Molecular Epidemiology and Lung Cancer: A Nested Case-Control Study

DESCRIPTION (provided by applicant): The broad objective of the proposed study is the early detection of lung cancer through the identification of patterns of protein expression with the potential to identify early disease. The research nests a case-control study within the prospective Harvard Physicians Health Study (PHS) to determine whether biomarkers in blood of healthy individuals can detect early disease that would manifest itself years or decades later. Specific Aims: Our aims are to test the following hypotheses: 1) The baseline blood samples from subjects who later develop lung cancer (cases) will differ significantly with respect to their patterns of protein expression from samples of subjects who are followed for the same period of time and have the same smoking status at enrollment but do not develop lung cancer (controls), such that an overall discriminating pattern can be identified that has adequate sensitivity and specificity (>80%) to serve as a potential biomarker of early disease. 2) The baseline blood samples from subjects who later develop Non-Small Cell Lung Cancer (NSCLC) will differ significantly with respect to their patterns of protein expression from samples of controls, such that a specific discriminating pattern can be identified that has higher sensitivity and specificity (>85%) to serve as a potential biomarker of early NSCLC. While a number of studies have demonstrated the potential of proteomics in early detection of lung cancer, the sample sizes have been small, none have involved well characterized prospective cohorts, and the specific markers, when identified, have been inconsistent from study to study. None has systematically assessed biomarker validity in terms of the sensitivity and specificity of the patterns under different circumstances of time from blood draw to diagnosis, histology, stage, or smoking status. This research is an essential step in their validation as tools in early detection of lung cancer. Research design and methods: Baseline blood samples from 350 cases and 700 matched controls will be selected from stored enrollment samples from the PHS and analyzed for patterns of protein expression. Detailed smoking, environmental, health and dietary histories (baseline and follow-up) have been obtained on each subject. Preliminary studies have demonstrated the feasibility of this collaboration between Columbia University, Harvard University, and the NCI/FDA and have generated data directly supporting all of the research aims. Lung cancer is the leading cause of death from cancer in the U.S. and worldwide and claims over 162,000 lives every year in the U.S. The proposal fits within the mission of the NIH, NCI, and NIEHS to reduce the burden of illness and within the NCI's and NIEHS' strategic areas highlighted for advancement: cancer prevention, early detection and prediction, molecular epidemiology, and validation of new biomarkers of preclinical risk.