CORRELATIONS BETWEEN CHEMICALLY RELATED SITE-SPECIFIC CARCINOGENIC EFFECTS IN LONG-TERM STUDIES IN RATS AND MICE

We examined a database of 379 long-term carcinogenicity studies in rat s and mice to evaluate sex and species correlations in site-specific c arcinogenic responses. Within a species, most target sites showed a st rong correlation between males and females. For example, chemicals pro ducing forestomach or liver tumors in males were likely to produce the se same types of tumors in females. There was also a significant corre lation between species for certain site-specific carcinogenic effects, most notably tumors of the forestomach, liver, and thyroid gland. In contrast, adrenal pheochromocytoma, preputial/clitoral gland neoplasms , and lung tumors showed no significant interspecies correlation. Many chemicals produced a syndrome of carcinogenic effects involving tumor s of the skin, Zymbal gland, preputial/clitoral gland, mammary gland, and/or oral cavity. Regarding different target sites, there appeared t o be a correlation between thyroid and liver tumors both within and be tween species. Further, all chemicals producing mesotheliomas in male rats also produced mammary gland neoplasms in female rats. In contrast , kidney and urinary bladder tumors showed no significant association with any other tumor type in rats or mice. If a chemical produced a si te-specific carcinogenic effect in female rats or mice, there was appr oximately a 65% probability that the chemical would also bc carcinogen ic at that same site in males. The interspecies correlation was somewh at lower: approximately 36% of the site-specific carcinogenic effects ob-served in one species (rats or mice) were also observed in the othe r species. The high correlation between males and females suggests tha t in some instances it may be appropriate to consider a reduced protoc ol such as male rats and female mice. This would result in substantial cost savings, and an examination of NCI/NTP studies suggests that whi le there will bc some sensitivity loss associated with this approach, the rodent carcinogens not detected by this reduced protocol are not t hose most likely to pose a carcinogenic threat to humans. Nevertheless , the decision of whether to use a reduced protocol is best made on a case-by-case basis.