Drug Metab. Pharmacokinet. 25 (1): 3–15 (2010). Review Drug Interaction Studies on New Drug Applications: Current Situations and Regulatory Views in Japan

Naomi NAGAI* Office of New Drug IV, Pharmaceuticals and Medical Devices Agency (PMDA), Tokyo, Japan

Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk

Summary: Drug interaction studies on new drug applications (NDAs) for new molecular entities (NMEs) ap- proved 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 mechanis- tic theory and new technologies. On the basis of incremental evidence of transporter roles in human phar- macokinetics, 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 in- formation 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 in- teraction studies in recent new drug development. This review summarizes current situations and further dis- cussionpointsondruginteractionstudiesinNDAsinJapan.

Keywords: drug interaction; drug development; new drug application; new molecular entity; phar- macokinetics; metabolism; transport; package insert

remarkably significant increase in the concentration of Introduction 5-FU.1–3) Between 1998 and 2001, there has also been Drug interactions change dose-response relationships, early termination of development or withdrawal from and result in low efficacy or high toxicity of drugs, which medical practice of some NMEs, mainly due to metabolic are important considerations especially in medical prac- drug interactions caused by the inhibition of cytochrome tice with multiple-drug therapies. In current clinical set- P450 3A4 (CYP3A4).4) Through these experiences, the tings, multiple-drug therapies are often prescribed, importance of investigating drug interaction potentials resulting in more difficult situations for healthcare for NME during drug development and providing ap- professionals to adequately monitor drug interactions. propriate drug interaction information in PIs has been re- Over the past 20 years, there have been several fatal drug emphasized. Regulatory authorities in Japan and overseas interactions that were only detected by serious adverse therefore issued documents for drug interaction studies reactions occurring after marketing. The life-threatening around 2000.5–8) drug interactions caused by taking sorivudine and tegafur Drug interactions are generally categorized as phar- became a serious problem heavily covered by the media macokinetic (PK)- and pharmacodynamic (PD)- related in Japan in 1993. Further investigations reported that the drug interactions. PK-related drug interactions include mechanism of this interaction could be inhibition of 5- those affecting absorption, distribution, metabolism and fluorourasil (5-FU) metabolism by sorivudine, resulting in excretion. PD-related interactions deal with additive, syn-

Received; December 27, 2009, Accepted; January 12, 2010 *To whom correspondence should be addressed: Naomi NAGAI Ph.D., Office of New Drug IV, Pharmaceuticals & Medical devices Agency, Shin- Kasumigaseki Bldg, 3-3-2 Kasumigaseki, Chiyoda-ku, Tokyo 100-0013, Japan. Tel. 81-3-3506-9487, Fax. 81-3-3506-9567, E-mail: nagai-naomi＠ pmda.go.jp This review is the author's current thinking on this topic and do not represent the official policy or the requirements of the Pharmaceuticals and Medical Devices Agency and the Ministry of Health, Labour and Welfare. The work was presented in part at the 2009 Annual Meeting and Disposition of the American Association of Pharmaceutical Scientists (AAPS). Table 1 and 3 were based on the regulatory document, ``Methods of Drug Interaction Studies5) and package inserts of HMG-CoA reductase inhibi- tors in Japan. Figure 3 was modified with the figure in the reference book of Clinical Pharmacokinetic Studies of Pharmaceuticals12).

3 4 Naomi NAGAI

Fig. 1. Regulatory documents and discussion on drug interaction studies in new drug development JPMA: Japan Pharmaceutical Manufacturer Association, JSSX: the Japanese Society for the Study of Xenobiotics, JSCPT: the Japanese Society of Clinical Pharmacology and Therapeutics, JSPHCT: Japanese Society of Pharmaceutical Health Care and Sciences, CBI: Chem- Bio Informatics Society, ISSX: International Society for the Study of Xenobiotics, AAPS: American Association of Pharmaceutical Scientists. Periods 1, 2 or 3 is a 4-years study period before and after publication of the ``Methods of Drug Interaction Studies''. ergistic and antagonistic action based on the pharmacolo- regulatory documents, related guidelines and informa- gy of both investigational and concomitant drugs. This tion of literatures.12) The United States Food and Drug review refers to only PK-related drug interactions. Administration (U.S. FDA) published a concept paper in 2004 and new draft guidance for drug interaction studies Regulatory Documents and Discussions in 2006, including study design, data analysis and label- on Drug Interaction Studies ing implications.13,14) Many conference meetings and wor- Figure 1 shows regulatory documents and related dis- kshops have also been held in the U.S. to discuss and up- cussion on drug interaction studies in Japan and over- date on this issue over the past 5 years. The European seas. In 1998, the Ministry of Health, Labour and Wel- Medicines Agency (EMEA) announced a concept paper in fare (MHLW) organized the Working Groups for Drug 200815) to recommend revising guidance on investiga- Interactions and Clinical Pharmacokinetic Studies and tions of drug interactions published in 1997. Recently, in started an intensive discussion for drug interaction stu- Japan, there have been many opportunities mainly dies during new drug development. Since then, an in- provided by academia to discuss drug interaction related creasing number of related conference meetings has topics especially on transporter-based drug interactions. been held and related regulatory documents have been Each of the 3 ICH regions (International Conference published in Japan. The regulatory document, ``Methods on Harmonization of Technical Requirements for Regis- of Drug Interaction Studies'' was issued on June 4, 2001 tration of Pharmaceuticals for Human Use) has published and covers fundamental points, in vitro and in vivo study its own regulatory documents for drug interaction stu- approaches concerning drug interactions. ``Clinical Phar- dies. These documents provide similar principles for macokinetic Studies of Pharmaceuticals'' and ``Guideline drug interaction studies in drug development and also on Non-clinical Pharmacokinetic Studies'' describe the reflecte the latest scientific knowledge, technologies and scope and basic principles of pharmacokinetic studies medical practices at the time of their publications, necessary for submitting new drug applications, both of although some differences are found in the scope of in which are useful to plan, conduct and evaluate drug in- vitro and in vivo studies and labeling implications. Figure teraction studies.9–11) In addition to these regulatory 2 shows the contents of ``Methods of Drug Interaction documents, the MHLW Working Groups for Drug Inter- Studies''. The Japanese document includes the following: actions and Clinical Pharmacokinetic Studies published a (1) focusing on not only metabolism-mediated drug inter- reference book in 2003 that includes English versions of actions as the most frequent mechanism for drug interac- Drug Interaction Studies in New Drug Applications in Japan 5

Fig. 2. Methods of drug interaction studies: table of contents

tions,16) but also on absorption-, distribution- and ex- mechanisms of drug interactions associated with inves- cretion-mediated drug interactions, therefore including tigationed drugs are identified by basic non-clinical and the points of consideration for all PK process, (2) brief clinical PK studies, non-clinical drug interaction studies labeling implications because official guidelines and and clinical drug interaction studies, (3) the most possible regulatory documents for PIs had been issued,17–20) (3) risk of interactions between investigationed drugs and primarily describing clinical drug interaction studies and other drugs in humans are assessed based on drug inter- in vitro studies using human tissue-derived samples and action studies in early-clinical development phase, (4) expression systems. possible drug interactions in humans are exploratorily Figure 3 shows a typical approach for investigating evaluated in late-clinical development phase, (5) labeling drug interaction potentials during new drug develop- information and drug interaction study data are ap- ment.12) Although the reference book shows a case in propriately provided in PIs and (6) risk reassessment of which a drug candidate may be a metabolic inhibitor and drug interactions, followed by feed back of this informa- developed in oral dosage form,12) basic principles, typical tion to medical practice is periodically conducted during approaches and general decision-making to investigate post-marketing period. drug interactions in new drug development are common Drug Interaction Studies in Recent NDAs in Japan and summarized as follows, (1) sequence of studies are rationally planned from pre-clinical development phase Purpose and methods: The purpose of this survey to clinical development phase, (2) possible factors and is to identify current situations of drug interaction stu- 6 Naomi NAGAI

Fig. 3. Typical approach for investigating drug interaction potentials in new drug development

A case in which an investigationed drug candidate may be a metabolic inhibitor and developed in oral dosage form. CuHi max(appr): an approxi- mation of maximum concentration of the unbound inhibitor in the liver. CP*/CP: extent of increase in plasma concentration (AUC or Css) of the inhibited drug. (modified with the figure in reference 12), in Japanese).

dies conducted and submitted for NDAs in Japan, then cal materials, cDNA expression systems, animals) and consider the impact of ``Methods of Drug Interaction marker drugs (substrates, inhibitors, inducers) Clinical Studies'' and further discussion points. drug interaction study: study design, concomitant drugs, The 300 NDAs of NMEs (Shin-yukoseibun ganyu- labeling information. iyakuhin, both chemical and biological NMEs) approved Current situation of drug interaction studies: in Japan between 1997 and 2008 were examined. The in- The trends of drug interaction studies for NMEs ap- formation sources used for analysis are mainly the review proved between 1997 and 2008 are shown in Figure 4. reports prepared by the PMDA and Pharmaceuticals and The number of approved NMEs were more than 100 in Medical Devices Evaluation Center of the National In- periods 1 and 3, although temporarily decreased in stitute of Health Sciences (PMDEC, predecessor organi- period 2 when PMDEC was reorganized to the present zation of PMDA), and summaries of data submitted by PMDA in 2004 (Fig. 4A). In period 1, 82% of the NDAs applicants. Both review reports and summaries of data (93/113 NDAs) were found to have at least one non-clini- are open to the public, and are mostly available on the cal drug interaction study, which has decreased to 71% Web site.21–22) For comparing situations before and after (55/77 NDAs) and 72% (79/110 NDAs) in periods 2 and publication of ``Methods of Drug Interaction Studies'' as 3(Fig. 4B), respectively. Those containing at least one well as considering effects of recent trends in science and clinical drug interaction study have slightly increased technology, NDAs were divided into three groups, NDAs from 54% (61/113 NDAs) in period 1 to 56% (43/77 approved between 1997 and 2001 (Period 1, included NDAs) and 58% (64/110 NDAs) in periods 2 and 3, 113), NDAs between 2001 and 2004 (Period 2, included respectively (Fig. 4C). 77) and NDAs between 2005 and 2008 (Period 3, includ- Figure 5 shows trends in administration routes of ed 110) (Figs. 1 and 4). 300 NMEs approved between 1997 and 2008. Among The following information regarding non-clinical and the approved NMEs, 54% (163/300 NDAs) have been de- clinical drug interaction studies was collected by six rev- veloped in oral dosage form, 28% (83/300 NDAs) in sys- iewers of the Office of New Drugs in the PMDA and temic injection dosage form (intravenous or intra-arteri- PMDEC, Non-clinical drug interaction study: mechanism al) and about 9% (26/300 NDAs) in local injection dosage of PK-related drug interactions (plasma protein binding, form (subcutaneous or intramuscular) (Fig. 5A). NDAs transport proteins, drug-metabolic enzyme-inhibition or containing at least one drug interaction study were induction), study systems and materials (human biologi- different in oral and other dosage forms (Fig. 5B). Drug Interaction Studies in New Drug Applications in Japan 7

Fig. 4. Drug Interaction studies in new drug applications (NDAs) for new molecular entities (NMEs) approved in Japan between 1997 and 2008 A: Numbers of approved NMEs. B: NDAs containing non-clinical drug interaction studies, C: NDAs containing clinical drug interaction stu- dies.

Fig. 5. Relationships between drug interaction studies in NDAs and administration routes (A and B) for new molecular entities (NMEs) approved between 1997 and 2008 Open column: numbers of approved NDAs between 1997 and 2008, light grey column: NDAs containing non-clinical drug interaction stu- dies, solid column: NDAs containing clinical drug interaction studies. * one NMEs was developed as both oral and intravenous injection dosage forms.

Regarding orally administered NMEs, most NDAs con- of 300 NMEs approved between 1997 and 2008. Ther- tained non-clinical drug interaction studies and four apeutic areas were classified into 10 disease groups: gas- fifths of NDAs contained clinical drug interaction stu- trointestinal disease, cardiorenal disease, neurophar- dies, while less drug interaction studies were submitted macologic disease, endocrinologic disease, infectious dis- for injection and other routes of administration. Drug in- ease, HIV, urologic and gynecologic diseases, allergic and teractions after oral administration could pre-systemical- immunologic diseases, cancer and others (medical diag- ly and systemically occur so that more drug interaction nosis, radiopharmacology and biological products other studies were conducted in the development of the oral than recombinant NMEs), according to the new drug rev- dosage form. Figure 6 shows trends in therapeutic areas iew teams of the Office of New Drugs in the PMDA as of 8 Naomi NAGAI

Fig. 6. Relationships between drug interaction studies in NDAs and therapeutic areas (A and B) for new molecular entities (NMEs) approved between 1997 and 2008 Open column: numbers of approved NDAs between 1997 and 2008, light grey column: NDAs containing non-clinical drug interaction stu- dies, solid column: NDAs containing clinical drug interaction studies.

2008. NDAs that contained at lease one drug interaction tally increased from 4% of total NDAs in the period 1 to study were not consistent in different therapeutic areas 15% and 25% in the periods 2 and 3, respectively. The (Fig. 6B). Drug interactions were the most frequently ``Methods of Drug Interaction Studies'' describes that it studied in the development of anti-HIV, cardiorenal and is difficult to uniformly apply points to consideration of anti-infective drugs, followed by neuropharmacologic this document in drug interaction studies on biological drugs. Cardiorenal and neuropharmacologic drugs are NMEs, because of different PK profiles between chemical usually administered in oral dosage form and prescribed and biological NMEs. This analysis shows that drug inter- as multiple and/or long-term drug therapies. During long- action information about biological NMEs is limited at term drug therapies, drug accumulation and enzyme in- present, probably because drug interaction studies have duction may occur and the number and frequency of not been conducted during current drug development. concomitant drugs may be increased. Compared to these Therefore, a case-by-case discussion is currently needed therapeutic areas, NDAs of hormones (therapeutic area, to study and evaluate drug interaction potentials for bio- mainly endocrinologic disease and cancer) and anti-can- logical NMEs. cer drugs contained fewer drug interaction studies, since Non-clinical drug interaction studies: Trends in in part anti-cancer drugs and hormones are usually deve- non-clinical drug interaction studies (plasma protein loped in injection dosage form. Almost all NDAs of HIV binding-, transporter- and drug-metabolizing enzyme- drugs contained various and many drug interaction stu- based drug interaction studies) are shown in Figures 7 dies, probably due to their oral dosage form and the style and 8. of medical practice (multiple-drug therapy). No drug in- 1) Plasma protein binding-based drug interaction studies teraction studies have been submitted for NDAs of diag- (Fig. 7A) nostic drugs and radiopharmaceuticals and fewer drug The percentage of NDAs containing plasma protein interaction studies have been submitted for biological binding-based drug interaction studies (displacement stu- NMEs. This analysis thus shows that drug interaction stu- dies) decreased from 40% in period 1 to about 20% in dies conducted in new drug development and submitted period 3. According to the ``Methods of Drug Interaction for NDAs reflect both PK profiles of each NME, dosege Studies'', displacement studies between NME and other form and current medical practice in each therapeutic drugs should be conducted primarily based on the extent area. of plasma protein binding ability of NME, with considera- In this analysis, the number of NDAs for biological tion to volume of distribution, elimination pathways and NMEs including recombinant NMEs has been incremen- rapeutic range. This analysis shows that displacement stu- Drug Interaction Studies in New Drug Applications in Japan 9

Fig. 7. Classification of targeted mechanism of drug interactions in non-clinical drug interaction studies A: plasma protein binding-based drug interaction studies, B: transporter-based drug interaction studies, C: drug-metabolic enzyme-based drug interaction studies (metabolic inhibition), D: drug-metabolic enzyme-based drug interaction studies (enzyme induction).

Fig. 8. Study systems using metabolic inhibition (A) and enzyme induction (B) studies A: Open column: numbers of approved NDAs containing metabolic inhibition studies, light grey column: microsome, solid column: human hepatocyte, dark grey column: cDNA expression system, B: Open column: numbers of approved NDAs containing enzyme induction stu- dies, solid column: human hepatocyte, dark grey column: in vivo animals. dies have recently tended to be conducted and submitted period 1. ``Methods of Drug Interaction Studies'' men- in NDAs only for NMEs that have more than 90% of in tion that it may be useful to examine in vitro inhibition vitro plasma protein binding, using human tissue-derived samples, cells, membrane 2) Transporter-based drug interaction studies (Fig. 7B) vesicles and transport protein expression systems and Transporter-based drug interactions have been in- study the extent of transporter's contribution to absorp- creasingly studied in vitro from about 6% in period 1, to tion, distribution and excretion process of NMEs, when about 12% and about 19% in periods 2 and 3, respective- active transport is greatly suspected. In this analysis, the ly. NDAs containing transporter-based drug interaction target transport proteins in the pre-clinical development studies in period 3 have tripled in number, compared to phase were primarily P-glycoprotein (P-gp), followed by 10 Naomi NAGAI

organic anion transporting polypeptide (OATP), organic of methodologies and the common understanding of the anion transporter (OAT), organic cation transporter mechanistic theory to examine in vitro CYP-based drug (OCT), multidrug resistance-associated protein (MRP) interaction studies has been achieved in recent new drug and breast cancer resistance protein (BCRP). development. Because in vitro studies for transporter- 3) Drug-metabolic enzyme-based drug interaction stu- based drug interaction have been increasingly conducted dies (Figs. 7C and 7D, 8A and 8B) in recent new drug development, standardization of The percentage of NDAs containing metabolic inhibi- materials and general methodologies including decision tion studies has increased throughout the 3 periods, criteria and prediction from in vitro data to clinical stu- reaching a maximum of 60% in period 3. On the other dies needs to be discussed. bend, the percentage of NDAs containing enzyme induc- Non-CYP enzymes, interaction potentials of quantita- tion studies has decreased from about 75% in period 1 to tively important metabolites in humans, and co or mul- about 53% in period 3. To predict clinically significant tiple-interactions between metabolism and transport drug-metabolic enzyme-based drug interactions caused such as CYP3A4 and P-gp-mediated interactions or be- by NME, the importance of in vitro studies using human tween inhibition and induction such as CYP3A4-mediat- tissue-derived samples and enzyme expression systems is ed interaction also need to be discussed. emphasized in ``Methods of Drug Interaction Studies''. Clinical drug interaction studies: Drug interac- Although the study systems using human liver micro- tion studies in the early clinical development phase are somes were the most common over the 3 periods, there usually conducted as clinical pharmacology studies using was a recent trend toward increase in the number and marker drugs which can assess most expected drug inter- variety of study systems such as human hepatocytes in actions in clinical situations. ``Methods of Drug Interac- period 3 (Fig. 8A). ``Methods of Drug Interaction Stu- tion Studies'' recommends that clinical studies must be dies'' describes that the case and points regarding the in- ethically and scientifically conducted. Therefore, it is duction study system using experimental animals, which desirable to reduce the number of clinical studies in the is at present, in vivo animal studies where levels of P450 drug development. To successfully understand clinical isoforms are monitored following the repeated dosing of drug interaction profiles for investigationed drugs, ap- NME, may be useful for screening in the early stage of propriate choices of concomitant drugs for early clinical drug development. The document also points out that pharmacology studies, results of in vitro studies and basic prediction of human outcomes with animal data must be human PK data are essential. Table 2 shows the top 5 carefully made, because of species differences. Cultured drugs used in clinical drug interaction studies. Although human hepatocytes can be useful. This analysis proves digoxin, warfarin and cimetidine have been the most fre- that study systems for examining CYP induction have quently used drugs, the number of clinical drug interac- remarkably changed over the past 12 years, showing in- tion studies with cimetidine which inhibits both P450 crease in the use of human hepatocytes, while the use of nonspecifically and renal tubular secretions, has been in vivo animals has been reduced (Fig. 8B). decreasing. Clinical drug interaction studies with 4) Interpretation of pre-clinical findings to clinical de- ketoconazole, one of the potent CYP3A4 inhibitors, have velopment phase increasingly been submitted for recent NDAs. The Although understanding mechanisms of drug interac- rationale for choice of ketoconazole as the marker drug tions at molecular levels helps designing appropriate clin- in clinical drug interaction studies has been noted in ical drug interaction studies, quantitative interpretation regulatory documents in Japan and overseas.5,13–14) The from in vitro results is still one of the most important con- FDA draft guidance suggests that if an investigationed siderations.23–28) Thechoiceofprobeinhibitors,inducers drug is shown to be metabolized by CYP3A4 and contri- and marker substrates for early in vitro studiesisthefirst bution of this enzyme to the overall elimination pathway important process to examine drug interaction potentials is either substantial (À25% of the clearance pathway) or in new drug development. ``Methods of Drug Interaction unknown, the choice of inhibitor and inducer may be Studies'' provides detailed information on typical sub- ketoconazole and rifampin, respectively.14) In this analy- strates, inhibitors and inducers for investigating drug in- sis, a wide range of drugs was selected for clinical drug teractions for in vitro and in vivo studies, calculation interaction studies, mainly since the choice of drugs was methods for decision making of whether an inves- determined based on possible co-prescription or combi- tigationed drug candidate should be further developed or nation therapy in current medical practice, in addition to not, and designs of clinical drug interaction studies PK based considerations. (Table 1 and Fig. 3). Based on this analysis, materials ``Methods of Drug Interaction Studies'' also points out and methods for in vitro drug-metabolic enzyme, espe- that population PK (PPK) approach may be useful when cially CYP-based drug interaction studies including probe data from many patients are available, depending on the substrates, other reagents and study systems have been target disease. PPK is used to evaluate a large number common in recent NDAs. Therefore, the standardization drugs in clinical phases 2 and 3, allowing more scientifi- Drug Interaction Studies in New Drug Applications in Japan 11

Table 1. Subsrates, inhibitors, inducers and reference compounds of major CYP enzymes for in vitro and in vivo drug interaction studies (in ``Methods of Drug Interaction Studies'')

P450s Substrates Inhibitors (in vitro) Inhibitors (in vivo)InducersMarkerdrugs(in vitro)Markerdrugs(in vivo)

Clopidogrel Charcoal-grilled beef Furafylline Fluvoxamine Flutamide Cigarette smoke a-Naphthofravone Furafylline Phenacetin CYP1A2 Caffeine Cruciferous vegetables Caffeine Ellipticine Ciprofloxacin Ethoxyresorufin (Phenacetin) Omeprazole Methoxsalen Methoxsalen Theophylline Griseofulvin

(Coumarin) R-(－)-Menthofuran Tegafur Tranylcypromine Methoxsalen SM-12502 SM-12502 CYP2A6 Nicotine Pilocarpine Ketoconazole Nicotine Coumarin SM-12502 Ellipticine

Paclitaxel Paclitaxel Diclofenac (5-OH) CYP2C8 Diclofenac Paclitaxel (6a-OH) Rosiglitazone Retinoic acid Fluvastatin

NSAID drugs S-Warfarin Phenytoin Sulfaphenazole Sulfaphenazole Rifampicin S-Warfarin Tolbutamide CYP2C9 Tolbutamide Dicoumarol Sulfinpyrazone Barbiturates Tolbutamide Indomethacin S-Warfarin Diclofenac (4?-OH)

(S-Mephenytoin) Diazepam Omeprazole (Mephenytoin) Hexobarbital S-Mephenytoin Rifampicin Omeprazole Mephenytoin CYP2C19 Imipramine Tranylcypromine Omeprazole Phenobarbital Proguanil Omeprazole Omeprazole Mephobarbital Diazepam Proguanil Paraverine Propranolol

Antidepressants Neuroleptics Ajimaline b-blockers Haloperidol Fluoxetine (Debrisoquine) Debrisoquine Antiarrhythmics CYP2D6 Quinidine Paroxetine Dextromethorphan Dextromethorphan Codeine Ritonavir Quinidine Metoprolol Bufuralol Dextromethorphan Ritonavir Ethylmorphine Nicotine

(Chlorzoxazone) Diethyldithiocarbamate Alcohols Ethanol p-Nitrophenol CYP2E1 1,1,1-Trichloroethane Dimethyl sulfoxide (Chlorzoxazone) Enflurane Isoniazid Chlorzoxazone Disulfiram Dapsone

Midazolam Erythromycin Cyclosporin Clotrimazole Felodipine Saquinavir Felodipine Ritonavir Dexamethasone Midazolam Carbamazepine Midazolam (Ketoconazole) Phenytoin Simvastatin Felodipine Simvastatin Ketoconazole Troleandomycin Rifampicin Dextromethorphan CYP3A4 Nifedipine Dextromethorphan Metyrapone Clarithromycin Troleandomycin Testosterone Triazolam Triazolam Glibenclamide Carbamazepine Dapsone Simvastatin [Cortisol-6OH Itraconazole Phenobarbital Diazepam Terfenadine excretion] Grapefruit juice Triazolam Dextromethorphan Verapamil Warfarin

There are cases where other enzymes or isoforms, not indicated in the table, are main metabolic enzymes. SM-12502: not approved for sale ( ): therapeutic indications are different from overseas [ ]: endogenous substances used as index of metabolic activity cally based screening of drug interaction potentials of in NDAs are conducted overseas. Foreign clinical NMEs.12) Recent new drug development uses the PPK ap- trials may be applicable for Japanese regulatory submis- proach for that purpose. If precise sample collection and sion.29–31) When foreign clinical drug interaction studies data analysis are successfully achieved the PPK approach are extrapolated to the patient population and medical probably can minimize the number of clinical drug inter- practice in Japan, intrinsic and extrinsic ethnic factors action studies during drug development, as well as obtain should be carefully considered based on ICH E5 guide- useful information for post marketing phase. lines, ``Ethnic Factors in the Acceptability of Foreign Approximately 90% of clinical drug interaction studies Clinical Data''. 12 Naomi NAGAI

Table 2. Concomitant drugs used in clinical drug interaction studies

2005–2008 (Period 3, 110NDAs) 2001–2004 (Period 2, 77 NDAs) 1997–2000 (Period 1, 113NDAs) 1992–1997 (US, 193 NDAs)*

1 Warfarin Warfarin Cimetidine Cimetidine 2 Digoxin Antacids Warfarin Digoxin 3 Ketoconazole Digoxin Digoxin Warfarin 4 Oral Contraceptive Ketoconazole Zidovudine Theophylline 5 Antacids/Cimetidine Cimetidine Antacids Antacids

Top 5 drugs in all clinical drug interaction studies in NDAs * reference 44)

Table 3. Drug interaction information in PIs: HMG-CoA reductase inhibitors-ciclosporin or digoxin

Drug interaction information Name of HMG-CoA Approval Metabolic enzyme with ciclosporin with digoxin reductase year Transport protein inhibitors Possible Possible Status Changes in PK Status Changes in PK mechanism mechanism

(not metabolized by Pravastatin 1989 precausion — — — — — CYP3A4)

inhibition Simvastatin 1991 CYP3A4 precausion — —— — (CYP3A4)

Cmax: increase not clearly Fluvastatin 1998 CYP2C9 (mainly) precausion — — precausion AUC: no change determined (digoxin)

inhibition Cmax: 1.10–1.20 inhibition Atorvastatin 2000 CYP3A4, P-gp precausion — (metabolism and precausion AUC: 1.04–1.15 (P-gp) biliary excretion) (digoxin)

OATP1B1 (OATP-C/OATP2), increase inhibition Pitavastatin 2003 (not metabolized by contraindication Cmax 6.6 fold —— — (hepatic uptake) CYP3A4), CYP2C9 (slightly) AUC 4.6 fold

increase OATP-C (OATP-2), AUC 7 fold inhibition no change Rosuvastatin 2005 contraindication — — CYP2C9, 2C19 (slightly) no chanage (hepatic uptake) (digoxin) (ciclosporin)

This table is based on information from PIs of HMG-CoA reductase inhibitors in Japan, —: not informed in PIs

``Methods of Drug Interaction Studies'' also recom- teraction studies based on the ICH E5 guideline and re- mends the followings, (1) If significant individual differ- cent advances in pharmacogenomics contribute to ra- ences in PK due to genetic polymorphism are expected, it tional planning of the number and kinds of clinical drug is desirable to conduct a study including or excluding interaction studies for approval. subjects with specific genotypes. (2) If a polymorphically Considerations of labeling: This analysis shows expressed enzyme is largely responsible for the that drug interaction information has been increasingly metabolism of an investigational drug, it is necessary to and systemically provided on PIs of recently approved discuss the possibility of drug interactions considering NMEs. When outcomes of clinical drug interaction stu- the phenotype and/or genotype of each patient. Most dies conducted in drug development gave significant im- clinical drug interaction studies are generally conducted pact, they were submitted for NDAs and informed as a on healthy volunteers. Therefore, for example, if PK labeling implication in the drug interaction section and study is conducted in poor metabolizers of a polymorph- data of clinical drug interaction study were usually ic enzyme, it can be useful instead of drug interaction in- presented in either the drug interaction section or PK formation with potent inhibitors. The official discussion section. Among NMEs in the same pharmacology class, opportunity for pharmacogenomic issue between spon- more recently approved NMEs tend to obtain detailed sor and regulatory agency is now available as formal con- and increasing amounts of drug interaction information sultation in PMDA.32) during drug development. Approved labeling tend to in- The appropriate use of PPK, global sharing of drug in- clude both mechanism and quantitative information of Drug Interaction Studies in New Drug Applications in Japan 13

Table 4. Transporter-based drug interaction information in PIs in Japan

Transport proteins Nonproprietary Name of NMEs approved between 1997 and 2008

Atorvastatin calcium1, Clarithromycin1, Etaravirine1&2, Everolimus1, Fexofenadine hydrochroride1&2, P-glycoprotein Ivelmectine2, Lanatocide1,Maraviroc2, Raltegravir pottassium2,*, Rosuvastatin calcium2,*, Saquinavir mesylate1

Organic anion transporter Adefovir pivoxil1

Organic anion transporting polypeptide Rosuvastatin calcium2, Pitavastatin calcium2

Peptide transporter Valaciclovir hydrochloride2, Valganciclovir hydrochrolide2

Bosentan hydrate2, Doripenem hydrate2,Emtricitabine2, Entecavir hydrate1&2, Tenofovir disoproxil Others (documented as active renal secretion etc) 1&2 fumarate

Section of statement: 1 Interaction, 2 Pharmacokinetics * no interaction suggested

drug interactions. For example, Table 3 shows drug than or in addition to P-gp were OATP and peptide trans- interaction information on PIs of HMG-CoA reductase porter (PEPT1). As the mechanism of drug interactions, inhibitors approved in Japan. In the NDAs of atorvasta- descriptions such as active renal secretion were also tin, pitavastatin and rosuvastatin been approved in Japan found in PIs of some NMEs, although no information of after 2000, information from many various drug in- the target transport proteins and quantitative impact on teraction studies including the studies with ciclosporin the PK was provided. and/or digoxin were submitted. Ciclosporin is a substrate The primary objectives of drug interaction studies dur- of CYP3A4 and an inhibitor of OATP.33,34) Some trans- ing new drug development is to obtain information about port proteins including OATP and/or CYP enzymes have whether dose adjustment of the drug itself or con- important roles in PK of HMG-CoA reductase inhibi- comitant drugs is needed. Although there are many tors.35–38) Several molecular based studies suggest that in- NMEs approved before the publication of ``Methods of hibition of hepatic uptake of HMG-CoA reductase inhibi- Drug Interaction Studies'', clinically important drug in- tors may be a mechanism of drug interactions between teractions for these drugs have been studied and reported HMG-CoA reductase inhibitors and ciclosporin.39) In PIs in the literature after approval. Some studies are informa- of pitavastatin and rosuvastatin, possible mechanisms of tion sources for drug interaction labeling in PIs, depend- drug interactions with ciclosporin and quantitative ef- ing on the quality of studies. Therefore, drug interaction fects on the pharmacokinetics of pitavastatin and information helpful for dose adjustment considerations rosuvastatin have been shown based on clinical and non- should be periodically reflected in PIs even after ap- clinical drug interaction studies. PIs of other HMG-CoA proval. Furthermore, how to appropriately provide drug reductase inhibitors provide no information on either interaction information and give adequate recommenda- possible mechanisms or PK change. The drug interaction tions including dose adjustment in PIs should be carefully possibility with digoxin was provided in the PI of ator- considered based on non-clinical and clinical drug inter- vastatin and fluvastatin. For atorvastatin, the mechanism action studies. would be inhibition of P-gp and digoxin PK change was Conclusions and Further Discussion Points marginal and for fluvastatin, the mechanism was not clearly determined. According to the literature and Drug interaction studies in new drug development Japanese PIs,40–43) Japanese PIs tend to include not were analyzed using NDAs of NMEs in Japan between enough quantitative information or description about 1997 and 2008. The situation of drug interaction studies drug interactions. Differences in drug interaction infor- conducted during new drug development has recently mation in PIs among HMG-CoA reductase inhibitors may changed, compared to situations when the present be time differences between drug development and pub- regulatory document was issued. The quality of drug in- lication of regulatory documents.40,41) teraction studies has improved and become more ration- Some recently approved NMEs in Japan include trans- al, resulting in more informative labeling in PIs of recent- porter-based drug interaction information in the ap- ly approved NMEs. Thus, the Japanese regulatory docu- proved PIs (Table 4). Of the various transport proteins, ment, ``Methods of Drug Interaction Studies,'' in addi- drug interactions with P-gp are the most well understood tion to recent advances in science and technology, has and investigated in new drug development; they are listed contributedtoplansandevaluationofdruginteraction in either the drug interaction section or PK section in PIs studies in recent new drug development as well as under- of 11 NMEs approved between 1997 and 2008. The standing of drug interaction studies by pharmaceutical transport proteins currently documented in PIs other companies and a regulatory agency. Based on this analy- 14 Naomi NAGAI sis, the following areas described as general points to analysis, and recommendations for dosing and labeling. Center consideration in the ``Methods of Drug Interaction Stu- for Drug Evaluation and Research/Center for Biologics Evalua- dies'', therefore need to be further discussed and updat- tion and Research. U.S. November 1999. ed, 8) European Medicines Agency: Note for guidance on the investi- (1) Standardization of materials and general methodo- gation of drug interactions, CPMP December 1997. logies for transporter-based drug interactions to 9) Ministry of Health, Labour and Welfare: Clinical pharmacoki- netic studies of pharmaceuticals, notification No. 796 (Kachou- provide more clear instructions for choosing the Tsuuchi, in Japanese), Pharmaceuticals Medical Safety Bureau, timing and types of drug interaction studies to be Japan, June 1, 2001. conducted during new drug development. 10) Ministry of Health, Labour and Welfare: Guideline on non-clini- (2) More reliable interpretation of pre-clinical data to cal pharmacokinetic studies, notification No. 496 (Kachou- design clinical drug interaction studies. Tsuuchi, in Japanese), Pharmaceuticals Medical Safety Bureau, (3) Points of consideration to evaluate co or multiple Japan, June 26, 1998. interactions such as multiple CYP inhibitions, inhi- 11) Nagai, N.: Japanese guidance on clinical pharmacokinetic stu- bition and induction, interactions of both CYP and dies and review of the new drug applications. Xenobio. Metabol. transporter, non-CYP enzyme mediated drug in- Dispos., 16: 187–192 (2001) teractions and drug interaction potentials of quan- 12) Ministry of Health, Labour and Welfare, Clinical Pharmacoki- titatively important metabolites. netic Studies and Drug Interaction Working Group: Reference book for ``Clinical Pharmacokinetic Studies of Pharmaceuticals'' (4) Basic principles and general approaches for ex- (in Japanese), Tokyo Jiho Ltd., 2003. amining drug interaction potentials of biological 13) Food and Drug Administration: Concept paper/drug interaction NMEs during new drug development. studies – study design, data analysis, and implications for dosing (5) More adequate description and reassessment sys- and labeling. Center for Drug Evaluation and Research/Center tem for drug interaction information on PIs for Biologics Evaluation and Research. U.S. 2004. 14) Food and Drug Administration: Guidance for industry/drug in- Acknowledgements: I thank co-investigators of this teraction studies – study design, data analysis, and implications survey,T.Arato,F.Kobayashi,S.HidakaandH. for dosing and labeling. Center for Drug Evaluation and Makimoto and T. Ito. I also thank M. Saito, D. Iwata, M. Research/Center for Biologics Evaluation and Research. U.S. Hirano,K.Chikazawa,M.Yamada,K.Mori,H.Akagawa September 2006. and S. Toyoshima for their comments on this work. 15) European Medicines Agency: Concept paper/recommendation on the need for revision of (CHMP) note for guidance on the in- References vestigation of drug interactions, July 2008. 16) Lin, J. H. and Lu, A. Y.: Inhibition and induction of cytochrome 1) Okuda, H., Nishiyama, T., Ogura, K., Nagayama, S., Ikeda, K., P450 and the clinical implications. Clin. Pharmacokinet., 35: Yamaguchi, S., Nakamura, Y., Kawaguchi, Y. and Watabe, T.: 361–390 (1998). Mechanism of lethal toxicity exerted by simultaneous adminis- 17) Ministry of Health and Welfare: Guidelines for package inserts tration of the new antiviral, sorivudine, and the antitumor a- for prescription drugs, notification No. 606 (Kyokuchou- gent, Tegafur. Xenobio. Metabol. Dispos., 10: 166–169 (1995). Tsuuchi, in Japanese), Pharmaceuticals Affairs Bureau, Japan, 2) Watabe, T.: Strategic proposals for predicting drug-drug interac- April 1997. tions during new drug development: based on sixteen deaths 18) Ministry of Health and Welfare: Guidelines for package inserts caused by interactions of the new antiviral sorivudine with 5- for prescription drugs, notification No. 59 (Kachou-Tsuuchi, in fluorouracil prodrugs. J. Toxicol. Sci., 21: 299–300 (1996). Japanese), Pharmaceuticals Affairs Bureau, Japan, April 1997. 3) Diasio, R. B.: Sorivudine and 5-fluorouracil; a clinically sig- 19) Ministry of Health and Welfare: Guidelines for precautions for nificant drug-drug interaction due to inhibition of di- prescription drugs, notification No. 607 (Kyokuchou-Tsuuchi, hydropyrimidine dehydrogenase. Br. J. Clin. Pharmacol., 46:1–4 in Japanese), Pharmaceuticals Affairs Bureau, Japan, April 1997. (1998). 20) Ministry of Health, Labour and Welfare: Guidelines for precau- 4) Alfaro, C. L.: Emerging role of drug interaction studies in drug tions for prescription drug, (Jimurenraku in Japanese), Phar- development : the good, the bad, and the unknown. Psy- maceuticals Affairs Bureau, Japan, December 2000. chopharm. Bull., 35: 80–93 (2001). 21) Information for approved drugs in Japan (since 2001, in 5) Ministry of Health, Labour and Welfare: Methods of Drug Inter- Japanese). Pharmaceuticals and Medical Devices Agency, Japan. action Studies, Notification No. 813 (Kachou-Tsuuchi, in Available from URL: http://www.info.pmda.go.jp/ shinyaku/ Japanese), Evaluation and Licensing Division, Pharmaceuticals shinyaku_index.html. Accessed January 7, 2010. Medical Safety Bureau, Japan, June 4, 2001. 22) Information for approved drugs in Japan (since September 6) Food and Drug Administration: Guidance for industry; drug 1999, in Japanese). Japan Pharmacists Education Center, Japan. metabolism/drug interaction studies in the drug development Available from URL: http://www.jpec.or.jp/contents /c01/ process: studies in vitro. Center for Drug Evaluation and link.html. Accessed January 7, 2010. Research/Center for Biologics Evaluation and Research, U.S. 23) Huang, S-M., Rwmple, R., Throckmorton, D. C. and Lesko, L. J.: April 1997. Drug-drug interaction studies: study design, data analysis, and 7) Food and Drug Administration: Guidance for industry; In vivo implications for dosing recommendations. Clin. Pharmcol. Ther, drug metabolism/drug interaction studies – study design, data Drug Interaction Studies in New Drug Applications in Japan 15

81: 298–304 (2008). P-glycoprotein: implications for drug delivery and activity in 24) Ito, K., Brown, H. S. and Houston, J. B.: Database analyses for cancer chemotherapy. Mol. Carcinogen., 13: 129–134 (1995). the prediction of in vivo drug-drug interactions from in vitro 35) Igel, M., Sudhop, T. and vonBergmann, K.: Metabolism and data. Br. J. Clin. Pharmacol., 57: 473–486 (2008). drug interactions of 3-hydroxy-3-methylglutaryl coenzyme A- 25) Tucker, G. T., Houston, J. B. and Huang, S. H.: Optimizing drug reductase inhibitors (statins). Eur. J. Clin. Pharmacol., 57: development: strategies to assess drug metabolism/transporter 357–364 (2001). interaction potential-toward a consensus. Clin. Pharmacol. Ther., 36) Giacomini, K. M. and Sugiyama, Y.: Membrane transporters and 70: 103–14 (2001). drugresponse.InBrunton,L.L.(ed.):Goodman & Gilman's The 26) Bjornsson,T.D.,Callaghan,J.T.,Einolf,H.J.,Fischer,V.,Gan, Pharmacological Basis of Therapeutics 11th edtion,NewYork, L.,Grimm,S.,Kao,J.,King,P.,Miwa,G.,Ni,L.,Kumar,G., McGraw-Hill, 2005, pp41–70. Mcleod, J., Obach, R. S., Roberts, S., Roe, A., Shar, A., Snikeris, 37) Lin, J. H. and Yamazaki, M.: Role of P-glycoprotein in phar- F., Sullivan, J. T., Tweedie, D., Vega., J. M., Walsh, J. and macokinetics: clinical implications. Clin. Pharmacokinet., 42: Wrighton, S. A.: The conduct of in vitro and in vivo drug-drug 59–98 (2003). interaction studies: a pharmaceutical research and manufac- 38) Shitara, Y. and Sugiyama, Y.: Pharmacokinetic and phar- turers of America (PhRMA) perspective. Drug. Met. Dispos., 31: macodynamic alterations of 3-hydroxy-3-methylglutaryl coen- 815–832 (2003). zyme A (HMG-CoA) reductase inhibitors: drug-drug interac- 27) Ito,K.,Iwatsubo,T.,Kanamitsu,S.,Ueda,K.,Suzuki,H.and tions and interindividual differences in transporter and metabol- Sugiyama, Y.: Prediction of pharmacokinetic alterations caused ic enzyme functions. Pharmacol Ther., 112: 71–105 (2006). by drug-drug interactions: metabolic interaction in the liver. 39) Shitara, Y., Itoh, T., Sato, H., Li, A. P. and Sugiyama, Y.: Inhibi- Pharm. Rev., 50: 387–411 (1998). tion of transporter-mediated hepatic uptake as a mechanism for 28) Maeda, K. and Sugiyama, Y.: In vitro – in vivo scale-up of drug drug-drug interaction between cerivastatin and cyclosporine A. transport activities. In You, G. and Morris, M. E. (ed.): Drug J. Pharmacol Exp. Ther., 304: 610–616 (2003). transporters, New Jersey, John Wiley & Sons, Inc., 2007, 40) Koizumi, H. M., Saito, M., Miyake, S. and Hasegawa, R.: The in- pp557–588. cremental effect and mechanism of cyclosporine on blood con- 29) Uyama, Y., Shibata, T., Nagai, N., Hanaoka, H., Toyoshima, S. centration of statins and statin package insert information in and Mori, K.: Successful bridging strategy based on ICH E5 Japan. Bull. Natl. Inst. Health Sci., 123: 37–40 (2005). guideline for drugs approved in Japan. Clin. Pharmacol. Ther., 41) Saito, M., Koizumi, H. M., Urano, T., Miyake, S. and Hasegawa, 78: 102–113 (2005). R.: A literature search on pharmacokinetic drug interactions of 30) Ministry of Health, Labour and Welfare: Ethnic factors in the statins and analysis of how such interactions are reflected in acceptability of foreign clinical data. notification No. 739 package inserts in Japan. J. Clin. Pharmacol. Ther., 30: 21–37 (Kyokuchou-Tsuuchi, in Japanese), Pharmaceuticals Medical (2005). Safety Bureau, Japan 1998. 42) Yoshida, N., Yamada, A., Mimura, Y., Kawakami, J. and Adachi, 31) Ministry of Health, Labour and Welfare: Ethnic factors in the I: Trends in new drug interactions for pharmaceutical products acceptability of foreign clinical data. notification No. 672 in Japan. Pharmacoepidemiol. Drug Saf., 15: 421–427 (2006). (Kachou-Tsuuchi, in Japanese), Pharmaceuticals Medical Safety 43) Saito,M.,Koizumi,H.M.,Miyake,S.andHasegawa,R.:Com- Bureau, Japan, August 1998. parison of information on the pharmacokinetic interactions of 32) Information for pharmacogenomics/biomarker consultation (in Ca antagonists in the package inserts from three countries Japanese). Pharmaceuticals and Medical Devices Agency, Japan. (Japan, USA and UK). Eur, J. Clin. Parmacol., 61: 531–536 Available from URL: http://www.pmda.go.jp/operations/shonin/ (2005). info/consult/taimen.html#2. Accessed January 7, 2010. 44) Marroum, P. J., Uppoor, R. S., Parmelee, T., Ajayi, F., Burnett, 33) Saeki, T., Ueda, K., Tanigawara, Y., Hori, R. and Komano, T.: A., Yuan, R., Svadjian, R., Lesko, L. J. and Balian, J. D.: In vivo Human P-glycoprotein transports cyclosporine A and FK506. J. drug-drug interaction studies – a survey of all new molecular Biol. Chem., 268: 6077–6080 (1993). entities approved from 1987–1997. Clin. Pharmacol. Ther., 68: 34) Wacher,V.J.,Wu,C.Y.andBenet,L.Z.:Overlappingsubstrate 280–285 (2000). specificities and tissue distribution of cytochrome P450 3A and