Introduction:

Arsenic is a naturally occurring carcinogen with the potential to contaminate drinking water. Exposure to arsenic through drinking water has been implicated in many malignancies; notably, a strong association exists between arsenic and bladder cancer, with prostate and kidney cancer being implicated as well. Therefore, in 2000, the U.S. Environmental Protection Agency (EPA) proposed a new limit to arsenic levels in water, a proposal that ultimately was accepted in 2001 and limited arsenic from 50 parts per billion (ppb) to 10 ppb.

Although this legislation has reduced levels of arsenic in drinking water, there remains concern that the 10ppb cutoff is not stringent enough. Many studies have suggested deleterious effects of arsenic exposure even at low levels. We therefore sought to describe the impact of the EPA legislation, and whether low levels (<10ppb) of arsenic are associated with incident bladder, prostate and kidney cancers within the state of Washington.  

Methods:

We conducted a retrospective ecological study to explore the association between arsenic levels in drinking water and the incidence of bladder, prostate and kidney cancers in Washington State. We reviewed all arsenic tests from public water systems (PWS) in Washington between 2000 – 2018, as documented by the Washington State Department of Health. These PWS serve approximately 85% of the state’s population and includes water systems serving more than two residences.

To estimate average arsenic levels by county, we compared the geographic regions served by each PWS with county boundaries. We then obtained county-specific cancer incidence data for bladder, prostate and kidney cancers diagnosed between 2014 – 2018 from the Washington State Department of Health. These years were selected to account for an adequate latency period following arsenic exposure.

We used multivariable Poisson regression to estimate adjusted incidence rate ratios, evaluating potential associations between arsenic exposure and cancer risk.

Results:

Between 2000-2018, we analyzed 46,476 unique arsenic measurements from drinking water across Washington State. County-level annual average arsenic concentrations ranged from 2 ppb - 7.5 ppb, with no counties exceeding the EPA standard of 10 ppb (Figure 1A). The weighted mean arsenic level across all counties was 3.76 ppb (SD 1.06). From 2014-2018, there were 19,342 prostate, 8,073 bladder, and 6,727 kidney cancer cases diagnosed (Figure 1B).

We found a statistically significant association between arsenic exposure and the incidence of bladder and prostate cancers (Table 1). Compared to the lowest arsenic exposure quartile, those in the 3rd and 4th quartiles had a 1.15x (95% CI 1.08-1.24) and 1.21x (95% CI 1.09-1.35) higher adjusted risk of bladder cancer, respectively, and a 1.11x (95% CI 1.06-1.16) and 1.20x (95% CI 1.12-1.29) higher risk of prostate cancer. No significant association was observed between arsenic exposure and kidney cancer.

Conclusion:

To our knowledge, this is the first study to assess the impact of the EPA’s arsenic regulation (<10 ppb) on the incidence of urologic cancers, and the first in Washington to explore the relationship between low arsenic levels and urologic cancer rates. Although all Washington counties remained below the 10 ppb threshold during the study period, we still observed a dose-dependent association between arsenic exposure and the incidence of prostate and bladder cancers. These findings suggest that the current EPA standard may not be stringent enough; even lower arsenic concentrations in drinking water may be necessary to mitigate the risk of these cancers.

Funding: N/A