Risk for Cancer in Denmark

Arsenic is a ubiquitous element in the environment, where it occurs in both organic and inorganic forms; it can be found in food, water, soil, and airborne particles, and humans are widely exposed through these sources (Tchounwou et al. 2004). Arsenic can cause fatal acute poisoning, and long-term exposure has been associated with various cancers, diabetes, skin disease, chronic cough, and toxic effects in the liver, kidney, cardiovascular system, and the peripheral and central nervous systems (Vahter et al. 2006). Organic arsenic, which is less harmful than the inorganic forms, is most abundant in food, whereas inorganic arsenic compounds are found mainly in aquifers (Abernathy et al. 2003), where they accumulate by natural processes such as weathering, erosion, and biological activity, or eventually from anthropogenic contamination (Smedley and Kinniburgh 2005). Consequently, most health-related problems associated with arsenic are derived from groundwater used for drinking (Farago et al. 1997; Smedley and Kinniburgh 2005).
Epidemiologic studies in Asia (Chen et al. 1986, 1988; Tsuda et al. 1995; Wu et al. 1989) and Latin America (Ferreccio et al. 2000; Hopenhayn-Rich et al. 1996, 1998; Marshall et al. 2007) have shown that high arsenic concentrations (up to several hundred micrograms per liter) in drinking-water are associated with various internal cancers and with cancer of the skin. Some of these studies also provide evidence of a dose–response relation (Chen et al. 1986, 1988; Wu et al. 1989). However, few studies, most of which were conducted in the United States, have addressed the adverse effects of exposure to low doses of arsenic, and their results are inconsistent. Some showed a positive association between relatively low doses of arsenic and cancers of the skin, prostate, and bladder (Knobeloch et al. 2006; Kurttio et al. 1999; Lewis et al. 1999), whereas others showed no such effects (Bates et al. 1995; Karagas et al. 2001; Steinmaus et al. 2003). One study showed a nonsignificant decreasing risk for bladder cancer with increasing exposure to arsenic in the range of 3–60 μg/L (Lamm et al. 2004), and Karagas et al. (2002) found a U-shaped dose–response relation between exposure to arsenic and non-melanoma skin cancer, with a decreased risk at low levels and increased risk at higher levels. The existence of a threshold for the carcinogenic effect of arsenic has been debated, especially in the United States (Abernathy et al. 1996; Schoen et al. 2004), and some studies have suggested an interaction between exposure to arsenic and smoking in the causation of cancers of the lung, bladder and skin (Bates et al. 1995; Ferreccio et al. 2000; Knobeloch et al. 2006; Steinmaus et al. 2003; Tsuda et al. 1995).
Recent animal models for inorganic arsenic carcinogenesis suggest that the carcinogenicity of arsenic involves several mechanisms and co-exposure to other carcinogens (Burns et al. 2004; Cohen et al. 2007; Rossman et al. 2004; Waalkes et al. 2007; Wanibuchi et al. 2004). In vitro low concentrations of arsenic protected against oxidative stress and DNA damage (Snow et al. 2005), in accordance with the results of some of the epidemiologic studies (Karagas et al. 2002; Lamm et al. 2004). More studies are needed, however, to evaluate the possible carcinogenic effect of exposure to low concentrations of arsenic. The aim of this large, population-based cohort study was to determine if individual exposure to low levels of arsenic in drinking-water in Denmark is associated with a risk for cancer.