Using a monoclonal antibody to bromodeoxyuridine, we studied the cell kinetics of human hepatocellular carcinoma, liver cirrhosis, chronic active hepatitis and alcoholic liver fibrosis. Specimens were taken either by biopsy or surgery and immediately incubated with 0.1% bromodeoxyuridine solution at 37°C for 45 min. After in vitro labeling, the bromodeoxyuridine taken up by the nuclei of S-phase cells was determined by the avidinbiotin-peroxidase complex method, using an anti-bromodeoxyuridine monoclonal antibody as the first antibody. The number of positive nuclei in 1,000 hepatic cells was counted, and the bromodeoxyuridine labeling index was expressed per thousand.
The mean bromodeoxyuridine labeling index ± S.D. of the cancerous portion of hepatocellular carcinoma, the noncancerous portion of hepatocellular carcinoma, liver cirrhosis, chronic active hepatitis and alcoholic liver fibrosis were 64.1 ± 31.3, 33.6 ± 14.4, 23.2 ± 20.8, 9.1 ± 6.1 and 21.6 ± 13.0, respectively. The mean bromodeoxyuridine labeling index of the hepatocellular carcinoma cancerous portion was statistically higher than that of any other group. There was no statistical difference by the t test or the Wilcoxon test between the noncancerous portion of hepatocellular carcinoma and liver cirrhosis, and these two groups were proved interdependent by χ2 test (Fisher's exact test), whether they were subdivided by bromodeoxyuridine labeling index ±10 or not.
Bromodeoxyuridine labeling index was not significantly correlated with the usual biochemical parameters such as serum AST, ALT, γ-GTP, alkaline phosphatase, lactate dehydrogenase, cholinesterase, albumin, and α-fetoprotein.
These results suggest that the bromodeoxyuridine labeling index of biopsied or resected specimens could give valuable information on cell kinetics when analyzing liver diseases and could be especially useful in distinguishing hepatocarcinogenesis from hepatic cirrhosis.