Abstract

The association between serum selenium concentration and risk of cancer was studied in a nested case control study. Case control pairs came from a population of 9364 people examined in 1979. During the six year follow up, 60 men and women aged between 20-54 at the time of blood sampling, who had been free of malignant disease, developed cancer. The mean serum selenium concentration of 1.56 mumol/l (123.2 micrograms/l) in patients was not significantly different from that in controls (1.63 mumol/l (128.7 micrograms/l]. The difference in mean selenium concentration was largest and most significant for haematological malignancies alone. The difference in selenium concentrations in cases of fatal cancer compared with controls was significant (p less than 0.01). The risk of developing adenocarcinomas does not seem to be influenced by serum selenium concentration.