The Laboratory of Human Carcinogenesis and the Pharmocogenetics Section of the Genetic Epidemiology Branch will conduct a lung cancer case-control study in Baltimore, Maryland. The primary hypothesis of the study is to determine if mutagen sensitivity, p53 induction and apoptosis in cultured lymphocytes will be predictive of lung cancer risk. While there are some studies that examine mutagen sensitivity, none of these assays has been well-studied in an epidemiological setting. Because of methodological issues described herein, and the proposed development of new assays, this study will be viewed as a pilot and therefore hypotheses generating. The design of this molecular epidemiology study has been specifically developed to test the reliability and validity of the mutagen sensitivity assay, where a case-control study is needed to assess the possibility of case bias (i.e., results vary due to the concurrent presence of lung cancer rather than risk). Importantly, this protocol will establish a resource that will allow for the validation of these assays and also for the study of other biomarkers and gene-environment interactions, especially those related to DNA repair. Th secondary goals of this study are to 1) demonstrate gene-neuro-behavioral interactions for smoking addiction in controls and 2) assess the relationship of sex-steroid metabolism an and estrogen exposure to lung cancer risk. Cases will have histologically confirmed lung cancer and reside in Baltimore an and surrounding areas. They will be identified through six hospitals in Baltimore. Cases will be recently diagnosed and blood will be collected prior to chemotherapy or radiation therapy. Because of this requirement to obtain samples before treatment (or for surgical cases at least two months after surgery), we recognize that case ascertainment will be reduced, but critical data to assess differences between eligible and ineligible subjects will be collected through tumor registries. Two control groups will be used, the first will be hospital-based (frequency matched by age, gender, race, smoking and hospital) and the second will be population-based (frequency matched by age-, gender and race). The first control group will allow us to examine risk factors for lung cancer independent of smoking (odds ration for smoking = 1.0), and the second will allow the results to be extrapolated to the general population and also will be used to validate the phenotyping assays. The strategy for recruitment will allow us to over-sample for women and African Americans, so that after examination of data for the entire study group, we can assess differences by these subgroups. Cases and controls will receive a structured, in person interview assessing prior medical and cancer history, tobacco use, alcohol use, current medications, occupational history, family medical history, menstrual history and estrogen use, recent nutritional supplements and caffeine intake, and socioeconomic status. The questionnaire also will include the Fagerstrom index for nicotine dependence (FTND), Center for Epidemiologic Studies Depression (CES-D) scale, and a modified version of the Horn-Waingrow Reasons for Smoking (RFS) Scale. The phenotypic markers to be studied will assess DNA repair with cellular response by using lymphocyte cultures exposed in vitro to radiation, bleomycin, benzo(a)pyrene-diol-expoxide and N-methyl-nitrosurea and then measuring induction of chromosomal aberrations, p53 induction and apoptosis. DNA from cases and controls also will be used for genetic polymorphism analysis of carcinogen metabolism, and those relating to the dopaminergic system and nicotinic receptors. Tumors from cases will be evaluated for estrogen and progesterone receptors. The target accrual number of total subjects will be 1,200 where there will be 100 cases for each combination of gender and race (Caucasian- and African Americans), matched to 100 each of the hospital-based and population-based controls.