531ISSN 1473-7159© 2009 Expert Reviews Ltd10.1586/ERM.09.38 www.expert-reviews.com

Editorial

Value of new biomarkers for safety testing in drug developmentExpert Rev. Mol. Diagn. 9(6), 531–536 (2009)

“Animal testing will always be required by regulators, but the use of more modern techniques ... should decrease our

dependence on large, cumbersome, expensive and sometimes meaningless animal studies.”

Molecular profiling in modern toxicologyToxicity testing and the assessment of safety of drugs, veterinary products and chemicals are essential in today’s society. The current accepted paradigm for non-clinical safety estimation is that different animal species are sufficient to predict effects in humans. However, these tests have not significantly progressed over the years, and millions of animals are used every year for toxicity studies globally. Legislation in modern societies is based on the assumption that animal testing is ethi-cally acceptable. The 3Rs concept, defined by Russel and Burch in 1959, states that all efforts to ‘replace’, ‘reduce’ and ‘refine’ ani-mal experiments must be undertaken [1]. This has obvious implications for toxic-ology testing, where such large numbers of animals are required, by law, to assure consumer safety.

In addition, such huge, long-running testing strategies cost the industry enor-mous amounts of resources. The total size of the toxicology space is approximately US$1.5 billion, and in vivo toxicology, consisting of animal testing, represents the largest fraction thereof (approxi-mately US$1.3 billion). Taken together, it is important that alternative meth-ods for assessing the safety of new (and old) compounds are found that will still ensure consumer safety but decrease the reliance on animals. Animal testing will always be required by regulators, but the use of more modern techniques, such as transgenic animals, ’omics technologies (e.g., toxicogenomics), in vitro assays, in silico ‘expert systems’ tools and novel

predictive/diagnostic safety biomarkers should decrease our dependence on large, cumbersome, expensive and sometimes meaningless animal studies.

ToxicogenomicsToxicogenomics represents a desire to step outside the boundaries of traditional toxic-ology. It is based on the measurement of thousands of genes simultaneously and has shown potential to revolutionize toxicity testing. It has been successfully used as a tool to elucidate mechanisms of toxicity, as well as having the potential to predict toxicities much earlier during drug devel-opment [2–4]. The advanced knowledge of gene- and protein-expression patterns, together with modern classification algo-rithms, has also demonstrated practical benefits for predicting pathological events and toxic end points [5–6]. Unfortunately, these early promises are only being realized after a period of relatively expensive and deliberate test validation and generation of large reference databases, which are still essential for the future of mechanism elu-cidation. Without adequate study design, appropriate use of controls, and multi-disciplinary development of standardized methods, acceptance has been slow. In the interim, the power of toxicogenomics is likely to be applied on a case-by-case basis, with specific and often proprietary toxico-genomics assays developed and applied within each individual company.

A positive example for understanding nephrotoxicity is that of cisplatin, a well-known inducer of nephrotoxicity, which causes specific gene deregulation. Genes involved in drug resistance (e.g., MDR1

Philip Hewitt Head of Molecular Toxicology, Merck Serono, Frankfurter Str. 250, 64285 Darmstadt, Germany

Thomas HergetAuthor for correspondence

Director New Technology Evaluation, Merck KGaA, Frankfurter Str. 250, 64285 Darmstadt, Germany

thomas.herget@merck.de

For reprint orders, please contact reprints@expert-reviews.com

Expert Rev. Mol. Diagn. 9(6), (2009)532

Editorial Hewitt & Herget

and P-gp), in cytoskeleton structure and function (Vim, Tubb5, Tmsb10, Tmsb4x and Anxa2), in cell adhesion (Spp1, Col1a1, Clu and Lgals3), in apoptosis (cytochrome c oxidase subunit I, BAR, heat-shock protein 70-like protein and Bax), in tissue remodeling (clusterin, IGFBP-1 and TIMP-1) and in detoxification (Gstm2 and Gstp2) are upregulated during cisplatin-induced injury. Genes downregulated by cisplatin include those that localize to the proximal tubules (Odc1, Oat, G6pc and Kap), those that con-trol intracellular calcium homeostasis (SMP-30), and those that encode growth factors or their binding proteins (Egf, Ngfg, Igfbp3 and Ghr). These gene expression changes can be well correlated with the known mechanisms of cisplatin damage to proximal tubules, tissue remodeling and regeneration [7].

Subsequent analyses, including pathway analysis of these extremely complex datasets, could lead to the discovery and vali-dation of other novel biomarkers, such as gene fingerprints or small subset of relevant, mechanistically based proteins. Therefore, this gain of information can potentially reduce the duration of toxic-ology studies for preclinical investigations in the near future, and consequently, the number of animals required for toxicity testing. In addition, this will have an enormous benefit financially. Owing to the nature of these studies and the necessity for adequate statisti-cal power, an individual company or institute would find it difficult to validate and qualify new biomarkers. Therefore, many consortia have been established over the past few years to push forward the use of ’omics technologies and the discovery of new safety biomark-ers. Some of those important international networks are listed in Table 1. To date, several of these groups have identified and reported new molecular biomarkers, for numerous target organs of toxicity.

The rapid evolution of genomics technology has resulted in a large number of biomarkers, and the guidance classifies those as either exploratory biomarkers, probable valid biomarkers or known valid biomarkers. This will allow companies to understand how these biomarkers will be used in decision-making in drug development and within the US FDA. An example of evaluating nephrotoxicity in this way will be given in the following section.

Other ‘omics techniquesTranscriptomics using microarrays can be combined and comple-mented with similarly powerful techniques, such as proteomics (2D electrophoresis mass spectrometry (MS), SELDI-TOF MS, HPLC MS/MS, for example, iTRAQ, or protein arrays) and metabonomics (liquid chromatography [LC] MS, gas chro-motography MS and NMR), thereby monitoring translational and post-translational events as well. Metabonomics may also be considered as a high-throughput extensive clinical chemistry technology, where hundreds of individual analytes can be moni-tored simultaneously [8]. Multianalyte profiling can be performed non invasively (e.g., using blood or urine), and the same tech-nique used for all species. A recent animal study demonstrated the presence of glucose, amino acids and tricarboxylic acid cycle metabolites in the urine 2 days after cisplatin exposure [9]. If this altered metabolic profile can be demonstrated and translated to human studies, they might be applicable as sensitive biomarkers for early cisplatin-induced nephrotoxicity.

It is very enticing when you want to consider the trans latability of new safety biomarkers across from standard tox species to humans, as availability of tissue is often extremely limited. These molecular ’omics technologies should allow for more sensitive and earlier detection of adverse effects in (animal) toxicity stud-ies. In the near future, human risk assessment will benefit from functional toxicogenomics (systems toxicology) owing to an improved extrapolation between ‘toxicology’ species and humans. Molecular methods, such as toxicogenomics, provide the oppor-tunity to reduce the uncertainty of extrapolating from laboratory animal models to the human, (e.g., by developing ‘bridging effect markers’). A bridging effect marker would be a set of key genes in a gene-expression fingerprint for a given exposure to a particular toxicant that is highly similar between a laboratory animal model and man (or human cells). In fact, regulatory agencies are encour-aging the use of modern molecular techniques, such as genomics. The FDA recognizes the importance of pharmacogenomics and encourages its use in drug development and has released guidance documents for submission [101]. With a greater understanding of toxicity mechanisms, smaller, more focused and reliable animal studies can be established, on a case-by-case basis (reduce and refine). In the future, in vitro systems using these ’omics tools may also help to further reduce the number of animals required for safety evaluation and maybe even replace them.

Novel rat nephrotoxicity markerTwo markers that can detect toxicity in the kidney and liver are alanine aminotransferase (ALT) and aspartate aminotransferase (AST). ALT and AST work well in liver and kidney when toxicity is already present and tissue has begun to break down. To assess renal injury in drug testing, nowadays two other well-accepted biomarkers are employed, namely serum creatinine and blood urea nitrogen (BUN), already described in 1904 and 1952, respec-tively. However, it has been observed that approximately 65% of kidney function must be lost before BUN and serum creatinine increase significantly. Furthermore, BUN level is highly influ-enced by protein metabolism, and creatinine is influenced by muscle breakdown. Thus, current kidney functional tests are either insensitive, variable or nonspecific to kidney injury [10,11].

By completing a major milestone under their Critical Path and Roadmap to 2010 initiatives, the FDA and the EMEA, respec-tively, have concluded that seven new urinary biomarkers of kid-ney injury submitted by the Critical Path Institute’s Predictive Safety Testing Consortium (PSTC) are considered qualified for particular uses in regulatory decision-making [102]. This consor-tium, a pharmaceutical industry public–private partnership led by the non-profit Critical Path Institute (C-Path) (Table 1), submitted the data for these seven urinary proteins (e.g., KIM-1, albumin, total protein, b2-microglobulin, cystatin C, clusterin and trefoil factor-3; see Table 2 for details of analytes) to both the FDA and the EMEA [103]. The submission included data from PSTC member companies, such as Novartis and Merck and Co, as well as lead-ing scientists at Harvard Medical School and the FDA’s research laboratories. Additional valuable input came from other PSTC members of the Nephrotoxicity Working Group.

www.expert-reviews.com 533

EditorialValue of new biomarkers for safety testing in drug development

The preclinical data identified thus far have demonstrated evidence for the superiority of the new renal biomarkers over the current standards used to assess renal injury in drug test-ing (serum creatinine and BUN). Further efforts will focus on the extended clinical qualification of biomarkers that could allow clinicians to detect kidney injury in patients earlier than

current clinical practice allows. As a service business, Rules-Based Medicine, Inc., (RBM) examines rat and human samples in its CLIA-certified laboratory [104] and made a Rat Kidney xMAP Testing Kit available through its worldwide distribu-tion partner, EMD Chemicals, Inc., at the beginning of this year [105].

Table 1. International initiatives and consortia.

International initiative or consortium

Comment Ref.

Innovative Medicines Initiative (EU)

The Seventh Framework Programme (FP7) bundles all research-related EU initiatives [108]

Critical Path Initiative (US FDA) The Critical Path Initiative is the FDA’s effort to stimulate and facilitate a national effort to modernize the scientific process through which a potential human drug, biological product, or medical device is transformed from a discovery or ‘proof of concept’ into a medical product. In a 2004 white paper, now known as the Critical Path Initiative, the FDA called attention to an alarming decline in the number of innovative medical products being submitted for FDA approval

[109]

Medicamentos Innovadores (Spain)

The Spanish Technological Platform for Innovative Medicines is an initiative that aims to encourage biomedical research into new medicines through cooperations between academia, industry and others

[110]

Top Institute Pharma (The Netherlands)

Top Institute Pharma aims to achieve leadership in research and education in areas that are critical for pharma industry to compete internationally. Focus is improving the efficiency of the entire process of drug discovery and development

[111]

European Clinical Research Infrastructure Network

This is the pan-European Infrastructure for clinical trials providing research infrastructures and services for multinational clinical research

[112]

Safety Biomarkers (UK) The Technology Programme has allocated funding in Safety Biomarkers for Pharmaceutical Development

[113]

Critical Path Institute (University of Arizona, USA)

Arizona Center for Education and Research on Therapeutics is a collaboration between C-Path and the College of Pharmacy of The University of Arizona. Its goal is to improve therapeutic outcomes and reduce adverse events caused by drug interactions

[102]

Center for Biomedical Innovation (MIT)

The Center for Biomedical Innovation has joined the MIT Engineering Systems Division, allowing the two to more closely align their efforts in tackling large-scale challenges in the healthcare industry

[114]

Toxicogenomics Project (Japan) The Toxicogenomics Project is a cooperative research project joining both the national and private companies to create a toxicology database that enables both forward and reverse toxicology

[115]

Proteome Factory Consortium (JPMA)

The Foundation aims to take the initiative in health science and as an active organization that has capacity to contribute internationally as well as domestically

[116]

Japan Medical Association Center for Clinical Trials (Japan)

Japan Medical Association Center for Clinical Trials funds and supports clinical trials and research through several grant schemes; conducts the ‘Large Scale Clinical Trial Network Project,’ subsidized by the Ministry of Health, Labour and Welfare

[117]

Chemical Effects in Biological Systems programme of the US National Institute of Environmental Health Sciences

National Institute of Environmental Health Sciences tries to define how environmental exposures affect our health, how individuals differ in their susceptibility to these exposures, and how these susceptibilities change over time, supported by Chemical Effects in Biological Systems

[118]

Consortium for Metabonomic Toxicology project (UK)

The utility of metabonomics in the evaluation of xenobiotic toxicity has beencomprehensively assessed by the Consortium for Metabonomic Toxicology, formed between five major pharmaceutical companies and Imperial College London, UK

[119]

Health and Environmental Sciences Institute Genomics Committee (USA)

The Health and Environmental Sciences Institute is a nonprofit institution of the International Life Sciences Institute to provide an international forum to advance the understanding of scientific issues related to human health, toxicology, risk assessment, and the environment

[120]

InnoMed Consortium (EU) InnoMed PredTox is a joint Industry and European Commission collaboration to improve drug safety. The consortium is composed of ten pharmaceutical companies, three academic institutions and two technology providers

[121]

In order to save resources and gain statistically significant data many partnerships are set up relying on international initiatives and consortia often publicly funded.

Expert Rev. Mol. Diagn. 9(6), (2009)534

Editorial Hewitt & Herget

The recent announcements signify the first-ever regulatory biomarker qualification decision under the FDA’s and EMEA’s joint Voluntary Exploratory Data Submission review process [101]. The FDA’s conclusions represent the completion of their process, while the EMEA statement provides the Committee for Medicinal Products for Human Use (CHMP) opinion and is being released for a short period of public consultation before the decision is finalized, as per the EMEA’s new draft Biomarker Qualification process. CHMP performs the scientific assessment of medicinal products within the EMEA [106].

However, regarding the translatability of these rat toxic-ity marker to human, there are still some open questions. Limitations exist in the data package submitted for all the iden-tified bio marker (including Kim-1, clusterin, trefoil Factor-3 and cystatin C). Although the PSTC provided information on dose- and time-dependent changes in the biomarkers and the appearance of histopathological alterations during periods of dosing, data are lacking to establish a clear correlation between the biomarker and the evolution of the nephrotoxic alterations over time, as documented by histopathology. The reversibility of the biochemical changes is unsatisfactorily correlated with kid-ney function recovery. Therefore, the use of these biomarkers in ‘monitoring’ renal toxicity at this stage is not adequately demon-strated. Additional drawbacks were that comprehensive data con-cerning bodyweight, food and water consumption were missing.

Histopathology scores were based on the evaluation of only one section of one kid-ney. Therefore, data were not satisfactory to establish a clear temporal correlation between lesions and biomarker levels or to demonstrate that specific biomarkers can establish the exact location of injury.

Although the use of animals has been essential to help define the molecular basis and the progression of model nephropathies, it may be inappropriate to extrapolate ani-mal toxicology data directly to man because of marked species, strain, dietary and sex differences. In addition, there may be dif-ferences in dosing levels and regimen and in the absorption, distribution, metabolism, and excretion of potential nephrotoxins. The drugs were given to rats for 2 weeks, which may reflect an acute treatment, but can the data be extrapolated to long-term treatment of chronically ill patients? What about patients with additional diseases, such as widespread diabetes? The vulnerability of the kidney to toxicity from exposure to a particular drug or chemical is the culmina-tion of several groups of risk factors. In any one person, multiple factors may contribute to their susceptibility to toxins. The holistic approach to nephrotoxicity assessment also demands that in vivo investigations are not

separated from in vitro studies, and that data continue to be derived from several different animal species and related to accurately conducted epidemiological and clinical studies.

“...the FDA and the EMEA … have concluded that seven new urinary biomarkers of kidney injury

submitted by the Critical Path Institute’s Predictive Safety Testing Consortium are considered qualified for particular uses in regulatory

decision‑making.”Nevertheless, it is recognised that it is worthwhile to further

explore in early clinical trials the potential of Kim-1, albumin, total protein, b2-microglobulin, clusterin, trefoil factor 3 and cystatin C as clinical biomarkers for acute drug-induced kidney injury (Table 2). Additional details of the science supporting the utility of these biomarkers in nonclinical and clinical settings, as well as recommended best practices for safety biomarker qualifi-cation, and thoughts from the EMEA and the FDA on working with the PSTC to evolve biomarker qualification are summarized in several manuscripts to be published in the upcoming months. Until further data are available to correlate the biomarker with the evolution of the nephrotoxic alterations, and their reversibil-ity, their general use for monitoring nephrotoxicity in clinical set-ting is so far questionable. The process of how to best implement

Table 2. Seven new nephrotoxicity markers.

Analyte Function Damaged region

b2m Small cell surface protein shed into the blood and normally reabsorbed by the proximal tubules of the kidney. High b2m levels result from lack of efficient reabsorption due to renal failure

Proximal tubule, glomerulus

KIM-1 Membrane protein expressed at elevated levels after injury of proximal tubule epithelial cells due to ischemic renal damage

Proximal tubule

TIMP-1 Regulates extracellular matrix synthesis and degradation and, along with matrix metalloproteinases, is essential for tumor growth and health

Proximal tubule, distal tubule

Clusterin(apolipoprotein J)

Conserved protein induced during tissue injury or remodeling

Proximal tubule, distal tubule

Cystatin C Extracellular inhibitor of cysteine proteases normally expressed in vascular wall smooth muscle cells

Glomerulus

Albumin Most abundant plasma protein essential for maintaining the osmotic pressure; acts also as a plasma carrier by nonspecific binding of hydrophobic macromolecules

Glomerulus

Total protein Multiple proteins of serum and kidney Unspecific

The results of the Predictive Safety Testing Consortium were submitted to the FDA and EMEA in 2008, leading to the listing of seven urinary kidney damage biomarkers: KIM-1, b2-microglobulin, cystatin C, clusterin, total protein and albumin. These markers are considered acceptable in the context of nonclinical drug development for the detection of acute drug-induced nephrotoxicity [107].b2m: b2-microglobulin; KIM-1: Kidney injury molecule 1; TIMP-1: Tissue inhibitor of matrix metalloproteinase-1.

www.expert-reviews.com 535

EditorialValue of new biomarkers for safety testing in drug development

these biomarkers in a further development program should be discussed on a case-by-case basis in the context of the new EMEA qualification advice [106,107].

Therefore, both the FDA and EMEA also recommend further investigation in clinical studies of these biomarkers, supplementary to the current standards BUN and serum creatinine. Additionally, the FDA has acknowledged that the new renal biomarkers could be used in certain circumstances to enable the regulatory advance-ment of certain highly promising drugs into human studies – drugs that, in the past, would otherwise have been abandoned because traditionally used markers were not able to detect early-onset renal injury. If a sponsor can show, in the preclinical set-ting, that the new biomarkers can be used to detect early toxicity, to monitor onset and reversibility, and to manage any potential renal adverse effects of a new drug with significant therapeutic potential, the sponsor could discuss this with the associated FDA review division and the EMEA about engaging in clinical stud-ies using these new biomarkers (in addition to current standard assessments) on a case by case basis. Finally, the FDA and EMEA endorse the submission of additional clinical data, to widen the context of the regulatory clinical use of these biomarkers.

This milestone represents a significant advance, not just for the FDA, EMEA and the pharmaceutical industry, but for public health in general, as new tools to enhance the safety evaluation of promis-ing new medicines are now available for use in drug development. The fact that the data submitted was the result of just over a year of consortium work demonstrates not only the value of such collabora-tive efforts but also the commitment of the member companies to advance new safety tests. Owing to the promise the new markers have shown in the data already generated, further translational stud-ies are planned by the PSTC in order to generate the needed data to expand the qualification of the biomarkers for broader clinical use.

ConclusionSystems biology to understand toxicologyRecent high-profile drug withdrawals increase the pressure on regulators and the pharmaceutical industry to improve preclini-cal safety testing. The new urinary kidney biomarkers (Table 2) are promising and considered acceptable in the context of non-clinical drug development for the detection, but less for monitor-ing, of acute drug-induced nephrotoxicity. Therefore, the FDA and EMEA acknowledge the benefits from the voluntary use of the new renal biomarkers.

Over the past few years, many researchers have been working towards a fully integrated ‘systems biology’ approach to science. This, in the log-term, will benefit toxicologists as more is under-stood about toxicity mechanisms and the relationship/concord-ance between animal data and toxicity in man. Understanding mechanisms of drug toxicity is an essential step toward this goal by providing the basis for mechanism-based risk assessments, and the current new technologies, including ‘omics technologies and new, novel safety biomarkers, will enhance this improvement. However, it is important that assurances are given that the model systems will be reliable and the results comparable with, if not better than, those from the current classical toxicological meth-ods, and are not being implemented solely on ethical grounds (i.e., the 3Rs).

Financial & competing interests disclosureThomas Herget and Philip Hewitt are employees of Merck KGaA (Germany). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

References1 Russell WMS, Burch RL. The Principles of

Humane Experimental Technique. Springfield, London, UK (1959).

2 Hamadeh HK, Bushel PR, Jayadev S et al. Prediction of compound signature using high density gene expression profiling. Toxicol. Sci. 67, 232–240 (2002).

3 Schena M, Shalon D, Davis RW, Brown PO. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270, 467–470 (1995).

4 Ganter B, Zidek N, Hewitt PR, Müller D, Vladimirova A. Pathway analysis tools and toxicogenomics reference databases for risk assessment. Pharmacogenomics 9, 35–54 (2008).

5 Waring JF, Jolly RA, Ciurlionis R et al. Clustering of hepatotoxins based on mechanism of toxicity using gene expression profiles. Toxicol. Appl. Pharmacol. 175, 28–42 (2001).

6 Steiner G, Suter L, Boess F et al. Discriminating different classes of toxicants by transcript profiling. Environ. Health Perspect. 112, 1236–1248 (2004).

7 Yao X, Panichpsal K, Kurtzman N, Nugent K. Cisplatin nephrotoxicity: a review Am. J. Med. Sci. 334, 115–124 (2007).

8 Nicholson JK, Lindon JC, Holmes E. “Metabonomics”: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR data. Xenobiotica 29, 1181–1189 (1999).

9 Xu EY, Perlina A, Vu H et al. Integrated pathway analysis of rat urine metabolic profiles and kidney transcriptomic profiles to elucidate the systems toxicology of model nephrotoxicants. Chem. Res. Toxicol. 21, 1548–1561 (2008).

10 Han WK, Bailly V, Abichandani R, Thadhani R, Bonventre JV. Kidney Injury Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury. Kidney Int. 62, 237–244 (2002).

11 Hewitt SM, Dear J, Star RA. Discovery of protein biomarkers for renal diseases. J. Am. Soc. Nephrol. 15(7), 1677–1689 (2004).

Websites

101 US FDA: Genomics www.fda.gov/Drugs/ScienceResearch/ResearchAreas/Pharmacogenetics/ default.htm

102 Critical Path Institute; Predictive Safety Testing Consortium (PSTC) www.c-path.org/pstc.cfm

103 EMEA: First EMEA-FDA joint biomarkers qualification process www.emea.europa.eu/htms/human/mes/biomarkers.htm

104 RBM, the Biomarker testing laboratory www.rulesbasedmedicine.com

105 Novagen: WideScreen™ Rat Kidney Toxicity Assays www.emdbiosciences.com/html/NVG/home.html

Expert Rev. Mol. Diagn. 9(6), (2009)536

Editorial Hewitt & Herget

106 EMEA: Committee for Medicinal Products for Human Use, Jan. 2009 www.emea.europa.eu/pdfs/human/biomarkers/7289408en.pdf

107 FDA, European Medicines Agency to Consider Additional Test Results When Assessing New Drug Safety Collaborative effort by FDA and EMEA expected to yield additional safety data www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/2008/ucm116911.htm

108 European Commission: CORDIS. Seventh framework programme www.cordis.europa.eu/FP7/home.html

109 US FDA. Critical Path Initiative www.fda.gov/oc/initiatives/criticalpath

110 Medicamentos Innovadores. Spanish platform www.medicamentosinnovadores.org

111 TI Pharma. www.tipharma.com/home.html

112 European Clinical Research Infrastructures Network www.ecrin.org

113 3Rs studies considered for DTI Safety Biomarker funding. Research into safety biomarkers may have implications for the 3Rs (reduction, refinement and replacement of animals in research) www.nc3rs.org.uk/news.asp?id=241

114 Center for Biomedical Innovation http://web.mit.edu/cbi/

115 Toxicogenomics Project in Japan Web Site wwwtgp.nibio.go.jp/index-e.html

116 Vision of Japan Health Science Foundation www.jhsf.or.jp/English/vision.html

117 Japan Medical Association Center for Clinical Trials www.jmacct.med.or.jp/english/index.html

118 Chemical effects in biological systems http://cebs.niehs.nih.gov

119 Metabolomics.net. The metabolic portal www.metabolomics.net

120 The Health and Environmental Sciences Institute www.hesiglobal.org

121 InnoMed PredTox www.innomed-predtox.com