Identifying mechanisms of harm and susceptibility in individual tobacco users can greatly assist in reducing disease from tobacco. To do harm, the carcinogens in tobacco smoke must be absorbed and activated by the body. Biomarkers in blood for specific tobacco carcinogens have been identified that measure the effective dose; other biomarkers measure the degree to which those carcinogens have become capable of damaging DNA and so indicate the individual phenotypic susceptibility. This proposed study would estimate the association of biomarkers of tobacco carcinogen exposure and metabolic activation in smokers to their risk of developing lung cancer. The study would be nested in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO), a randomized trial of multi-site cancer screening in over 150,000 men and women nationwide, about 7,000 of whom are current smokers. The PLCO is an ideal epidemiologic setting for this activity, because it provides a well characterized smoking population with rigorously collected data, has an associated, extensive biorepository of blood samples collected at screening visits, and eliminates a major source of bias found in partially screened populations. From a source cohort of PLCO bio-repository participants who reported smoking on their baseline questionnaire, a sample of 300 incident lung cancer cases and 300 controls will be drawn and their demographic and baseline data obtained. After pooling their blood samples into triplets of similar risk (yielding 100 case and 100 control samples), each pooled sample will be analyzed to measure the specified metabolites. The biomarker levels in lung cancer subjects will be compared to those in non-lung cancer subjects and regression estimates will be made of the odds ratios of developing lung cancer associated with these biomarkers. In order to evaluate the variability of biomarker levels in individual smokers over time, a separate, longitudinal study will recruit 25 men and 25 women who are current smokers to give blood samples every two months for a year. These data will be used to strengthen the results of the case-control study by allowing a sensitivity analysis based on the within subject variance and to provide improved estimates. If successful, these data will improve 1) understanding of lung cancer etiology and 2) predictive models for lung cancer. We project that carcinogen metabolite phenotyping will enhance prevention, early detection, and risk reduction. Moreover, this study will validate the relationship to disease of biomarkers, thus supporting their use in other projects of this TTURC.
Showing the most recent 10 out of 132 publications
