The Center for Alternatives to Animal Testing is an academic center affiliated with the Division of Toxicological Sciences in the Department of Environmental Health Sciences of the Johns Hopkins University Bloomberg School of Public Health.
Text Size:
TestSmart Pharmaceuticals
May 7-9, 2001
PIER 5 HOTEL
711 Eastern Avenue
Baltimore, Maryland
Sponsors: 3M, Bristol-Meyers Squibb Company, Huntingdon Life Sciences, Schering-Plough Research Institute, and Taconic Farms, Inc.
Abstract for TestSmart--Pharmaceuticals: An Efficient and Humane Approach to Predictors of Potential Toxic Effects of Drugs
Use of Alternative Models for Assessment of Carcinogenic Potential: Results of ILSI Collaborative Studies
Denise Robinson
ILSI Health and Environmental Sciences Institute
During discussions of the Safety Expert Working Group of the International Conference on Harmonization, the approach to assessment of human carcinogenic risk from pharmaceuticals was considered. Specifically, the value of the two standard rodent bioassays was questioned. As a result of these discussions, it was concluded that, under certain circumstances, the information from a second species (usually the mouse) was not conclusive and that data from alternative shorter-term assays may provide more useful information. At the time these discussions were being held (1996), there was very little information available on how some of the newer transgenic or knockout mouse models performed with pharmaceuticals. As this was the focus of the ICH discussants, it was decided to undertake a collaborative study to evaluate a series of prototype compounds in several of these newer assays. To facilitate this effort, the Health and Environmental Sciences Institute (HESI) of the International Life Sciences Institute (ILSI) was selected to enable the participation of a broad array of interested parties. Twenty-one chemicals were tested in 5 in vivo and 1 in vitro assays. These chemicals represented a broad range of activity from non-genotoxic to genotoxic human carcinogens. Fifty-three sponsors conducted a total of 138 assays on these chemicals. The resultant database developed from this effort has enabled several conclusions to be made regarding the value of the alternative models evaluated. The responses seen in the in vivo models demonstrated that these models are not overly sensitive. The available data suggest that these models can be used effectively as part of a comprehensive, weight-of-evidence approach to assessment of human risk of cancer from chemical exposure. It is also clear that these models have limitations and cannot be used as a definitive determinant of human risk. As such, they provide a useful addition to the risk assessment process and enable us to begin to move away from inappropriately weighting the results of the traditional rodent bioassay.
Testing for Carcinogenicity -- Expectations
- Identify chemicals having potential to cause cancer in humans
- Provide indication of carciongenic potency —
- Dose response relationship
- Applicable to all use categories
- Industrial chemicals, food additives, drugs, etc.
- New methods should improve testing
Development of Alternative Models
- ICCVAM-Congressional mandate to develop and validate alternative models
- NIEHS/NTP development of short term transgenic models, NCTR evaluation of the neonatal model
- ICH Topic on Testing of Pharmaceuticals for carcinogenicity
- FDA acceptance of an alternative model for one of the 2 year carcinogenicity studies
Using Transgenic Models to Assess Carcinogenicity
PREMISE-
- Alteration in genes related to human cancer development
- Genetic event insufficient to spontaneously induce cancer
- Potential to predispose animal to tumor development upon exposure to carcinogens
- Tumor induction may be more relevant in assessing human cancer risk
p53-deficient model -- gene knockout
- p53 mutated or lost in many (50%) human cancers
- Li-Fraumeni syndrome — increased susceptibility for many cancers
- p53 protein — gatekeeper for chromosomal integrity/ cell cycle control
- p53 heterozygote - single wild type allele
- Increased susceptibility to tumors
- Responds mostly to genotoxics
TG:AC model -- v-Ha-ras transgene
- ras oncogene mutated in approx. 30% of human cancers
- Tg.AC — v-Ha-ras gene inserted; expression regulated by zeta-globin promoter
- Dermal reporter system — skin papillomas
- Low spontaneous incidence of skin tumors
- Dermal application of carcionogens activates transgene and induces papillomas
- Also responds via oral route of exposure
- Responds to genotoxics and tumor promoters
rasH2 model -- c-Ha-ras transgene
- ras oncogene mutated in about 30% of human cancers
- rasH2 — insertion of multiple copies of the human c-Ha-ras proto-oncogene
- Transgene expressed in normal and tumor tissues
- Low incidence of spontaneous tumors
- Responds to genotoxics and tumor promoters
XPA repair deficient model
- Human syndromes involving DNA repair defects are predisposed to cancer
- Nucleotide excision repair — major route of DNA repair
- XPA double knockout mouse — completely deficient in DNA repair
- Murine equivalent of human Xeroderma Pigmentosa
- XPA/p53 double transgenic — enhanced tumor response
- No tissue restriction
- Responds to genotoxic stimuli
ICH Guidance for Testing for Carcinogenicity of Pharmaceuticals
Issues with Implementation of Guideline
- Inadequate database on performance of alternative assays
- Rationale for selection of model for 2nd in vivo test
- Acceptability of models in different regulatory regions
ILSI Health and Environmental Sciences Institute
- Non-profit research and educational organization
- Provides organizational structure and international forum for collaborative scientific activities
- Participation from industry, government, academia
- Reputation for scientific focus and objectivity
HESI ACT Committee
Steering Committee
- Dr. Sid Aaron, Pharmacia, Inc.
- Dr. Gerald Long, Eli Lilly and Company
- Dr. R. Michael McClain, Hoffmann-La Roche Inc.
- Dr. James MacDonald, Schering-Plough Corp.
- Dr. James Popp, DuPont Pharmaceuticals
- Dr. Denise Robinson, ILSI HESI
- Dr. Ray Stoll, Boehringer Ingelheim
- Dr. Richard Storer, Merck & Company Inc.
- Dr. Toshimi Usui, Central Institute for Exp. Animals
HESI ACT Committee
Scientific Advisors
- Dr. Dan Casciano, FDA/NCTR
- Dr. Sam Cohen, Univ. of Nebraska Medical Center
- Dr. Jay Goodman, Michigan State University
- Dr Yuzo Hayashi, Kitasato University
- Dr. James Swenberg, Univ. of North Carolina
- Dr. Norikazu Tamaoki, Tokai University
- Dr. Ray Tennant, NIEHS
- Dr. Coen van Kreijl, RIVM
Alternative models
Validation vs Evaluation
- No model can be validated as classically defined
- Validated against what? 2 year bioassay not "validated"
- Comparisons to the 2 year bioassay are not very meaningful
- More than 50% are "positive" in the 2 species test
- Number of positive findings considered not relevant to human risk
- Need to have understanding of how each model responds to ensure that results are interpretable
HESI Alternatives to Carcinogenicity Testing (ACT) Committee
Scope and Objectives
- Through a global partnership, expand database on new methods
- Establish common experimental approach
- Contribute to sound scientific basis for understanding value of new assays
- Provide public forum for discussion of evolving state of science
Models to be Evaluated
- p53+/- knockout mouse model
- rasH2 transgenic mouse model
- Tg.AC transgenic mouse model
- xpa-/- knockout mouse model
- neonatal mouse model
- Syrian Hamster Embryo cell transformation assay
Government and Research Laboratories
- National Toxicology Program, US
- National Institute of Environmental Health Sciences, US
- Center for Drug Evaluation and Research, US FDA
- National Center for Toxicological Research, US FDA
- National Institute of Public Health and the Environment (RIVM), Netherlands
- Veterinary and Food Administration, Denmark
- Central Institute for Experimental Animals, Japan
Industry Laboratories
- AstraZeneca Pharmaceuticals
- Aventis Pharma
- Bayer Corporation
- Boehringer Ingelheim Pharmaceuticals
- Bristol-Myers Squibb Co.
- Centre International de Toxicologie (CIT)
- Covance Laboratories.Daiichi Pharmaceutical Co., Ltd.
- Dainippon Pharmaceutical Co., Ltd.
- Dow Chemical Company
- DuPont Pharmaceuticals
- Eisai Co., Ltd.
- Eli Lilly & Co.
- Fujisawa Pharmaceutical Co., Ltd.
- G.D. Searle/Monsanto
- Genzyme Transgenics
- Glaxo Wellcome, Inc.
- Hoechst Marion Roussel
- Hoffmann-La Roche Inc.
- Huntingdon Life Sciences
- Janssen Pharmaceutica NV
- Lederle (Japan) Ltd.
- Merck and Co., Inc.
- Mitsubishi-Tokyo Pharmaceuticals, Inc.
- Nippon Shinyaku Co., Ltd.
- Novartis Pharmaceuticals Corp.
- Novo Nordisk Pharma Co., Ltd.
- NV Organon.Pfizer Inc.
- Pharmacia & Upjohn Company
- The Procter & Gamble Company
- Purdue Pharma LP
- RW Johnson Pharmaceutical Research Institute
- Rhône-Poulenc Rorer
- Sankyo Co., Ltd.
- Sanofi Pharmaceuticals, Inc.
- Scantox AS
- Schering-Plough Corporation
- Shionagi & Co., Ltd.
- SmithKline Beecham Pharmaceutical
- Solvay Pharmaceuticals
- Springborn Laboratories
- Synthelabo Recherche
- Takeda Chemical Industries Ltd.
- Tanabe Pharmaceutical Co., Ltd.
- Teikoku Hormone Mfg. Co., Ltd.
- TNO Toxicology
- Yamanouchi Pharmaceutical Co., Ltd.
- Wyeth-Ayerst Research.
Rationale for Compound Selection
- Supplement existing database (primarily industrial, genotoxic chemicals)
- Focus on pharmaceuticals - well-studied compounds with extensive human experience
- Established toxicology database (in vitro and in vivo)
- Representative of broad array of mechanisms
- Non-proprietary, multi-source
21 Compounds Selected for Study
- genotoxins
- immunosuppressants
- hormonal agents
- enzyme inducers
- cell proliferators
- peroxisome proliferators
- receptor active agents
- non-carcinogens
Compounds Under Assessment through HESI ACT Committee's Collaborative Research Program
| Class | Compound |
| Genotoxic Human Carcinogens | Cyclophosphamide, Melphalan, Phenacetin |
| Immunosuppressant Human Carcinogen | Cyclosporin A |
| Hormones | Diethylstilbestrol, Estradiol |
| Rodent Carcinogens/Putative Human Non-Carcinogens (based on Human Data) | Phenobarbital, Clofibrate, Reserpine, Dieldrin, Methapyrilene |
| Rodent Carcinogens/Putative Human Non Carcinogens (by Mechanism) | Haloperidol, Chlorpromazine, Chloroform, Metaproterenol, WY-14643, DEHP, Sulfamethoxazole |
| Non-Carcinogens | Ampicillin, D-Mannitol, Sulfisoxazole |
Development of Common Experimental Approach
- Permit establishment of robust database which supports comparisons
- Coordination across all participating laboratories
- Protocol development and dose selection
- Evaluate reproducibility of findings
Standardized Protocols -- General Parameters
- 4-week dose range finding study
- Standard positive controls
- 15 animals/dose group
- 3 dose groups plus vehicle control
- Vehicle and high dose control in wild type animals
- Full histopathology on control and high dose animals
- Tissues archived from all dose groups
Functions of Assay Working Groups
- Establish common format for data collection/reporting
- Ensure uniformity of approach across different laboratories
- Review/approve dose selection based on results from 4 week studies
- Resource for protocol questions
- Establish interpretive criteria to evaluate assays results
- Focal point for data review and analyses
General Criteria for Evaluation of Responses
- Background incidence of tumor in model
- Multiple tissues involved
- Dose-response relationship
- Evidence of biological effect of compound
- Pharmacology/biochemistry of compound
- Compound-related toxicity/mortality
- Statistical significance
- All studies accompanied by rationale statement for conclusions
Non-Carcinogens
- Ampicillin
- D-Mannitol
- Sulfisoxazole
- Negative in All Models
Genotoxic Carcinogens
| Cyclophosphamide | Positive in all |
| Melphalan | Positive in all |
| Phenacetin | Positive in rasH2 Negative in p53, TgAC, XPA and neonatal |
Phenacetin
- Aromatic Amide
- Mutagenic
- Rat - Renal Cell, Urothelial, Nasal Cavity
- Mouse - Renal cell, Urothelial
- Human - Urothelial Carcinoma
Immunosuppressant (Human Carcinogen)
Cyclosporin A
- Positive in p53, XPA, XPA/p53
- Equivocal in TgAC
- Negative in rasH2, Neonatal
Hormones (Human Carcinogens)
Diethylstilbestrol
- Positive in p53, rasH2, TgAC(D), XPA, XPA/p53
- Negative in neonatal, TgAC (oral)
Estradiol
- Positive in TgAC(D), XPA/p53
- Negative in p53, rasH2, TgAC (oral), XPA, Neonatal
Rodent Carcinogens — Putative Human Non-Carcinogens (Epidemiology or Mechanism)
Negative in All
- Phenobarbital
- Methapyrilene
- Reserpine
- Dieldrin
- Haloperidol
- Chlorpromazine
- Chloroform (Chloroform equivocal in p53)
- Metaproterenol
- Sulfamethoxazole
Peroxisome Proliferators
| Clofibrate | Positive in rasH2, TgAC(D) Equivocal in XPA Negative in p53, Neonatal |
| DEHP | Positive in ras H2 Equivocal in p53 Negative in TgAC, XPA, XPA/p53, Neonatal |
| WY-14643 | Positive in rasH2, XPA Equivocal in TgAC (diet) Negative in p53, TgAC (oral) |
Comparison of Models
| Neonatal | Genotoxic agents 1 Year |
| TgAC | Picks up estrogenic and immunosuppressive chemicals |
| p53, rasH2, XPA | Similar results XPA requires 9 months |
| SHE | Not an In Vivo endpoint Positive for Genotoxic and nongenotoxic |
Conclusions
- Alternative tests are not 100% predictive of risk to humans
- Alternative Tests can provide useful information for weight-of-evidence evaluation
- Mechanistic basis for models needs further study
Alternative Carcinogenesis Tests
Pro
- Shorter duration
- Have genes that may be relevant to human cancer
- Cheaper (overall)
- Fewer animals
Con
- Expensive (mice)
- Still left with the interspecies extrapolation problem
Weight of Evidence
- Genotoxicity - DNA Reactivity
- Structure Activity Relationship Alerts
- Rat Bioassay
- Alternative Tests
- Mechanistic Information
- Pharmacokinetics
Outcome of ILSI Research Program
- Robust dataset generated on new models
- Data from compounds from various mechanistic and chemical classes
- Historical databases on spontaneous lesions for each model
- Database on positive control responses
- Characterization of metabolism of transgenic vs parental strains
- Protocol evaluation
Issues for Continued Discussion
- Protocols
- number of animals per dose group
- use of positive controls
- tissues to be evaluated
- exposure duration
- route of exposure
- wild type controls
- Histopathology
- macroscopic vs. microscopic lesions
- evaluation of rare tumors
- malignant vs. benign vs. other proliferative lesions
Issues for Continuing Discussion
- Strategies for use of alternative assays
- pre-screening
- substitutes for bioassays
- adjunct, mechanistic investigations
- Selection of one model over another
- How to integrate data from alternative models with other traditional toxicity data
- Validation process - how to evaluate future models
Continuing Activities for 2001
- Remaining studies undergoing final review by AWGs
- Publication of workshop proceedings and data compendium — Toxicologic Pathology
- ILSI ACT Database
- to be completed by end 2001
- Meeting in Fall 2001 to consider potential next steps