Atrazine use and cancer risk within the Agricultural Health Study (AHS) cohort: An update

Abstract

In the United States, atrazine is a common agricultural herbicide. Few epidemiologic studies have evaluated cancer risks associated with this ubiquitous chemical. A previous analysis within the Agricultural Health Study (AHS) in 2011 found limited evidence of cancer risk based on 3,146 exposed cases. Here, we updated that analysis with additional follow-up time and exposure information. Information about lifetime pesticide use was reported on questionnaires at enrollment (1993-1997) and follow-up (1999-2005). Among 53,508 pesticide applicators in Iowa (IA) and North Carolina (NC), there were 6,401 incident exposed cancer cases identified through linkage to state cancer registries through 2017 (IA) and 2014 (NC). We evaluated cumulative intensity-weighted lifetime days of atrazine use. Atrazine exposures were categorized into quartiles among users and compared to a non-exposed referent group. Poisson regression was applied to estimate rate ratios (RR) and 95% confidence intervals (CI) adjusting for age, smoking status and pack-years, alcohol use, family cancer history, and correlated pesticides. We also stratified by age at cancer diagnosis. Approximately 71% of applicators reported ever using atrazine. Greater use of atrazine was associated with increased risk of cancers of the lung (RRQ4=1.21, CI=0.95-1.55, p-trend=0.19, n=551 cases), and prostate (RRQ4=1.12, CI=0.98-1.28, p-trend=0.10, n=2,521); the association was stronger for aggressive prostate cancer (RRQ4=1.37, CI=1.02-1.84, p-trend=0.07, n=553). The association with prostate cancer was most pronounced among those diagnosed <50 years of age (RRQ4=3.53, CI=1.34-9.29, p-interaction=0.012). Associations with lung cancer did not vary by age. Our findings with an average of 8.5 years of additional follow-up and twice as many cases compared to a previous analysis, suggest increased risk of prostate and lung cancers is associated with atrazine use. The associations with prostate cancer were most notable among those diagnosed at younger ages and those with aggressive disease. Further work is needed to investigate the reasons for these observations.