Airborne dioxin exposure and breast cancer risk in a case-control study nested within the French E3 N prospective cohort

Abstract

Introduction Dioxins are a mixture of related chemicals emitted by industrial chlorinated combustion processes, including chemical manufacturing of pesticides, and activities from metallurgy, steel and municipal solid waste incineration. TCDD (2,3,7,8-tetrachlorodibenzo-p-dioxin) is the most potent dioxin congener and has been classified as a group 1 carcinogen by the International Agency for Research on Cancer, with sufficient evidence for all cancers combined. As a persistent endocrine disrupting chemical, TCDD is suspected to be involved in breast cancer (BC) etiology and may influence estrogen- and progesterone-mediated pathways. The long-term nature of airborne dioxin exposure may imply variations in exposure intensities over time and given their tumor promoting properties, the impact of dioxin exposures close to the time of diagnosis needs to be considered. We aimed to estimate BC risk associated with airborne dioxin exposure in a case-control study nested within the E3 N cohort (“Etude Epidemiologique aupres de femmes de la Mutuelle Generale de l’Education Nationale”), improving the method for the assessment of low-dose airborne dioxin exposure and considering temporal dimensions of exposure in the risk estimates. Methods We designed a case-control study nested within the French E3 N prospective cohort and restricted to the Rhone-Alpes region, France. Between 1990 and 2008, 429 invasive BC cases were diagnosed and matched to 716 controls on relevant factors. Assessment of airborne dioxin exposure was based on a detailed inventory of dioxin emitting sources and residential history of the study subjects. Exposure was evaluated at the individual address level with a geographic information system (GIS)-based exposure metric that included proximity to dioxin emitting sources and their technical characteristics, exposure duration and wind direction. We first estimated odds ratios (OR) and 95% confidence intervals (CI) for BC in relation to cumulative airborne dioxin exposure using conditional logistic regression models adjusting for main BC risk factors. We then estimated time-dependent effects of annual airborne dioxin exposure on overall BC risk according to time prior to diagnosis with a flexible time-dependent weight function. Results We observed no linear trend across quintiles of airborne dioxin exposure (P = 0.81) and no increased risk of overall BC for higher dioxin exposure levels (OR for Q5 versus. Q1: 1.12, 95% CI: 0.69–1.82). We however observed a statistically significant OR for Q2 vs. Q1 (OR: 1.61, 95% CI: 1.04–2.49). For an increase of 0.1 μg-TEQ/m2 in annual airborne dioxin exposure, risk estimates for overall BC according to time prior to diagnosis did not vary from 1.00 at each time point. No association was observed between airborne dioxin exposure and BC risk defined according to status of estrogen-receptor and progesterone-receptor. Conclusions No association was observed between airborne dioxin exposure and overall BC risk in our study population, as well as for hormone-receptor defined BC tumors. Our results suggested an increased risk in overall BC for Q2 vs Q1; this might be explained by the non-monotonic effect of dioxins, which has been suggested for other endocrine disruptor chemicals, meaning that the effect would be different depending on the dose and that stronger effects would be observed for lower levels of exposure. Confirmation of our findings is required in larger populations. Our study may provide a new tool for the assessment of long-term exposure with the GIS-based metric, as well as new ways of considering temporal dimensions for environmental exposures in disease risk assessment.