The strengths and limitations of the mouse as a model for cancer prevention research dominated discussions at a retreat last fall held by the NCI Cancer Prevention Faculty.

Molecular biologists, mouse geneticists, epidemiologists, and clinical oncologists from diverse laboratories and branches took a step back and contemplated the work of the last two decades in preclinical in vivo models. Have they been useful in modeling human disease and in intervening in the process of cancer progression?

The faculty opened the retreat by agreeing to a broader definition of the scope of cancer prevention activities—not just preventing the disease but also delaying progression at any stage, thereby resulting in delayed cancer mortality (or mortality from other causes). They then assessed the track record of the mouse model within this context.

Mouse Cons and Pros

Specific limitations to the applicability of research in inbred rodents to a diverse human population include:

Different metastatic patterns—common human cancers often metastasize to bone, lung, brain, and liver, whereas metastases are found predominantly in lung in mice.

Some pathways of cancer in rodents, such as P450-mediated xenobiotic metabolism, are different in humans.

On the other hand:

Genetically engineered mice susceptible or resistant to cancer have proved to be relevant models for testing hypotheses generated by epidemiological, laboratory, and treatment-based observations and in staging interventions in human disease.

Mouse and human gene expression profiling is revealing that mouse models can be predictive for scoring tumor progression or regression in humans.

New technologies—such as gene expression arrays and proteomics analysis, along with comparative genomic hybridization and fluorescence imaging—are accelerating the pace of discovery of new molecular targets and of early and intermediate cancer endpoints to be used in scoring outcomes of interventions.

These technologies, together with conditional mouse models engineered for susceptibility or resistance and the Molecular Targets pipeline for drug development, are among the resources NCI offers interested investigators.

Some Mouse Studies

Nutrition. Points made on the role of nutrition in cancer included:

Reduction of energy consumption by 20 percent in a mouse model was found to have an anticancer effect.

The APC 1638 mouse model of intestinal cancer recapitulates the human situation of 60 years ago when the majority of intestinal tumors were located in the small intestine; that the administration of a "Western diet" to APC mutant mice shifts the cancer incidence to the colon suggests that changes in the American diet are contributing to this common cancer.

Combinations. Combinations of interventions, each targeting a different pathway, present an attractive possibility for circumventing the development of drug-resistant cancer cells. Current colon cancer prevention trials are targeting cyclooxygenase-2 and ornithine decarboxylase. Investigators emphasized the importance of context and its influence on whether a given combination of drugs is synergistic or antagonistic.

A Possible Extrapolation

It was noted that studies of cancer incidence ought to be undertaken in patients being treated for other diseases with potentially carcinogenic agents—for example, patients with type 2 diabetes being treated with PPAR-g agonists, which have been implicated in cancer in animal models.

A future forum will focus on human cancers directly and formulate recommendations for molecular-targeted approaches to carcinogenesis prevention.

Return to Table of Contents