A molecular epidemiological case-control study of lung cancer.

Polycyclic aromatic hydrocarbon-DNA adducts were measured by ELISA in peripheral leukocytes from 119 non-small cell lung cancer patients and 98 controls at the Columbia-Presbyterian Medical Center. Thirty-one cases had adduct measurements in leukocytes, lung tumor, and nontumor specimens collected at surgery, and 34 had paired leukocyte and tumor specimens. Information on smoking, diet, and occupational exposure was collected. After adjustment for age, gender, ethnicity, season, and smoking, adducts in leukocytes were significantly higher in cases (P < 0.01) than controls; the odds ratio was 7.7 (95% confidence interval = 1.7-34; P < 0.01). Adducts in leukocytes were increased significantly in smokers and ex-smokers compared to nonsmokers among cases and controls (separately and combined) after adjusting for age, gender, ethnicity, and season (P < 0.05). The cases and controls differed in several respects: (a) adducts increased with the number of cigarettes smoked among the 51 cases who were current smokers (P = 0.05) but not among the current smokers in the controls; and (b) a seasonal variation in DNA binding, corresponding to that reported for aryl hydrocarbon hydroxylase inducibility, was observed in cases but not in controls. Among the cases, adducts in leukocytes were correlated more strongly with adducts in the lung tumor tissue than with those in nontumor lung tissue. The results in leukocytes are consistent with a constitutional susceptibility to lung cancer, which results in greater DNA damage from carcinogens in cigarette smoke. They suggested that it may ultimately be possible to use biomarkers such as adducts to identify individuals who would benefit most from early intervention.