Air pollution and cancer: the GENAIRr study

Introduction: The overall aim of the study is to quantitatively assess the effects of air pollution and Environmental Tobacco Smoke (ETS) on cancers of the lung, bladder, pharynx, and larynx in non-smokers in nine European countries (UK, Sweden, Denmark, Germany, The Netherlands, France, Spain, Italy and Greece). This has been done with a nested case-control design in the EPIC investigation, by the means of biomarkers of exposure and susceptibility) (N=1100 cases and 3000 controls). Methods: Exposure assessment has been made by experts on the basis of the already available questionnaires plus objective information on air pollution in European cities. In addition, we measure cotinine in the blood for ETS exposure assessment, and we assess the effect of dose and length of exposure to environmental pollutants on biological endpoints, such as DNA and hemoglobin adducts. We also evaluate the relationships between cancer and mutations in relevant genes (p53, ras), as measured in the blood DNA according to a newly developed technique. Results: The procedure for assessing exposure to outdoor air pollution has been completed. We estimated indicators of long-term average exposure to a number of relevant air pollutants, including PM10, Black Smoke, NO2, NO and SO2. Exposure assessment methods made use of the existing spatial variation of air pollution concentrations between and within countries and/or cities. Specifically, air pollution exposure has been mainly characterized by background air pollution concentration and traffic intensity near the home address. The method assigns concentration data from the monitoring site(s) of the routine monitoring network in the community where the subject lives to the home address, taking into account site type (urban/rural). We will present: (a) validation of exposure assessment through different sources of information; and (b) analyses on the relationship between air pollution levels and the risk of cancer at different sites, after adjustment for confounders.