Levels of DNA adducts vary greatly in vivo, attributable to individual differences in enzymatic bioactivation of benzo(a)pyrene. We developed an assay to measure the levels of DNA adducts induced in vitro by benzo(a)pyrene diol epoxide (BPDE), a bioactivated form of benzo(a)pyrene. In this large molecular epidemiological study of lung cancer, we tested the hypothesis that the level of in vitro BPDE-induced adducts is associated with risk of lung cancer. This hospital-based case-control study included 221 newly diagnosed lung cancer cases and 229 healthy controls frequency matched on age, sex, ethnicity, and smoking status. Short-term cultured peripheral blood lymphocytes from each subject were exposed in vitro to BPDE (4 microm) for 5 h, and the 32P-postlabeling method was then used to measure BPDE-induced DNA adducts in the host cells. Overall, the patients had significantly higher levels of BPDE-DNA adducts than did the controls (mean +/- SD per 107 nucleotides, 93.2+/-89.3 for cases versus 63.7+/-61.1 for controls; P = 0.001). Univariate and multivariate logistic regression analyses were performed to calculate the crude and adjusted odds ratios and their 95% confidence intervals. When the median adduct level of controls (46/10(7) nucleotides) was used as the cutoff point, 64% of cases had higher levels (odds ratio, 2.15; 95% confidence interval, 1.39-3.33, adjusted for age, sex, ethnicity, body mass index, recent weight loss, pack-years smoked, smoking in the last 24 h, and family history of cancer). Stratified analyses showed consistently higher levels of BPDE-induced adducts in cases than in controls, regardless of subgroup of age, sex, ethnicity, body mass index, recent weight loss, pack-years smoked, smoking in the last 24 h, and family history of cancer. A significant dose-response relationship between the quartile levels of BPDE-induced DNA adducts and the risk of lung cancer was observed (trend test, P < 0.001). The significant association between the level of in vitro BPDE-induced DNA adducts and risk for lung cancer suggests that subjects very sensitive to BPDE-induced DNA damage may have a suboptimal ability to remove the BPDE-DNA adducts and so are susceptible to tobacco carcinogen exposure and, therefore, may be at increased risk of lung cancer.