This page has been archived and is no longer updated
Making Better Drugs: Decision Gates in Non-Clinical Drug Development
Author: J. Fred Pritchard
Keywords
Keywords for this Article
Add keywords to your Content
Save
|
Cancel
Share
|
Cancel
Revoke
|
Cancel
Rate & Certify
Rate Me...
Rate Me
!
Comment
Save
|
Cancel
Flag Inappropriate
The Content is
Objectionable
Explicit
Offensive
Inaccurate
Comment
Flag Content
|
Cancel
Delete Content
Reason
Delete
|
Cancel
Close
Full Screen
"� 2003 Nature Publishing Group 542 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS The process of drug development The search for successful therapeutic interventions to treat disease and improve quality of life is evolving from managed serendipity to engineered selection. Recent advances in combinatorial chemistry, high-throughput screening, functional genomics and proteomics have fuelled an explosive proliferation of new chemical enti- ties (NCEs) that possess promising pharmacological properties. The challenge is to select and advance one or two compounds with properties that are predictive of good efficacy and safety in humans. Despite improve- ments in the number and quality of potential drug can- didates, drug development remains a highly difficult, costly and risky business. Success rests not only in the intrinsic qualities of the molecule, but also in how the development of the drug is planned and executed, and in the effective management of key resources: effort, time and expenditure. Drug development is a process that proceeds through several key go/no-go ?decision gates?, from the identification of a potential therapeutic candidate through to marketing a drug product (FIG. 1). A ?decision gate? is a useful analogy, because to pass through a deci- sion gate, an NCE must successfully meet a series of predetermined criteria, followed by another, usually larger, set of studies to answer questions relevant to the next decision gate. Two of the most important decision gates in the development of a therapeutic agent ask whether the drug works in humans, and whether the drug can be marketed. The first question is answered by a clinical proof-of-concept study in a small, but targeted, group of patients. Although it is not essential to establish statistical proof at this stage, sufficient evidence is needed to provide confidence that the therapeutic product is working as intended. Ultimately, the safety of human subjects is the predominant focus of early clinical research. The decision to allow a new therapeutic entity to be given to human subjects in a clinical research setting, or to allow marketing of a drug to a large population of patients, is shared by many stakeholders, including government health regulators, clinical investigators, ethics review committees, healthcare providers and the MAKING BETTER DRUGS: DECISION GATES IN NON-CLINICAL DRUG DEVELOPMENT J. Fred Pritchard*, Malle Jurima-Romet ? , Mark L. J. Reimer ? , Elisabeth Mortimer � , Brenda Rolfe ? and Mitchell N. Cayen || Drug development is a risky business. Success or failure often depends on selecting one or two molecules for development from many choices offered by the engines of high-throughput discovery. A lead candidate needs to possess adequate bioactivity, appropriate physical?chemical properties to enable formulation development, the ability to cross crucial membranes, reasonable metabolic stability and appropriate safety and efficacy in humans. Predicting how a drug will behave in humans before clinical testing requires a battery of sophisticated in vitro tests that complement traditional in vivo animal safety assessments. This review discusses how to strategically identify which non-clinical studies should be performed to provide the required guidance and comfort to stakeholders involved in clinical drug testing. *MDS Pharma Services, 3504 Proprietor Way, Raleigh, North Carolina 27612, USA. ? MDS Pharma Services, 2350 Cohen Street, St. Laurent, Montreal, Quebec, Canada. � MDS Pharma Services, Les Oncins, BP118, 69593 l?Arbresle, Cedex, France, || Cayen Pharmaceutical Consulting, 98 Autumn Ridge Road, Bedminster, New Jersey 07921, USA. Correspondence to J. F. P. e-mail: fred.pritchard@mdps.com doi:10.1038/nrd1131 A GUIDE TO DRUG DISCOVERY � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 543 REVIEWS and to increase the chance of picking a winning thera- peutic molecule. When resources are limited, the ques- tion of what knowledge is needed and when ? the ?decision gate? analogy ? becomes the principal strategic consideration in determining the priority of activities during early drug development. Once a compound has been selected for develop- ment, regulatory guidances have a large influence on the conduct of non-clinical studies. As the administration of experimental drugs carries real risks to the human study subjects, certain non-clinical studies must be done according to government-regulated standards so that the data can be trusted by all of the stakeholders involved in clinical research 1 . In recent years, through the efforts of the International Conference on Harmonization (ICH), common global requirements have been largely agreed on 2 . Even though regulatory guidances define the normal expecations of what knowledge is needed to proceed to human testing or marketing, flexibility exists in the actual planning and sponsoring organization itself. The primary goal must be to establish what is best for the patient. Deciding if an NCE can be administered to humans for the first time is profoundly influenced by preclinical studies performed in animals that address the toxicology and exaggerated pharmacology of a drug product. During the latter stages of drug development, longer-term safety studies in animals are necessary before marketing approval can be obtained. The science of non-clinical drug development has been dramatically affected by the emergence of sensi- tive analytical instrumentation and by the application of molecular genetics to enable the detection of changes in the expression of human proteins that could affect, or be affected by, new drug candidates. These new tools generate vast amounts of information and data that require active management. Increasingly, more knowledge about the characteristics of a drug is expected by decision makers at each phase of drug development in order to reduce risk to human subjects Therapeutic target identification Are targets validated? Clinical product manufacturing Lead identification Are leads identified? Lead optimization studies (non-GLP toxicology, metabolism, pharmacology) Is lead developable? Preclinical studies (GLP toxicology, ADME, safety pharmacology Give to humans? Clinical safety, tolerance, PK and proof of concept studies Scale-up chemical synthesis and formulation process Clinical dose optimization, safety and efficacy studies Genomics/ genetics Proteomics Natural products Biotechnology Synthetic and combinatorial chemistry Approved for marketing? Product launch Postmarketing studies/long-term evaluation Optimize product life cycle/new formulations and indications Does drug sell? Does drug work in humans? Clinical proof of concept Figure 1 | Key decision gates in drug development. Gear symbols represent the key go/no-go decisions for development of a pharmaceutical product. Boxes represent key activities or disciplines involved in the research necessary to answer the questions generated at each decision gate. Asking the right questions at the right time is the key to strategic drug development planning. ADME, absorption, distribution, metabolism and excretion; GLP, Good Laboratory Practice; PK, pharmacokinetics. � 2003 Nature Publishing Group 544 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS Decision gates in drug development The balance of this paper provides an overview of the types of approaches taken today to provide the right information for the decisions that need to be made in non-clinical drug development. Three decision gates are addressed: lead optimization, first-in-human studies and marketing decisions. In addition, the dis- cussion focuses primarily on small-molecule drugs (that is, drugs with a molecular mass (MM) < 1,000), but also recognizes that larger therapeutic modalities alter the types of questions that need to be addressed. To be truly strategic in considering non-clinical drug development studies, regardless of the size of the mol- ecule, the question that must always be addressed is what real predictive value do these studies add to our understanding of the efficacy and safety of drug can- didates in humans? Decision gate: can this drug be developed? Smart drug development means eliminating potential drug candidates that are likely to fail in clinical trials early in their development. Typical questions addressed during the lead optimization phase are focused on determining whether there are any properties of the drug candidate that would make future development very difficult (TABLE 1). ?Fail fast and cheap? are the watchwords at this stage. Although each drug develop- ment plan is unique, a typical flowchart of lead opti- mization activity is shown in FIG. 2, and key aspects of these studies are discussed below. To this end, several approaches have been devel- oped to help understand how drugs behave at the sev- eral barriers that exist between dosing a drug and the target of drug action 6,7 . These include permeation across epithelial membranes such as the gastrointesti- nal mucosa, drug metabolism, plasma protein bind- ing and transport into and out of tissues, especially organs that eliminate drug products, such as the kidney design of a drug development programme. No one per- son can bring focus and expertise to all of the scientific and regulatory aspects of drug development. Therefore, the planning and execution of a programme of work is usually overseen by a team of committed scientists led by a ?champion? who is able to leverage the resources needed to complete the work in a timely way. It is estimated that only 1 out of 5,000 screened com- pounds is approved as a new medicine 3 . A significant number of potential drugs in development fail because of unacceptable animal toxicity. Toxicity can be defined as any unwanted effect on normal structural or functional integrity, including unwanted or overly exaggerated phar- macological effects. Although most drugs will express unacceptable toxicity at some level of dosing, acceptabil- ity is determined by the degree of separation between the targeted pharmacological dose and the doses or expo- sures in animals when the first truly toxic signs develop. It is not sufficient just to know what the drug does to the body (that is, its pharmacology and toxicology); it is also crucial to know what the body does to the drug. Knowledge of the absorption, distribution, metabolism and excretion (ADME) properties of the drug and its metabolites in humans, and animals used in toxicology assessments, is crucial to understanding differences in effect among species and for optimizing drug dosing 4,5 . For orally administered drugs, adequate absorption and bioavailability must be achieved. Absorption refers to the amount of total drug-derived material that crosses the gastrointestinal epithelium, whereas bioavailability addresses the amount of biolog- ically active material that reaches the systemic circula- tion. Duration of drug action is often dependent on how rapidly the body eliminates the active molecules, either through chemical modification (metabolism) by the action of drug-metabolizing enzymes or by binding and transport away from the biologically active sites and excretion from the body. Table 1 | Drug candidate selection: key questions and pivotal studies Questions Studies that provide answers Can the drug candidate be measured and is it stable Bioanalytical assay development. in biological matrices? Does the drug candidate have reasonable metabolic In vitro metabolism studies using animals and human stability? What are the metabolites and are they hepatocytes, microsomes or human expressed enzyme active, possibly even a better drug candidates? systems and analysis of incubates using LC/MS/MS Are there species differences in metabolism? or LC/NMR/MS. Synthesis and pharmacology testing of metabolites. Does the drug have sufficient oral bioavailability and Pharmacokinetic studies in rodent and non-rodent persistence in the bodies of animal models? species after oral versus intravenous administration. Is the drug mutagenic or cytotoxic in vitro? Ames bacterial mutagenicity assay. Mammalian cell (for example, mouse lymphoma) mutagenicity assay. What is the maximum tolerated dose (MTD) and Rising single- and repeat-dose escalation toxicology study dose-limiting toxicity? in rodent and non-rodent species until limiting toxicity is observed. Can active pharmaceutical ingredient (API) be Chemical synthesis process assessment. Assay synthesized at reasonable cost? Is the API stable after development for API purity. Chemical stability synthesis? assessment. Generation of API certificate of analysis. Can the drug be formulated for use in animal toxicology Pre-formulation development testing. Assay development studies and early human studies? for purity and content formulated product. LC, liquid chromatography; MS, mass spectrometry; NMR, nuclear magnetic resonance. � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 545 REVIEWS cells are a human intestinal cell line that are grown to form a monolayer on a filter separating two microwell chambers. Drug candidates can be added to one side of the membrane and their permeation followed. Active and passive transport processes can be studied. The Madin?Darby canine kidney (MDCK) cell model is a practical alternative to Caco-2 cells, in that continuous monolayers of cells can be grown in a just a few days 13,14 , rather than the three weeks required for Caco-2 cells. However, MDCK cells are less reflective of the intestinal epithelium and cannot be used for bi-direc- tional transport studies. Increasingly, non-cell-based approaches, such as the parallel artificial membrane permeability assay (PAMPA), are gaining popularity for assessing passive permeability 15,16 . and liver. Arguably, the most important of these tech- niques in the prediction of the safety and efficacy of a drug are metabolism and oral absorption, whereas protein-binding characteristics are less significant 8 . In vitro ADME approaches are being applied to drug candidates both in the pre-selection (that is, discov- ery) process and earlier in development, as they are essential for identifying and eliminating compounds likely to present safety challenges in the later stages of drug development. Is the drug absorbed? The Caco-2 cell permeability assay is one approach that has been widely adopted for understanding the gastrointestinal drug absorption process 9?12 . Caco-2 Pharmaceutical development assessment: solubility, stability and synthesis MDCK and/or Caco-2 cell permeability Ames bacterial mutation Bioanalytical method validation Inhibition of CYP450 enzymes Crossvalidation ? rat plasma Single intravenous/oral dose pharmacokinetics in rat Crossvalidation ? dog plasma hERG-K + conductance Single intravenous/oral dose pharmacokinetics in dog Single (acute) dose intravenous in mouse (MTD) Single (acute) dose intravenous in rat (MTD) Identify new lead compound Repeat intravenous dose range finder and toxicokinetics in rat Repeat intravenous dose range finder and toxicokinetics in dog Is lead developable? Progress preclinical development Mouse lymphoma assay In vitro metabolism, metabolic stability in hepatocytes Yes No Toxicology studies Bioanalysis Metabolism Pharmaceutical development Safety pharmacology Lead identified Figure 2 | Potential flowchart for assessing whether a small-molecule new drug candidate is developable. CYP450, cytochrome protein 450; hERG, human ether-a-go-go; MDCK, Madin?Darby canine kidney; MTD, maximum tolerated dose. � 2003 Nature Publishing Group 546 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS TORSADES DE POINTES This is a form of polymorphic ventricular tachycardia that is preceded by a prolongation of the QT interval. Although this condition is found in many clinical settings, it is mostly induced by drugs and drug interactions that prompt a long QT syndrome. enzymes 20,21 . In vitro preparations of human liver, including microsomes, hepatocytes and liver slices, as well as human complementary-DNA-expressed enzymes, now complement similar preparations for the principal toxicology species 22,23 . Therefore, metabolic stability can be readily tested in vitro across several species, enabling the comparison of the metabolism of the drug in animals with that in humans. In addition, by employing modern liquid chromatography/tandem mass spectrometry (LC/MS/MS) and LC/MS nuclear magnetic resonance (LC/MS/NMR) technology (BOX 1), metabolites can be quickly identified from the incubates of these experiments 24,25 . The next crucial question to be addressed early in drug development is which human drug-metabolizing enzymes are responsible for the metabolism of the drug. Many enzymes can contribute to the overall metabo- lism of a drug, and some are genetically polymorphic and/or associated with high inter-individual variability in expression levels 26?28 . From a safety perspective, it is important to identify, and perhaps eliminate, drugs from further development if they are subject to poly- morphic metabolism or to extensive metabolism by key human enzymes, such as CYP3A4/5, CYP2D6, CYP2C9, CYP1A2 and CYP2C19 (REF. 29). These five enzymes are the predominant drug-metabolizing enzymes in human liver, and CYP3A4 is particularly significant as it has a role in the metabolism of approximately 50% of all small-molecule drugs in use today 30 . Extensive metabolism by a single enzyme such as CYP3A4 can lead to adverse drug?drug interactions when concomi- tant drug therapies require the same enzyme for their metabolic clearance. Individuals with inherited defi- ciencies in one or more key CYP enzymes are at increased risk for developing drug-related toxicities, or in the case of drugs requiring metabolic activation, experiencing a lack of therapeutic efficacy. Some of these genetic deficiencies in drug-metabolizing enzymes are relatively common: for example, 5?10% of Caucasians are poor metabolizers of CYP2D6, and 13?23% of people from South East Asia and Japan lack functional CYP2C19 enzyme. In vitro studies employ- ing human liver microsomes or expressed enzymes, and selective chemical inhibitors or antibodies, are valuable tools for providing early information on human metab- olism of a new drug that is crucial in predicting poten- tial clinical drug?drug interactions and the impact of genetic polymorphisms. Is the drug too toxic? Not so long ago, the toxicologist would only become involved in drug development when the discovery research scientist handed over one precious candidate drug and got busy discovering the next. Nowadays, toxicology and safety pharmacology are involved much earlier and play an increasingly important role in the decision of whether to move a lead candidate into reg- ulated in vivo toxicology studies 31 . In vitro genotoxicity assays are mandatory regulatory studies designed to detect potential mutagens and/or carcinogens. These assays are often performed during Is the drug metabolized? Metabolism is a key determinant of the in vivo fate of a drug candidate molecule. Too rapid or extensive metabolism to pharmacologically less active or inert metabolites and a compound will lose its therapeutic efficacy. Metabolism can lead to pharmacologically active metabolites, frequently with properties different from those of the parent drug. For example, the non- sedating antihistaminic drug terfenadine (Seldane; Marion Merrell Dow), was associated with TORSADES DE POINTES ventricular arrhythmias and was consequently withdrawn from the market, whereas its pharmacologi- cally active metabolite fexofenadine (Allegra; Aventis), is devoid of cardiotoxicity. It is often important to define active metabolites early in development, to understand their relevance to toxicology and to fully protect the intellectual property value of the drug. Owing to the importance of metabolism in interpret- ing toxicology findings 17 , in vitro metabolism studies normally precede in vivo preclinical safety assessment 18,19 . Not surprisingly, inter-species differences exist in drug metabolism, in particular among the cytochrome P450 (CYP450) superfamily of drug-metabolizing Box 1 | LC/MS/MS analytical technologies enable early ADME testing Liquid chromatography coupled with tandem mass spectrometric detection (LC/MS/MS) is the analytical method of choice in early absorption, distribution, metabolism and excretion (ADME) testing. The atmospheric pressure ionization interface now standard on modern LC/MS/MS systems enables rapid method development coupled with excellent sensitivity, specificity and high sample throughput. The quantitative selectivity afforded by selective reaction monitoring on a triple quadrupole instrument precludes the need for high chromatographic resolution or extensive sample clean up. Analytical throughput can be further maximized by using automated sample-processing techniques, such as on-line column switching 72 ,combined with high-sample-density microtiter plates. Modern LC/MS/MS also offers limits of detection extending down to the sub-nanogram per ml range using only minimal quantities of biological matrix (for example, 0.025 ml of plasma). Not only does LC/MS/MS enable rapid and sensitive quantitation of new drug candidates, but it also provides important structural information on metabolites 73 .A full- scan LC/MS analysis can initially suggest possible oxidative and/or conjugative metabolic transformations on the basis of the ionic species observed. In the MS/MS mode, the instrument can be tuned to a selected precursor ion of interest, which is then further fragmented to form product ions that uniquely identify the metabolite (product ion scan). Selectivity has been further enhanced by the quadrupole ion trap, a device that ?traps? ions in a space bounded by a series of electrodes 74 .The unique feature of the ion trap is that an MS/MS experiment (or, in fact, multi-step MS n experiments) can be performed sequentially in time within a single mass analyser, yielding a wealth of structural information. Hybrid quadrupole-time-of-flight (Q-TOF) LC/MS/MS systems are also popular for the characterization of metabolite profiles 75 . The configuration of a Q-TOF results in an instrument capable of high sensitivity, mass resolution and mass accuracy in a variety of scan modes. Looking ahead, liquid chromatography coupled with nuclear magnetic resonance spectroscopy (LC-NMR) 76 provides a way of confirming absolute molecular configurations. A new linear ion-trap mass spectrometer possessing significantly enhanced product ion-scanning capabilities, while retaining all of the scan functions of a triple quadrupole MS, has recently been introduced 77 . The ultra-high resolution and sensitivity of Fourier transform ion-cyclotron resonance MS (FT-ICR MS) 78 holds great promise for the analysis and characterization of biological mixtures. Finally, continued advancements in data processing and interpretation software packages should enable the analytical scientist to more efficiently identify and quantify metabolites throughout the drug development process. � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 547 REVIEWS MINI-PIG A small species of pig weighing about 20?45kg. primate, on the basis of relevant pharmacological knowledge (the species should respond pharmacologi- cally to the drug), MTD, and metabolism information (as close as possible to human). The MINI-PIG is becoming a useful toxicology species for dermal products and car- diovascular agents, just as pig skin and heart are good models of the corresponding human organs. What happens to the drug after dosing? Oral bioavailability, plasma half-life, clearance and, to a lesser extent, volume of distribution and mean residence time, are pharmacokinetic parameters that are typically determined from plasma drug concentration profiles in a rodent and non-rodent species following oral and intravenous administration 42 . The area under the ?plasma concentration versus time? curve (AUC), which measures time-averaged systemic exposure of the body to the drug, is very helpful in comparing systemic expo- sure of animals used in toxicology testing with humans. Finally, measurement of a drug and its principal metabolites in urine can help identify the contribution that renal clearance and metabolic clearance make to the overall elimination of the drug from the body. Pharmacokinetic studies, properly planned and exe- cuted, are rich sources of information and are therefore among the first studies performed in drug development. Pharmacokinetic analysis is dependent on the devel- opment of a reliable bioanalytical method to measure parent drug and relevant active metabolite(s) in plasma, and possibly urine, samples. As these assays are usually used to support systemic exposure relationships in the species used for toxicology studies, the methods must meet agreed guidelines that support GLP 43 . A fully vali- dated assay has a defined quantitative analytical range, meets precision and accuracy expectations for the type of analysis chosen (for example, high-pressure LC (HPLC), LC/MS, radioimmuno assay (RIA), enzyme- linked immunosorbent assays (ELISA)), and has proven ruggedness across a period of repeated use. In vivo pharmacokinetic studies are resource inten- sive. Therefore, to speed up the process during lead selection and optimization, methods such as ?cassette? dosing have been used 44,45 . This approach involves sev- eral potential lead compounds that are simultaneously administered to each animal, and pooling of samples to reduce analysis times using LC/MS/MS. Choice of com- pounds is important to avoid isobaric mass ions that would complicate bioanalysis. Also, results can be skewed by unintended drug?drug interactions. Nevertheless, this is a popular approach for gathering good pharmacokinetic information on several com- pounds quickly in early drug development. Decision gate: safe for humans? Timing and costs are always issues in drug develop- ment planning. Typically, the execution of the studies in FIG. 2 could take four to six months and cost US $500,000. However, costs can triple, and timing dou- ble, in order to answer the questions that need addressing before making the decision to proceed to human testing (TABLE 2, FIG. 3). The rate-limiting step is lead optimization, but mini-versions of these have also been applied as early drug selection screens. Any statisti- cally significant increase in mutation rate during such studies is often a reason to abandon further develop- ment of the drug candidate. There is growing use of the human ether-a-go-go K + (hERG-K + ) conductance assay using Chinese hamster ovary (CHO) cells during lead optimization 32,33 . This in vitro electrophysiology study can be used to screen out those candidates likely to cause QT interval prolonga- tion. Prolongation of cardiac repolarization has been associated with the specific and potentially fatal tachycar- dia Torsade de Pointes in humans, and usually precludes further development of the compound 34 . Increased understanding of individual differences and disease-related mechanisms will undoubtedly improve the selectivity of lead candidates. As relational databases reveal genomic and proteomic differences between the disease versus normal states, and between genomic and proteomic responses to different drugs and reference xenobiotics, this should lead to better biomarkers that will predict relevant human toxicities during non-clinical drug development. The term ?toxi- cogenomics? now refers to the application of genomics and proteomics to drug-safety assessment and has been the subject of several recent scientific work- shops 35,36 . In the future, drug developers will attempt to match the genomic responses of lead candidates to the individual genotypes. For example, trastuzumab (Herceptin; Genetech) is only effective for women who carry multiple copies of the ERBB2 (also know as HER-2/neu) gene 37 . What is the best species for toxicology? The primary goal of early toxicology studies is to find a dose that produces toxicity in one or more rodent or non-rodent species 38?40 . Acute toxicity studies in two species (often mouse and rat) inform what happens after a single, very high dose. This knowledge is useful not only to help select a safe starting dose for humans, but also in providing worker-safety data for those employees faced with handling a new chemical sub- stance. It also forms the basis for overdose information for future human trials. Various dose-escalation study designs exist to quickly predict what would be the maximum tolerated dose (MTD) in the putative toxicology species. In addition, it is important to identify early on the target organs of toxicity, as this can affect future study design. Finally, there are often large inter-species differences in MTD. For example, the dog is known to be very sensi- tive to compounds that cause gastrointestinal irrita- tion, and non-human primates are better predictors of effects in humans. On the basis of studies of the metabolic profile of the drug across species, the rodent and large animal species are selected for formal toxicology evaluations under Good Laboratory Practice (GLP) procedures 41 . As a large body of data exists for the albino white rat, this species is virtually always chosen as the rodent species. The large animal species is typically the dog or a non-human � 2003 Nature Publishing Group 548 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS the maximum no-effect level dose or exposure is at least tenfold that of the expected pharmacological exposure in rodents, and at least sixfold that in non- rodents, then it is generally believed that the compound can be safely managed in humans during clinical dose- escalation safety and tolerance testing. Related consid- erations are whether unique toxicological conse- quences of drug administration can be separated from exaggerated pharmacology produced by overdosing animals, and whether there are inter-species differ- ences in the toxicological response. A final focus, espe- cially for larger-molecule drugs, is whether the drug affects the immune system 51 . New technology (such as surface plasmon resonance and cell-based bioassays) is available to determine whether antibodies have formed or if immune system biomarkers have changed during toxicology testing. Toxicology criteria can be relaxed if the new drug entity is designed to treat a previously unmet medical need or will be administered to a very ill patient popula- tion in which there is little, if any, hope for alternative therapy. Also, there is no point studying repeated-dose toxicity of a drug in male animals when it is a product for women?s health and vice versa. Toxicology testing for products that are delivered directly to their site of action (for example, inhalation of an antibronchospasmodic agent, or dermal application of a treatment for a skin ailment such as psoriasis) require special thinking in order to design an appropriate test for both potential systemic as well as local exposure. What is the systemic exposure of the drug? Blood and sometimes urine samples are collected at vari- ous times during a study from either toxicology animals on study or from ?satellite? groups of animals dosed and housed with the toxicology animals. These are analysed for drug and any known active metabolites and pharma- cokinetic parameters (such as C max , AUC and half-life) determined as measures of systemic exposure over time. As these values are then associated with any observed often the completion of the enabling toxicology stud- ies; however, sourcing drug substance or preparing an appropriate formulation can also determine the pace of this phase of drug development. Full ADME char- acterization in the toxicology species is not required before human testing, although an understanding of the fate of the parent drug and known active metabo- lites and their distribution is helpful, and even crucial, for certain drugs. Toxicokinetics (pharmacokinetics applied to samples taken from animals dosed during GLP toxicology testing) can provide key evidence of exposure versus response that can sometimes be extrapolated to man 46 . There are several questions that are optional for this decision gate, depending on the intent of the early clinical research programme (TABLE 3). Before repeated doses can be administered to women of child-bearing age, Segment I and II reproductive toxicology studies need to have been completed 47 . In vitro studies that evaluate enzyme inhibi- tion or induction potential can also be crucial elements of a plan, especially if drug?drug interactions of this nature plague other drugs in the same therapeutic class 48 . Finally, in vivo genotoxicity testing is done if there are any questions arising from earlier in vitro work. What toxicity occurs following repeated dosing? Repeated-dose toxicology studies are designed to mimic the human dosing regimen for the first repeat- dose clinical tolerance and efficacy trials, and, depending on the planned initial clinical programme, a minimal length of two weeks and a maximum of three months are required. Studies in both rodent and non-rodent species are usually required by regulatory reviewers. It is important to know how the toxicity changes with dose: does it change with time during the study, and does toxicity differ between male and female animals? Some studies include a design to establish whether the toxicity is reversible 49,50 . The ultimate question that must be answered by the toxicology studies is what the safety margin is. If Table 2 | Human safety indicators: key questions and pivotal studies Questions Studies that provide answers Can the drug be reliably prepared and formulated according Formulation development. Define release GMP to Good Manufacturing Practice (GMP) guidelines that testing criteria for content and purity. Define regulate drug material prepared for human use? formulation stability. What is the maximum no-effect dose following repeated Fourteen-day to three-month toxicology studies in a dosing? Is there dose-related exposure? What organs rodent and non-rodent species with toxicokinetics. are affected by repeated dosing? What is the safety margin? Ratio of maximum no-effect dose or exposure over expected pharmacological dose or exposure. What enzymes are involved in the drug?s metabolism? In vitro drug metabolism studies using human micro- somes and/or cytochrome protein expression systems. How long and by what routes is parent drug eliminated? Pharmacokinetics in rodent and non-rodent species. Radiolabelled drug-excretion-balance studies Does the drug produce any cardiovascular effects or In vivo telemetry studies in animals evaluating alterations cardiac conductance? in cardiac electrophysiology and cardiovascular vital signs. Action potential duration using isolated rabbit Purkinje fibers. Are there any effects on behavior or pulmonary function? Irwin behaviour test in rats. Rat pulmonary function evaluation. � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 549 REVIEWS There is renewed interest in sparse sampling paradigms that can be applied to assess systemic exposure more intensively in the actual animals used in toxicology assess- ments. Usually, removal of anything but small amounts of blood will compromise the integrity of the toxicology study, hence the need for satellite groups of animals. Modern LC/MS methodology has dramatically improved the lower limits of assay sensitivity to low picogram per millilitre ranges, thereby making small sample volumes (10?50 �l) practical. Sparse sampling data sets fit nicely with population-based pharmacokinetic analyses. toxicology findings, the term toxicokinetics is now applied to this form of analysis. C max and AUC can be used instead of dose to estimate the safety margin and provide guidance for early dose-escalation studies in humans. Toxicokinetic assessments can also show whether systemic exposure is affected by the presence of food at or near the dosing time, by the administration of drug relative to the light/dark cycle (rodents are nocturnal creatures) and by the duration of administration or by the route of dosing. Yes Plasma protein binding Excretion balance and metabolite identification ? rat Initiate early clinical program Is lead safe to give to humans? Excretion balance and metabolite identification ? dog or monkey Identifiy cytochrome P450 enzymes Toxicity assessment in rats after four weeks exposure and toxicokinetics Formulation development and stability ? establish clinical trial material release specifications Toxicity assessment in dogs or monkeys after four weeks exposure and toxicokinetics Quantitative tissue distribution In vivo bone marrow micronucleus test in rodents Reproductive toxicology studies No Cardiovascular safety ? telemetry in conscious animal and pharmacokinetics/pharmacodynamics Irwin test of behaviour and body temperature Normal pulmonary function ? rat Lead is developable Toxicology studies Metabolism Formulation Safety pharmacology Stop Figure 3 | Potential flowchart for assessing whether a small-molecule new drug candidate is safe to give to humans. � 2003 Nature Publishing Group 550 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS GAMMA SCINTIGRAPHY Gamma scintigraphy is a non- invasive method of examining the deposition of a compound in the test subject's body using standard radiolabelling techniques. A computerized image of ?-emissions displays overall product deposition, including differentiation of test article concentration. POSITRON EMISSION TOMOGRAPHY (PET). A method for imaging that measures changes in blood flow associated with brain function by detecting positrons emitted by radioactively labelled substances that have been injected into the body. tissue-distribution characteristics of a drug in humans. Although this information is not essential before early clinical studies, it is an expectation for marketing drug applications (for example, New Drug Applications or Marketing Authorization Applications). These studies are conducted by administering the radioactive drug, and the measurement of radioactivity in up to 60 tissues by either direct counting or sample combustion or by a whole-body scanning technique called whole-body autoradiography. GAMMA SCINTIGRAPHY and POSITRON EMIS- SION TOMOGRAPHY (PET) have also been applied to animal tissue distribution studies 55,56 . By including such radiosensitive tissues as gonads, thyroid and eyes, radioactive dosimetry is calculated that estimates the radiation exposure to each type of tissue. Dosimetry calculations are required before a radiolabelled drug can be given to humans, an approach that is often consid- ered to define human ADME for market registration. Plasma-protein and red-blood-cell binding is another key determinant of drug distribution and dispo- sition, although it is rarely a factor in assessing safety and efficacy 9,57 . High plasma-protein binding is an issue for some drugs, such as anti-infective agents, for which a target concentration of unbound drug sufficient to kill the microorganism is the goal of therapy. However, high binding can also stabilize or solubilize a drug product in blood. Moreover, if the affinity for the receptor or enzyme target is higher than the affinity for the binding proteins, then an effective system exists for delivering the drug to the target without it dispersing throughout the entire body fluid compartment. Efflux and uptake transporters present in many tissues of the body, including the intestine, liver, kidney and brain, can significantly influence the absorption and distribution of many drugs. One efflux protein in particular, P-glycoprotein (P-gp or MDR1), has been well studied, and in vitro models have been developed for identifying compounds that are P-gp substrates 58 . Because the multidrug resistance that develops towards many cancer chemotherapeutic agents is attributed to overexpression of P-gp in tumour cells, Does the drug affect vital functions? Safety pharmacology studies are designed to assess any effect on vital functions. The three studies expected by regulators, review boards and investigators involved in approving early clinical studies address the cardiovascu- lar, central and peripheral nervous systems, and the respiratory system. Recent regulatory consensus places a great deal of emphasis on the in vivo cardiovascular pharmacology safety study to provide assurance that the NCE does not cause QT interval prolongation 52 . Nevertheless, although in vitro screens are reasonably effective in eliminating compounds that change cardio- electrophysiological properties, these can no longer form the sole preclinical assessment of cardiovascular safety. What are the ADME characteristics of the drug? How does the body dispose of the drug? To collect infor- mation about how much and how fast a small-molecule drug product gets into and out of the body of animal species used in toxicology testing, an appropriately radio- labelled (and sometimes stable-labelled) drug is admin- istered and the amounts of radioactive excreta and blood quantified 53,54 . Ideally, if >90% of the originally adminis- tered radioactive dose can be recovered, then the profile of how fast and by what routes a drug and its metabolites are excreted in that species will be known. The samples can also be further analysed for metabolites that would better define how much of the total radioactivity mea- sured in each sample is attributable to the parent drug or its various metabolites. These data can be extremely use- ful in associating presence of parent drug or metabolite with toxicological or pharmacological effects. Where is the drug distributed? An important feature to know for many new drug candidates is where the drug goes in the body and how long it stays in specific tissues or organs. These data can then be related to tissue-specific pharmacology or toxicology. In vivo rodent models are usually used to quantify amounts of drug distributed into tissues, as results from such studies are generally believed to reflect Table 3 | Requirements for additional work: key questions and pivotal studies Questions Studies that provide answers Where does the drug go in the body; how long Tissue-distribution studies in rats using radiolabelled drug. does it or its metabolites stay there, and by which Whole-body autoradiography, usually in rats. routes are it and its metabolites excreted? Is there a potential for untoward clinical Cytochrome protein inhibition kinetic studies. Induction studies drug-drug interactions (for example, enzyme with human cultured hepatocytes. Effects on reporter gene inhibition or induction)? constructs. Does the drug bind to plasma proteins and In vitro studies in plasma or serum of relevant species including erythrocytes? If so, how does this affect human that has been seeded with radiolabelled drug and free drug interpretation of pharmacokinetic data separated from bound drug using equilibrium dialysis, ultrafiltration, derived from concentration measurements or chromatography. In vitro studies in blood seeded with of parent drug in plasma? radiolabeled drug and erythrocytes separated by centrifugation. Does the drug affect reproductive performance Segment I reproductive toxicology studies in rats. in female rats? Is there evidence of teratogenicity, mutagenicity Segment II reproductive toxiciology studies in rats and rabbits. or embryo toxicity in vivo in rodents? Does the drug irritate the gastrointestinal tract Injection site irritation studies. Gastrointestinal motility and gastric or other sites of administration? irritation studies in rats. � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 551 REVIEWS particular surrogate markers of toxicity to monitor during the clinical trials. For example, a common sce- nario is to pay particular attention to increases in cir- culating liver enzymes in the human studies on the basis of hepatotoxicity observed at high doses in ani- mals. Once the clinical trials are underway, the animal toxicologist is often ignored in setting up and evaluat- ing clinical safety databases for a new drug product. This crucial dialogue has the potential for significant learning and synergy between the non-clinical and clinical drug development scientists who specialize in drug safety assessment. Once the compound enters into longer-term clinical studies, additional highly regulated toxicology studies become necessary (TABLE 4). These include longer-term repeat-dose studies lasting up to six months in rodents, and nine or twelve months in non-rodents, again mim- icking the duration of dosing planned for clinical trials. Standardized reproductive toxicology tests (Segments I and II if not done earlier and then Segment III) are required to enable long-term administration to women of child-bearing potential. Often, the significance of metabolites is not appreci- ated in early studies but is known by the time these long-term toxicology studies are initiated. It is expected that the contribution of any major and potentially active metabolite will be assessed during toxicokinetic evalua- tions of these studies and submitted as part of the marketing registration package 67 . Carcinogenicity studies examine whether the drug is carcinogenic when given for a lifetime 68,69 . These ?bioassay? studies typically involve rat and mouse species, dosed for their lifetime of around two years. These studies are large, long and expensive, and tend to be performed later in the life of a drug candidate. Despite best efforts at trying to eliminate potentially carcinogenic compounds as early as possible, some still occasionally prove positive at this stage. Recently, a compound targeting the peroxisome proliferator- activated receptor-? (PPAR-?) family was shown to be positive in the carcinogenicity bioassay study 70 , caus- ing the drug to be dropped during Phase III trials after many millions of dollars had been spent on the drug development programme. (In July 2002, Novo Nordisk suspended clinical development of its dual- acting insulin senzitiser ragaglitazar (NN622) because of findings of urine bladder tumours in mice and rats treated with the drug.) carcinoma-derived cell lines such as Caco-2 can be used for examining the interaction of P-gp with drug candi- date compounds 59 . Stably transfected cell lines express- ing important efflux and uptake transporters are being developed for implementation in drug discovery and development screening programmes. Will the drug affect or be affected by other drugs? Compounds that rely on only one metabolic pathway for elimination, or which are potent inhibitors (or, less importantly, inducers) of a key CYP enzyme, would be at high risk for adverse drug?drug interactions in human clinical studies. The kinetics of human CYP enzyme inhibition can be readily measured in vitro. Coupled with information on the putative therapeutic drug concentration range in plasma, the impact of CYP inhibition in vivo can be estimated 60 . Although a drug might inhibit metabolizing enzymes, it can also induce the synthesis of a new func- tional protein, resulting in increased clearance of co- administered drugs, which reduces their efficacy. For example, the herbal remedy St John?s Wort contains hyperforin, a chemical that induces the CYP3A4 enzyme and which has been associated with reduced efficacy of many clinically important drugs 61 . Rat enzyme induction is not predictive of human induction, so in vivo or ex vivo animal studies are waste- ful unless they explain changes in animal toxicity after repeated dosing. Human-based in vitro models designed to investigate enzyme induction use longer- term monolayer cultures of human hepatocytes, and the subsequent measurement of messenger RNA, pro- tein expression or enzyme activity after exposure to the test drugs 62 . However, limited availability of fresh human liver, and complex isolation procedures, has led to poor reproducibility with this approach. Recent advances in the understanding of the induction mecha- nisms for the CYP3A and CYP1A families in particular have enabled the development of higher-throughput and more reliable induction assays that use reporter- gene constructs 63?66 . Are additional studies required? An understanding of the safety margin in animals and of the pharmacokinetic properties of the drug in these species forms the rationale for identifying a suitably safe starting dose for first-time-in-human studies. This is also a good time to advise the clinician of any Table 4 | Market readiness: key questions and pivotal studies Questions Studies That Provide Answers What toxicology arises following long-term Six-month to one-year toxicology studies in a rodent and administration of the drug? non-rodent species. Is the drug carcinogenic? Lifetime exposure studies in rat and/or mouse. What is the complete metabolic fate of the drug? Is Characterization of complete metabolic profile in humans there a contribution to efficacy or toxicity by active and animal species used in toxicology testing. Radiolabelled drug metabolites? drug often used to trace drug products. What potential drug?drug interactions can be excluded Identification of drug-metabolizing enzymes responsible based on knowledge of the drug?s interaction with human for new chemical entity metabolism. In vitro drug?drug drug-metabolizing enzymes or membrane transporters? interaction studies using human enzymes. � 2003 Nature Publishing Group 552 | JULY 2003 | VOLUME 2 www.nature.com/reviews/drugdisc REVIEWS the therapies in development, and over a quarter of new medicines approved for marketing in 2002, are macro- molecules largely from biological sources, rather than small-molecule synthetic drugs 71 . Sometimes, the rationale for preclinical testing does not conveniently fit the properties of these larger molec- ular agents. For example, immunogenicity is a key safety issue for larger molecules, but how relevant are animal toxicity models when the antibody, peptide or recombi- nant protein has been specifically engineered to affect human targets? Moreover, the metabolism of protein products often results in the incorporation of amino acids into endogenous molecules. It is difficult to trace the fate of metabolites that themselves are normal body constitutents. Building a consensus on what predictive tests are appropriate at the preclinical stage for these molecules will be a future challenge for drug developers. Drugs are getting bigger In addition to the impact of technology on the process of building predictable and timely knowledge about a new drug candidate, new drug entities themselves have undergone a significant renaissance, for several reasons. First, as most relatively small molecules have already been patented, the majority of newer drug candidates have higher molecular masses, leading to greater chal- lenges in formulation and bioavailability as a result of their inherent lower solubility. Second, advances in our knowledge of receptors and enzymes have made drug targets more specific; for example, many drugs targeting the central nervous system are designed to bind to a specific brain receptor subtype without affecting other similar receptors. Third, significant advances in genetics and proteomics have led to a wide variety of novel potential therapeutic products. Approximately half of 1. Guidance for Industry: S7A Safety Pharmacology Studies for Human Pharmaceuticals. US Food and Drug Administration, July 2001 <http://www.fda.gov/cder/4461fnl.pdf> (2001). 2. ICH Final Draft Guidance for Non-Clinical Safety Studies for the Conduct of Human Clinical Trials for Pharmaceuticals: Proposal for revision. 8 November 2000. 3. Pharmaceutical Research and Manufacturers of America (PhRMA), Phrma FAX of the Week, May 2, 2003, R&D Process: Long, Expensive, Risky Road. 4. Lin, J. H. & Lu, A. Y. Role of pharmacokinetics and metabolism in drug discovery and development. Pharmacol. Rev. 49, 403?449 (1997). 5. Gunaratna, C. Drug metabolism & pharmacokinetics in drug discovery: a primer for bioanalytical chemists, part I. Curr. Sep. 19, 17?23 (2000). This articles provides a good introduction to the fundamentals of drug metabolism 6. Kapur, R. Smarter lead optimization enabled. Drug Discov. Dev. 61. (1999). 7. Lead optimization strategies. Genet. Eng. News 19, 16 (1999). 8. Benet, L. Z. & Hoener, B. A. Changes in plasma protein binding have little clinical relevance. Clin. Pharmacol. Ther. 71, 115?121 (2002). 9. Artursson, P. Epithelial transport of drugs in cell culture I: A model for studying the passive diffusion of drugs over intestinal absorptive (Caco-2) cells. J. Pharm. Sci. 79, 476?482 (1990). 10. Irvine, J. D. et al. MDCK (Madin?Darby canine kidney) cells: A tool for membrane permeability screening. J. Pharm. Sci. 88, 28?33 (1997). 11. Inui, K.-I., Yamamoto, M. & Saito, H. Transepithelial transport of oral cephalosporins by monolayers of intestinal epithelial cell line Caco-2: specific transport systems in apical and basolateral membranes. J. Pharmacol. Exp. Ther. 261, 195?201 (1992). 12. Lu, S., Guttendorf, R. J. & Stewart, B. H. The mechanism of cefdinir transport in CaCO-2 cells. Pharm. Res. 11, S?258 (1994). 13. Artursson, P. & Borchardt, R. T. Intestinal drug absorption and metabolism in cell cultures: Caco-2 and beyond. Pharm. Res. 14, 1655?1658 (1997). 14. Balkovetz, D. F. Hepatocyte growth factor and Madin?Darby canine kidney cells: in vitro models of epithelial cell movement and morphogenesis. Microsc. Res. Tech. 43, 456?463 (1998). 15. Kansey, M. et al. Physiochemical high throughput screening: parallel artificial membrane permeation assay in the description of passive absorption processes. J. Med. Chem. 41, 1007?1010 (2001). 16. Sugano, K., Hamada, H., Machida, M. & Ushio, H. High throughput prediction of oral absorption: improvement of the composition of the lipid solution used in parallel artificial membrane permeation assay. J. Biomol. Screen. 16, 189?196 (2001). 17. Baillie, T. A. et al. Drug metabolites in safety testing. Toxicol. Appl. Pharmacol. 182, 188?196 (2002). A good paper the emphasizes the importance of understanding the safety of the products of drug metabolism. 18. Hawksworth, G. M. Advantages and disadvantages of using human cells for pharmacological and toxicological studies. Hum. Exp. Tox. 13, 568?573 (1994). 19. Roberts, S. A. Drug metabolism and pharmacokinetics in drug discovery. Curr. Opin. Drug Discov. Devel. 6, 66?80 (2003). 20. de Groene, E. M. et al. Development of human cytochrome P450-expressing cell lines: application in mutagenicity testing of ochratoxin A. Cancer Res. 56, 299?303 (1996). 21. de Groene, E. M. in Cytochrome P450 Protocols Methods in Molecular Biology (eds Shephard, E. A. & Phillips, E. A.) 219?226 (1998) 22. Li, A. P. et al. Present status of the application of cryopreserved hepatocytes in the evaluation of xenobiotics: consensus of an international expert panel. Chem. Biol. Interact. 121, 117?123 (1999). 23. Li, A. P. et al. Cryopreserved human hepatocytes: characterization of drug-metabolizing enzyme activities and applications in higher throughput screening assays for hepatotoxicity, metabolic stability, and drug?drug interaction potential. Chem. Biol. Interact. 121, 17?35 (1999). 24. White, R. E. High-throughput screening in drug metabolism and pharmacokinetic support of drug discovery. Annu. Rev. Pharmacol. Toxicol. 40, 133?157 (2000). 25. Roberts, S. A. High-throughput screening approaches for investigating drug metabolism and pharmacokinetics. Xenobiotica 31, 557?589 (2001). 26. Eichelbaum, M. Polymorphic drug oxidation in humans. Fed Proc. 43, 2298?2392 (1984). 27. Dayer, P. et al. Characterization of a common genetic defect of cytochrome P-450 function (debrisoquine-sparteine type polymorphism) ? increased Michaelis is Constant (Km) and loss of stereoselectivity of bufuralol 1?-hydroxylation in poor metabolizers. Biochem. Biophys. Res. Commun. 125, 374?380 (1984). 28. Pettipher, R. The role of genomics in the development of new and improved therapies: An appreciation of the importance of genetic diversity will have a profound impact on how drugs are discovered, developed and prescribed. Innov. Pharm. Technol. 62?67 (2000). 39. Kuehl, P. et al. Sequence diversity in CYP3A promoters and characterization of the genetic basis of polymorphic CYP3A5 expression. Nature Genet. 27, 383?391 (2001). 30. Guengerich, F. P. Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu. Rev. Pharmacol. Toxicol. 39, 1?17 (1999). 31. Guidance for Industry: Drug Metabolism/Drug Interaction Studies in the Drug Development Process: Studies In Vitro (1997) Food and Drug Administration: <http://www.fda.gov/cder/guidance/clin3.pdf> (1997) 32. Clancy, C. E. & Rudy, Y., Linking a genetic defect to its cellular phenotype in a cardiac arrhythmia. Nature 400, 566?569 (1999). 33. Bian, J., Cui, J. & McDonald, T. V. HERG K(+) channel activity is regulated by changes in phosphatidyl inositol 4,5- bisphosphate. Circ. Res. 89, 1168?1176 (2001). 34. Zhou, Z., Gong, Q., Epstein, M. L. January, C. T. HERG channel dysfunction in human long QT syndrome. Intracellular transport and functional defects. J. Biol. Chem. 273, 21061?21066 (1998). 35. Ulrich, R. & Friend, S. H. Toxicogenomics and drug discovery: will new technologies help us produce better drugs? Nature Rev. Drug Discov. 1, 85?88 (2002). This is an opinion article that explores the changes that toxicogenomics might have on the cost and speed of drug development. 36. Schmidt, C. W. Toxicogenomics: an emerging discipline. Environ. Health Perspect. 110, A750?A755 (2002). This article explains the how toxicogenomics could impact drug development and environmental health assessments of low-level exposure limits. 37. Thayer, A. M. Genomics moves on. Chem. Eng. News 80, 25?36 (2002). 38. Therapeutic Products Directorate Guidance: Drug?Drug Interaction Studies: In Vivo and In Vitro. Health Canada Website: http://www.hc-sc.gc.ca (2000). 39. Guidance for Industry: In Vivo Drug Metabolism/Drug Interaction Studies ? Study Design, Data Analysis, and Recommendations for Dosing and Labelling. US Food and Drug Administration <http://www.fda.gov/cder/guidance/2635.pdf> (1999). 40. Huang, S.-M. et al. Assessment of the quality and quantity of drug?drug interaction studies in recent NDA submissions: Study design and data analysis issues. J. Clin. Pharmacol. 39, 1006?1014 (1999). 41. Guidance for Industry: Good Laboratory Practice for Nonclinical Laboratory Studies (21 CFR Part 58) US Food and Drug Administration. http://www.fda.goc/cder 42. Eddershaw, P. J., Beresford, A. P. & Bayliss, M K. ADME/PK as part of a rational approach to drug discovery. Drug Discov. Today 5, 409?414 (2000). 43. Guidance for Industry: Good Laboratory Practice for Nonclinical Laboratory Studies (21 CFR Part 58) Food and Drug Administration Homepage. <http://www.fda.goc/cder> 44. Wilding, I. Injecting innovation into the drug development process. Scripps Magazine 15?16, October (2002). 45. Bayliss, M. K. & Frick, L. W. High-throughput pharmaco- kinetics: Cassette doisng. Curr. Opin. Drug Discov. Devel. 2, 20?25 (1999). 46. Cayen, M. N. Considerations in the design of toxicokinetic programs. Tox. Path. 23, 148?157 (1995). 47. Guideline for Industry: Detection of Toxicity to Reproduction for Medicinal Products. Food and Drug Administration. 48. Gonzalez, F. J. Overview of experimental approaches for study of drug metabolism and drug-drug interactions. Adv. Pharmacol. 43, 255?277 (1997). 49. Guideline for Industry: Pharmacokinetics: Guidance for Repeated Dose Tissue Distribution Studies. US Food and Drug Administration <httpp://www.fda.gov/cder/guidance/ichs3b.pdf> (1995) 50. Guidance for Industry: S4ADuration of Chronic Toxicity Testing in Animals (Rodent and Non Rodent Testing). US Food and Drug Adminsitration. <www.fda.gov/cder/guidance/62599.pdf> (1999). 51. Guidance for Industry: Immunotoxicology Evaluation of Investigational New Drugs, Issued October 2002, Food and Drug Administration. <http://www.fda.gov/cder/guidance/ 4945fnl.doc> (2002) 52. ICH Draft Consensus Guideline: Safety Pharmacology Studies For Assessing The Potential For Delayed Ventricular � 2003 Nature Publishing Group NATURE REVIEWS | DRUG DISCOVERY VOLUME 2 | JULY 2003 | 553 REVIEWS Repolarization (Qt Interval Prolongation) By Human Pharmaceuticals, February 2002. 53. Beckwith-Hall, B. M. et al. Nuclear magnetic spectroscopic and principal components analysis investigations into biochemical effects of three model hepatotoxins. Chem. Res. Tox. 11, 260?272 (1998). 54. Scarfe, G. B. et al. 19 F NMR and directly coupled HPLC- NMR-MS investigations into the metabolism of 2-bromo-4- trifluoromethylaniline in the rat: a urinary excretion balance study without the use of radiolabelling. Xenobiotica 28, 373?388 (1998). 55. Cherry, S. R. et al. MicroPET: a high resolution PET scanner for imaging small animals. IEEE Trans. Nucl. Sci. 44, 1161?1166 (1997). 56. Paans, A. M. J. & Vaalburg, W. Positron emission tomography in drug development and drug evaluation. Curr. Pharm. Des. 6, 1583?1591 (2000). This paper explores the emerging use of PET imaging to support traditional ADME/PK and toxicology investigations supporting drug development. 57. Benet, L. Z. & Hoener, B. A. Changes in plasma protein binding have little clinical relevance. Clin. Pharmacol. Ther. 71, 115?121 (2002). 58. Spahn Langguth, H. et al. P-glycoprotein transporters and the gastrointestinal tract: evaluation of the potential in vivo relevance of in vitro data employing talinolol as a model compound. Int. J. Clin. Pharmacol. Ther. 36,16?24 (1998). 59. Ramaswamy, M. et al. Influence of phytostanol phosphoryl ascorbate, FM-VP4, on pancreatic lipase activity and cholesterol accumulation within Caco-2 cells. J. Pharm. Pharm. Sci. 5, 29?38 (2002). 60. Tucker, G. T., Houston, J. B. & Huang, S.-M., Optimizing drug development: strategies to assess drug metabolism/transporter interaction potential ? toward a consensus. Clin. Pharmacol. Ther. 70, 103?114 (2001). 61. Moore, L. B. et al. St. John?s wort induces hepatic drug metabolism through activation of the pregnane X receptor. Proc. Natl Acad. Sci. USA 97, 7500?7502 (2000). 62. Li, A. P. Primary hepatocyte cultures as an in vitro experimental system for the evaluation of pharmacokinetic drug?drug interactions. Adv. Pharmacol. 43, 103?130 (1997). 63. Lehmann, J. M. et al. The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J. Clin. Invest. 102, 1016?1023 (1998). 64. Goodwin, B., Hodgson, E. & Liddle, C. The orphan human pregnane X receptor mediates the transcriptional activation of CYP3A4 by rifampicin through a distal enhancer module. Mol. Pharmacol. 56, 1329?1339 (1999). 65. Ogg, M. S. et al. A reporter gene assay to assess the molecular mechanisms of xenobiotic-dependent induction of the human CYP3A4 gene in vitro. Xenobiotica 29, 269?279 (1999). 66. Luo, G. et al. CYP3A4 induction by drugs: correlation between a pregnane X receptor reporter gene assay and Cyp3A4 expression in human hepatoyctes. Drug Metab. Dispos. 30, 795?804 (2002). 67. Baillie, T. A. et al. Drug metabolites in safety testing. Toxicol. Appl. Pharmacol. 182, 188?196 (2002). 68. Guideline for Industry: The Need for Long?term Rodent Carcinogenicity Studies of Pharmaceuticals, ICH S1A. US Food and Drug Administration. <http://www. fda. gov/cder/guidance/ichs1a. pdf> (1996). 69. Guidance for Industry: S1B Testing for Carcinogenicity of Pharmaceuticals. US Food and Drug Administration. <http://www.fda.gov/cder/guidance/1854fnl.pdf> (1997). 70. Press Release, Novo Nordisk suspends the clinical development of ragaglitazar (NN622), Bagsvaerd, Denmark (July 22, 2002). 71. Press Release, January 30, 2003, Pharmaceutical Research and Manufacturers of America (PhRMA), Pharmceutical Companies Receive FDA Approval for 26 NME and Biologics, 63 Other New Medicines and 172 New Uses for Medicines in 2002. 72. Ackerman, B. L., Murphy, A. T. & Berna, M. J. The resurgence of column switching techniques to facilitate rapid LC/MS/MS based bioanalysis in drug discovery. Amer. Pharm. Rev. 5, 54?63 (2002). 73. Clarke, N. J., Rindgen, D., Korfmacher, W. A. & Cox, K. A. Systematic LC/MS metabolite identification in drug discovery. Anal. Chem. 73, 430A?439A (2001). 74. March, R. E. An introduction to quadrupole ion trap mass spectrometry. J. Mass Spectrom. 32, 351?369 (1997). 75. Chernushevich, I. V., Loboda A. V. & Thomson, B. A. An introduction to quadrupole-time-of-flight mass spectrometry. J. Mass Spectrom. 36, 849?865 (2001). 76. Albert, K. Liquid chromatography ? nuclear magnetic resonance spectroscopy. J. Chromatogr. A. 856, 199?211 (1999). 77. Hagar, J. W. A new linear ion trap mass spectrometer. Rapid Commun. Mass Spectrom. 16, 512?526 (2002). 78. Marshall, A. G., Hendrickson, C. L. & Shi, D.-H. S. Scaling MS plateaus with FTICR MS. Anal. Chem. 74, 252A?259A (2002). Online links DATABASES The following terms in this article are linked online to: LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/ CYP1A2 | CYP2C9 | CYP2C19 | CYP2D6 | CYP3A4 | CYP3A5 | P-glycoprotein FURTHER INFORMATION National Center for Toxicogenomics: www.niehs.nih.gov/nct US Food and Drug Administration: www.fda.gov/cder International Conference on Harmonization: www.ich.org Indiana University School of Medicine Cytochrome P450 Drug Interaction Table: http://medicine.iupui.edu/flockhart/ Access to this interactive links box is free online. "
Add Content to Group
|
Bookmark
|
Keywords
|
Flag Inappropriate
share
Close
Digg
Facebook
MySpace
Google+
Comments
Close
Please Post Your Comment
*
The Comment you have entered exceeds the maximum length.
Submit
|
Cancel
*
Required
Comments
Please Post Your Comment
Posted By
Date Posted
Comment
William Umbenhauer
04/22/2018
How do I participate
Save Note
Note
View
Public
Private
Friends & Groups
Friends
Groups
Save
|
Cancel
|
Delete
Please provide your notes.
Next
|
Prev
|
Close
|
Edit
|
Delete
Genetics
Gene Inheritance and Transmission
Gene Expression and Regulation
Nucleic Acid Structure and Function
Chromosomes and Cytogenetics
Evolutionary Genetics
Population and Quantitative Genetics
Genomics
Genes and Disease
Genetics and Society
Cell Biology
Cell Origins and Metabolism
Proteins and Gene Expression
Subcellular Compartments
Cell Communication
Cell Cycle and Cell Division
Scientific Communication
Career Planning
Loading ...
Scitable Chat
Register
|
Sign In
Visual Browse
Close
Comments
CloseComments
Please Post Your Comment