So far, in most animal experimental studies isolated food components h
ave been tested. However,as components may interact with each other at
different mechanistic levels, testing complex food mixtures more repr
esentative for human consumption patterns may better predict the ultim
ate carcinogenic risk. Studies were performed in Wistar rats using hum
an and rat control diets to assess the effect of relevant food factors
such as heat processing and the presence of non-nutrients in vegetabl
es and fruit. The complete human diets, containing meat, bread and egg
s, with or without vegetables and fruit, were composed according to me
an consumption figures, balanced for macro- and micronutrients. Experi
ments were performed with spontaneous as well as with chemical-induced
tumor models. Heat processing had no effect on tumor induction, while
vegetables and fruit only exerted a protective effect on chemically i
nduced tumors in rats fed low-fat animal diets. Data suggest interacti
on between major food factors in the human diet on colon carcinogenesi
s. (C) 1997 Elsevier Science Ireland Ltd.