Correlation between medium-term liver bioassay system data and results of long-term testing in rats.

The effects of hepatocarcinogens (ethionine, thioacetamide, phenobarbital), non-hepatocarcinogens [N-ethyl-N-(4-hydroxybutyl)nitrosamine (EHBN), N-methyl-N'-nitro-N-nitrosoguanidine (MNNG)] and a hepatoinhibitor [(butylated hydroxyanisole (BHA)] were compared in medium- and long-term in vivo systems. In experiment I, 2 weeks after a single injection of diethylnitrosamine (DEN) groups of male F344 rats received chemical administration for 6 weeks, combined with partial hepatectomy at week 3 and were killed at the end of week 8. In experiment II, animals were treated in the same manner and then given basal diet and tap water (group 1) or chemical continuously (group 2) until the 2 year timepoint. Numbers and areas of glutathione S-transferase placental form (GST-P)-positive foci developing in the liver under medium-term bioassay conditions (experiment I) were found to closely correlate with eventual hepatocellular carcinoma incidences after continuation of test chemical administration (experiment II). Thus all of the hepatocarcinogens enhanced both the induction of GST-P-positive focal lesions and liver tumors. While non-hepatocarcinogens exerted no such effects, their influence being limited to inducing lesions in their own respective target organs such as urinary bladder cancers in the EHBN case and glandular stomach adenocarcinomas with MNNG, BHA demonstrated inhibition potential in both experiments. The observed correlation between long- and medium-term results strongly indicates the applicability of our medium-term bioassay system for detection of liver carcinogens.