Ethylene oxide emissions and risk of breast cancer and Non-Hodgkin lymphoma in a large U.S. cohort

Abstract

BACKGROUND AND AIM: Ethylene oxide (EtO) is a carcinogenic gas used to sterilize medical equipment; emissions increased during the COVID-19 pandemic. We investigated the relationship between environmental EtO exposure and risk of two {a priori} suspect cancers based on limited occupational studies. METHODS: Using the U.S. Environmental Protection Agency’s Toxic Release Inventory (TRI), we estimated historical exposures (1987-1995) at the address of NIH-AARP Diet and Health Study participants enrolled in 1995-1996. We computed distance-weighted average emissions indices (AEI=∑[lbs EtO/km²]) reflecting the sum of annual EtO emissions for all TRI facilities within 3,5, or 10km of the residence, and further weighted AEIs by the proportion of time the home was downwind of each facility. We evaluated incident Non-Hodgkin lymphoma (NHL) in the full cohort (N=451,945) and breast cancer (invasive and {in situ}) among postmenopausal women (N=173,670), overall and by subtype and disease characteristics. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated for AEI categories versus no exposure in Cox models adjusting for demographic and site-specific potential confounders. RESULTS:A total of 3,796 breast cancers and 1,671 NHL cases were diagnosed through 2011. We observed a non-significantly increased risk of breast cancer overall among those in the 95th percentile of the AEI within 3km (HR[P95.vs.0]=1.48,CI=0.93-2.35). Risk of breast cancer {in situ} increased with the AEI (p-trend0.03 at 3,5, and 10km), and the strength of the association declined with increasing distance from the home (HRs[Q4.vs.0]=1.67,1.55,1.29, respectively). No statistically significant associations were found by extent of disease or by hormone receptor status. At 5km, we found an elevated risk of NHL driven by a non-monotonic increase in follicular lymphoma (p-trend=0.11). CONCLUSIONS:Our novel study suggests a positive association between EtO emissions and risk of {in situ}, but not invasive, breast cancer; differences by stage may have mechanistic implications. Inconsistent patterns of risk for NHL subtypes warrant additional evaluation. KEYWORDS: cancer and cancer precursors, long-term exposure, chemical exposures, environmental epidemiology