Abstract
BACKGROUND: Drinking water can be an important source of exposure to nitrate and disinfection by-products, including trihalomethanes (THMs) and haloacetic acids (HAAs). N-nitroso compounds formed endogenously after nitrate ingestion are animal carcinogens, and THM and HAA exposures increase the risk of some cancers. Our objectives were to evaluate associations of drinking water nitrate and disinfection byproducts with total and aggressive (distant stage, poorly differentiated grade, fatal, or Gleason score ≥7) prostate cancer in the Agricultural Health Study cohort. METHODS: Male participants who were cancer free and used private wells or public water supplies (PWS) for drinking water at enrollment (1993-1997, n = 40 403) were followed through 2021 (mean = 21.9 years). Average nitrate-nitrogen (nitrate-N) concentrations were estimated for private well users based on state-specific geologic and meteorologic factors. We used monitoring data to compute average nitrate-N, THMs, and HAAs for PWS users. We estimated hazard ratios (HRs, 95% CIs) per doubling and categories of exposure for total (n = 3625) and aggressive (n = 2200) prostate cancer using Cox proportional hazards regression. RESULTS: Median (interquartile range) average water nitrate-N was 1.49 (0.76-3.01) mg L-1; 6% >10 mg L-1 (PWS maximum contaminant level). Compared to nitrate-N ≤ 1 mg L-1, exposures >10 mg L-1 were significantly positively associated with total (1.16, 1.01-1.35; P = .10 for trend) and aggressive disease (1.22, 1.02-1.47; P = .03 for trend). We observed weak associations between higher nitrate-N (Q4 vs Q1) and total (1.05, 0.95-1.16) and aggressive (1.13, 0.99-1.27) disease. We did not observe associations with total THMs or HAAs. CONCLUSIONS: These findings suggest that drinking water nitrate-N exposure, at average levels > 10 mg L-1, is a risk factor for prostate cancer, particularly aggressive disease.