JAMA Netw Open. 2026 Jul 1;9(7):e2624391. doi: 10.1001/jamanetworkopen.2026.24391.
ABSTRACT
IMPORTANCE: Some drinking water contaminants, including disinfection byproducts, are known or suspected carcinogens with endocrine-disrupting properties. Few epidemiologic studies have evaluated associations with uterine cancer.
OBJECTIVE: To investigate prospective associations between multiple regulated contaminants in drinking water and the risk of uterine cancer.
DESIGN, SETTING, AND PARTICIPANTS: The California Teachers Study is a prospective cohort of female California educators. Participants who were cancer-free and had not undergone hysterectomy at enrollment (1995-1996) were followed up through December 31, 2020. Data were analyzed from November 2024 to October 2025.
EXPOSURES: The 15-year mean concentrations (1990-2005) of individual and total trihalomethanes (THMs) and haloacetic acids (HAAs), nitrate, arsenic, and uranium were calculated. The mixture effects of individual THMs, arsenic, uranium, and nitrate were also evaluated using quantile-based g-computation.
MAIN OUTCOMES AND MEASURES: Associations with incident uterine cancer overall, endometrioid tumors, and nonendometrioid tumors were analyzed categorically by tertile and continuously per log2 increase in water contaminant exposures, using Cox models adjusted for age, body mass index, and smoking status.
RESULTS: This study included 53 100 women (median [IQR] age, 50 [43-60] years) with enrollment addresses linked to a community water supply among women with residential duration at enrollment of 10 years or more. Higher levels of total THMs were associated with a higher risk of uterine cancer (hazard ratio [HR], 1.18 [95% CI, 1.02-1.38]; P for trend = .06) and endometrioid tumors specifically (HR, 1.23 [95% CI, 1.05-1.46]; P for trend = .01). Positive associations were found for the individual THM chloroform. Higher levels of the sum of 5 HAAs were associated with nonendometrioid tumors (HR, 1.86 [95% CI, 1.06-3.25]; P for trend = .01), with similar associations for the HAA monochloroacetic acid. No associations were observed with nitrate, arsenic, or uranium levels. Associations per IQR increase in the drinking water contaminant mixture composed of individual THMs, nitrate, arsenic, and uranium were 1.21 (95% CI, 0.96-1.53) and 1.34 (95% CI, 1.04-1.74) for uterine and endometrioid histotypes, respectively. The THMs chloroform and dibromochloromethane were the major contributors.
CONCLUSIONS AND RELEVANCE: In this cohort study of female California educators, associations between drinking water THMs and uterine cancer overall and with endometrioid tumors specifically were consistent with findings from the only prior study, conducted in the midwestern US. Further investigation of drinking water contaminants as modifiable risk factors for this common gynecologic cancer is warranted.
PMID:42485040 | DOI:10.1001/jamanetworkopen.2026.24391
