Animal Tumour Models: The Intrusion of Artefacts

Abstract

Such are the technical and ethical limitations imposed on the clinical observer that our knowledge of the mechanisms of metastasis and our development and assessment of methods of restraining disseminated disease are very dependent on the use of experimental animal tumour systems. They are ‘models’ in the sense that they are intended to represent the clinical situation. The similar pathogenesis of cancer in different mammalian species gives the experimental cancer researcher great scope in confidently extrapolating his findings to clinical practice. But his confidence in interspecies analogy is merited only when there is scrupulous avoidance of the many laboratory contingencies which can and do compromise the validity of an animal model, By ‘artefact’ we here refer to features or conditions which are peculiar to the model and do not prevail in the clinical situation. Artefacts may be intrinsic to the model or can result from its technical handling. The commonest and most influential artefacts are avoidable, and should be avoided. Others are unavoidable using the small rodent models to which most researchers are confined by their need to use isotransplants; they mostly relate to the large differences of body size and tumour growth rate which distinguish man and laboratory rodents. These scale differences have to be prudently accommodated in attaching clinical relevance to the results of animal experiments, For example, occult metastases in the mouse will usually merit the description ‘micrometastases’ whereas occult clinical metatases commonly given this description are usually quite sizeable tumours which have not yet attained detection by non-invasive imaging techniques.