New Hampshire public health officials are conducting several highly publicized cancer related investigations. Community frustration centers around real difficulties analyzing small clusters of disease, and the problems of assessing whether those are related to unknown environmental pollutants. We propose use kidney cancer to demonstrate an interdisciplinary cancer etiology pipeline for environmental contaminant investigations, since an environmental etiology is well established. Exposure to the volatile solvent trichloroethylene (TCE) can cause kidney cancer via inhalation or consumption. There are numerous sites with environmental TCE contaminating the groundwater in New Hampshire and Vermont. The objective of this project is to perform an individual-level analysis of TCE and kidney cancer, illustrating the potential value of linking detailed residential histories with temporally and geospatially referenced environmental contaminant levels for the identification of cancer risk factors. We will process the residential addresses of cancer cases and controls obtained from commercial vendors into sequential residential histories. Cancer Registry residential address and epidemiologic study residence calendar information will validate the vendor data. We will take advantage of a large contaminant database already compiled through our prior work to map the geospatial distribution of TCE contaminated point-sources. We will create smoothed maps by interpolating dispersed contaminant levels. We will assess our estimated TCE levels by comparing with measured levels for a subset of locations. We will then model the association between the contaminant exposure level and cancer risk with adjustment for confounding factors. Finally, we will validate our geospatial TCE exposure assessment by using a unique TCE- linked hotspot mutation in the Von Hippel Lindau (VHL) gene as a biomarker of effect. Our project will demonstrate use of temporally- and geospatially- matched residence and contaminant data to assess environmental contaminant - cancer associations. This demonstration will create a potentially important workflow that would allow dozens of chemicals to be linked to cancer, or other diseases, at the population level, over time. This linkage is critical for identifying environmental risk factors for exposure mitigation.
Communities are frustrated because local variations in cancer rates appear related to the distribution of environmental pollutants, yet robust scientific evidence is lacking. We will model environmental pollutant distribution geospatially and temporally and obtain long-term residential history data of cancer patients. Beginning with trichloroethylene and kidney cancer, we will develop a workflow to systematically identify chemical pollutants linked to cancer risk.
