The aim of the present study was to evaluate the relation between iron status and cancer in a population of middle-aged adults living in France where iron supplementation and iron-fortified foods are rarely used. The SU.VI.MAX study is a randomized, double-blind, placebo-controlled primary prevention trial evaluating the effect of antioxidant supplementation on chronic diseases in women aged 35-60 and men aged 45-60 y. At baseline, concentrations of hemoglobin, serum transferrin and serum ferritin were measured in 10,197 subjects. Data on dietary intake were estimated from six 24-h dietary records completed during the first 2 study years and available for 5287 subjects. All cancer cases that occurred during the 7.5-y follow-up were validated. In men, baseline serum transferrin and serum ferritin concentrations did not differ between subjects with cancers (n = 467) and those without. In women, serum ferritin was higher (P < 0.0001) and serum transferrin tended to be lower (P < 0.08) in cancer cases. Iron status was not related to cancer risk in men, but women with serum ferritin concentrations > 160 microg/L had an increased risk of cancer (odds ratio = 1.88, 95% CI: 1.05,3.35). No relation was found between dietary iron intake and risk of all cancer sites combined for either men or women. Our results suggest that iron status is not a predictor of cancer risk in men, whereas a serum ferritin concentration > 160 microg/L may be associated with an increase in cancer risk in women.