Objective: To evaluate the association of prediagnostic serum antioxidants
and lung cancer risk we conducted a case-control study nested in an occupat
ional cohort of tin miners.
Methods: Male workers free of cancer enrolled in the cohort. During up to 6
years of follow-up, 339 lung cancer cases were diagnosed and, among these
cases, those who donated blood prospectively (n = 108) were eligible for th
is study. For each case, two controls alive and free of cancer at the time
of case diagnosis were matched on age and date of blood collection.
Results: Overall, we observed no association between serum alpha-tocopherol
, gamma-tocopherol or selenium levels and lung cancer risk. However, a sign
ificant gradient of decreasing lung cancer risk with increasing serum alpha
-tocopherol was apparent for men less than 60 years old (odds ratio by tert
ile: 1.0, 0.9, 0.2; trend p = 0.002). Alpha-tocopherol was also protective
in men who reported no alcohol drinking (OR by tertile: 1.0, 0.6, 0.3; tren
d p = 0.008).
Conclusion: Although there were no significant overall associations between
prospectively collected serum alpha-tocopherol, gamma-tocopherol or seleni
um and incidence of lung cancer, results from this study suggest that highe
r alpha-tocopherol levels may be protective in men less than 60 years old a
nd in those who do not drink alcohol.