Brain tumours--problems and perspectives.

The interpretation of carcinogenesis bioassay data is a complex process that requires a 'weight of evidence' approach rather than a rigid decision based solely on statistics. This is particularly true for rat brain tumours, since recent bioassays have demonstrated rather marked variations in the incidence of these neoplasms. The objective of this paper is to point out some of the considerations that should be included in such a 'weight of evidence' approach and to provide a few examples relevant to brain tumours. Areas for future research on mechanisms of neuro-oncology should provide additional information relevant to determining the chemical causes of brain tumours.