Water arsenic levels and breast cancer risk in a United States-wide prospective cohort.

In the U.S., drinking water is an important source of exposure to inorganic arsenic, an established human carcinogen. Chronic exposure to low levels of inorganic arsenic (≤10 μg/L) has been linked to bladder and kidney cancers; however, the association with other cancers remains inconsistent. We investigated the association between arsenic from private wells and community water systems and breast cancer incidence in a large U.S.-wide prospective cohort. Among 33,864 Sister Study participants residing in states with available arsenic data and reporting eligible water source types at enrollment (2003-2009), geocoded residential addresses were linked to county-level water arsenic concentrations. For community water system (CWS) users (n = 27,290), addresses were linked to county-level water arsenic concentrations using the EPA National Contaminant Occurrence database (2006-2008). For well water users (n = 6574), addresses were linked to county-level modeled probabilities that arsenic levels in private well water exceeded 10 μg/L. Cox proportional hazards regression was used to estimate adjusted hazard ratios (HR) and 95% confidence intervals (CI) for the association between water arsenic and incident breast cancer, with subgroup analyses by tumor subtypes and menopausal status. Over 13.2 years (mean) of follow-up, 3309 breast cancer cases (invasive and ductal carcinoma in situ) were diagnosed. We observed non-monotonic but elevated associations between increasing levels of water arsenic and invasive breast cancer risk (CWS HR Q5vsQ1 = 1.11, 95%CI 0.98 - 1.27, P-trend = 0.16; Well HR Q5vsQ1 = 1.11, 95%CI 0.84 - 1.47, P-trend = 0.88). When stratified by hormone receptor status and menopausal status, the associations did not meaningfully differ between subgroups. In a nationwide prospective cohort, we observed a suggestive positive association between relatively higher water arsenic levels and breast cancer incidence; however, the non-monotonic trends and exposure assessment limitations warrant further investigation. These results suggest that levels of arsenic in water in the U.S. may contribute to breast cancer risk.