Suggested procedures for reducing the pathology workload in a carcinogen bioassay program, Part I

Large test programs have been set up to help assess the carcinogenic potential of chemicals introduced into the environment. Most test protocols used in these programs require large numbers of animals and extensive pathologic evaluation of all test and control animals. In this paper we describe modifications to the present pathology protocol which will reduce the pathology workload of the program and make only small changes in the operational character of the screen. The first modification derives from the fact that a tumorigenic effect may be ruled out by looking at the treated animals only. The second procedure suggested uses simple random samples to provide information about the treated groups with still less pathology work. It is emphasized that these proposed modifications should be regarded as a starting point for further discussion and that we encourage comments from all interested scientists.