Background
Few studies have evaluated environmental chemical exposures in relation to ovarian cancer. We previously found an increased risk of ovarian cancer among postmenopausal women in Iowa associated with higher nitrate levels in public water supplies (PWS). However, elevated nitrate levels may reflect the presence of other agricultural chemicals, such as atrazine, one of the most commonly detected pesticides in Iowa PWS.

Methods
We evaluated the association between atrazine in drinking water and incident ovarian cancer (N=145, 1986–2010) among 13 041 postmenopausal women in the Iowa Women's Health Study who used their PWS for ≥11 years as reported in 1989. Average levels of atrazine (1986–1987), nitrate-nitrogen (NO₃-N, 1955–1988) and estimated levels of total trihalomethanes (TTHM, 1955–1988) from PWS monitoring data were linked to the participants' cities of residence. We computed HRs and 95% CIs by categories of the average atrazine level (not detected, ≤ or >0.37 parts per billion=median) using Cox proportional hazards regression adjusting for ovarian cancer risk factors.

Results
Atrazine was detected in water samples from 69 cities where 4155 women (32%) lived and levels were moderately correlated with NO₃-N (ρ=0.35) and TTHM (ρ=0.24). Atrazine levels were not associated with ovarian cancer risk with or without adjusting for NO₃-N and TTHM levels (p-trend=0.50 and 0.81, respectively). Further, there was no evidence for effect modification of the atrazine association by NO₃-N or TTHM levels.

Conclusions
In our study with low atrazine detection rates, we found no association between atrazine in PWS and postmenopausal ovarian cancer risk