Long-term exposure to trihalomethanes in drinking water and breast cancer in the Spanish multicase-control study on cancer (MCC-SPAIN)

Highlights

• Exposure to trihalomethanes (THMs) in drinking water increases bladder cancer risk.

• We assessed life-time exposure to THMs and breast cancer (BC) risk in Spain.

• A case-control study with extensive exposure assessment was conducted in 2461 women.

• Residential chloroform appeared to be associated with BC.

• This epidemiological study overcomes several limitations of previous studies.

Abstract

Background

Exposure to trihalomethanes (THMs) in drinking water has consistently been associated with an increased risk of bladder cancer, but evidence on other cancers including the breast is very limited.

Objectives

We assessed long-term exposure to THMs to evaluate the association with female breast cancer (BC) risk.

Methods

A multi case-control study was conducted in Spain from 2008 to 2013. We included 1003 incident BC cases (women 20–85 years old) recruited from 14 hospitals and 1458 population controls. Subjects were interviewed to ascertain residential histories and major recognized risk factors for BC. Mean residential levels of chloroform, brominated THMs (Br-THMs) and the sum of both as total THM (TTHMs) during the adult-lifetime were calculated.

Results

Mean adult-lifetime residential levels ranged from 0.8 to 145.7 μg/L for TTHM (median = 30.8), from 0.2 to 62.4 μg/L for chloroform (median = 19.7) and from 0.3 to 126.0 μg/L for Br-THMs (median = 9.7). Adult-lifetime residential chloroform was associated with BC (adjusted OR = 1.47; 95%CI = 1.05, 2.06 for the highest (> 24 μg/L) vs. lowest (< 8 μg/L) quartile; p-trend = 0.024). No association was detected for residential Br-THMs (OR = 0.91; 95%CI = 0.68, 1.23 for > 31 μg/L vs. < 6 μg/L) or TTHMs (OR = 1.14; 95%CI = 0.83, 1.57 for > 48 μg/L vs. < 22 μg/L).

Conclusions

At common levels in Europe, long-term residential total THMs were not related to female breast cancer. A moderate association with chloroform was suggested at the highest exposure category. This large epidemiological study with extensive exposure assessment overcomes several limitations of previous studies but further studies are needed to confirm these results.

Introduction

Breast cancer (BC) is the first cancer in incidence and mortality among women world-wide (GLOBOCAN, 2012), with an increasing incidence during the last decades, also in Spain (Pollán et al., 2009). BC is more common in western countries and among favoured socioeconomic status (Brody and Rudel, 2003). Main recognized risk factors affect endogenous estrogenic levels (Hankinson et al., 2004) and include sex, age, body mass index, early age at menarche, advanced age at first delivery and at menopause, life-style factors such as alcohol consumption and low physical activity, and drugs with estrogenic action before or after menopause (Hankinson et al., 2004). Established risk factors explain approximately 50% of the variability in BC incidence, and other environmental factors may partly explain the remaining variation (Brody et al., 2007). Toxicological studies, and to a lesser extend epidemiological studies, have related some environmental exposures to BC (Macon and Fenton, 2013), mainly through endocrine disruption (Brody and Rudel, 2003). Drinking water disinfection by-products (DBP) are among the chemicals suggested by toxicologic research as potentially related to BC that have not been investigated enough in epidemiologic studies (Brody et al., 2007).

DBPs are a mixture of hundreds of chemicals formed in water during the disinfection process. This is a ubiquitous exposure through ingestion of tap water, inhalation and dermal exposure during showering, bathing or washing dishes (Villanueva et al., 2015). The most prevalent DBP in drinking water are trihalomethanes (THM), which are the only DBP group regulated in the EU with a maximum contaminant level of 100 μg/L. Several DBPs have been shown to be genotoxic in in vitro assays and carcinogenic in animal experiments (Richardson et al., 2007) and the WHO International Agency for Research on Cancer (IARC) classifies chloroform and other widespread DBP as possible humans carcinogens (Villanueva et al., 2015). Several epidemiological studies have related exposure to DBPs and cancer risk, being the most consistent evidence for bladder cancer and in a lower extent for colon and rectal cancer (Villanueva et al., 2015). Only sporadic epidemiological studies have assessed the impact of DBPs on other cancer sites including the breast (Villanueva et al., 2015).

Among the few epidemiological studies on DBP exposure and BC, some detected a positive association (Doyle et al., 1997, Gottlieb et al., 1982, Koivusalo et al., 1997, Wilkins and Comstock, 1981), while others did not (Kanarek and Young, 1982, Marcus et al., 1998, Vinceti et al., 2004, Young et al., 1981, Zierler et al., 1986). These are studies conducted 20 years ago (Doyle et al., 1997, Koivusalo et al., 1997, Marcus et al., 1998) or 30 (Gottlieb et al., 1982, Kanarek and Young, 1982, Wilkins and Comstock, 1981, Young et al., 1981, Zierler et al., 1986) with only one exception (Vinceti et al., 2004) and had important methodological limitations, including an ecological design (Marcus et al., 1998, Vinceti et al., 2004, Wilkins and Comstock, 1981), a poor control for confounding and a very limited exposure assessment based on surrogates of DBP exposure (Doyle et al., 1997, Gottlieb et al., 1982, Kanarek and Young, 1982, Koivusalo et al., 1997, Young et al., 1981). Furthermore, evidence that 3-Chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), a major mutagenic constituent of DBP, causes mammary tumours in rats (Komulainen et al., 1997) also suggests that the association between DBP exposure and BC should be further investigated in epidemiological studies overcoming the limitations of previous studies (Brody et al., 2007).

We aim to provide new epidemiological evidence on the association between lifetime exposure to DBPs and female BC risk in a large case-control study, including areas with contrasting THM concentrations in Spain and evaluating different routes of exposure and THM species.