*Corresponding Author:
Tongzhang Zheng
Department of Epidemiology, School of Public Health, Brown University, 121 South Main Street, Providence, RI 02903, USA, Tel: 401-863-6365;
E-mail: tongzhang_zheng@brown.edu
Key Words: Breast cancer, fetal origin hypothesis, in utero exposures, windows of susceptibility.
INTRODUCTION
Breast cancer is the world’s most prevalent cancer and the leading cause of cancer death among women worldwide [1-3]. While tremendous strides have been made, a substantial portion of breast cancer occurrence cannot be adequately explained by known risk factors [4-7]. Almost 30 years ago the eminent epidemiologist, Dr. Trichopoulos, hypothesized that breast cancer originates in-utero [8]. Interest in this hypothesis has increased markedly [9-23] including a very recent call for the specific examination of endocrine-disrupting chemicals that act during “windows of susceptibility” [9]. The fetal origin hypothesis of breast cancer proposes that in utero exposures [e.g., diethylstilbestrol use] alter fetal cell development and, thereby, trigger a chain of biologically linked events [impaired fetal growth, adverse birth outcomes, and postnatal accelerated growth] that ultimately leads to increased breast cancer risk among offspring [24-26]. Because of the lack of long-term longitudinal studies that start during pregnancy, most research testing this hypothesis has relied upon indirect measures of adverse pregnancy outcomes, such as maternal and infant anthropometry [e.g. pre-eclampsia, birth size] or perinatal factors [gestational age, twin birth] [26-35]. However, a recent report from a long-term cohort study [36] that directly measured levels of in utero exposure to Dichlorodiphenyltrichloroethane [DDT] strongly support the fetal origin hypothesis of breast cancer. Cohn et al. [36] based on the California Child Health and Development Study [CHDS] show that in-utero exposure to DDT is associated with breast cancer risk in female offspring, with a nearly 4-fold increased risk for those exposed to o,p’-DDT in the highest quartile. Notably, earlier studies of DDT/DDE exposure levels among adult women showed no such increased risk – evidence that in utero exposure may be a susceptibility window of breast cancer. Since adult exposures fail to explain a substantial part of breast cancer occurrence, ascertaining whether there is, in fact, a fetal ‘window of vulnerability’ is an urgent scientific goal, essential to identifying new opportunities to potentially prevent breast cancer.
1. Indirect evidence that in utero exposure increases breast cancer risk
Because of the long induction period from birth to breast cancer occurrence in adulthood, early studies were forced to largely rely on indirect evidence using various proxies of in utero exposures (such as birth weight and length, and maternal weight gain) [26-29]. This evidence has been extensively reviewed [20, 21, 26-29, 37, 38] and is briefly summarized below.
1.1 Pregnancy complications as indicators of prenatal exposures and breast cancer in offspring: The pregnancy complications most intensively studied for their association with breast cancer risk among daughters include maternal pre-eclampsia or eclampsia. Studies showed that daughters of women who develop toxemia during pregnancy (preeclampsia/eclampsia) have a significant 10 to 60 percent lower risk of breast cancer than daughters of normotensive mothers. Perhaps similarly, daughters of mothers with a lifetime history of diabetes have also been found to be at a decreased risk of breast cancer, especially for premenopausal breast cancer [28, 29, 39]. These studies have viewed toxemia as a surrogate for low estrogen exposure in utero, supporting the hypothesis of a direct relationship between in utero estrogen levels and breast cancer risk among offspring.
1.2 Perinatal factors as surrogates of prenatal exposures and breast cancer in offspring: Various perinatal factors have been associated with breast cancer risk suggesting that in utero exposures are important in breast carcinogenesis. A relatively consistent positive association of breast cancer risk associated with birth weight and birth length strongly suggests in utero exposure influences on subsequent breast cancer risk [26, 27, 40-43]. Among positive reports, a U-shaped or J-shaped association was reported between birth weight and breast cancer risk in some studies [44-51] while a positive linear relationship has been found in other studies [24, 52-56]. Preterm birth is reported to be associated with an increased risk of breast cancer in some [13, 49, 56-58] but not all studies [44, 51-53, 59, 60]. Preterm birth female infants were found to have 10 to 20 times higher serum follicle-stimulating hormone concentrations compared with full term female infants [61], the raised serum concentrations of gonadotropins in preterm female infants might lead to ovarian hyper stimulation that increases estradiol concentrations and increase breast cancer risk [57, 62]. Twinning [29] particularly dizygotic twins [7, 63-65] that is associated with higher levels of pregnancy estrogens [66, 67] has been associated with an increased risk of breast cancer among offspring. Higher maternal age at delivery has been inconsistently associated with an increased risk of breast cancer in daughters [27-29] and it is known that concentrations of estrogen in maternal blood during pregnancy are higher in older women [68, 69]. These perinatal factors have been associated with pregnancy estrogens or insulin and insulin-like growth factors [8, 16] which are important regulators of somatic growth during fetal life and childhood [70]. It should be noted that high levels of IGFs (like estrogens) may also result in an increased number of stem cells and/or increased mitosis in the developing mammary gland [24].
1.3 Postnatal accelerated growth and risk of breast cancer: Evidence strongly suggests that accelerated postnatal or early childhood growth predicts later breast cancer and modifies the relationship between birth size and breast cancer risk [24, 71-73]. A UK national cohort study showed that the effect of birth weight on breast cancer differ significantly based on height at age 7 years, suggesting a significant interaction between birth weight and childhood growth on breast cancer risk [24]. Another UK report showed that women who grow faster in childhood were at particularly increased risk of breast cancer [72]. Studies have shown that rapid childhood growth increases breast cancer risk and in utero effects may be modified by childhood growth velocity. A cohort study of 117,415 Danish women showed that high birth weight, earlier age of peak growth [between ages 8 – 14 years], high stature at 14 years, low BMI at 14 years were independent risk factors for breast cancer [73]. A systematic review [74] showed that rapid catch-up growth of LBW neonates following reduced intrauterine development is a more important factor than LBW alone for adult diseases.
1.4 Intrauterine diethylstilbestrol [DES] exposure and breast cancer in offspring: DES is a synthetic estrogen widely used by pregnant women in the 1950s and 1960s to reduce the risk of fetal loss [75]. Studies showed that women exposed to DES in utero are at an increased risk of developing breast cancer [76-79]. Daughters of mothers who took DES during pregnancy have two times higher breast cancer risk than women who were not exposed to it [78-81]. This risk may extend to their granddaughters as well [76]. The reason for the increased breast cancer risk of DES in utero exposure may not only be due to elevated pregnancy estrogenic environment, but may also be attributable to epigenetic alterations that target genes regulating stem cells and prevent differentiation of their daughter cells [5]. Strikingly, DES use during pregnancy has also been associated with a higher risk of SGA and preterm birth among offspring [82] comparable to the outcomes observed following in utero DDE exposure by Longnecker et al. [35] and several smaller studies [30-34].
2. Direct and quantitative evidence that in utero DDT exposure increases breast cancer risk
Cohn’s cohort study [36] is the only direct, quantitative study to date to prospectively link in utero DDT exposure to increased risk of breast cancer in offspring. In studying DDT exposure and breast cancer risk, Cohn and colleagues creatively used blood samples largely collected 1-3 days after delivery from the California Child Health and Development Study [CHDS] to investigate breast cancer risk in two generations: a] the female offspring; and b] the pregnant mothers themselves.
2.1 Study of offspring linking in utero DDT exposure to breast cancer risk: In a nested case control study, Cohn et al. [36] quantitatively assessed o,p’-DDT, p,p’- DDT, and p,p’-DDE levels in blood samples collected 1-3 days after delivery for 118 breast cancer cases and 354 controls among the female offspring. Out of the 3 measured DDT isomers, the study found that elevated maternal perinatal serum o,p’-DDT significantly predicted a nearly 4-fold increase in the daughter’s risk of breast cancer. Cohn’s nested case control study is the only direct, quantitative study to date to prospectively link in utero DDT exposure to increased risk of breast cancer in offspring and thus filled a major gap by directly measuring DDT exposures soon after birth and breast cancer risk in offspring.
2.2 Studies of mothers by age of first DDT exposure and age at diagnosis of breast cancer: Using the same blood samples collected 1-3 days after delivery, Cohn et al. conducted two prospective, nested case–control studies of DDT exposure and breast cancer among the mothers themselves. The blood samples were collected when female study participants were, on average, approximately 26 years old [83, 84]. Accordingly, due to the long half-life of DDT, presence of DDT during pregnancy presumably reflected exposure at earlier ages. The authors combined information on levels of DDT [during pregnancy] and secular trends in DDT exposure in the US to generate several new findings. The first nested case control study [83] involving 129 cases and 129 controls found that, out of the 3 measured DDT isomers [o,p’-DDT, p,p’- DDT, and p,p’-DDE], high levels of serum p,p´-DDT during pregnancy predicted a statistically significant 5-fold increased risk of breast cancer, for a specific subset of women. That is, women who were born after 1931, and who were under 14 years of age in 1945, when DDT came into widespread use. The authors concluded that exposure to p,p´-DDT early in life [before puberty], but not later, may increase breast cancer risk. The recently published second nested case control study [84] involving 153 new cases and 432 controls presented the joint effects of p, p’-DDT, age at first exposure, and age at diagnosis on breast cancer risk. Among women estimated to have first been exposed before age 3 years [based on age in 1945], p, p’-DDT was associated with increased risk of early breast cancer [< 50 years] but not later breast cancer [ages 50–54 years]. Among women estimated to have been first exposed from ages 3 to 13 years, p, p’-DDT was associated with increased risk of both early and late breast cancer. Among women first exposed after age 13 years, p, p’- DDT was associated only with increased risk of later breast cancer [ages 50–54 years]. Based on these results, the study concluded that risk of breast cancer associated with DDT exposure depended on timing of first exposure and diagnosis age, supporting the hypothesis that early life is a critical vulnerability window for mammary cancer [85-87].
In conclusion, these nested case control reports are significant because previous studies of body levels of DDT/DDE and other organochlorines in adult women and breast cancer in the US, including our own, were largely negative [88-94] Cohn’s findings [36, 83, 84] suggest that risk of breast cancer associated with DDT exposure depends upon timing of exposure – with exposure that starts in utero and possibly very early life (but not in adulthood) determining breast cancer effects.
3. Direct evidence showing that in utero exposure to DDT/DDE may trigger a cascade of biologically linked events that ultimately contributes to increased risk of breast cancer among offspring
Using a South African birth cohort, Murray et al. [95] recently reported that p,p’-DDT and p,p’-DDE serum concentrations were associated with hypertensive disorders of pregnancy while an earlier study by Savitz et al. [96] did not find an association of DDT exposure with preeclampsia . Several studies have also reported an increased risk of hypertension associated with higher levels of DDT exposure [97-102].
A large cohort study by Longnecker et al. [35] measured serum levels of p,p ́-DDT and p,p ́-DDE and found that in utero exposure to p,p ́-DDE significantly increased SGA risk. A significant dose response relationship was reported between maternal p,p ́-DDE levels during pregnancy and adverse birth outcomes. The adjusted SGA ORs for each increasing quintile of DDE were 1, 1.9, 1.7, 1.6, 2.6, respectively (trend p=0.04). Longnecker’s findings have since been replicated by several smaller studies [30, 34, 103-111] including our own [112] although some discrepancies remain [113-115].
Importantly, a recent study from Berkeley using the South Africa VHEMBE birth cohort showed that maternal serum p,p’-DDT concentration was consistently and positively associated with accelerated postnatal growth, including higher BMI-for-age in girls aged at 1 and 2 years of age [116]. A recent small cohort study from California involving 240 children showed that in utero exposure to DDT and DDE affected early childhood growth and significantly increased the risk of obesity at 12 years among boys, while non-significantly increasing the obesity risk among girls [117]. A study from China [118] showed that women exposed to the highest quartile of DDT/DDE had a significantly younger age at the time of menarche than those in the lowest quartile. A recent report from the Child Health and Development Studies by Cirillo et al. [119] showed that grand maternal perinatal serum DDT is associated with granddaughter’s early menarche and adult obesity, the risk factors for female breast cancer.
This collective evidence begins to illuminate a general model for the fetal origins of breast cancer associated with DDT exposure in utero. Namely, that in utero exposure to DDT impairs fetal development, results in subsequent early childhood accelerated growth and, eventually, increased risk for breast cancer among adult offspring.
4. Biological Plausibility
Experimental studies have shown that fetal life is an important window of susceptibility for mammary cancer [120-125]. In humans, however, the underlying mechanisms linking in utero exposures to subsequent breast cancer among offspring are not fully understood. In utero exposures may increase the number of breast stem cells during the fetal period and, therefore, the number at risk of malignant transformation. The likelihood of breast cancer occurrence depends on the number of mammary tissue specific stem cells and this is determined early in life, notably in utero or during immediate postnatal life [17-19]. In utero exposures can modify the epigenome and epigenetic modifications in the fetus might lead to changes in mammary gland development, such as increased vulnerability of epithelial targets [perhaps stem cells] for malignant transformation that alter the susceptibility to factors that can initiate breast cancer [19]. Breast tissue undergoes rapid proliferation and development in utero, which creates a fertile soil for cancer initiation representing a critical window of increased susceptibility for mammary carcinogenesis [8, 10, 17, 18, 126-128]. Thus, breast cancer may be diagnosed over the lifetime but begin in utero.
5. Major gaps in studying the fetal origin hypothesis
• Studies of fetal origin hypothesis to date have largely relied upon indirect evidence. Most previous studies of the fetal origin hypothesis of breast cancer have been based on indirect evidence of breast cancer risk associated with various proxies of in utero exposures (such as birth weight, birth length) as summarized previously. This is mainly due to the fact that the induction period is long from in utero exposure to breast cancer in adulthood taking about 50-60 years, but there is a general lack of long-term longitudinal prenatal cohort studies with direct prospective measures (bio specimens) of the prenatal environmental milieu.
• Studies that explicitly test the “fetal origin of breast cancer” hypothesis in the same cohort population have been sorely lacking. No prior study, including the prior DDT-breast cancer study [36], the DDT-adverse birth outcome study [30-35] and the DDT-early childhood growth and menarche studies [116-119] has directly examined, in the same cohort population, the extent to which in utero exposure to DDT is associated with intrauterine growth restriction (e.g., SGA), subsequent accelerated growth and –ultimately- increased risk of breast cancer.
• Cohn’s DDT-breast cancer study [36] is the only published longitudinal study so far with the ability to investigate whether in utero DDT exposure significantly increases the risk of breast cancer. It is important to both extend these findings and test DDT-related fetal origin hypotheses. Millions of current adult women were heavily exposed in utero and are potentially at increased risk of breast cancer [36].
• Need to test in utero exposure to multiple environmental pollutants and breast cancer risk. It is needed to determine the relationship between in utero exposure to multiple environmental pollutants and breast cancer risk in offspring, including persistent organic pollutants (POPs), metals and other environmental pollutants. These pollutants could jointly produce their impact or confound each other while their effects are evaluated in testing the fetal origins hypothesis of breast cancer.
CONCLUSION
As stated previously, although tremendous efforts have been made, a substantial portion of breast cancer occurrence cannot be adequately explained by known risk factors. If we are to make headway in preventing breast cancer, we must further enhance our understanding of the etiology of this disease, critically highlighting its modifiable risk factors. Only in this way can we positively impact the occurrence of the disease that is currently the leading cause of cancer death among women worldwide. Both the direct and indirect evidence strongly supports the hypothesis of fetal origin of breast cancer. It is critically important to advance our understanding of the hypothesis by ascertaining whether there is a fetal ‘window of vulnerability’ for breast cancer as proposed, which is essential to identifying new opportunities to prevent breast cancer considering the fact that adult exposures fail to explain a substantial part of breast cancer occurrence.
We are well aware that there are not many prenatal cohorts available for directly testing the fetal origin hypothesis by linking prenatal exposure to fetal development, pregnancy complication, adverse birth outcomes, postnatal and early childhood growth and adiposity, age at menarche, and ultimately breast cancer risk in offspring. Until the current prenatal cohort studies mature and are available for studying the hypothesis, efforts could still be made by testing the relationship between prenatal exposures and various indirect measures involving early life events (such as adverse birth outcomes, postnatal and early childhood growth and adiposity, and age at menarche) since these events have been linked to the risk of breast cancer, and are in fact likely to be the intermediate steps in the causal pathways—reflecting the chain of biologically linked events from the same in utero exposures that lead to increased susceptibility of the tissue to further insults – ultimately, leading to increased risk of breast cancer in the offspring. These early life events may mediate the association between in utero exposures and breast cancer risk. By advancing the fetal origin hypothesis, these efforts may lead to refine strategies and reach the ambitious goal of identifying new opportunities to prevent and control this deadly disease.
Acknowledgements
Funding for this project was provided in part by ES032876, ES029082, and ES031391.
REFERENCES
PEER REVIEW
Not commissioned. Externally peer reviewed.
The Roles of Choline in Maintaining Optimal Health
The Effects of Intake of Bread with Treated Corn Bran Inclusion on Postprandial Glyceamic Response
Food Waste throughout the Food Production Continuum – Water Food and Energy Nexus
Use of Lightly Potassium-Enriched Soy Sauce at Home Reduced Urinary Sodium-to-Potassium Ratio
A Different Type of Critical Migration
PRECEDE: A Conceptual Model to Assess Immigrant Health
Walking Together: Supporting Indigenous Student Success in University
Cosmetic Surgery and Body Image in Race/Ethnic Minorities
Our articles most useful
Beth Turner, Charlotte Garrett and Anja Wittkowski
Published : February 13, 2018
Women's Health Research
Yinjia Guo, Min Zhao, Victoria Rutten, Yifei Gao, Arier Lee, Qi Chen
Published : November 22, 2017
Women's Health Research
Jessi Hanson-DeFusco*
Published : December 07, 2023
Women's Health Research
Dana R. Riedy, Sarah M. Ghose, Sahar M. Sabet, Morgan P. Reid, Claire M. Williams, Ashley MacPherson and Natalie D. Dautovich
Published : October 11, 2023
Women's Health Research
Peggy B. Smith MA*, Reginald Hatter, Jasmine King, Anne Van Horn and Sophie McCollum
Published : November 16, 2023
Women's Health Research
Shanette M. Harris
Published : August 22, 2022
Women's Health Research