ABSTRACT
All residents aged 40 years or more in Oyabe City, Toyama Prefecture, Japan were involved in an annual medical check-up between 1987 and 1988. The cohort was followed and death certificates from cancers were confirmed prospectively. During follow-up to December 31st, 1994, 100 deaths (28 gastric, 17 lung and 55 other cancers) from cancers occurred, and these subjects were included in this study as the case group. Subjects in the control group, matched for gender and age with the cases, were selected randomly from participants whose serum samples had been stocked during annual medical check-up. The concentration of serum thiocyanate in all (79.8 μmol/l), gastric (86.7 μmol/l) and lung (90.0 μmol/l) cancer patients were significantly higher than that of relevant controls (64.3 μmol/l, 59.0 μmol/l and 61.0 μmol/l, respectively; and p<0.001, p<0.001 and p<0.05, respectively). After adjusting for BMI, blood pressure and total serum cholesterol, the results of multiple logistic regression analysis showed that the risk of all cancers (OR=3.40, 95% confidence interval (95% CI): 1.67−6.96, p<0.01), gastric cancer (OR=7.98, 95% CI: 1.91−33.34, p<0.05) and lung cancer (OR=8.83, 95% CI: 1.19−65.65, p<0.05) were elevated significantly with logarithm transformed values of serum thiocyanate increased. The present findings suggested that in epidemiological studies confirmation of smoking status with biomarkers such as serum thiocyanate may be important, although considering the small sample size, a relatively weaker risk to interested factors rather than the strong relationship between smoking and cancer was noted.