Formerly regarded as an experimental curiosity–used sparingly in drug development programs by a handful of brave souls–pharmaceutical companies are embracing the 6-month transgenic mouse carcinogenicity study as an alternative to the 2-year mouse carcinogenicity bioassay, with the rasH2 model used most often.   

For decades, drug developers have used the 2-year carcinogenicity bioassay in two rodent species to assess a pharmaceutical’s carcinogenicity. But the assay has its drawbacks, including expense (> $1 million per bioassay), the long duration between study initiation and final report (~3 years), the large number of animals required (≥ 480 animals/study), the high background tumor incidence in control groups, and the general lack of useful mechanistic data.

Issued in 1997, The International Conference on Harmonization (ICH) Guidance S1B specifies assessment of carcinogenicity in one long-term study in a rodent (typically the rat) as well as another short-term or medium-term in vivo rodent study or long-term carcinogenicity study in a second rodent species. The S1B lists transgenic mouse models as a short- or medium-term rodent test system.

From 1996 to 2001, The International Life Sciences Institute/Health and Environmental Sciences Institute Alternatives to Carcinogenicity Testing Project, a multinational consortium of over 50 academic, government and industrial laboratories, evaluated six transgenic mouse models as well as the neonatal mouse for carcinogenicity testing. The results of this effort, in conjunction with the directives of ICH S1B, led to the initial acceptance by regulatory agencies of three primary transgenic mouse models for carcinogenicity testing:  the p53+/- model, the Tg.AC model, and the rasH2 model. Each model has a shorter testing duration (6 months), uses fewer animals (25/sex/group), and costs much less than a standard 2-year mouse carcinogenicity bioassay.

Background of the p53+/-, Tg.AC and rasH2 Models

Both proto-oncogenes and tumor suppressor genes are required for normal cellular growth and differentiation; proto-oncogenes are generally positive regulators of cell growth and tumor suppressor genes are negative. The cell can be transformed to a neoplastic phenotype by the accumulation of genetic changes through oncogene activation via point mutation, chromosomal translocation, overexpression or post-transcriptional/post-translational alterations as well as tumor suppressor gene inactivation through chromosomal loss/deletion or point mutation. A transgenic mouse with an activated/overexpressed oncogene or an inactivated tumor suppressor gene should be much more susceptible to carcinogens than a normal mouse, resulting in a more rapid induction of tumorigenesis.

The p53 tumor suppressor gene, a critical gatekeeper for preserving chromosomal integrity and maintaining cell cycle control, is mutated or lost in ~50% of human cancers.  The heterozygous p53+/- transgenic mouse contains a single functional p53 allele in each cell, has a low spontaneous tumor incidence, and responds primarily to genotoxic carcinogens. The US FDA and Japan’s Ministry of Health, Labor and Welfare (MHLW) consider the p53+/- model appropriate for the testing of genotoxic compounds, whereas the EMEA’s Committee for Proprietary Medicinal Products (CPMP) accepts the model for testing of both genotoxic and nongenotoxic compounds. Because the majority of the studies submitted for regulatory review have been negative, there is some concern that the p53+/- model may be relatively insensitive compared to other transgenic models and standard 2-year bioassays.

Both the Tg.AC mouse and the rasH2 mouse models are Ha-ras-derived models. At least one member of the ras family of proto-oncogenes (Ha-ras, Ki-ras, or N-ras) has been found to be mutated in approximately one-third of human cancers. The Tg.AC transgenic mouse has a v-Ha-ras gene with activating mutations fused to a zeta-globin promoter. Expression of the vector appears to be restricted to only a few tissues (e.g., skin and forestomach). Dermal application of genotoxic or nongenotoxic carcinogens in Tg.AC mice results in the production of papillomas. The FDA and CPMP have accepted the Tg.AC model for use in testing dermal products, whereas MHLW has not accepted the model, as it considered the vector too unstable. However, as most of the Tg.AC studies submitted for regulatory approval have had positive results, interest in the Tg.AC model as an alternative to the 2-year bioassay has dropped precipitously and few recommend its continued use.

The rasH2 mouse contains multiple copies of the human c-Ha-ras proto-oncogene as well as its native murine Ha-ras gene. In contrast to the Tg.AC model, the rasH2 model transgene is expressed in a larger number of tissues. The rasH2 model has a very low incidence of spontaneous tumors, responds to both genotoxic and nongenotoxic carcinogens and is accepted by the FDA, the MHLW, and the CPMP as appropriate for testing of compounds that fall into either category. In addition, the rasH2 model is considered neither insensitive nor prone to false positive results. For these reasons, the rasH2 transgenic mouse model has become the transgenic model of choice most often used as a replacement for a traditional 2-year mouse carcinogenicity study.

Transgenic Mouse Alternatives Conform with the 3Rs

Under the 3Rs (Replacement, Reduction, and Refinement), the use of a transgenic mouse alternative to the 2-year mouse bioassay qualifies as reduction, as the experimental group size is typically 25/sex/group in a transgenic study in comparison to ≥ 60/sex/group in a 2-year bioassay. One can also argue that the use of a transgenic mouse alternative constitutes refinement; scheduled euthanasia in a 6-month transgenic mouse study occurs long before the senescence-related morbidity that invariably results in the humane euthanasia of a relatively high proportion of study animals near the end of 2-year studies.

Use of a Transgenic Mouse Model

Before embarking on the use of a transgenic mouse model as a replacement for a conventional 2-year mouse bioassay, a company should secure regulatory agency approval of the plan, as regulatory agencies such as the FDA tend to approve transgenic mouse alternatives on a case-by-case basis. A 28-day range-finding study with full histopathology should be performed prior to the actual carcinogenicity study. Range-finding studies are normally conducted in wild-type mice, because the presence of transgenes (or inactivated genes) is not expected to affect either the subchronic toxicity or pharmacokinetics of the test article under normal circumstances. However, the appropriate parental strain (e.g. CByB6F1) of the transgenic mouse is expected to be used, rather than a more commonly used mouse strain (e.g. CD-1). For studies intended for FDA submission, the results of the 28-day range-finding study and the proposed protocol for the 6-month transgenic mouse carcinogenicity study should be submitted to the FDA Carcinogenicity Assessment Committee for prior review.


  1. French, J.E., et al.  Panel Discussion: Alternative Mouse Models for Carcinogenicity Assessment.  Toxicol. Pathol. 38: 72-75, 2010.
  2. ICH Harmonized Tripartite Guideline S1B. Testing for Carcinogenicity of Pharmaceuticals.  July, 1997.
  3. Jacobson-Kram, D., et al.  Use of transgenic mice in carcinogenicity hazard assessment. Toxicol. Pathol. 32 (Suppl 1): 49–52, 2004.
  4. Morton, D., et al.  The Tg rasH2 Mouse in Cancer Hazard Identification.  Toxicol.  Pathol.  30: 139-146, 2002.
  5. Russell, W.M.S., and Burch, R.L. The Principles of Humane Experimental Techniques. Methuen & Co., London, 1959.
  6. Storer, R.D., et al.  An industry perspective on the utility of short-term carcinogenicity testing in transgenic mice in pharmaceutical development.  Toxicol. Pathol.  38:51-61, 2010.