Drug interaction studies on new drug applications (NDAs) for new molecular entities (NMEs) approved in Japan between 1997 and 2008 are examined in the Pharmaceuticals and Medical Devices Agency (PMDA). The situations of drug interaction studies in NDAs have changed over the past 12 years, especially in metabolizing enzyme and transporter-based drug interactions. Materials and approaches to study drug-metabolizing enzyme-based drug interactions have improved, and become more rational based on mechanistic theory and new technologies. On the basis of incremental evidence of transporter roles in human pharmacokinetics, transporter-based drug interactions have been increasingly studied during drug development and submitted in recent NDAs. Some recently approved NMEs include transporter-based drug interaction information in their package inserts (PIs). The regulatory document “Methods of Drug Interaction Studies,” in addition to recent advances in science and technology, has also contributed to plan and evaluation of drug interaction studies in recent new drug development. This review summarizes current situations and further discussion points on drug interaction studies in NDAs in Japan.
The ability of a drug to cause clinically significant drug-drug interactions due to direct or metabolism-dependent inhibition of cytochrome P450 (CYP) can generally be predicted from in vitro studies with human liver microsomes (HLM) or recombinant CYP enzymes, as recommended by the FDA and other regulatory agencies. This review highlights some examples of system-dependent inhibition of CYP and uridine diphosphate glucuronosyltransferase (UGT) enzymes. In the case of CYP enzymes, examples are presented where in vitro studies with HLM under-predict or over-predict the degree of inhibition observed in the clinic and where the correct prediction comes from studies with human hepatocytes. Studies with HLM under-predict the ability of gemfibrozil and bupropion to cause clinically significant inhibition of CYP2C8 and CYP2D6, respectively, and over-predict the ability of ezetimibe to cause clinically significant inhibition of CYP3A4. Gemfibrozil and bupropion represent examples of glucuronidation-dependent and reduction-dependent activation to metabolites that inhibit CYP2C8 and CYP2D6, respectively, whereas ezetimibe represents an example of glucuronidation-dependent protection against metabolism-dependent inhibition of CYP3A4. This article illustrates why, when drug candidates are extensively metabolized by non-CYP enzymes, it would be prudent to use human hepatocytes in addition to HLM or recombinant enzymes to evaluate their ability to inhibit CYP enzymes.
Current drug-drug interaction (DDI) prediction models incorporate intestinal interaction as the ratio of the intestinal availability in the presence and absence of an inhibitor/inducer (FG′ and FG, respectively). The incorporation of the gut is commonly associated with a reduced number of false negative predictions; however, in some instances a trend for over-prediction is apparent. This differential success is highly dependent on the initial model assumptions and parameter estimates used (often inconsistent between the datasets) and cannot be associated exclusively with the incorporation of the intestine. The current review provides an assessment of the contribution of intestinal inhibition and induction in conjunction with different perpetrator and victim drug-related properties, focusing in particular on victim drugs with high intestinal first-pass extraction (>75%). Recommendations are given in order to avoid significant over-estimation of true positives and increased number of false positive predictions. This review discusses advantages and limitations of different in vitro and in vivo methods for assessing intestinal availability and associated inter-individual variability, due to the sensitivity of the DDI prediction models to the FG.
The prediction of drug-drug interactions (DDIs) associated with change in clearance for metabolism is reviewed, particularly focusing on pharmacokinetic theories for prediction based on in vitro and in vivo observation. First, there is discussion about how quantitative determination of the contribution of major clearance pathways is fundamental for the accurate prediction of DDIs. Secondly, the concentrations of causative drugs at sites of interactions are discussed. Although DDIs have been predicted from in vitro pharmacokinetic parameters based on predicted hepatic unbound concentrations of inhibitors and inducers, there are noticeable discrepancies between predicted and observed magnitudes of these DDIs. To solve these issues, a method for the prediction of unbound hepatic concentration is proposed based on theoretical considerations. Finally, a pharmacokinetic model to describe the intestinal first pass metabolism is considered, particularly focusing on the importance of the Qgut model. Although this Qgut model was proposed as an empirical model, theoretical considerations suggest that the model is regarded as a physiologically-based pharmacokinetic model that can predict significance of intestinal DDIs. Theoretical considerations proposed in the present article may be helpful for future analysis of DDIs.
The complex transporter-cytochrome P450 (CYP) enzyme interplay in the disposition of drugs makes it very challenging to address the safety of clinical drug interactions. Thus, two major subjects are discussed herein. First, the concept of an intravenous drug interaction study (where the perpetrator is administered intravenously or orally while the drug candidate administered intravenously) to facilitate prediction of maximal possible systemic exposure of a substrate drug when oral drug interactions occur is explored. If a substrate drug with oral bioavailability is equal to or less than 80%, an intravenous drug interaction study at low dose along with a few key oral drug interaction studies could be useful for achieving this objective with the aid of modeling and simulations. Along with the clinical safety of the drug, one could then attempt to predict the safety margin when the worst drug interactions occur. Secondly, the efficacy and safety disparity of clopidogrel, statins and irinotecan each among races and genetic variants are discussed to illustrate that pharmacogenetic knowledge is important for the interpretation and prediction of drug interaction-induced adverse events, whereas drug interaction -induced adverse events are equally informative for identifying genes-based mechanisms involved.
Protein kinases are potential drug targets for the treatment of a variety of diseases, including cancer. In particular, specific tyrosine kinase inhibitors are rapidly being developed as new drugs for the inhibition of malignant cell growth and metastasis formation. Most of these newly developed tyrosine kinase inhibitors are hydrophobic and thus rapidly penetrate the cell membrane to reach intracellular targets. However, intracellular accumulation of a drug is regulated by multiple factors, including influx and efflux as well as metabolism. In cancer chemotherapy, overexpression of drug efflux transporters in cancer cells is a major cause of multidrug resistance that reduces the efficacy of anticancer drugs. Thus, the transport mechanism-based molecular design strategy would provide an effective tool for chemotherapy against cancer. To develop a platform for molecular modeling to circumvent multidrug resistance and reduce drug-induced adverse effects, we established methods for high-speed screening for human ABCG2-drug interactions, quantitative structure-activity relationship (QSAR) analysis, and quantum chemical calculation for lead optimization. This review addresses recent advances in the strategy of transport mechanism-based molecular design.
Liver kidney microsomal antibody type 1 (LKM-1) is a diagnostic marker for autoimmune hepatitis type 2 (AIH-2). Characterization of LKM autoantibodies of patients with AIH-2 demonstrated that a proportion of LKM sera contains autoantibodies which recognize one or more small linear epitopes on cytochrome P450, CYP2D6, an enzyme of drug metabolism pathways. The identification and epitope mapping of antigens involved in autoimmune diseases are important in understanding the mechanisms triggering autoimmunity and providing guidance for designing immunomodulatory therapy. In this study, several proteins recognized by LKM-1-positive sera in rat and human hepatic microsomes were analyzed by MALDI-TOF-MS after separation with ion-exchange chromatography or two-dimensional polyacrylamide gel electrophoresis. We identified these proteins as ERp57 and carboxylesterase 1 (CES1) as well as CYP2D6. Epitopes in ERp57 and CES1 recognized by LKM-1-positive serum were investigated by enzyme-linked immunosorbent assay (ELISA) with protease-digested peptides of ERp57 and CES1. The peptides comprising amino acids 105-129 of ERp57 and 558-566 of CES1 were specifically recognized by the serum. The epitopes in EPp57 and CES1 recognized by LKM-1-positive sera were homologous with those in hepatitis C virus (HCV). Viral infection of such as HCV may thus possibly trigger autoimmune hepatitis.
Mechanism-based inactivation (MBI) of cytochrome P450 3A (CYP3A) often causes serious drug-drug interactions. To examine species differences in MBI of CYP3A between humans and rodents, we compared MBI potencies of five representative CYP3A inhibitors in human, rat and mouse liver microsomes. Among the inhibitors studied, erythromycin and clarithromycin exhibited markedly weaker MBI effects on CYP3A activity in rat and mouse liver microsomes compared to human liver microsomes. Results of spectroscopic experiments showed that erythromycin and clarithromycin form a metabolic intermediate complex with human liver microsomes but not with rat or mouse liver microsomes. In contrast, troleandomycin, diltiazem and nicardipine form a metabolic intermediate complex with rat and mouse liver microsomes, although some differences in MBI potency among species were observed. Parameters for MBI potency (kinact/KI ratio) and reversible inhibition (IC50) were negatively correlated (r=-0.820, p=0.0003), suggesting that the different affinities of CYP3A inhibitor for CYP3A may partly contribute to the different MBI potencies of inhibitor among species. Taken together, the results suggest that there are species differences in MBI of CYP3A in humans, rats and mice, which should be considered when rodents are used as in vivo models for MBI-mediated drug-drug interaction study.
Human α1-acid glycoprotein (AGP), a serum glycoprotein, is thought to have anti-inflammatory effects by a mechanism that is not well understood. In this study, we investigated the pharmacokinetics of AGP in mice using enzymatically modified AGP (AGP with the sialic acids removed, asialo-AGP, and with both sialic acids and galactose removed, agalacto-AGP). It was observed that AGP was eliminated from the circulation slowly, and was mainly taken up by the liver. The elimination of labeled AGP, asialo-AGP and agalacto-AGP from the circulation was suppressed in the presence of excess unlabeled AGP, asialo-AGP and agalacto-AGP, respectively, suggesting the receptor-mediated uptake of these AGPs. Interestingly, the uptake of AGP by the liver competed with agalacto-AGP, but not with asialo-AGP, while agalacto-AGP competed with asialo-AGP. These results suggest that agalacto-AGP binds to at least two types of receptors, namely asialoglycoprotein receptor (ASGPR) and an as yet unidentified receptor that is shared with AGP, and that AGP is directly taken up by the liver through such a receptor and not via ASGPR. These findings help improve our understanding of the clearance mechanism of AGP.
The constitutive androstane receptor (CAR; NR1I3) is a key transcriptional factor that regulates genes encoding drug-metabolizing enzymes and drug transporters. However, studies on regulation of CAR target genes via up- or down-regulation of CAR are limited. In this study, we examined the effects of PPARα agonists (ciprofibrate, bezafibrate, fenofibrate and WY14643) on the expression of CAR and its target gene CYP2B1/2 in rat primary hepatocytes. Results from real-time PCR analysis showed that CAR and CYP2B1/2 mRNAs exhibit increases in response to all PPARα agonists studied (5 to 10-folds of control). Pretreatment of cells with cycloheximide, an inhibitor of protein synthesis, completely suppressed increase in CYP2B1/2 mRNA in response to ciprofibrate, suggesting that protein synthesis is required in this process. In addition, the induction of CAR by ciprofibrate on the protein level was observed with nuclear extracts as well as total cell lysates. These results indicate that CYP2B1/2 mRNAs are induced by PPARα agonists and that this effect is accompanied by increase in the expression of CAR gene at both mRNA and nuclear protein levels. Activated PPARα may increase functional CAR protein, which can induce the expression of CAR target genes such as CYP2B.
Novel organic cation transporter 2 (OCTN2) is a multispecific, bidirectional, pH-dependent organic cation transporter. It can function as a carnitine co-transporter with higher affinity for carnitine than OCTN1 but also functions as a uniporter for other cations. Drugs such as verapamil, pyrilamine and β-lactam antibiotics have been characterized as substrates of OCTN2 and/or inhibitors of carnitine transport. This study identified variants of the SLC22A5 gene in two distinct ethnic groups of the Singaporean population (n=192) by DNA sequencing. Twenty-eight genetic variants of SLC22A5, including 13 that were novel, were found: 14 were located in the coding exons, 10 in the introns, 1 in the promoter region, 2 in the 5′-untranslated region and 1 in the 3′-untranslated region. Among the novel nonsynonymous variants, Asp122Tyr was predicted to be functionally significant. Functional nonsynonymous variants detected include Ser467Cys and Arg254X; the latter resulted in a premature stop codon and is predicted to result in a truncated protein that is less than half the molecular mass of wild-type OCTN2. These data constitute fundamental information of value for future pharmacogenetic studies in Asian populations on drugs that are substrates of OCTN2.