration of the interplay of multiple enzymes or transporters and use of plasma concentration as a surrogate of the perpetrator concentration at the site of interaction16. The majority of metabolism-mediated DDI predictions focus on CYP-mediated interactions due to their clinical significance. Prediction methods for other enzymes such as UGTs have not been well established. For reversible inhibition, the basic model uses a simple comparison of the inhibitor concentration and the Ki for inhibitor-enzyme of interest. There are differences in the recommendations by regulatory agencies on inhibitor concentrations and cut-off values to use: e.g., Imax/Ki ≥ 0.1 (FDA recommendation in 2012 draft guidance2 and MHLW recommendation in 2014 draft guideline4) and Imax,u/Ki ≥ 0.02 (EMA recommendation in 2013 guideline3), where Imax and Imax,u represent the maximum total (unbound plus bound) and unbound inhibitor concentrations in plasma, respectively. A recent analysis suggested that these two criteria performed similarly in predicting reported DDIs23. The above criteria are intended for the prediction of DDIs after drug reaches systemic circulation. The concentration of a potential inhibitor in the intestine following oral administration can 5
ACCEPTED MANUSCRIPT
RI PT
be much higher than the systemic concentration24. Therefore, the prediction based on systemic concentration of inhibitor can underestimate interaction potential for orally administered drugs. The underestimation is especially of concern for substrates of CYP3A enzymes (and transporters such as P-gp or BCRP). Using dose/250mL as a surrogate for intestinal inhibitor concentration, a cut-off criterion of 10 for the (dose/250mL)/Ki was proposed by regulatory agencies2-4. This criterion was deemed appropriate based on analyses where performance of in vivo DDI prediction was assessed for CYP3A and P-gp using criteria similar to those in the guidances23,25-27. The prediction criteria were recently evaluated using receiver operating characteristic analysis resulting in a higher intestinal [I2]/IC50 cut-off, while tightening the criterion for systemic P-gp inhibition 28.
SC
Prediction of the time-dependent inhibition (TDI) is more challenging than the reversible inhibition. The inherent difficulty is that dynamic changes in enzyme function due to inactivation are approximated with a simple, static equation. The parameters for basic models include degradation rate constant of CYP enzyme (kdeg), apparent inactivation constant (KI), and maximal inactivation rate constant (kinact). There are differences in the concentrations and cut-off criteria used for predictions among regulatory agencies. The following two equations are used for the basic model prediction of TDI of hepatic CYP enzymes: (kobs+kdeg)/kdeg ≥ 1.1 where kobs=kinact×Imax /(KI + Imax) (FDA 2012 draft guidance and MHLW 2014 draft guideline 2,4)
•
(kobs+kdeg)/kdeg ≥ 1.25 where kobs=kinact×50×Imax,u /(KI + 50×Imax,u) (EMA 2013 guideline3)
M AN U
•
TE D
where kobs represents the apparent inactivation rate constant. A retrospective analysis demonstrated that neither equation resulted in any false-negative predictions23. However, both equations led to positive predictions of inhibition for the majority of tested drugs, regardless of observed clinical DDI outcomes. Because of the time-dependent nature of TDI, a dynamic model-based approach provides a more reliable prediction of the extent of DDI with TDI. However, recent analyses highlight challenges in the prediction of complex metabolic DDIs, e.g., combined TDI and induction interactions, even when using mechanistic PBPK models29,30. Furthermore, confidence in predicting complex transporter-metabolism interplay and associated DDIs is still generally low29,31,32.
EP
Several basic methods have been proposed for the prediction of enzyme induction30. Proposed approaches with their respective threshold criteria include: 1) Fold-change method: mRNA increase ≥ a predefined threshold based on positive/negative controls and exceeds a certain % in comparison to positive controls. For example, EMA recommends a concentration-dependent mRNA increase of ≥100% (2-fold) and ≥20% of the response of the positive control in the presence of an investigational drug, at relevant concentrations stipulated in the guidance, as positive induction (EMA 2013 guideline3);
AC C
33
2) Basic model: R3=1/(1+d×Emax×[I]/(EC50 +[I])) ≤ 0.8 as positive induction, where EC50 is the concentration of inducer associated with half-maximum induction, Emax is the maximum induction 34 effect, [I] is the inducer concentration, and d is the induction calibration factor ; 3) Correlation method using Imax,u / EC50: Calculate Imax,u / EC50 values with cut-off values defined with known inducers and non-inducers; and 4) Correlation method using relative induction score, RIS: calibration of RIS (= Emax×[I]/(EC50 +[I])) with known inducers35. Each regulatory guidance includes one or more of the above four methods for induction predictions. In addition to mRNA expression levels, induction can also be studied by assessing enzyme activities, with the caveat that induction may be masked by concurrent inhibition. 6
ACCEPTED MANUSCRIPT 4.2
Prediction of transporter-mediated DDI
Basic models are the main prediction method used to extrapolate in vitro assessment of investigational drugs as potential in vivo transporter inhibitors. Because different transporters are expressed in different organs and different side of membranes, the accurate estimation of inhibitor concentrations at the site of inhibition is key for accurate predictions.
SC
Iu,in,max = fu,p × (Imax + ka × Dose × FaFg/Qh/RB),
RI PT
For OATP1B1 and OATP1B3, which are expressed at the basolateral membrane of hepatocytes, the basic prediction method uses the unbound hepatic inlet plasma concentration as the inhibitor concentrations. It is important to note that hepatic plasma concentration can be higher than systemic concentration after oral drug administration, because drugs absorbed from small intestine first reach the liver via portal vein. In order to account for the contribution of drug concentration from absorption, the following equation was proposed to estimate the theoretical maximum unbound hepatic inlet plasma concentration, Iu,in,max36:
TE D
M AN U
where ka, FaFg, fu,p, Qh, and RB represent absorption rate constant, intestinal bioavailability, fraction unbound in plasma, hepatic blood flow rate, and blood-to-plasma concentration ratio, respectively. Prediction accuracies of OATP-mediated DDIs using different inhibitor concentrations and different “cut-off” criteria have been compared in recent studies 21,37 (Fig 1). In the study by Vaidyanathan et al, Iu,in,max/Ki ≥ 0.1, Imax,u/Ki ≥ 0.02, and Iu,in,max/Ki ≥ 0.25 alone (MHLW recommendation in 2014 draft guideline4), or combined criterion Imax/Ki ≥ 0.1 and Iu,in,max/Ki ≥ 0.25 (FDA recommendation in 2012 draft guidance2) were all considered reasonable to determine the need to consider conducting clinical DDI studies. Although there were small differences in numbers of false predictions among these criteria, it is important to note that different criteria led to different types of false predictions. For example, prediction using maximum plasma total (bound + unbound) concentration in the systemic circulation tends to cause false-positive predictions of highly protein-bound drugs. Therefore, it is important to consider differences between methods when interpreting prediction results, as described in detail in the publication by Vaidyanathan et al21.
AC C
EP
For renal transporters, OAT1/3, OCT2, and MATE1/2-K, proposed basic prediction methods use unbound maximum systemic concentrations of inhibitors. Compared to CYP enzymes or OATPs, there are fewer cases of renal transporter-mediated DDIs, and the degree of change in systemic concentration is relatively small, leading to a limited dataset for validation of basic prediction methods for renal transporters. Furthermore, OCT2 and MATEs often share the same substrates and inhibitors, therefore the net DDI effect may be affected by both transporters at the basolateral and apical side of the renal tubule cells. Thus, an accurate prediction of DDI is more challenging for substrates for OCT2 and MATEs and inhibition of both transporters may need to be considered simultaneously. It is also important to note that inhibition of MATEs may cause an increase of intracellular concentrations, which would not be reflected in the systemic exposure of substrate drugs32. According to studies by Dong et al38,39, for OAT1/3-mediated DDIs, two of the recommended criteria showed comparable prediction outcomes: Imax,u/Ki ≥ 0.1 (FDA recommendation in 2012 draft guidance2) and Imax,u/Ki ≥ 0.25 (MHLW recommendation in 2014 draft guideline4). For OCT2 and MATEs, they proposed to use a lower cut-off for MATEs (i.e., Imax,u/Ki ≥ 0.02) than for OCT2 (i.e., 0.1, the same as that for OATs) because a lower cut-off appeared to avoid false negative prediction. Using a lower cut-off for MATEs also accounts for possible higher intracellular concentration of the inhibitor than its systemic concentration in the situation where the inhibitor is a substrate of an uptake transporter such as OCT2. Overall, because of the limited number of reported
7
ACCEPTED MANUSCRIPT DDIs and complex transporter interplays, the criteria for renal transporter-mediated DDIs need further evaluation.
SC
RI PT
P-gp and BCRP are expressed at various organs, but the main focus of DDI prediction for these transporters is on their inhibition in the intestinal lumen. One reason is that changes in oral bioavailability of substrate drugs by the inhibition of intestinal transporters can be reflected in the absorption profile of substrate drugs (e.g., Cmax change). Another reason is that the inhibitor concentration is in general much higher in the intestine than in systemic circulation after oral dosing, thus more prone to the perpetrator effect18. In fact, modulation of these transporters at the systemic level is considered unlikely in clinical situations, because unbound systemic concentrations of marketed drugs are generally too low to exhibit significant inhibition40. For intestinal transportermediated DDI prediction, the same cut-off criterion as intestinal CYP3A (dose/250mL/Ki ≥10) is proposed by regulatory agencies2-4, as described earlier in the section 4.1.
5 Conclusions
EP
TE D
M AN U
In vitro experimental systems and basic prediction models have evolved as useful tools to evaluate the risk of metabolism- and transporter-mediated DDIs for new molecular entities. Depending on the site or mechanisms of interactions, various equations and decision criteria have been proposed for DDI predictions. Considering the importance of this first-step of DDI prediction, there has been an ongoing effort among FDA, EMA, and MHLW to harmonize the recommendations for DDI prediction involving both metabolizing enzymes and transporters23. It is important to note, however, that the focus of these simple prediction methods is to eliminate the need to further evaluate DDIs that are unlikely to cause significant PK changes by using conservative assumptions to minimize falsenegative predictions. For a more accurate and quantitative prediction of DDIs based on in vitro observations, use of sophisticated model-based predictions such as PBPK modeling that incorporate mechanistic considerations of victim and perpetrator drug disposition should be explored, especially for DDIs involving multiple elimination pathways18,41.
Acknowledgements
AC C
This research was supported in part by the US Food and Drug Administration’s (FDA) Medical Countermeasures initiative. Drs. Aleksandra Galetin and Kenta Yoshida were supported in part by an appointment to the Research Participation Program at the Center for Drug Evaluation and Research, administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the FDA.
Conflict of interest/disclosure The authors declared no conflicts of interest. The contents of this article reflect the views of the authors and should not be construed to represent the FDA's views or policies. No official support or endorsement by the FDA is intended or should be inferred. The mention of commercial products, their sources, or their use in connection with material reported herein is not to be construed as either an actual or implied endorsement of such products by the FDA.
8
ACCEPTED MANUSCRIPT
Figure Legend Figure 1. Relationships between observed AUCR and different concentration of inhibitors divided by Ki or IC50. (a) Imax/ Ki or IC50, (b) Imax,u/ Ki or IC50, and (c) Iu,in,max/ Ki or IC50.
RI PT
Each symbol represents one DDI case. Horizontal lines represent cut-off values for 5 basic prediction methods evaluated in the original study21: (1) Imax/ Ki or IC50 ≥ 0.1, (2) Imax,u/ Ki or IC50≥0.02, (3) R (= 1 + Iu,in,max / Ki or IC50) ≥1.04 (EMA 3), (4) R (= 1 + Iu,in,max / Ki or IC50) ≥1.1, (5) R (= 1 + Iu,in,max / Ki or IC50) ≥1.25 (MHLW 2014 draft 4), and (6) Imax/ Ki or IC50 ≥0.1 AND R ≥ 1.25 (combination of (1) and (5); FDA 2012 draft 2). AUCR: area under the concentration-curve ratio, DDI: drug-drug
SC
interactions, IC50: half maximal inhibitory concentration, Imax: maximum systemic plasma concentration, Imax,u: maximum unbound systemic plasma concentration, Iu,in,max: estimated maximum
M AN U
inhibitor concentration at the inlet to the liver, Ki: inhibition constant, and OATP: organic anion
AC C
EP
TE D
transporting polypeptide. Figure from Vaidyanathan, et al., 2016 21.
9
ACCEPTED MANUSCRIPT
References 1. Giacomini KM, Krauss RM, Roden DM, Eichelbaum M, Hayden MR, Nakamura Y 2007. When good drugs go bad. Nature 446(7139):975-977.
RI PT
2. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER) 2012. Draft Guidance / Guidance for Industry. Drug Interaction Studies--Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM2 92362.pdf Accessed Mar 3, 2017. 3. European Medicines Agency 2013. Guideline on the Investigation of Drug Interactions. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/07/WC50012960 6.pdf Accessed Mar 3, 2017.
SC
4. Ministry of Health, Labour and Welfare, Japan 2014. Guideline of drug interaction studies for drug development and appropriate provision of information. https://www.pmda.go.jp/files/000206158.pdf Accessed Mar 23, 2017.
M AN U
5. Wienkers LC, Heath TG 2005. Predicting in vivo drug interactions from in vitro drug discovery data. Nature reviews Drug discovery 4(10):825-833. 6. Takashima T, Kitamura S, Wada Y, Tanaka M, Shigihara Y, Ishii H, Ijuin R, Shiomi S, Nakae T, Watanabe Y, Cui Y, Doi H, Suzuki M, Maeda K, Kusuhara H, Sugiyama Y, Watanabe Y 2012. PET imaging-based evaluation of hepatobiliary transport in humans with (15R)-11C-TIC-Me. J Nucl Med 53(5):741-748. 7. He J, Yu Y, Prasad B, Link J, Miyaoka RS, Chen X, Unadkat JD 2014. PET imaging of Oatpmediated hepatobiliary transport of [(11)C] rosuvastatin in the rat. Mol Pharm 11(8):2745-2754.
TE D
8. Giacomini KM, Huang SM, Tweedie DJ, Benet LZ, Brouwer KL, Chu X, Dahlin A, Evers R, Fischer V, Hillgren KM, Hoffmaster KA, Ishikawa T, Keppler D, Kim RB, Lee CA, Niemi M, Polli JW, Sugiyama Y, Swaan PW, Ware JA, Wright SH, Yee SW, Zamek-Gliszczynski MJ, Zhang L 2010. Membrane transporters in drug development. Nature reviews Drug discovery 9(3):215-236.
EP
9. Hillgren KM, Keppler D, Zur AA, Giacomini KM, Stieger B, Cass CE, Zhang L, International Transporter C 2013. Emerging transporters of clinical importance: an update from the International Transporter Consortium. Clin Pharmacol Ther 94(1):52-63.
AC C
10. Huang SM, Zhang L, Giacomini KM 2010. The International Transporter Consortium: a collaborative group of scientists from academia, industry, and the FDA. Clin Pharmacol Ther 87(1):32-36. 11. Giacomini KM, Huang SM 2013. Transporters in drug development and clinical pharmacology. Clin Pharmacol Ther 94(1):3-9. 12. Lu C, Liao M, Cohen L, Xia CQ 2010. Emerging in vitro tools to evaluate cytochrome P450 and transporter-mediated drug-drug interactions. Curr Drug Discov Technol 7(3):199-222. 13. Brouwer KL, Keppler D, Hoffmaster KA, Bow DA, Cheng Y, Lai Y, Palm JE, Stieger B, Evers R, International Transporter C 2013. In vitro methods to support transporter evaluation in drug discovery and development. Clin Pharmacol Ther 94(1):95-112. 14. Khojasteh SC, Prabhu S, Kenny JR, Halladay JS, Lu AY 2011. Chemical inhibitors of cytochrome P450 isoforms in human liver microsomes: a re-evaluation of P450 isoform selectivity. Eur J Drug Metab Pharmacokinet 36(1):1-16. 15. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers. 10
ACCEPTED MANUSCRIPT http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/DrugInteractionsLa beling/ucm093664.htm Accessed Feb 3, 2017. 16. Zamek-Gliszczynski MJ, Lee CA, Poirier A, Bentz J, Chu X, Ellens H, Ishikawa T, Jamei M, Kalvass JC, Nagar S, Pang KS, Korzekwa K, Swaan PW, Taub ME, Zhao P, Galetin A, International Transporter C 2013. ITC recommendations for transporter kinetic parameter estimation and translational modeling of transport-mediated PK and DDIs in humans. Clin Pharmacol Ther 94(1):64-79.
RI PT
17. Scotcher D, Jones C, Posada M, Rostami-Hodjegan A, Galetin A 2016. Key to Opening Kidney for In Vitro-In Vivo Extrapolation Entrance in Health and Disease: Part I: In Vitro Systems and Physiological Data. The AAPS journal 18(5):1067-1081. 18. Gertz M, Cartwright CM, Hobbs MJ, Kenworthy KE, Rowland M, Houston JB, Galetin A 2013. Cyclosporine inhibition of hepatic and intestinal CYP3A4, uptake and efflux transporters: application of PBPK modeling in the assessment of drug-drug interaction potential. Pharm Res 30(3):761-780.
SC
19. Shitara Y, Sugiyama Y 2017. Preincubation-dependent and long-lasting inhibition of organic anion transporting polypeptide (OATP) and its impact on drug-drug interactions. Pharmacol Ther.
M AN U
20. Bentz J, O'Connor MP, Bednarczyk D, Coleman J, Lee C, Palm J, Pak YA, Perloff ES, Reyner E, Balimane P, Brannstrom M, Chu X, Funk C, Guo A, Hanna I, Heredi-Szabo K, Hillgren K, Li L, Hollnack-Pusch E, Jamei M, Lin X, Mason AK, Neuhoff S, Patel A, Podila L, Plise E, Rajaraman G, Salphati L, Sands E, Taub ME, Taur JS, Weitz D, Wortelboer HM, Xia CQ, Xiao G, Yabut J, Yamagata T, Zhang L, Ellens H 2013. Variability in P-glycoprotein inhibitory potency (IC(5)(0)) using various in vitro experimental systems: implications for universal digoxin drug-drug interaction risk assessment decision criteria. Drug Metab Dispos 41(7):1347-1366. 21. Vaidyanathan J, Yoshida K, Arya V, Zhang L 2016. Comparing Various In Vitro Prediction Criteria to Assess the Potential of a New Molecular Entity to Inhibit Organic Anion Transporting Polypeptide 1B1. J Clin Pharmacol 56 Suppl 7:S59-72.
TE D
22. Lee CA, Kalvass JC, Galetin A, Zamek-Gliszczynski MJ 2014. ITC commentary on the prediction of digoxin clinical drug-drug interactions from in vitro transporter assays. Clin Pharmacol Ther 96(3):298-301.
EP
23. Vieira ML, Kirby B, Ragueneau-Majlessi I, Galetin A, Chien JY, Einolf HJ, Fahmi OA, Fischer V, Fretland A, Grime K, Hall SD, Higgs R, Plowchalk D, Riley R, Seibert E, Skordos K, Snoeys J, Venkatakrishnan K, Waterhouse T, Obach RS, Berglund EG, Zhang L, Zhao P, Reynolds KS, Huang SM 2014. Evaluation of various static in vitro-in vivo extrapolation models for risk assessment of the CYP3A inhibition potential of an investigational drug. Clin Pharmacol Ther 95(2):189-198.
AC C
24. Galetin A, Gertz M, Houston JB 2010. Contribution of intestinal cytochrome p450-mediated metabolism to drug-drug inhibition and induction interactions. Drug Metab Pharmacokinet 25(1):2847. 25. Tachibana T, Kato M, Watanabe T, Mitsui T, Sugiyama Y 2009. Method for predicting the risk of drug-drug interactions involving inhibition of intestinal CYP3A4 and P-glycoprotein. Xenobiotica 39(6):430-443. 26. Zhang L, Zhang YD, Strong JM, Reynolds KS, Huang SM 2008. A regulatory viewpoint on transporter-based drug interactions. Xenobiotica 38(7-8):709-724. 27. Agarwal S, Arya V, Zhang L 2013. Review of P-gp inhibition data in recently approved new drug applications: utility of the proposed [I(1) ]/IC(50) and [I(2) ]/IC(50) criteria in the P-gp decision tree. J Clin Pharmacol 53(2):228-233. 28. Ellens H, Deng S, Coleman J, Bentz J, Taub ME, Ragueneau-Majlessi I, Chung SP, HerediSzabo K, Neuhoff S, Palm J, Balimane P, Zhang L, Jamei M, Hanna I, O'Connor M, Bednarczyk D, Forsgard M, Chu X, Funk C, Guo A, Hillgren KM, Li L, Pak AY, Perloff ES, Rajaraman G, Salphati L, Taur JS, Weitz D, Wortelboer HM, Xia CQ, Xiao G, Yamagata T, Lee CA 2013. Application of 11
ACCEPTED MANUSCRIPT receiver operating characteristic analysis to refine the prediction of potential digoxin drug interactions. Drug Metab Dispos 41(7):1367-1374. 29. Jones H, Chen Y, Gibson C, Heimbach T, Parrott N, Peters S, Snoeys J, Upreti V, Zheng M, Hall S 2015. Physiologically based pharmacokinetic modeling in drug discovery and development: A pharmaceutical industry perspective. Clin Pharmacol Ther 97(3):247-262.
RI PT
30. Einolf HJ, Chen L, Fahmi OA, Gibson CR, Obach RS, Shebley M, Silva J, Sinz MW, Unadkat JD, Zhang L, Zhao P 2014. Evaluation of various static and dynamic modeling methods to predict clinical CYP3A induction using in vitro CYP3A4 mRNA induction data. Clin Pharmacol Ther 95(2):179-188. 31. Wagner C, Zhao P, Pan Y, Hsu V, Grillo J, Huang SM, Sinha V 2015. Application of Physiologically Based Pharmacokinetic (PBPK) Modeling to Support Dose Selection: Report of an FDA Public Workshop on PBPK. CPT: pharmacometrics & systems pharmacology 4(4):226-230. 32. Galetin A, Zhao P, Huang SM 2017. PK modelling of drug transporters to facilitate individualized dose prediction. J Pharm Sci Accepted.
SC
33. Fahmi OA, Kish M, Boldt S, Obach RS 2010. Cytochrome P450 3A4 mRNA is a more reliable marker than CYP3A4 activity for detecting pregnane X receptor-activated induction of drugmetabolizing enzymes. Drug Metab Dispos 38(9):1605-1611.
M AN U
34. Fahmi OA, Hurst S, Plowchalk D, Cook J, Guo F, Youdim K, Dickins M, Phipps A, Darekar A, Hyland R, Obach RS 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(8):1658-1666. 35. Fahmi OA, Ripp SL 2010. Evaluation of models for predicting drug-drug interactions due to induction. Expert Opin Drug Metab Toxicol 6(11):1399-1416.
TE D
36. Ito K, Iwatsubo T, Kanamitsu S, Ueda K, Suzuki H, Sugiyama Y 1998. Prediction of pharmacokinetic alterations caused by drug-drug interactions: metabolic interaction in the liver. Pharmacol Rev 50(3):387-412. 37. Yoshida K, Maeda K, Sugiyama Y 2012. Transporter-mediated drug--drug interactions involving OATP substrates: predictions based on in vitro inhibition studies. Clin Pharmacol Ther 91(6):1053-1064.
EP
38. Dong Z, Yang X, Arya V, Zhang L 2016. Comparing various in vitro prediction criteria to assess the potential of a new molecular entity (NME) to inhibit Organic Anion Transporter 1 and 3 (OAT1 and 3). Clin Pharmacol Ther 99(S1):S94.
AC C
39. Dong Z, Yang X, Arya V, Zhang L 2016. Comparing various in vitro prediction criteria to assess the potential of a new molecular entity (NME) to inhibit OCT2 and MATE transporters in vivo. Clin Pharmacol Ther 99(S1):S94. 40. Kalvass JC, Polli JW, Bourdet DL, Feng B, Huang SM, Liu X, Smith QR, Zhang LK, ZamekGliszczynski MJ, International Transporter C 2013. Why clinical modulation of efflux transport at the human blood-brain barrier is unlikely: the ITC evidence-based position. Clin Pharmacol Ther 94(1):80-94. 41. Varma MV, Pang KS, Isoherranen N, Zhao P 2015. Dealing with the complex drug-drug interactions: towards mechanistic models. Biopharm Drug Dispos 36(2):71-92.
12
Figure 1
10 2
10 2
10 1
10 0
10 -1
I / Ki or IC50
10 0
Method 1
10 -1 0.02 10 -2
Method 2
TE D
10 -2
10 -5
10 -5
11.25 2
AUCR
EP
10
11.25 2
10 0 0.25
Method 5
0.1
Method 4
0.04
Method 3
10 -4 10
AUCR
10 1
10 -3
10 -4
10 -4
10 2
10 -2
10 -3
10 -3
AC C
I / Ki or IC50
10 1
I / Ki or IC50
10 3
(c) Iu,in,max/Ki or IC50 ≥ 0.04, 0.1, or 0.25
SC
(b) Imax,u/Ki or IC50 ≥ 0.02
M AN U
(a) Imax/Ki or IC50≥0.1
RI PT
ACCEPTED MANUSCRIPT
11.25 2
10
AUCR
S0022-3549(17)30281-2
DOI:
10.1016/j.xphs.2017.04.045
Reference:
XPHS 768
To appear in:
Journal of Pharmaceutical Sciences
Received Date: 7 February 2017 Revised Date:
10 April 2017
Accepted Date: 20 April 2017
Please cite this article as: Yoshida K, Zhao P, Zhang L, Abernethy DR, Rekić D, Reynolds KS, Galetin A, Huang SM, In Vitro-In Vivo Extrapolation of Metabolism- and Transporter-Mediated Drug-Drug Interactions – Overview of Basic Prediction Methods, Journal of Pharmaceutical Sciences (2017), doi: 10.1016/j.xphs.2017.04.045. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
In Vitro-In Vivo Extrapolation of Metabolism- and TransporterMediated Drug-Drug Interactions – Overview of Basic Prediction Methods Kenta Yoshida1,2,3, Ping Zhao1, Lei Zhang1, Darrell R. Abernethy1, Dinko Rekić1, Kellie S Reynolds1, Aleksandra Galetin1,2,4, Shiew-Mei Huang1 1
2
Oak Ridge Institution for Science and Education (ORISE) Fellow
3
RI PT
Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD, USA
Current affiliation: Clinical Pharmacology, Genentech Research and Early Development, 1 DNA Way, South San Francisco, CA, USA
4
SC
Centre for Applied Pharmaceutical Research, School of Health Sciences, University of Manchester, Manchester, UK
Shiew-Mei Huang, PhD
M AN U
Corresponding author:
Office of Clinical Pharmacology, Office of Translational Sciences, Center of Drug Evaluation and Research, U.S. Food and Drug Administration Silver Spring, MD 20993 Phone: 301-796-1541
Figures and Tables: 1 Words in Abstract: 119
TE D
Email: [email protected]
EP
Words in Manuscript: 4632 (including abstract, tables, figures and references)
AC C
Keywords: Drug interaction, Cytochrome P450s, Membrane transport/transporters, Regulatory, Guidance/Guideline
1
ACCEPTED MANUSCRIPT List of abbreviations: ADME: absorption, distribution, metabolism and excretion BCRP: breast cancer resistance protein CYP: cytochrome P-450 d: induction calibration factor DDI: drug-drug interaction
RI PT
EC50: concentration of inducer associated with half-maximum induction EMA: European Medicines Agency Emax: maximum induction effect FaFg: intestinal bioavailability fu,p: fraction unbound in plasma [I]: perpetrator (inhibitor or inducer) concentration
SC
FDA: Food and Drug Administration
Imax: maximum total (unbound plus bound) inhibitor concentration in plasma ITC: International Transporter Consortium
M AN U
Imax,u: maximum unbound inhibitor or inducer concentration
Iu,in,max: theoretical maximum unbound hepatic inlet concentration IVIVE: in vitro-in vivo extrapolation ka: absorption rate constant
kdeg: degradation rate constant of CYP enzyme
TE D
Ki: inhibition constant KI: apparent inactivation constant
kinact: maximal inactivation rate constant
kobs: the apparent inactivation rate constant
MATE: multidrug and toxin extrusion protein
EP
OAT: organic anion transporter
OATPs: organic anion transporting polypeptides
AC C
OCT: organic cation transporter
PET: positron emission tomography P-gp: P-glycoprotein
PK: pharmacokinetic
PMDA: Pharmaceuticals and Medical Devices Agency of Japan Qh: hepatic blood flow rate
RB: blood-to-plasma concentration ratio RIS: relative induction score UGT: UDP-glucuronosyltransferase
2
ACCEPTED MANUSCRIPT
Abstract
RI PT
Evaluation of drug-drug interaction (DDI) risk is vital to establish benefit-risk profiles of investigational new drugs during drug development. In vitro experiments are routinely conducted as an important first step to assess metabolism- and/or transporter-mediated DDI potential of investigational new drugs. Results from these experiments are interpreted, often with the aid of in vitro-in vivo extrapolation (IVIVE) methods, to determine whether and how DDI should be evaluated clinically to provide the basis for proper DDI management strategies, including dosing recommendations, alternative therapies, or contraindications under various DDI scenarios and in different patient population. This article provides an overview of currently available in vitro experimental systems and basic IVIVE methodologies for metabolism- and transporter-mediated DDIs.
SC
1 Background
TE D
M AN U
Many factors can affect pharmacokinetic (PK) profiles of drugs after administration to different patients, including concomitant medications, disease status, and genetic polymorphisms. For optimal use of medications, it is important to understand which factors are relevant to the pharmacokinetics of each drug. DDIs are one of the important factors that alter the pharmacokinetics of drugs, and they have in some cases resulted in fatal adverse events leading to withdrawal of drugs from the market1. Because of their importance, the U.S. Food and Drug Administration (FDA), European Medicines Agency (EMA), and Ministry of Health, Labour and Welfare of Japan (MHLW) have published a draft guidance or guideline (FDA and MHLW) or a final guideline (EMA) on methodologies to evaluate DDIs for investigational new drugs during drug development2-4. The evaluation of DDIs generally involves multiple processes, including 1) prediction of clinical DDIs with preclinical or in vitro systems, 2) evaluation/confirmation of the extent of DDIs with clinical studies, and/or 3) modelbased simulations of DDIs. This review article focuses on the first element, prediction of DDI using data from preclinical and in vitro systems, especially focusing on basic IVIVE methodologies.
EP
2 Clinically relevant metabolizing enzymes and transporters for DDI evaluation
AC C
The most well-known and important proteins involved in PK-based DDIs are enzymes of the cytochrome P-450 (CYP) family. The inhibition or induction of CYP enzymes can directly affect the exposures to substrate drugs and result in unexpected changes in efficacy and/or safety profiles without appropriate dose adjustment. Many of CYP enzymes have broad substrate specificity and they mediate the metabolism of the majority of marketed small molecule drugs5. Clinically important CYP enzymes relevant for PK-based DDIs include CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, and CYP3A. Many of these CYP enzymes are predominantly expressed in the liver. CYP3A is also expressed significantly in the intestine and it can contribute to the intestinal first-pass extraction of orally administered drugs. Other drug metabolizing enzymes, such as UDPglucuronosyltransferase (UGT), are less frequently evaluated for potential PK-based DDIs, mainly due to the lack of selective inhibitor and inducer drugs that can significantly influence PK of their substrates. In contrast to drug metabolizing enzymes, the alteration of transporter function can not only affect drug elimination, but also drug distribution. For example, inhibition of organic anion transporting polypeptides (OATPs) by rifampin significantly reduces radioactivity measured in the 3
ACCEPTED MANUSCRIPT
3 In vitro experimental systems
RI PT
liver with positron emission tomography (PET) after the administration of (15R)-11C-TIC-Me in humans6 or rosuvastatin in rats7. A list of drug transporters that may alter drug disposition has been recommended by the International Transporter Consortium (ITC)8-11. Drug transporters that may be involved in clinically significant DDIs include OATP1B1/OATP1B3, organic anion transporter 1/3 (OAT1/OAT3), organic cation transporter 2 (OCT2), breast cancer resistance protein (BCRP), Pglycoprotein (P-gp), and multidrug and toxin extrusion protein 1/2-K (MATE1/MATE2-K). The regulatory agencies from the U.S., Europe, and Japan adopted this recommended list in their recently published DDI guidance/guidelines2-4.
M AN U
SC
The gold standard for preclinical DDI evaluation has been the use of enzyme- or transporterexpressing systems or human cellular and subcessular fractions. Because of species differences in metabolizing enzymes and transporters between human and animals, results of DDI studies in animals are not frequently used to predict clinical DDIs. In general, the purpose of using human expression systems is to evaluate specific interaction of one particular human protein with the drug of interest. Use of holistic cellular systems (e.g., hepatocytes) allows characterization of the relative contribution of individual metabolizing enzymes or transporters to the drug clearance in the presence of all other proteins. In this section, we briefly explain available in vitro experimental systems for DDI evaluation, focusing predominantly on the inhibition type interactions. More detailed descriptions of individual experimental systems can be found in other review articles12,13.
TE D
Human liver microsomes and human hepatocytes are the most commonly utilized human in vitro systems to evaluate metabolism-mediated DDI. In many cases, in vivo metabolism pathways of a drug can be effectively evaluated with these systems. Probe substrates, inhibitors and inducers for individual CYP enzymes have been well established 14,15 allowing the evaluation of DDI potential for an investigational drug with these systems with the drug acting either as a substrate or as a perpetrator. Microsomes prepared from cells overexpressing a specific isoform of enzyme (e.g., recombinant CYP enzymes) are useful to determine the involvement of a specific enzyme to the metabolism of the investigational drug and can be applied for evaluation of DDIs (hepatic and intestinal).
AC C
EP
Compared to metabolism-mediated DDIs, cell lines transftected with individual transporters (e.g., HEK-OATP1B1) are used more frequently for the evaluation of transporter-mediated DDIs. The system of choice depends on questions of interest, function and directional characteristics of transporters13,16,17. Transfected cell lines expressing individual transporters of interest, coupled with probe substrates, are useful for studying a drug as a potential transporter inhibitor. When evaluating the drug as a potential substrate, these systems address a “yes/no” question to determine the involvement of a particular transporter. Human hepatocytes express multiple transporters and are useful tools to determine quantitative contributions of a transporter, investigate rate limiting processes, and complex metabolism-transporter interplay13. Certain methodological aspects of transporter in vitro inhibition studies are still challenging and may affect ability to predict transporter-mediated DDI risk. For example, selective probe substrates and inhibitors for each transporter have not been well established. Occurrence of substratedependent inhibition, associated with possible multiple binding sites, highlights the need to use relevant substrate co-medications/clinical probe in the in vitro inhibition studies16. Some OATP inhibitors (e.g., cyclosporine) have shown increased inhibitory potency following a pre-incubation step 18,19. In addition, quantitative estimation of inhibition parameters may require mathematical modeling of in vitro data. This is of particular relevance for the efflux transporters, as the interaction 4
ACCEPTED MANUSCRIPT of these transporters with the unbound intracellular drug concentration is considered in the model, rather than the nominal incubation concentration16. Furthermore, large inter-laboratory variations have been reported for in vitro inhibition parameters (IC50 values) for certain transporters (e.g, OATP1B1 and P-gp) 20,21. This may require additional effort for successful IVIVEs, such as standardization of in vitro methods or calibration of cellular systems using a standard set of model inhibitors, as discussed in digoxin case22 to generate “customized” decision criteria for DDI prediction.
4 In vitro-in vivo extrapolation of DDI
SC
RI PT
The next step of DDI evaluation is the extrapolation of in vitro parameters to in vivo interaction predictions. For a new drug as a substrate, knowing the contribution of a specific metabolizing enzyme or transporter to the overall absorption, distribution, metabolism and excretion (ADME) processes, based on IVIVE and clinical PK data, is required for an effective evaluation of clinical DDI potential. For example, in the case of metabolism-based DDI, it is recommended to conduct a clinical DDI study with strong index inhibitors and/or inducers if an enzyme of interest is responsible for ≥25% of the drug’s elimination based on in vitro phenotyping and human PK data.
Prediction of metabolism-mediated DDI
TE D
4.1
M AN U
For a new drug as a perpetrator, the first step of extrapolation is often a simple comparison of in vitro parameters with in vivo exposure to the drug (also known as a basic model). If a basic model shows the potential of clinical DDI according to a predefined cut-off criterion, the interaction potential may be usefully investigated via a clinical DDI study or with the use of mechanistic models such as physiologically-based pharmacokinetic (PBPK) models2-4. In the following sections, we provide an overview of the basic models for evaluating metabolism- and transporter-mediated DDIs involving a new drug as a perpetrator.
AC C
EP
The essential components of any metabolism- and transporter-mediated DDI predictions are the estimation of perpetrator concentration at the site of interaction and the in vitro potencies of the perpetrator (defined by inhibition constant, Ki). A basic model uses a constant value of perpetrator concentration to simplify prediction calculations. In many cases, the maximum steady-state concentration of the perpetrator at the highest clinical dose is used to minimize the likelihood of false negative prediction during this first step of evaluation. Basic models include other simplifications, such as assumption of 100% involvement of the pathway of interest, no consideration of the interplay of multiple enzymes or transporters and use of plasma concentration as a surrogate of the perpetrator concentration at the site of interaction16. The majority of metabolism-mediated DDI predictions focus on CYP-mediated interactions due to their clinical significance. Prediction methods for other enzymes such as UGTs have not been well established. For reversible inhibition, the basic model uses a simple comparison of the inhibitor concentration and the Ki for inhibitor-enzyme of interest. There are differences in the recommendations by regulatory agencies on inhibitor concentrations and cut-off values to use: e.g., Imax/Ki ≥ 0.1 (FDA recommendation in 2012 draft guidance2 and MHLW recommendation in 2014 draft guideline4) and Imax,u/Ki ≥ 0.02 (EMA recommendation in 2013 guideline3), where Imax and Imax,u represent the maximum total (unbound plus bound) and unbound inhibitor concentrations in plasma, respectively. A recent analysis suggested that these two criteria performed similarly in predicting reported DDIs23. The above criteria are intended for the prediction of DDIs after drug reaches systemic circulation. The concentration of a potential inhibitor in the intestine following oral administration can 5
ACCEPTED MANUSCRIPT
RI PT
be much higher than the systemic concentration24. Therefore, the prediction based on systemic concentration of inhibitor can underestimate interaction potential for orally administered drugs. The underestimation is especially of concern for substrates of CYP3A enzymes (and transporters such as P-gp or BCRP). Using dose/250mL as a surrogate for intestinal inhibitor concentration, a cut-off criterion of 10 for the (dose/250mL)/Ki was proposed by regulatory agencies2-4. This criterion was deemed appropriate based on analyses where performance of in vivo DDI prediction was assessed for CYP3A and P-gp using criteria similar to those in the guidances23,25-27. The prediction criteria were recently evaluated using receiver operating characteristic analysis resulting in a higher intestinal [I2]/IC50 cut-off, while tightening the criterion for systemic P-gp inhibition 28.
SC
Prediction of the time-dependent inhibition (TDI) is more challenging than the reversible inhibition. The inherent difficulty is that dynamic changes in enzyme function due to inactivation are approximated with a simple, static equation. The parameters for basic models include degradation rate constant of CYP enzyme (kdeg), apparent inactivation constant (KI), and maximal inactivation rate constant (kinact). There are differences in the concentrations and cut-off criteria used for predictions among regulatory agencies. The following two equations are used for the basic model prediction of TDI of hepatic CYP enzymes: (kobs+kdeg)/kdeg ≥ 1.1 where kobs=kinact×Imax /(KI + Imax) (FDA 2012 draft guidance and MHLW 2014 draft guideline 2,4)
•
(kobs+kdeg)/kdeg ≥ 1.25 where kobs=kinact×50×Imax,u /(KI + 50×Imax,u) (EMA 2013 guideline3)
M AN U
•
TE D
where kobs represents the apparent inactivation rate constant. A retrospective analysis demonstrated that neither equation resulted in any false-negative predictions23. However, both equations led to positive predictions of inhibition for the majority of tested drugs, regardless of observed clinical DDI outcomes. Because of the time-dependent nature of TDI, a dynamic model-based approach provides a more reliable prediction of the extent of DDI with TDI. However, recent analyses highlight challenges in the prediction of complex metabolic DDIs, e.g., combined TDI and induction interactions, even when using mechanistic PBPK models29,30. Furthermore, confidence in predicting complex transporter-metabolism interplay and associated DDIs is still generally low29,31,32.
EP
Several basic methods have been proposed for the prediction of enzyme induction30. Proposed approaches with their respective threshold criteria include: 1) Fold-change method: mRNA increase ≥ a predefined threshold based on positive/negative controls and exceeds a certain % in comparison to positive controls. For example, EMA recommends a concentration-dependent mRNA increase of ≥100% (2-fold) and ≥20% of the response of the positive control in the presence of an investigational drug, at relevant concentrations stipulated in the guidance, as positive induction (EMA 2013 guideline3);
AC C
33
2) Basic model: R3=1/(1+d×Emax×[I]/(EC50 +[I])) ≤ 0.8 as positive induction, where EC50 is the concentration of inducer associated with half-maximum induction, Emax is the maximum induction 34 effect, [I] is the inducer concentration, and d is the induction calibration factor ; 3) Correlation method using Imax,u / EC50: Calculate Imax,u / EC50 values with cut-off values defined with known inducers and non-inducers; and 4) Correlation method using relative induction score, RIS: calibration of RIS (= Emax×[I]/(EC50 +[I])) with known inducers35. Each regulatory guidance includes one or more of the above four methods for induction predictions. In addition to mRNA expression levels, induction can also be studied by assessing enzyme activities, with the caveat that induction may be masked by concurrent inhibition. 6
ACCEPTED MANUSCRIPT 4.2
Prediction of transporter-mediated DDI
Basic models are the main prediction method used to extrapolate in vitro assessment of investigational drugs as potential in vivo transporter inhibitors. Because different transporters are expressed in different organs and different side of membranes, the accurate estimation of inhibitor concentrations at the site of inhibition is key for accurate predictions.
SC
Iu,in,max = fu,p × (Imax + ka × Dose × FaFg/Qh/RB),
RI PT
For OATP1B1 and OATP1B3, which are expressed at the basolateral membrane of hepatocytes, the basic prediction method uses the unbound hepatic inlet plasma concentration as the inhibitor concentrations. It is important to note that hepatic plasma concentration can be higher than systemic concentration after oral drug administration, because drugs absorbed from small intestine first reach the liver via portal vein. In order to account for the contribution of drug concentration from absorption, the following equation was proposed to estimate the theoretical maximum unbound hepatic inlet plasma concentration, Iu,in,max36:
TE D
M AN U
where ka, FaFg, fu,p, Qh, and RB represent absorption rate constant, intestinal bioavailability, fraction unbound in plasma, hepatic blood flow rate, and blood-to-plasma concentration ratio, respectively. Prediction accuracies of OATP-mediated DDIs using different inhibitor concentrations and different “cut-off” criteria have been compared in recent studies 21,37 (Fig 1). In the study by Vaidyanathan et al, Iu,in,max/Ki ≥ 0.1, Imax,u/Ki ≥ 0.02, and Iu,in,max/Ki ≥ 0.25 alone (MHLW recommendation in 2014 draft guideline4), or combined criterion Imax/Ki ≥ 0.1 and Iu,in,max/Ki ≥ 0.25 (FDA recommendation in 2012 draft guidance2) were all considered reasonable to determine the need to consider conducting clinical DDI studies. Although there were small differences in numbers of false predictions among these criteria, it is important to note that different criteria led to different types of false predictions. For example, prediction using maximum plasma total (bound + unbound) concentration in the systemic circulation tends to cause false-positive predictions of highly protein-bound drugs. Therefore, it is important to consider differences between methods when interpreting prediction results, as described in detail in the publication by Vaidyanathan et al21.
AC C
EP
For renal transporters, OAT1/3, OCT2, and MATE1/2-K, proposed basic prediction methods use unbound maximum systemic concentrations of inhibitors. Compared to CYP enzymes or OATPs, there are fewer cases of renal transporter-mediated DDIs, and the degree of change in systemic concentration is relatively small, leading to a limited dataset for validation of basic prediction methods for renal transporters. Furthermore, OCT2 and MATEs often share the same substrates and inhibitors, therefore the net DDI effect may be affected by both transporters at the basolateral and apical side of the renal tubule cells. Thus, an accurate prediction of DDI is more challenging for substrates for OCT2 and MATEs and inhibition of both transporters may need to be considered simultaneously. It is also important to note that inhibition of MATEs may cause an increase of intracellular concentrations, which would not be reflected in the systemic exposure of substrate drugs32. According to studies by Dong et al38,39, for OAT1/3-mediated DDIs, two of the recommended criteria showed comparable prediction outcomes: Imax,u/Ki ≥ 0.1 (FDA recommendation in 2012 draft guidance2) and Imax,u/Ki ≥ 0.25 (MHLW recommendation in 2014 draft guideline4). For OCT2 and MATEs, they proposed to use a lower cut-off for MATEs (i.e., Imax,u/Ki ≥ 0.02) than for OCT2 (i.e., 0.1, the same as that for OATs) because a lower cut-off appeared to avoid false negative prediction. Using a lower cut-off for MATEs also accounts for possible higher intracellular concentration of the inhibitor than its systemic concentration in the situation where the inhibitor is a substrate of an uptake transporter such as OCT2. Overall, because of the limited number of reported
7
ACCEPTED MANUSCRIPT DDIs and complex transporter interplays, the criteria for renal transporter-mediated DDIs need further evaluation.
SC
RI PT
P-gp and BCRP are expressed at various organs, but the main focus of DDI prediction for these transporters is on their inhibition in the intestinal lumen. One reason is that changes in oral bioavailability of substrate drugs by the inhibition of intestinal transporters can be reflected in the absorption profile of substrate drugs (e.g., Cmax change). Another reason is that the inhibitor concentration is in general much higher in the intestine than in systemic circulation after oral dosing, thus more prone to the perpetrator effect18. In fact, modulation of these transporters at the systemic level is considered unlikely in clinical situations, because unbound systemic concentrations of marketed drugs are generally too low to exhibit significant inhibition40. For intestinal transportermediated DDI prediction, the same cut-off criterion as intestinal CYP3A (dose/250mL/Ki ≥10) is proposed by regulatory agencies2-4, as described earlier in the section 4.1.
5 Conclusions
EP
TE D
M AN U
In vitro experimental systems and basic prediction models have evolved as useful tools to evaluate the risk of metabolism- and transporter-mediated DDIs for new molecular entities. Depending on the site or mechanisms of interactions, various equations and decision criteria have been proposed for DDI predictions. Considering the importance of this first-step of DDI prediction, there has been an ongoing effort among FDA, EMA, and MHLW to harmonize the recommendations for DDI prediction involving both metabolizing enzymes and transporters23. It is important to note, however, that the focus of these simple prediction methods is to eliminate the need to further evaluate DDIs that are unlikely to cause significant PK changes by using conservative assumptions to minimize falsenegative predictions. For a more accurate and quantitative prediction of DDIs based on in vitro observations, use of sophisticated model-based predictions such as PBPK modeling that incorporate mechanistic considerations of victim and perpetrator drug disposition should be explored, especially for DDIs involving multiple elimination pathways18,41.
Acknowledgements
AC C
This research was supported in part by the US Food and Drug Administration’s (FDA) Medical Countermeasures initiative. Drs. Aleksandra Galetin and Kenta Yoshida were supported in part by an appointment to the Research Participation Program at the Center for Drug Evaluation and Research, administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the FDA.
Conflict of interest/disclosure The authors declared no conflicts of interest. The contents of this article reflect the views of the authors and should not be construed to represent the FDA's views or policies. No official support or endorsement by the FDA is intended or should be inferred. The mention of commercial products, their sources, or their use in connection with material reported herein is not to be construed as either an actual or implied endorsement of such products by the FDA.
8
ACCEPTED MANUSCRIPT
Figure Legend Figure 1. Relationships between observed AUCR and different concentration of inhibitors divided by Ki or IC50. (a) Imax/ Ki or IC50, (b) Imax,u/ Ki or IC50, and (c) Iu,in,max/ Ki or IC50.
RI PT
Each symbol represents one DDI case. Horizontal lines represent cut-off values for 5 basic prediction methods evaluated in the original study21: (1) Imax/ Ki or IC50 ≥ 0.1, (2) Imax,u/ Ki or IC50≥0.02, (3) R (= 1 + Iu,in,max / Ki or IC50) ≥1.04 (EMA 3), (4) R (= 1 + Iu,in,max / Ki or IC50) ≥1.1, (5) R (= 1 + Iu,in,max / Ki or IC50) ≥1.25 (MHLW 2014 draft 4), and (6) Imax/ Ki or IC50 ≥0.1 AND R ≥ 1.25 (combination of (1) and (5); FDA 2012 draft 2). AUCR: area under the concentration-curve ratio, DDI: drug-drug
SC
interactions, IC50: half maximal inhibitory concentration, Imax: maximum systemic plasma concentration, Imax,u: maximum unbound systemic plasma concentration, Iu,in,max: estimated maximum
M AN U
inhibitor concentration at the inlet to the liver, Ki: inhibition constant, and OATP: organic anion
AC C
EP
TE D
transporting polypeptide. Figure from Vaidyanathan, et al., 2016 21.
9
ACCEPTED MANUSCRIPT
References 1. Giacomini KM, Krauss RM, Roden DM, Eichelbaum M, Hayden MR, Nakamura Y 2007. When good drugs go bad. Nature 446(7139):975-977.
RI PT
2. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER) 2012. Draft Guidance / Guidance for Industry. Drug Interaction Studies--Study Design, Data Analysis, Implications for Dosing, and Labeling Recommendations. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM2 92362.pdf Accessed Mar 3, 2017. 3. European Medicines Agency 2013. Guideline on the Investigation of Drug Interactions. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2012/07/WC50012960 6.pdf Accessed Mar 3, 2017.
SC
4. Ministry of Health, Labour and Welfare, Japan 2014. Guideline of drug interaction studies for drug development and appropriate provision of information. https://www.pmda.go.jp/files/000206158.pdf Accessed Mar 23, 2017.
M AN U
5. Wienkers LC, Heath TG 2005. Predicting in vivo drug interactions from in vitro drug discovery data. Nature reviews Drug discovery 4(10):825-833. 6. Takashima T, Kitamura S, Wada Y, Tanaka M, Shigihara Y, Ishii H, Ijuin R, Shiomi S, Nakae T, Watanabe Y, Cui Y, Doi H, Suzuki M, Maeda K, Kusuhara H, Sugiyama Y, Watanabe Y 2012. PET imaging-based evaluation of hepatobiliary transport in humans with (15R)-11C-TIC-Me. J Nucl Med 53(5):741-748. 7. He J, Yu Y, Prasad B, Link J, Miyaoka RS, Chen X, Unadkat JD 2014. PET imaging of Oatpmediated hepatobiliary transport of [(11)C] rosuvastatin in the rat. Mol Pharm 11(8):2745-2754.
TE D
8. Giacomini KM, Huang SM, Tweedie DJ, Benet LZ, Brouwer KL, Chu X, Dahlin A, Evers R, Fischer V, Hillgren KM, Hoffmaster KA, Ishikawa T, Keppler D, Kim RB, Lee CA, Niemi M, Polli JW, Sugiyama Y, Swaan PW, Ware JA, Wright SH, Yee SW, Zamek-Gliszczynski MJ, Zhang L 2010. Membrane transporters in drug development. Nature reviews Drug discovery 9(3):215-236.
EP
9. Hillgren KM, Keppler D, Zur AA, Giacomini KM, Stieger B, Cass CE, Zhang L, International Transporter C 2013. Emerging transporters of clinical importance: an update from the International Transporter Consortium. Clin Pharmacol Ther 94(1):52-63.
AC C
10. Huang SM, Zhang L, Giacomini KM 2010. The International Transporter Consortium: a collaborative group of scientists from academia, industry, and the FDA. Clin Pharmacol Ther 87(1):32-36. 11. Giacomini KM, Huang SM 2013. Transporters in drug development and clinical pharmacology. Clin Pharmacol Ther 94(1):3-9. 12. Lu C, Liao M, Cohen L, Xia CQ 2010. Emerging in vitro tools to evaluate cytochrome P450 and transporter-mediated drug-drug interactions. Curr Drug Discov Technol 7(3):199-222. 13. Brouwer KL, Keppler D, Hoffmaster KA, Bow DA, Cheng Y, Lai Y, Palm JE, Stieger B, Evers R, International Transporter C 2013. In vitro methods to support transporter evaluation in drug discovery and development. Clin Pharmacol Ther 94(1):95-112. 14. Khojasteh SC, Prabhu S, Kenny JR, Halladay JS, Lu AY 2011. Chemical inhibitors of cytochrome P450 isoforms in human liver microsomes: a re-evaluation of P450 isoform selectivity. Eur J Drug Metab Pharmacokinet 36(1):1-16. 15. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers. 10
ACCEPTED MANUSCRIPT http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/DrugInteractionsLa beling/ucm093664.htm Accessed Feb 3, 2017. 16. Zamek-Gliszczynski MJ, Lee CA, Poirier A, Bentz J, Chu X, Ellens H, Ishikawa T, Jamei M, Kalvass JC, Nagar S, Pang KS, Korzekwa K, Swaan PW, Taub ME, Zhao P, Galetin A, International Transporter C 2013. ITC recommendations for transporter kinetic parameter estimation and translational modeling of transport-mediated PK and DDIs in humans. Clin Pharmacol Ther 94(1):64-79.
RI PT
17. Scotcher D, Jones C, Posada M, Rostami-Hodjegan A, Galetin A 2016. Key to Opening Kidney for In Vitro-In Vivo Extrapolation Entrance in Health and Disease: Part I: In Vitro Systems and Physiological Data. The AAPS journal 18(5):1067-1081. 18. Gertz M, Cartwright CM, Hobbs MJ, Kenworthy KE, Rowland M, Houston JB, Galetin A 2013. Cyclosporine inhibition of hepatic and intestinal CYP3A4, uptake and efflux transporters: application of PBPK modeling in the assessment of drug-drug interaction potential. Pharm Res 30(3):761-780.
SC
19. Shitara Y, Sugiyama Y 2017. Preincubation-dependent and long-lasting inhibition of organic anion transporting polypeptide (OATP) and its impact on drug-drug interactions. Pharmacol Ther.
M AN U
20. Bentz J, O'Connor MP, Bednarczyk D, Coleman J, Lee C, Palm J, Pak YA, Perloff ES, Reyner E, Balimane P, Brannstrom M, Chu X, Funk C, Guo A, Hanna I, Heredi-Szabo K, Hillgren K, Li L, Hollnack-Pusch E, Jamei M, Lin X, Mason AK, Neuhoff S, Patel A, Podila L, Plise E, Rajaraman G, Salphati L, Sands E, Taub ME, Taur JS, Weitz D, Wortelboer HM, Xia CQ, Xiao G, Yabut J, Yamagata T, Zhang L, Ellens H 2013. Variability in P-glycoprotein inhibitory potency (IC(5)(0)) using various in vitro experimental systems: implications for universal digoxin drug-drug interaction risk assessment decision criteria. Drug Metab Dispos 41(7):1347-1366. 21. Vaidyanathan J, Yoshida K, Arya V, Zhang L 2016. Comparing Various In Vitro Prediction Criteria to Assess the Potential of a New Molecular Entity to Inhibit Organic Anion Transporting Polypeptide 1B1. J Clin Pharmacol 56 Suppl 7:S59-72.
TE D
22. Lee CA, Kalvass JC, Galetin A, Zamek-Gliszczynski MJ 2014. ITC commentary on the prediction of digoxin clinical drug-drug interactions from in vitro transporter assays. Clin Pharmacol Ther 96(3):298-301.
EP
23. Vieira ML, Kirby B, Ragueneau-Majlessi I, Galetin A, Chien JY, Einolf HJ, Fahmi OA, Fischer V, Fretland A, Grime K, Hall SD, Higgs R, Plowchalk D, Riley R, Seibert E, Skordos K, Snoeys J, Venkatakrishnan K, Waterhouse T, Obach RS, Berglund EG, Zhang L, Zhao P, Reynolds KS, Huang SM 2014. Evaluation of various static in vitro-in vivo extrapolation models for risk assessment of the CYP3A inhibition potential of an investigational drug. Clin Pharmacol Ther 95(2):189-198.
AC C
24. Galetin A, Gertz M, Houston JB 2010. Contribution of intestinal cytochrome p450-mediated metabolism to drug-drug inhibition and induction interactions. Drug Metab Pharmacokinet 25(1):2847. 25. Tachibana T, Kato M, Watanabe T, Mitsui T, Sugiyama Y 2009. Method for predicting the risk of drug-drug interactions involving inhibition of intestinal CYP3A4 and P-glycoprotein. Xenobiotica 39(6):430-443. 26. Zhang L, Zhang YD, Strong JM, Reynolds KS, Huang SM 2008. A regulatory viewpoint on transporter-based drug interactions. Xenobiotica 38(7-8):709-724. 27. Agarwal S, Arya V, Zhang L 2013. Review of P-gp inhibition data in recently approved new drug applications: utility of the proposed [I(1) ]/IC(50) and [I(2) ]/IC(50) criteria in the P-gp decision tree. J Clin Pharmacol 53(2):228-233. 28. Ellens H, Deng S, Coleman J, Bentz J, Taub ME, Ragueneau-Majlessi I, Chung SP, HerediSzabo K, Neuhoff S, Palm J, Balimane P, Zhang L, Jamei M, Hanna I, O'Connor M, Bednarczyk D, Forsgard M, Chu X, Funk C, Guo A, Hillgren KM, Li L, Pak AY, Perloff ES, Rajaraman G, Salphati L, Taur JS, Weitz D, Wortelboer HM, Xia CQ, Xiao G, Yamagata T, Lee CA 2013. Application of 11
ACCEPTED MANUSCRIPT receiver operating characteristic analysis to refine the prediction of potential digoxin drug interactions. Drug Metab Dispos 41(7):1367-1374. 29. Jones H, Chen Y, Gibson C, Heimbach T, Parrott N, Peters S, Snoeys J, Upreti V, Zheng M, Hall S 2015. Physiologically based pharmacokinetic modeling in drug discovery and development: A pharmaceutical industry perspective. Clin Pharmacol Ther 97(3):247-262.
RI PT
30. Einolf HJ, Chen L, Fahmi OA, Gibson CR, Obach RS, Shebley M, Silva J, Sinz MW, Unadkat JD, Zhang L, Zhao P 2014. Evaluation of various static and dynamic modeling methods to predict clinical CYP3A induction using in vitro CYP3A4 mRNA induction data. Clin Pharmacol Ther 95(2):179-188. 31. Wagner C, Zhao P, Pan Y, Hsu V, Grillo J, Huang SM, Sinha V 2015. Application of Physiologically Based Pharmacokinetic (PBPK) Modeling to Support Dose Selection: Report of an FDA Public Workshop on PBPK. CPT: pharmacometrics & systems pharmacology 4(4):226-230. 32. Galetin A, Zhao P, Huang SM 2017. PK modelling of drug transporters to facilitate individualized dose prediction. J Pharm Sci Accepted.
SC
33. Fahmi OA, Kish M, Boldt S, Obach RS 2010. Cytochrome P450 3A4 mRNA is a more reliable marker than CYP3A4 activity for detecting pregnane X receptor-activated induction of drugmetabolizing enzymes. Drug Metab Dispos 38(9):1605-1611.
M AN U
34. Fahmi OA, Hurst S, Plowchalk D, Cook J, Guo F, Youdim K, Dickins M, Phipps A, Darekar A, Hyland R, Obach RS 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(8):1658-1666. 35. Fahmi OA, Ripp SL 2010. Evaluation of models for predicting drug-drug interactions due to induction. Expert Opin Drug Metab Toxicol 6(11):1399-1416.
TE D
36. Ito K, Iwatsubo T, Kanamitsu S, Ueda K, Suzuki H, Sugiyama Y 1998. Prediction of pharmacokinetic alterations caused by drug-drug interactions: metabolic interaction in the liver. Pharmacol Rev 50(3):387-412. 37. Yoshida K, Maeda K, Sugiyama Y 2012. Transporter-mediated drug--drug interactions involving OATP substrates: predictions based on in vitro inhibition studies. Clin Pharmacol Ther 91(6):1053-1064.
EP
38. Dong Z, Yang X, Arya V, Zhang L 2016. Comparing various in vitro prediction criteria to assess the potential of a new molecular entity (NME) to inhibit Organic Anion Transporter 1 and 3 (OAT1 and 3). Clin Pharmacol Ther 99(S1):S94.
AC C
39. Dong Z, Yang X, Arya V, Zhang L 2016. Comparing various in vitro prediction criteria to assess the potential of a new molecular entity (NME) to inhibit OCT2 and MATE transporters in vivo. Clin Pharmacol Ther 99(S1):S94. 40. Kalvass JC, Polli JW, Bourdet DL, Feng B, Huang SM, Liu X, Smith QR, Zhang LK, ZamekGliszczynski MJ, International Transporter C 2013. Why clinical modulation of efflux transport at the human blood-brain barrier is unlikely: the ITC evidence-based position. Clin Pharmacol Ther 94(1):80-94. 41. Varma MV, Pang KS, Isoherranen N, Zhao P 2015. Dealing with the complex drug-drug interactions: towards mechanistic models. Biopharm Drug Dispos 36(2):71-92.
12
Figure 1
10 2
10 2
10 1
10 0
10 -1
I / Ki or IC50
10 0
Method 1
10 -1 0.02 10 -2
Method 2
TE D
10 -2
10 -5
10 -5
11.25 2
AUCR
EP
10
11.25 2
10 0 0.25
Method 5
0.1
Method 4
0.04
Method 3
10 -4 10
AUCR
10 1
10 -3
10 -4
10 -4
10 2
10 -2
10 -3
10 -3
AC C
I / Ki or IC50
10 1
I / Ki or IC50
10 3
(c) Iu,in,max/Ki or IC50 ≥ 0.04, 0.1, or 0.25
SC
(b) Imax,u/Ki or IC50 ≥ 0.02
M AN U
(a) Imax/Ki or IC50≥0.1
RI PT
ACCEPTED MANUSCRIPT
11.25 2
10
AUCR