In a search for novel circulating biomarkers for pancreatic cancer, we examined the association between serum soluble Fas (sFas) levels and superoxide dismutase (SOD) activity and the risk of death from pancreatic cancer in a nested case-control study within the Japanese Collaborative Cohort Study. Case subjects were 68 persons who were free of morbidity, had provided a blood sample at baseline (1988-1990), and subsequently died from pancreatic cancer before December 31, 1997. Control subjects were 199 matched persons who were selected from the remaining participants in the cohort. Conditional logistic regression models were used to estimate age-adjusted and multivariate-adjusted odds ratios (ORs) and their 95% confidence intervals (CIs). No statistically significant differences were noted in mean sFas levels (p=0.11) and SOD activity (p=0.42) between cases and controls. Overall, neither serum sFas levels nor SOD activity were associated with the risk of pancreatic cancer deaths, after adjustment for area, BMI, cigarette smoking, and history of diabetes. Furthermore, no significant risk trends were noted. Our results do not support the hypothesis that serum sFas levels and SOD activity are associated with pancreatic cancer risk.