Environment/Climate Change
Drinking water contaminants and risk of lymphohematopoietic malignancies in the Iowa Women’s Health Study Rena R. Jones* Luz Villanueva Villanueva Villanueva Villanueva Villanueva Villanueva Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute
Background: Some drinking water disinfection byproducts and nitrate/nitrite that endogenously form carcinogenic N-nitroso compounds are suspected lymphomagens and leukemogens, but studies are limited.
Methods: Among 15,577 postmenopausal women in the Iowa Women’s Health Study, followed 1986-2016, using public water supplies (PWS) >10 years, we computed pre-enrollment, residential duration-based average concentrations of total trihalomethanes (TTHM), the sum of five haloacetic acids (HAA5), and nitrate-nitrogen (NO3-N) from PWS monitoring data. We estimated hazard ratios (HR) and 95% confidence intervals (CI) for continuous exposures (per log2) and quartiles (Q1 referent) with leukemia (N=127 cases) and non-Hodgkin lymphoma (NHL; N=344) and its major subtypes: chronic lymphocytic leukemia (CLL); multiple myeloma; follicular lymphoma; diffuse large b-cell lymphoma. Models adjusted for age, body mass index, and smoking; we also evaluated mutual adjustment of contaminants.
Results: We observed no clear associations for leukemia and NHL overall with TTHM (HRper-log2[CI]=1.01[0.96-1.06]; 1.01[0.96-1.05], respectively) and HAA5 (1.01[0.99-1.04]; 1.02[0.99-1.06]). Although we found no exposure-response in categorical analyses for follicular lymphoma (N=48) with TTHM (p-trend=0.53) or HAA5 (p-trend=0.49), risk was elevated for continuous exposures HRper-log2=1.10[1.00-1.20]; 1.12[0.99-1.26]). We observed non-significant positive associations with continuous HAA5 for other NHL subtypes. CLL risk (N=78) was elevated non-monotonically across HAA5 categories (p-trend=0.34), with the strongest association in Q3 (HRQ3vsQ1=2.18[1.11-4.29]), and weaker and imprecise associations in Q2 (HRQ2vsQ1=1.56[0.82,2.96]) and Q4 (HRQ4vsQ1=1.11[0.53,2.30]). Mutual adjustment did not change these associations. We found no associations with nitrate.
Conclusion: We report the first, albeit modest, evidence of relationships between HAA5 and risk of some NHL subtypes. Larger study replication is needed.