In 1995, 234 adults from Qidong, People's Republic of China, were enrolled and followed in a Phase IIa 4-methyl-5-(N-2-pyrazinyl)-1,2-dithiole-3-thione (oltipraz) chemoprevention trial. Residents of this area are at high risk for development of hepatocellular carcinoma, in part due to consumption of aflatoxin-contaminated foods. The intervention was a randomized, placebo-controlled, double-blind study. Elements of the study design and clinical outcomes have been recently published (Jacobson et al, Cancer Epidemiol. Biomark. Prev., 6: 257-265, 1997). The primary objective was to conduct a preliminary assessment of the ability of oltipraz to modulate levels of a validated biomarker of aflatoxin exposure and of the risk of hepatocellular carcinoma by determining levels of aflatoxin-albumin adducts in sera. Healthy eligible individuals were randomized into three arms to receive p.o. 125 mg of oltipraz daily, 500 mg of oltipraz weekly, or placebo for 8 weeks. There were no consistent changes in biomarker levels in the placebo arm over the 16-week observation period, nor was any apparent effect observed in the arm receiving 125 mg of oltipraz each day. However, individuals receiving 500 mg of oltipraz once a week for 8 weeks showed a triphasic response to oltipraz. No effect was observed during the 1st month of the intervention, whereas a significant (P = 0.001) diminution in adduct levels was observed during the 2nd month of active intervention and during the lst month of follow-up. A partial rebound in adduct levels toward baseline values was observed during the 2nd month postintervention. Linear regression models up to week 13 confirmed a significant (P = 0.008) weekly decline of biomarker levels in the group receiving 500 mg of oltipraz once a week. However, despite these effects relative to baseline values within the 500-mg weekly arm, there were no statistically significant differences in biomarker trajectories between treatment arms. The genotype for glutathione S-transferase M1, an oltipraz-inducible isoform involved in the detoxification of aflatoxin B1, did not appear to affect either baseline levels or rates of decline in the biomarker. A follow-up Phase IIb trial with a longer intervention period will be necessary to determine the full extent to which aflatoxin biomarker burden can be reduced and whether diminution of biomarkers can be sustained over the long term.