Can risk assessment and chemoprevention research rely on surrogates for tumour yield?

The main areas of interest within toxicological pathology relevant to cancer prevention are risk assessment, mechanisms of carcinogenesis and chemoprevention. All rely to a large extent on studies in experimental animals. In the present issue of the APJCP, Subapriya and Nagini ( 2003) report on the chemopreventive potential of an ethanol extract of neem leaves in rats, providing evidence of reduction in gastric tumour yield as well as diminished lipid peroxidation within lesions and change in antioxidant parameters in background tissue. The paper thus provides a good example of a study encompassing both mechanisms and preventive potential. Fukushima et al (2003) earlier argued for the necessity of taking mechanisms into account in assessing hazard risk with animal models and many of their points are pertinent to a wider ongoing debate regarding the applicability of surrogate end point biomarkers (SEBs). In a recent issue of Cancer Epidemiology Biomarkers and Prevention a point/counterpoint discussion of this theme was included. While Armstrong et al (2003) concluded only limited usefulness of SEBs, Kelloff and others (2003) argued for high utility in the development of cancer chemopreventive agents against sporadic cancers. Clearly, distinction can be made between those biomarkers which are actual focal lesions with some potential to give rise to cancers and others which reflect processes underlying neoplastic development (see Moore et al., 1999, Ito, 2000 and Brewer et al., 2001, for further discussion of this point). Typical examples of each are listed in the Table. We are now in a position to identify very early lesions in many of the organs of the body which are important in terms of human cancer burden, using either morphological characteristics, histochemistry or immunohistochemistry.