Replacing Animals in Absorption, Distribution, Metabolism & Excretion Studies

Home / New Perspectives / Pharmacokinetics / Replacing Animals in Absorption, Distribution, Metabolism & Excretion Studies

Pharmacokinetics

Replacing Animals in Absorption, Distribution, Metabolism & Excretion Studies

Gill Langley, Dr Hadwen Trust

Published: December 6, 2007

About the Author(s)
Dr. Gill Langley is the Dr. Hadwen Trust’s Science Director. She has an honours degree in Natural Sciences from Cambridge University and a doctorate in neurochemistry, also from Cambridge. As a research fellow at Nottingham University she studied neurochemistry in vitro. She has led the Dr Hadwen Trust’s science and outreach programmes since 1981.

Dr. Langley served for eight years as a member of the British government’s Animal Procedures Committee, which advises the Home Secretary on animal experimentation issues. She was an adviser to the government on the introduction of the new EU chemicals legislation, REACH. She has been an invited expert at several initiatives of the European Commission and of the Organisation for Economic Co-operation and Development (OECD). Dr. Langley is a member of the Replacement Advisory Group of the British government’s National Centre for Replacement, Reduction and Refinement of animals in research.

Dr. Gill Langley
Dr Hadwen Trust
84a Tilehouse Street
Hitchin
Herts
SG5 2DY
UK
E-mail: gill@drhadwentrust.f2s.com

The absorption, distribution, metabolism and excretion (ADME) characteristics of chemicals are central to interpreting hazard data and making risk assessments; and in pharmaceuticals, the ADME properties of medicines are critical to their success or failure. Improving human welfare and gaining economic advantage are powerful drivers for accomplishing the best possible ADME studies with drugs and chemicals. And, of course, there’s a third equally important motivation: preventing animal suffering. Until tests for ADME can be achieved without using animals, it will be very difficult to replace a range of systemic toxicity tests on animals. Progress will also be delayed in advancing in vitro and in silico methods from the status of screening tests to full replacements. Therefore predicting human ADME more effectively and without using laboratory animals is a pivotal goal.

This review provides a historical perspective for today’s toxicology; looks at some key events in acceptance of non-animal testing strategies; summarises some recently proposed toxicology paradigms for non-animal ADME approaches; and highlights some exciting tools for replacing animals in ADME studies.

Toxicology evolving

The paradigms and processes of toxicology have evolved in a highly ad hoc manner over many decades. Each development has been ‘bolted on’ to the previous structure, resulting in the unbalanced tottering edifice we see today. Nobody is willing to defend toxicology, in detail, as being fit for purpose. But while some argue that the basic structure remains sound and the form can be adapted yet again to meet current demands, others call for new foundations and a rebuild to modern standards.

In human health studies, including assessing the ADME characteristics of xenobiotics, animal testing has been placed at centre-stage since the 1960s. Early on, it became clear that animal-based toxicology had severe limitations. In 1980, for example, Professor George Teeling-Smith summarized the situation for pharmaceuticals: “There is at present no hard evidence to show the value of more extensive and more prolonged laboratory testing as a method of reducing eventual risk in human patients. In other words the predictive value of studies carried out in animals is uncertain. The statutory bodies such as the Committee on Safety of Medicines which require these tests do so largely as an act of faith rather than on hard scientific grounds” [1].

And for chemicals, the British Royal Society had reached a similar conclusion at the same time: “Unfortunately, this use of animal models for predicting risks for man is beset with difficulties… it is rarely possible to be sure that the animal model properly represents the relationship in man. …Even if it were possible to improve the accuracy of the present-day test procedures so that the risk to the test animals were known with greater precision, this would not necessarily bring about a corresponding improvement in the assessment of potential risk to man because of the uncertainty regarding the relevance of the animal data for man” [2].

Time for change

ADME studies use a considerable number of animals in tests that may inflict pain, distress and suffering, not only by the nature of the procedures used but also as a result of the compounds being tested. In Britain alone, for example, 32,381 animals were used in 2006 in ADME and residue studies for pharmaceuticals, including mice, rats, guinea pigs, hamsters, rabbits, dogs (234), cats (12), pigs (332), primates (119), ferrets, cattle and chickens [3].

Everyone knowledgeable in this field is aware of the deficiencies of animal testing, but a familiar system is more comfortable than change. Traditional toxicologists and regulators have ploughed on, arguing that the animal-centred approach is the best for the time being, even in the absence of systematic data on the validity of animal tests. They have tried to adapt the tottering edifice of regulatory toxicology to each new challenge, and in recent years there have been plenty. In chemicals regulation, the challenges include endocrine disrupters and nanomaterials. In medicine, an example is the safety testing of therapeutic proteins, brought starkly to light by the British disaster with the humanised monoclonal antibody TGN1412, which nearly killed several volunteers in clinical trials conducted in March 2006 [4].

In the meantime, non-animal toxicology has been taking root. Originated by the British anti-vivisection movement in the late 1960s, the concept of conducting biomedical research and testing without animal experiments was not quickly accepted by the scientific, regulatory or policy-making communities. Novel scientific approaches often take many years to become mainstream; examples include the development of cognitive neuroscience in the 1980s, and of neuroimaging technologies in the 1990s. With toxicology, in contrast to biomedical research, the regulatory context adds an additional dimension. So it isn’t surprising that for many years, replacement tests were developed only by a minority of progressive researchers and were published mainly in niche journals.

Replacement goes mainstream

It is only in the last few years that non-animal toxicology has made a genuine breakthrough into mainstream science. Increased support at European Union (EU) level, together with legislative initiatives and visionary reports from highly influential agencies, have combined to take the approach to a new level. The work of the European Centre for the Validation of Alternative Methods (ECVAM) has consolidated the strategy, and produced a number of validated non-animal assays. In 2002, the European Union’s Sixth Framework Programme on Research and Development offered unprecedented opportunities for funding to replace animal-based toxicology. More than 20 research projects, some of them tackling the ADME properties of xenobiotics, were funded to a total level of about 92 million euros. From integrated projects, such as ACuteTox (9 million euros, 35 partners in 13 European countries) to networks of excellence such as BioSim (40 partners, 10.7 million euros), a new creative energy to develop novel in vitro and in silico approaches was released in the European Union research arena.

Legislative initiatives have also been highly significant in sharpening the focus on non-animal methods. The 7th amendment to the EU Cosmetics Directive, which introduces bans on the animal testing of cosmetics ingredients, has stimulated new resarch efforts. The REACH legislation (Registration, Evaluation, Authorisation and Restriction of Chemicals) included intelligent testing strategies, incorporating non-animal methods, which have minimised the numbers of animals to be used in the REACH testing program [5].

A major step was achieved with the publication of the US Food and Drug Administration’s 2004 “Critical Path” report on medicines development [6]. This highlighted how the current drug development process is increasingly difficult, costly and inefficient. The report identified as a major problem the failure to create and use novel scientific tools to deliver “…fundamentally better answers about how the safety and effectiveness of new products can be demonstrated, in faster time frames, with more certainty, and at lower costs”.

In 2007 another seminal report emerged, this time commissioned by the US Environmental Protection Agency (EPA) [7]. The EPA recognized the need for a full review of toxicity testing frameworks for drugs, food additives, pesticides and industrial and other chemicals, and asked the National Research Council (NRC) to conduct this review and propose a new vision and strategy for toxicology. A key feature of the proposed vision is to move towards human-relevant in vitro and computational techniques that offer faster, more efficient, less expensive, mechanistically based and more comprehensive assays. As the NRC report said, “Toxicity testing is approaching [such] a scientific pivot point. It is poised to take advantage of the revolutions in biology and biotechnology. Advances in toxicogenomics, bioinformatics, systems biology, epigenetics, and computational toxicology could transform toxicity testing from a system based on whole animal testing to one founded primarily on in vitro methods that evaluate changes in biologic processes using cells, cell lines, or cellular components, preferably of human origin.”

Non-animal ADME test strategies

The traditional, aging paradigms of toxicology and risk assessment reflect neither current thinking nor the availability of exciting new non-animal technologies, which cannot fulfil their potential if they are constrained by a 20th-century system of hazard identification and risk assessment [8]. The paradigms must change and, clearly, non-animal ADME data are central both to improving risk assessment and to replacing systemic animal tests such as carcinogenicity and chronic studies. Equally, it is obvious that ADME studies cannot be replicated in a single assay in vitro.

A successful strategy will need conceptually to ‘break down’ ADME processes into the critical components that can be modelled using in vitro and in silico systems. These systems may need to be applied firstly in ways that give quick answers via a mass screening of drugs and chemicals, and can also identify compounds which need further detailed assessment. This approach is already exploited by the pharmaceutical industry and has been adopted in the EU’s REACH legislation. These critical component data will then be integrated to ‘re-create’ the whole picture (i.e. data interpretation and extrapolation), by means of computational techniques such as physiologically based pharmacokinetic (PBPK) modeling, aided by microdosing studies in human volunteers where these are safe and ethical.

The US NRC’s approach envisages three modules to achieve this process, which they call chemical characterization, toxicity testing, and dose-response and extrapolation modeling [7]. Chemical characterization would provide data such as a molecule’s physical and chemical properties, likely routes of exposure, possible routes of metabolism, and likely interactions of the compound and its metabolites at the cell level. Much of this information would be obtained using various computational methods, including (quantitative) structure-activity relationship ((Q)SAR) models which predict biologic activity on the basis of molecular structure.

The NRC’s second module is predicting cellular toxicity pathways, using high-throughput human cell and cell-line assays, which will progressively replace all animal tests in this area. Compounds would be further evaluated in “targeted testing”, and the NRC sees a challenge to develop new ways of identifying metabolites in vitro rather than in animals. Possibilities include tissue bioreactors, especially a “liver bioreactor” or co-cultures of human liver cells with other tissues. The NRC’s third module is dose-response and extrapolation modeling. Dose-response models would be based mainly on mechanistic in vitro assays, and extrapolated to human chemical exposures and intakes using PBPK models.

The acquisition and application of human data are a key component of the NRC’s strategy. Animal protection scientists have long argued that an over-emphasis on animal toxicology has obscured the importance of obtaining and using human information of this kind [9]. The EPA sees human exposure data being “pivotal as toxicity testing shifts from the current apical end-point whole-animal testing to cell-based testing” [7]. Workplace and environmental exposure information, including measures of chemicals and metabolites in blood, urine, saliva and other tissues, would direct the choice of dose ranges for in vitro toxicity testing and of concentrations to use in human PBPK models. New tools to measure human exposures could include environmental and biological sensors, geographic information systems, toxicogenomics and body burden measurements [10].

The pharmaceutical sector has much more experience of in silico and in vitro strategies than does the chemicals industry. It also has the advantage that, during drug development, molecules are assessed in non-animal, animal and human trials, providing a huge potential data resource which should play an essential role in validating novel approaches. Drug regulatory agencies, such as the US Food and Drug Authority (FDA) and the European Medicines Agency (EMEA) hold parallel test data for hundreds of novel drugs. Data-mining will inevitably reveal information of direct importance to the development and validation of non-animal techniques to assess xenobiotic ADME, as acknowledged by Sir Michael Rawlins, Chair of the British National Institute of Health and Clinical Excellence: “Clarity can be provided by analyzing the hundreds and hundreds of studies locked in the files of the regulatory authorities in both the U.S. and the EU. …In particular, it’s worthwhile looking at drugs that failed or have been withdrawn” [11].

Other cross-fertilization between the experience and information in the pharmaceutical industry and the chemical sector could powerfully facilitate advances with non-animal methods. One concept envisages that safety information from non-animal models could be interpreted or “translated” by means of comparable clinical data [12]. This would involve the dual application of 21st-century technologies, such as genomics, metabonomics or sensitive analytical techniques, both to non-animal assays and to human clinical studies (where safe and ethical). The novel data from the clinical studies would thus assist the interpretation of similar data generated by non-animal tests, by allowing a “translation” of the significance of the in vitro results in terms of genuine human health effects. This concept offers substantial benefits over the current paradigm of comparing the outcomes of new non-animal assays with old animal-based assays: it would eliminate the problem of species variations, and would allow the performance of new tests to be compared against genuine, gold-standard human health data.

In the pharmaceutical arena, the US FDA acknowledges the substantial limitations of animal testing. Its Deputy Commissioner for Operations, Janet Woodcock, has said: “We believe the applied sciences have not kept up with the basic science in drug development. Therefore, when discoveries made using advanced sciences are brought up to the preclinical stage, we start applying technologies that have been used for 50-60 years, such as animal tox, which is completely empirical. We just dose the animals and cut them up and see what happens” [11].

The FDA’s Critical Path report [6] called for new toxicology tools and concepts, emphasizing the importance of correctly predicting human handling of novel drugs early in their development, so that candidate molecules with poor human ADME characteristics can be discarded sooner rather than later. This will pre-empt these ‘failing’ molecules from proceeding either to further animal testing or to clinical trials, thus saving time, costs and animal lives, and protecting the health of volunteers. It has been estimated that a 10% improvement in predicting failing drug candidates before clinical trials could save $100 million in development costs per drug [13].

The Critical Path report [6] envisages more emphasis on in vitro, computational and clinical tools and stated [page 7] that recommendations on “…the use of human cell lines to characterize drug metabolic pathways provide a straightforward in vitro method for prediction of human metabolism, allowing developers to eliminate early on compounds with unfavorable metabolic profiles (e.g. drug-drug interaction potential). Failures in the clinic due to drug interaction problems are now far less likely”.

As well as in vitro approaches, the FDA recognizes the importance of computational modeling. Modeling is increasingly being used to simulate human diseases; to predict the pharmacokinetics of novel drugs; and to conduct ‘virtual’ clinical trials. In terms of ADME studies, their report stated [page 9] that “Model-based drug development involves building mathematical and statistical characterizations of the time course of the disease and drug using available clinical data to design and validate the model. The relationship between drug dose, plasma concentration, biophase concentration (pharmacokinetics), and drug effect or side-effects (pharmacodynamics) is characterized, and relevant patient covariates are included in the model. Systematic application of this concept to drug development has the potential to significantly improve it”.

New tools for ADME

Today a tremendous creative energy is focused on advancing new techniques and technologies to replace animals in drug and chemical hazard identification and risk assessment, including the prediction of ADME characteristics. Some of these tools are based on cells, tissues and sub-cellular components, ideally from human sources; some are ‘virtual’ tools, harnessing the power of modern computational systems; some are allowing safe and powerful studies of human volunteers, whether patients or healthy subjects. Some approaches use combinations of these techniques. There are many exciting opportunities to address ADME issues using these new tools; here are just a few examples.

The ADME properties of compounds are strongly influenced by their physicochemical properties (such as solubility, lipophilicity, molecular weight and protein binding). These have traditionally been measured singly, but advances in automation and data-handling combined with technology such as liquid chromatography with mass spectral detection, are generating multiple information more quickly [14]. In vitro assays also contribute to a more streamlined ADME assessment of drug candidates, earlier in pre-clinical development [15]. The ability to study several processes simultaneously in vitro will greatly improve confidence in the predictive reliability of these assays. Increasingly sophisticated cell models should provide new ways to study complex interactions between drug or chemical uptake, accumulation and metabolism in tissues [16].

More data on physicochemical properties and ADME properties has, in turn, prompted the development of better ‘virtual’ predictions. Aspects such as permeability, solubility and cytochrome P450 metabolism predictions by means of (Q)SARs and structure-property relationships; and better computational estimates of drug half-life values and passage through the blood-brain barrier, are being actively progressed [7]. Researchers are working towards quicker and earlier decision-making about drug ADME characteristics, by integrating SAR models with ‘pattern’ databases of tissue responses to drugs (compiled from high-throughput experiments), and with ‘systems biology’ databases of metabolic pathways, genes and regulatory networks [17]. As larger, high-quality in vitro or human datasets become available, existing models can be refined and improved.

Techniques of microfabrication, using miniaturized devices, may make a significant contribution. Isolated primary hepatocytes rapidly lose their metabolic capacities in standard cell culture conditions. A microreactor system has sustained primary hepatocytes for longer than other in vitro methods, and with more normal metabolizing and enzyme induction activities [18]. Primary human hepatocytes have been cultured on collagen-coated silicon sensor chips in chemically defined specialist medium allowing them to retain their normal characteristics for several weeks in vitro [19]. Enabling human hepatocytes to function for longer in vitro is a key advance in their wider use to replace animal metabolism studies. Some groups are developing micromodels of human metabolism, using cells from liver, lung and adipose tissue in interconnected chambers, all on a one-inch-square silicon chip. Drug metabolites formed in the liver-cell microchambers then perfuse through the chambers containing lung or other cells, to assess their potential toxicity to target tissues [20].

An ECVAM project has assessed the possibility of obtaining high-quality human hepatocytes in the USA and the EU [21]. Guidelines were produced for obtaining liver samples suitable for use as suspension or monolayer cultures. The limited supply of human hepatocytes is also being tackled, and should be facilitated by the establishment of EU networks of human tissue banks for research purposes.

ACuteTox is an integrated research project of the EU Sixth Framework Programme on Research and Development, and its goal is to produce a strategy for predicting acute oral toxicity without laboratory animals. It has developed a multiprocessor computer system for estimating oral drug uptake and passage through the blood-brain barrier, based on data from more than 20 reference chemicals using the Caco-2 cell line system and in vitro models of the blood-brain barrier [22]. Information is being generated on protein-binding and the partitioning of chemicals in cell culture, which will assist the modeling of interactions between chemicals. ACuteTox partners are also studying the toxicity of 21 reference compounds in parallel in metabolically active and inactive cell models; and are assessing the performances of computer programs that predict the toxicity and metabolic fate of chemicals.

The BioSim Network of Excellence, also funded by the European Union, is developing computer models to improve drug discovery, selection and testing, including the likely metabolic fates of pharmaceuticals while also replacing animal studies [23]. Simulation developments elsewhere include predicting drug metabolism in human populations from in vitro data [24] and the development of advanced, easier-to-use PBPK models. PBPK models are mathematical descriptions of the complex interactions affecting the disposition of compounds in the body, and are used to predict drug or chemical ADME properties. The models are based on compound-specific data, such as physicochemical and in vitro information (e.g. plasma protein-binding, cell membrane permeability, and microsomal or hepatocyte intrinsic clearance); and on species-specific physiological data (such as tissue perfusion rates and fractional blood flow). Software is used to estimate features such as rate and extent of absorption and hepatic clearance, which can then be ‘slotted’ into a whole-body PBPK model to predict the time course of the compound in the plasma and other tissues.

One way to obtain gold-standard human ADME data is by means of microdosing studies in volunteers [25]. Increasingly utilised by the pharmaceutical industry, and backed by recent guidelines from both the US FDA and the European Medicines Agency (EMEA), microdosing involves giving sub-pharmacologic and sub-toxic doses of a test compound to volunteers. New techniques of analysis that are extraordinarily sensitive, such as accelerator mass spectrometry [26], allow the measurement of ultra-low doses of new drugs and their known, as well as unknown, metabolites. PET scans can be used to ensure a drug microdose has penetrated to the target organ (e.g. the brain). Microdosing can pre-empt some animal tests and reduce the costs and timescales of drug development, through predicting early in the development process which drugs will ultimately fail in humans. If it fulfils its early promise, human microdosing will also provide data, in the species of ultimate interest, to assist in the validation of new in vitro and in silico ADME techniques.

A pivotal moment

We are witnessing an exciting and challenging paradigm shift in toxicology, a pivotal moment when non-animal techniques are going mainstream. There is a confident expectation that this will benefit chemical hazard and risk assessment as well as drug development. Novel tools applied intelligently in a 21st century toxicology framework will help save time, resources, and human and animal lives. Such a win/win solution should be pursued with the maximum enthusiasm, support and leadership from the scientific, regulatory, political and animal protection communities.
©2007 Gill Langley

References
    1. Teeling-Smith, G. (1980). A Question of Balance; the Benefits and Risks of Pharmaceutical Innovation. Publ. Office of Health Economics, London.
    2. The Royal Society. (1983). Risk Assessment — Report of a Royal Society Study Group. Publ. Royal Society, London.
    3. Home Office. (2007). Statistics of Scientific Procedures on Living Animals Great Britain 2006. Cm. 7153. Publ. The Stationery Office, London.
    4. Expert Group on Phase One Clinical Trials: Final Report, 7 December 2006. Publ. The Stationery Office, London.
    5. Regulation (EC) No 1907/2006 of the European Parliament and of the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/EC.
    6. The Food and Drug Administration. (2004). Innovation and Stagnation: Challenge and Opportunity on the Critical Path to New Medical Products.
    7. National Research Council (2007). Toxicity Testing in the Twenty-first Century: Vision and a Strategy. ISBN: 0-309-10989-2.
    8. Blaauboer, B. & Andersen, M.E. (2007). The need for a new toxicity testing and risk analysis paradigm to implement REACH or any other large scale testing initiative. Arch. Toxicol.81, 385-387.
    9. Langley, G. (2004). Endocrine Disrupting Chemicals: A Non-Animal Testing Approach. Publ. BUAV, London.
    10. Brown, D., Ellisman, M., Lioy, P., Omenn, G., Potter, J.D., Smith, M.T., Sohn, L., Suk, W.A., Sumner, S., Swenberg, J., Walt, D.R., Watkins, S., Thompson, C. & Wilson, S.H. (2005). Personalized exposure assessment: promising approaches for human environmental health research. Environ. Health Perspect.113, 840-848.
    11. BioCentury. (2005). The Bernstein Report on Biobusiness, September 5, 2005, pages A10-A12.
    12. Fentem, J., Chamberlain, M. & Sangster, B. (2004).The feasibility of replacing animal testing for assessing consumer safety: a suggested future direction. Altern. Lab. Anim. 32, 617-623.
    13. Parexel Pharmaceutical R&D Statistical Sourcebook 2002/2003. Boston Consulting Group: A Revolution in R&D – How Genomics and Genetics will affect Drug Development Costs and Times.
    14. Sheppard, G.S. & Bouska, J.J. (2005). Why optimize cancer drugs for ADMET? Drug Discov. Today.2, 343-349.
    15. Kassel, D.B. (2004). Applications of high-throughput ADME in drug discovery. Curr. Opin. Chem. Biol.8, 339-345.
    16. Riley R.J. & Kenna J.G. (2004). Cellular models for ADMET predictions and evaluation of drug-drug interactions. Curr. Opin. Drug. Discov. Devel.7, 86-99.
    17. Bugrim, A., Nikolskaya, T. & Nikolsky, Y. (2004). Early prediction of drug metabolism and toxicity: systems biology approach and modeling. Drug. Discov. Today.9, 127-135.
    18. Sivaraman, A., Leach, J.K., Townsend, S., Iida, T., Hogan, B.J., Stolz, D.B., Fry, R., Samson, L.D., Tannenbaum, S.R. & Griffith, L.G. (2005). A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction. Curr. Drug. Metab.6, 569-591.
    19. Thedinga, E., Ullrich, A., Drechsler, S., Niendorf, R., Kob, A., Runge, D., Keuer, A., Freund, I., Lehmann, M. & Ehret, R. (2007). In vitro system for the prediction of hepatotoxic effects in primary hepatocytes. ALTEX.24, 22-34.
    20. Viravaidya, K., Sin, A. & Shuler, M.L. (2004). Development of a microscale cell culture analog to probe naphthalene toxicity. Biotechnol. Prog.20, 316-323.
    21. Coecke, S., Ahr, H., Blaauboer, B.J., Bremer, S., Casati, S., Castell, J., Combes, R. Corvi, R., Crespi, C.L., Cunningham, M.L., Elaut, G., Eletti, B., Freidig, A., Gennari, A., Ghersi-Egea, J.F., Guillouzo, A., Hartung, T., Hoet, P., Ingelman-Sundberg, M., Munn, S., Janssens, W., Ladstetter, B., Leahy, D., Long, A., Meneguz, A., Monshouwer, M., Morath, S., Nagelkerke, F., Pelkonen, O., Ponti J., Prieto, P., Richert, L., Sabbioni, E., Schaack, B. Steiling, W., Testai, E., Vericat, J.A. & Worth, A. (2006). Metabolism: A Bottleneck in In Vitro Toxicological Test Development: The Report and Recommendations of ECVAM Workshop 541. Altern. Lab. Anim. 34, 49-84.
    22. ACuteTox
    23. BioSim
  • Rostami-Hodjegan, A. & Tucker, G.T. (2007). Simulation and prediction of in vivo drug metabolism in human populations from in vitro data. Nat. Rev. Drug Discov.6, 140-148.
  • Combes, R.D., Berridge, T., Connelly, J., Eve, M.D., Garner, R.C., Toon, S. & Wilcox, P. (2003). Early microdose drug studies in human volunteers can minimise animal testing: Proceedings of a workshop organised by Volunteers in Research and Testing. Eur. J. Pharm. Sci.19, 1-11.
  • Beumer, J.H., Garner, R.C., Cohen, M.B., Galbraith, S., Duncan, G.F., Griffin, T., Beijnen, J.H. & Schellens, J.H. (2007). Human mass balance study of the novel anticancer agent ixabepilone using accelerator mass spectrometry. Invest. New Drugs 25, 327-334.