Timing of Environmental Exposures as a Critical Element in Breast Cancer Risk

Abstract

Objective:

The role of the chemical environment in disease initiation or progression is becoming more evident. Endocrine disruption via environmental chemicals is now well documented in humans, rodent research models, and wildlife. Breast cancer is an endocrine-based disease whose risk may be modified by environmental exposures. Our purpose is to encourage more investigation into early life environmental exposures as they relate to breast cancer risk factors and disease over a lifetime.

Evidence:

The 2009 President's Cancer Panel, 2012 Institute of Medicine, 2013 Interagency Breast Cancer and the Environment Research Coordinating Committee reports, and research publications dated ≥2012 in PubMed were used to inform our perspective.

Consensus Process:

Literature was reviewed and evidence gathered on the effects of the environment on risk of breast cancer or mammary tumor development in animal research models as it pertained to the influence of timing of exposure on later-life outcomes.

Conclusions:

Evidence has accumulated for several chemicals that environmental factors have a stronger effect on breast cancer risk when exposure occurred early in life. The insecticide, dichlorodiphenyltrichloroethane, is an excellent example and is just one of several chemicals for which there seems to be both animal and human evidence for the developmental basis of adult disease. The developing breast undergoes many changes in early life, leaving it vulnerable to the effects of epigenetic marks, endocrine disruption, and carcinogens. More research is needed in the area of early beginnings of breast cancer, with prevention of the disease as the ultimate goal.

The theory behind the developmental basis of adult disease suggests that exposure to environmental factors may alter fetal development during vulnerable periods and thereby increase the risk of adverse health outcomes in adult life. Although the hypothesis, initially proposed in the mid 1980s (1) and later coined “fetal origins hypothesis” (2), first reported a link between poor prenatal nutrition and adult coronary heart disease, numerous observations have later been published for a range of disease outcomes in human and animal offspring exposed to environmental factors during critical periods (2) of development. A recent overview of the literature on this subject stated that, “… the heart of the hypothesis—that environmental influences during gestation have an effect on later development—should be considered a major insight and constitutes a complement to a focus on genetic and more proximal factors (such as adult lifestyle) as causes of adult disease” (3). The developmental basis of adult disease has been proposed as an important mechanism for adverse health outcomes in humans—the exact means by which early life changes in the environment may affect adult health are not completely understood.

Much of the initial research in this field focused on fetal nutrition as a source of risk for several of the major diseases of adulthood, including coronary heart disease, hypertension, and type 2 diabetes (4). It is believed that these diseases originate in impaired intrauterine growth and development and that the fetus/infant “programming” has gone awry, such that “a stimulus or insult at a critical, sensitive period of early life has permanent effects on structure, physiology, and metabolism” (4). Similar theories have been circulated about other diseases, including polycystic ovarian syndrome (5), obesity (6), and breast cancer (7–9). A well-powered study of women in their 40s from the United Kingdom discovered that two specific types of polycystic ovarian syndrome are associated with unique in utero hormonal exposures—specifically, excess androgens and maternal obesity (5). Studies in rodents confirmed these associations. For breast cancer there are several strong associations of endogenous hormones/metabolic shifts leading to increased or decreased later life breast cancer risk. For example, pre-eclampsia may lead to decreased risk of breast cancer in both the mother who developed the condition and the child born under such conditions (10). Pubertal timing and obesity are other variables shown to modify breast cancer risk over a lifetime (11). Associations of early life hormonal disruption with later life cancers have created a major public health concern, with breast cancer as the most prevalent cancer among women in the United States and other countries (12).

Endocrine-disrupting compounds (EDCs), which are “substances in our environment, food, and consumer products that interfere with hormone biosynthesis, metabolism, or action resulting in a deviation from normal homeostatic control or reproduction” (13), are associated with alterations in several breast cancer risk factors (14). It is important to understand that EDCs may cause organizational perturbations that are irreversible; it is not just a matter of receptor activation or binding. Approximately 20 EDCs are now reported to alter adipocyte differentiation/function and result in increased adipogenesis or lipogenesis, and potentially obesity over a lifetime (6). Some EDCs are reported to modify the developmental programming of the rodent mammary gland (15–17) with long-lasting and potentially permanent effects. There is growing evidence that this may also happen in humans. Early breast development has been reported in girls from numerous countries, including the United States, without a concomitant shift in time to first menses (18–19). The shift toward early breast development may be the result of direct or indirect EDC exposures, especially given that EDCs are reported to act as obesogens in some research models, and body composition is one of the strong variables associated with accelerated breast development (18). Data from 2011–2012 report 20.5% of 12–19-year-old and 17.7% of 6–11-year-old U.S. children are obese (20).

Rodent and human mammary glands demonstrate many similarities across the developmental time line [see pictorial comparison, Figure 5.1 (11)]. Therefore, much information on the potential EDC effect in altered breast development and cancer risk over time in women has been gleaned from rodent toxicology/nutrition studies. An important finding in rodents, which essentially changed the way epidemiologists developed their hypotheses, was that critical periods of mammary gland development exist and EDCs may exert their effects at these times at lower doses (suggesting a shift in sensitivity of the tissue), with persistent effects (suggesting programming), or more robust outcomes (7–9, 15–17); all suggest enhanced susceptibility and increased disease risk.

EDCs and Breast Cancer Risk over a Lifetime

There are numerous examples of environmental factors, including EDCs, affecting breast development and cancer risk later in life when exposures happen during critical periods of development in rodent models. Table 1 highlights three chemicals for which there are both human and animal data on breast impacts.

Table 1.

Examples of EDCs Affecting the Breast

Endocrine-Disrupting Compound	Properties and Uses	Animal Study Findings (in vivo, in vitro)	Human Exposure and Health Effects
Dioxin	Pollutant formed during uncontrolled combustion	Increased mammary tumor incidence and shorter latency in female rats exposed to carcinogen during development	Slowed breast development in the highest exposed girls in two countries
	Bioaccumulative, lipophilic contaminant	Alters pubertal end points in rodents, including delayed mammary gland development in multiple rat strains	Suggestive data for breast cancer from industrial accident in Seveso, Italy; not conclusive
	Binds/activates the AhR		Increased breast cancer risk in Hamburg cohort
	Known carcinogen
DDT (mixture)	Insecticide that controls insect-borne disease	DDT and metabolites are known to exhibit anti-androgenic and estrogenic activity	Use peaked in the United States in 1959 and was banned by EPA in 1972
	Degrades to p,p'-DDE, the most prevalent and persistent metabolite in the environment	Limited evidence for the chemical acting as a promoter of mammary tumors in adult exposed rats	No associations in pooled and meta-analyses evaluating adult serum levels
		Accelerated mammary tumor development in HER2/neu mice exposed to p,p'-DDE in utero	Two studies showing early life exposure associated with strong increase in breast cancer risk in women; correlates with HER2+ status and advanced tumor stage
PFOA	Possesses long half-life in humans (2–4 y) and mice	Effects on mammary glands of mice include altered development, poor lactation, obesity in young adults (developmental exposure), and changes in gene expression	Delayed pubertal timing in girls
	Used in fire-fighting foams, electronics, and to make products that are grease- and waterproof	Delays mammary gland development and obesity at body burdens that overlap with human exposure burden in contaminated parts of the United States	Low-powered case-control study of Greenlandic Inuit women demonstrated significant correlation of serum-perfluorinated chemicals and breast cancer risk
	Final degradation product of other >8-carbon perfluorinated materials		May transfer to infants in breast milk

Abbreviations: AhR, aryl hydrocarbon receptor; PFOA, perfluorooctanoic acid.

This table was adapted from Table 6.2 in Interagency Breast Cancer and the Environment Research Coordinating Committee Report, 2013 (11).

Diethylstilbestrol

Diethylstilbestrol (DES), a synthetic estrogen used globally to prevent miscarriage in women, is the most well-studied example of an EDC affecting breast cancer risk in later life (21). In utero DES exposure was associated with an increase in neoplastic lesions of the reproductive tract and incidence of benign reproductive problems in both sons and daughters. Women who were exposed in utero to DES had a significantly increased incidence of breast cancer (22), as did mothers who took the medication (23). The highest incidence of later-onset breast cancer was directly correlated with the cumulative dose of DES during pregnancy (24).

In studies of mice, rats, or hamsters, early exposure to DES increased the number of mammary tumors, numbers of tumors per animal, and/or the severity of malignancy [reviewed in Reed et al (21)]. Importantly, these studies suggested that in addition to being a complete carcinogen, DES increased the sensitivity of the mammary gland to other chemicals. In a recent review of the potential mechanisms underlying life-long DES effects following fetal exposures (25), epigenetic alterations such as DNA methyltransferase expression changes, altered methylation of the promoter region of estrogen target genes that may control cell fate, histone modification, and silencing of specific miRNAs were identified as having potential roles in the latent effects of DES. Most of the supporting mechanistic data for DES is in rodent uterine tissue (26) or breast cancer cell lines (27), but discoveries following prenatal ethinyl estradiol exposure in rats [another synthetic estrogen; Hilakivi-Clarke (25)] may lead to new understanding of potential mechanisms of action in the mammary gland and whether such mechanisms are also pertinent for DES.

Dioxin

Dioxin, or 2,3,7,8-tetrachlorodibenzo-p-dioxin, is a product of uncontrolled combustion that acts as an endocrine disruptor and carcinogen by activating the aryl hydrocarbon receptor (http://www.epa.gov/pbt/pubs/dioxins.htm). This EDC induces adverse effects on development and reproduction, including the mammary glands. In fact, rat toxicology studies were informative in demonstrating that the fetal mammary gland is highly sensitive to dioxin, and effects from postnatal exposure are not as severe/persistent as those earlier, during bud development (16). These observations suggest that tissue sensitivity in utero could influence the long-term development of disease in the adult tissue.

The effect of dioxin exposure on mammary development has been compared in the female offspring of three different rat strains (16, 28, 29). In each study, severe and persistent mammary-gland developmental abnormalities were evident when exposed to a single dose of dioxin during mammary bud development. The effects included decreased ductal branching, delayed epithelial migration into the fat pad, and fewer differentiated terminal end buds. Two of those studies suggested that the effects were regulated by signals from the stromal compartment of the gland (16, 28). In addition, two epidemiologic studies that evaluated early life exposures to dioxins and related compounds reported delayed breast development in young girls with the highest blood dioxin (Seveso, Italy) (30) or prenatal/lactational dioxin levels (The Netherlands) (31). More recently, a study of nearly 400 women employed in a Hamburg, Germany pesticide plant in which they were exposed to dioxin, found a significant increase in breast-cancer mortality (32).

Some rodent studies using prenatal dioxin exposure have evaluated later life risk for mammary tumors. Prenatal dioxin exposure to rats followed with a carcinogen challenge in early adulthood doubled the incidence of mammary tumors and decreased tumor latency compared with controls (29). Another carcinogen-induced tumor study also reported a doubling of mammary tumor incidence following prenatal dioxin exposure in mice fed a high-fat diet (33), suggesting that high-fat diet increased sensitivity to the effects of dioxin. Taken together, these results suggest that dioxin causes permanent changes in the mammary glands during gestation, which results in a heightened risk for tumors later in life.

An epidemiologic study in Seveso, Italy (the site of a chemical plant explosion) found that those having higher serum dioxin concentrations had an associated 2-fold increase in risk of breast cancer (34). At the time of evaluation, women in the study with the highest measured blood dioxin levels had not yet reached the age of highest breast cancer risk. A follow-up study in 2008 reported that individual serum dioxin measurements were significantly positively correlated with overall cancer incidence in women, but the risk for breast cancer had not reached significance during the 30-year follow-up period (35). Because this cohort of women were young when exposed to dioxin (between age 0 and 40 y), some of them had not reached menopause, after which the greatest numbers of invasive breast cancers are typically diagnosed (36).

DDT and Related Chemicals

Dichlorodiphenyltrichloroethane (DDT) is an insecticide that was used to control insect-borne disease around the world. It reached peak use in the United States in 1959 and was banned by the U.S. Environmental Protection Agency (EPA) in 1972 (37). DDT is a mixture of a dozen or so congeners (DDTs); some are estrogenic (o,p'-DDT; ∼15–23% of the mixture), whereas the main congener, p,p'-DDT, is ∼77% of the mixture and degrades to form p,p'-dichlorodiphenyldichlororethylene (DDE), which has antiandrogenic activity (38). p,p'-DDE is the most prevalent and persistent metabolite. The DDTs vary in their ability to affect the breast. o,p'-DDT can support the growth of estrogen-dependent breast tumors in rats, whereas DDTs that do not bind to the estrogen receptor are without effect. Most the rodent literature on DDTs as promoters of mammary tumors showed limited evidence of effect (37), but most of these studies exposed adult rodents.

There have been multiple meta-analyses of the epidemiological data on DDTs and breast cancer risk. Individual nested case-control studies conducted between 1996 and 2012 failed to report a significant positive relationship between serum or adipose tissue levels of DDE or DDT in women with breast cancer. A pooled analysis of five case-control studies from the Northeastern United States (1400 cases; 1642 controls) demonstrated no association between breast cancer risk and p,p'-DDE levels (39). Substitive meta-analyses as recent as 2014 failed to find an association of concurrent DDT or DDE exposures with breast cancer risk (40–42), although cases did have a significantly higher DDE serum level in one analysis compared with controls (41). These studies all had one thing in common: they were measuring DDTs in women diagnosed with breast cancer at or near the time of diagnoses. However, studies in Colombia (43), Tunisia (44), and Mexico City (45) have demonstrated an elevated risk of breast cancer in women with higher serum DDE.

Cohn and colleagues (46) sparked an important shift in thinking when they reported a significant 5-fold increase in risk of breast cancer among women exposed to p,p'-DDT prior to age 14 years and especially in those less than 4 years of age. Their prospective nested case-control study suggested that early life exposures are most relevant for breast cancer etiology, as there was no association between breast cancer risk and p,p'-DDT in women exposed after age 20 years. This was the first suggestion that further research in breast cancer risk associated with DDT exposure was warranted, especially among sensitive subpopulations and considering exposures during biologically relevant time periods.

Since that time, additional research has been performed in cancer-prone mouse models to evaluate developmental exposure to p,p'-DDE on mammary tumor risk (47). The authors reported an accelerated time to spontaneous tumor onset in those animals, with blood and lipid levels of p,p'-DDE nearing human exposure. The authors suggest that additional studies should assess risk of early-onset breast cancer in women with early life exposures.

In this issue (48), Cohn and coworkers found a >4-fold significant association between serum o,p'-DDT levels in mothers during pregnancy and advanced-stage and HER2-positive breast cancer outcomes in their daughters. They found that the association of prenatal DDT exposure and HER2-positive tumors was especially strong. Cohorts such as the Child Health and Development Studies (CDHS), used by Cohn et al (48), take not only scientists with determination, funding, and local support, but also study participants that see the value in long-term enrollment in the study.

Conclusion

Future studies within the National Institute of Environmental Health Sciences/National Cancer Institute–sponsored Breast Cancer and the Environment Research Program (http://epi.grants.cancer.gov/BCERP/#funded) will take lessons from this cohort, as they follow participants enrolled as children through puberty and into their reproductive years, and address important chemical/nutrition specific questions in rodent models using susceptible life-stage approaches. As a followup to the DDT effects reported by Cohn and coworkers (48), a recent publication (49) assessing temporal changes in mortality rates of Taiwanese women reported peak breast cancer mortality from the 1951 birth cohort, coinciding with high use of DDT to prevent malaria. Although it is still not clear how DDTs program the breast for later life disease, the reports of a strong correlation between prenatal/perinatal o,p'-DDT and HER2-positive tumors (48), evidence that o,p'-DDT may affect HER2 signaling in breast cell culture systems (50), and accelerated tumor development following p,p'-DDE in a HER2-neu mouse model (47), all suggest that growth factor signaling may be a critical component of the later life effects.

These studies, taken together, demonstrate that environmental chemicals, as with poor nutrition, may alter fetal developmental patterns such that the tissue is at a higher risk of disease in later life—in this case, increased risk of breast cancer in adult woman. These studies demonstrate that certain chemical exposures put affected women at a higher risk over a lifetime than women without such exposures. Additional investigation into early life environmental exposures as they relate to breast cancer risk factors and disease risk over a lifetime is encouraged. This is not easy to do in humans, but multigenerational cohorts like that of the CDHS, and further investigations in rodent models of breast cancer, will make it possible. Timing of environmental exposures is critical for disease risk. Each person will have their own mixture of environmental exposures over their lifetime, but the developing fetus should be protected from chemical exposures thought to increase susceptibility to late-life disease.