A Follow-Up Study of Urinary Markers of Aflatoxin Exposure and Liver Cancer Risk in Shanghai, People's Republic of China

A cohort of 18,244 mostly middle-aged (45-64 years) men residing in four small geographically defined areas of Shanghai was accrued between January 1986 and September 1989. In addition to an in-person interview regarding dietary and other past exposures, each subject donated a single void urine sample at recruitment so that the presence of aflatoxins in urine could be assessed. In addition, a 1-year survey of market foods in Shanghai was conducted to quantitatively estimate the extent of aflatoxin exposure in the study population. After close to 70,000 person-years of follow-up, 55 incident cases of hepatocellular carcinoma (HCC) had been identified. Levels of urinary aflatoxin B₁ and the oxidative metabolites, including the major aflatoxin nucleic acid adduct, aflatoxin-N⁷-guanine, were determined for 50 of the 55 identified cases of HCC. Two hundred sixty-seven controls were chosen randomly from the cohort; they were matched to the 50 cases by age (within 1 year), time of specimen collection (within 1 month), and residence. After integrating the high-pressure liquid chromatography chromatograms to measure aflatoxin-N⁷-guanine, aflatoxin M₁ aflatoxin P₁, and aflatoxin B₁ 49, 67, 53, and 71 of the urine samples had detectable levels of these compounds, respectively. The aflatoxin metabolite detected at the highest concentration was aflatoxin P₁; the range was 0.59-16.0 ng/ml. The range of aflatoxin M₁ in the urine was 0.17-5.2 ng/ml. The aflatoxin-N⁷-guanine adduct range was 0.3-1.81 ng/ml in the 49 positive samples. A nested case-control analysis showed highly significant associations between the presence of urinary aflatoxins, serum hepatitis B surface antigen positivity, and HCC risk. Risk was especially elevated in individuals who were positive for both of these biomarkers (relative risk = 59.4; 95% confidence limit, 16.6, 212.0 after adjustment for cigarette smoking, a potential confounder). On the other hand, a cohort analysis using all 55 cases of HCC revealed no strong or statistically significant association between HCC risk and dietary aflatoxin consumption as determined from the in-person food frequency interview combined with the survey of market foods in the study region. Our results underline the importance of biomarker measurements in assessing the aflatoxin-HCC association in epidemiological studies.