Review of experiments on multigeneration carcinogenicity: of design, experimental models and analyses.

It is widely accepted that parental exposure to a carcinogen may result in an increased risk for cancer in otherwise untreated offspring. Results of a new experiment and a review of published studies on multigeneration carcinogenesis are presented and discussed on the basis of the animal model used and of the design and analysis of the individual experiments. Several of the studies indicate an effect of multigenerational carcinogenesis, characterized by one or more of the following: an increase in tumour incidence (either specific to the carcinogen used or spontaneous, or neither spontaneous nor specific); an increase in tumour multiplicity; or transmission of the initiated state through germ cells. In several other well-conducted experiments, either no multigeneration effect was observed or the results were ambiguous. The clearest, most consistent results are obtained in experiments on mice. The published results, however, cannot preclude the possibility that the observed increases are due to a genetic predisposition rather than to induction of an oncogenic mutation in the germ cells. The studies reviewed here indicate that mothers and fathers may differ in transmission efficiency, but it would be premature to draw definitive conclusions. More work is needed in this field, but it appears that mice may be a more sensitive animal model for the study of multigeneration carcinogenesis experiments than rats. In future studies of this effect, experimental designs which allow for assessment of the relative efficiency of transmission through the mother and the father should be implemented. Care should also be taken to plan experiments that are large enough, and to retain all pertinent information, in particular litter information, in order that clear conclusions can be drawn from the results.