Reevaluation of the mutagenicity and carcinogenicity of chemicals previously identified as "false positives" in the Salmonella typhimurium mutagenicity assay.

An accurate determination of the correlation between the carcinogenicity and the mutagenicity of chemicals has been hampered by the lack of a well-documented list of noncarcinogens. To overcome this problem, Shelby and Stasiewicz (Environ Mutagen 6:871-878, 1984) published a list of 70 chemicals that showed no evidence of carcinogenicity in the National Cancer Institute (NCI) or National Toxicology Program (NTP) rodent carcinogenesis bioassays. More recently, Tennant et al. (Science 236:933-941, 1987) published a list of chemicals, including 29 noncarcinogens, that had been adequately tested for carcinogenicity by the NTP. Of the chemicals listed by Shelby and Stasiewicz or by Tennant and co-workers as noncarcinogenic, the NTP has evaluated 25 as mutagenic in Salmonella typhimurium; 48 of the noncarcinogens were evaluated as nonmutagenic. Thus, of the 73 noncarcinogens that have been evaluated as either positive or negative for mutagenicity, 34% (25/73) were "false positives" (mutagenic noncarcinogens) in the S. typhimurium assay. We re-evaluated the same mutagenicity and carcinogenicity data to determine whether the frequency of "false positives" is really as high as it appears to be. Our reevaluation of the mutagenicity data used more stringent criteria for calling a compound mutagenic than those used by the NTP, resulting in a substantial reduction in the frequency of "false positives" in the S. typhimurium mutagenicity assay. However, application of these same stringent criteria also substantially reduced the frequency of "true positives" (mutagenic carcinogens). Thus, it is concluded that modification of the evaluation criteria for the mutagenicity test can increase the specificity of the assay for the detection of carcinogens, but only at the cost of a corresponding reduction in sensitivity. We also performed a separate reevaluation of the NCI/NTP carcinogenicity data for the 25 S. typhimurium "false positives," assuming that the NTP evaluations of the mutagenicity data were correct. These reevaluations were based on the methodologies and findings of Griesemer and Cueto (In Montesano R, Bartsch H, Tomatis L (eds): Molecular and Cellular Aspects of Carcinogen Screening Tests.(ABSTRACT TRUNCATED AT 400 WORDS)