Epidemiological data suggests a role of dietary cholesterol in the etiology of lung cancer without having a clear biological hypothesis. Although smoking as the outstanding risk factor for lung cancer may enhance lipid peroxidation reactions, this study was planned to assess smoking-independent associations between the extent of cholesterol oxidation and the risk for lung cancer. In the frame of a nested case-control study in European Prospective Investigation on Cancer-Heidelberg, six cholesterol oxidation products (COPs) were determined in plasma samples of 20 incident lung cancer patients obtained 1.9 +/- 0.6 years before diagnosis and in 40 matched (including smoking habits) controls. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by conditional logistic regression. Among all COP compounds tested, plasma 7beta-hydroxycholesterol was associated with lung cancer risk. The high crude risk estimate (OR approximately equals 5) became significant after adjustment for sports activity (OR = 6.83, CI = 1.08-43.01, 3rd versus 1st tertile). For the independent effect of 7beta-hydroxycholesterol, i.e., adjusted for other COP compounds, an OR of 8.08 (CI = 1.12-58.54, 3rd versus 1st tertile) was calculated (P = 0.04 for trend). Lung cancer risk adjusted for sports activity significantly increased by 26% (CI = 1.050-1.506, P = 0.01) per unit (1 nmol/mmol plasma cholesterol) of 7beta-hydroxycholesterol. No dietary factor had a significant effect in the regression model, but the dietary intake of meat, eggs, animal fat, cholesterol, and fruits (inversely) correlated with plasma COP concentrations. In this small study, plasma 7beta-hydroxycholesterol appeared to be a smoking-independent predictor of lung cancer risk and might therefore be used as a biomarker. Because of the rather high-risk estimate, research on possible intrinsic biological effects of this compound should be encouraged.