Studies were undertaken to compare outcomes when four chemicals were evaluated under typical NTP bioassay conditions as well as by protocols employing dietary restriction. Four Chemicals, using three different routes of exposure (in utero [accomplished by feeding the dam dosed feed], dosed feed, and gavage) were used to 1) evaluate the effect of diet restriction on the sensitivity of the bioassay toward chemically-induced chronic toxicity and carcinogenicity; and 2) evaluate the effect of weight-matched control groups on the sensitivity of the bioassays. Control and chemical exposed F344 rats and B6C3F1 mice (50-60/group) were fed NIH-07 diet either ad libitum or at restricted levels such that body weights were similar to 80 % of ad libitum control weights. The dietary restricted groups were either sacrificed at the end of two or 3-years. Results consistently show that feed restriction decreased the incidence of neoplastic and non-neoplastic lesions at a variety of anatomic sites in both control and chemical exposed animals. Furthermore, the sensitivity of the bioassay to detect chemical carcinogenic response were altered by dietary restriction: three of the four chemicals were found to increase the incidence of neoplastic lesions at four sites when evaluated under standard ad libitum conditions for 104 weeks. When unexposed and exposed groups were both subjected to dietary restriction, none of these 4 sites were detected as a target for carcinogenesis after two or three years. Rather, two different sites of carcinogenesis were detected. When the top dosed ad libitum fed animals were compared against their weight-matched control groups, a total of 10 sites were identified as targets for carcinogenesis. These included all four sites identified under the ad libitum protocol, both sites identified unter the feed restricted protocol, and an additional four sites that were not identified under the other two protocols. These studies show that dietary restriction of all animals can be expected of decrease the sensitivity of carcinogenesis bioassays. However, restricting only unexposed groups (weight matching) of control for non-specific weight loss in chemical exposed groups yielded the most sensitivity among our comparisons.
