EDCs: The Endocrine Society’s Second Scientific Statement

Endocrine-Society banner
Five years after the Endocrine Society’s first Scientific Statement in 2009, a substantially larger body of literature has solidified our understanding of plausible mechanisms underlying EDC actions and how exposures in animals and humans-especially during development-may lay the foundations for disease later in life.


Executive Summary to EDC-2: The Endocrine Society’s Second Scientific Statement on Endocrine-Disrupting Chemicals, The Endocrine Society, dx.doi.org/10.1210/er.2015-1093, September 28, 2015.
Full study: EDC-2: The Endocrine Society’s Second Scientific Statement on Endocrine-Disrupting Chemicals, The Endocrine Society, dx.doi.org/10.1210/er.2015-1010, November 06, 2015.

The Endocrine Society’s first Scientific Statement in 2009 provided a wake-up call to the scientific community about how environmental endocrine-disrupting chemicals (EDCs) affect health and disease.

Five years later, a substantially larger body of literature has solidified our understanding of plausible mechanisms underlying EDC actions and how exposures in animals and humans—especially during development—may lay the foundations for disease later in life. At this point in history, we have much stronger knowledge about how EDCs alter gene-environment interactions via physiological, cellular, molecular, and epigenetic changes, thereby producing effects in exposed individuals as well as their descendants. Causal links between exposure and manifestation of disease are substantiated by experimental animal models and are consistent with correlative epidemiological data in humans. There are several caveats because differences in how experimental animal work is conducted can lead to difficulties in drawing broad conclusions, and we must continue to be cautious about inferring causality in humans.

In this second Scientific Statement, we reviewed the literature on a subset of topics for which the translational evidence is strongest:

  1. obesity and diabetes;
  2. female reproduction;
  3. male reproduction;
  4. hormone-sensitive cancers in females;
  5. prostate;
  6. thyroid;
  7. and neurodevelopment and neuroendocrine systems.

Our inclusion criteria for studies were those conducted predominantly in the past 5 years deemed to be of high quality based on appropriate negative and positive control groups or populations, adequate sample size and experimental design, and mammalian animal studies with exposure levels in a range that was relevant to humans. We also focused on studies using the developmental origins of health and disease model. No report was excluded based on a positive or negative effect of the EDC exposure. The bulk of the results across the board strengthen the evidence for endocrine health-related actions of EDCs. Based on this much more complete understanding of the endocrine principles by which EDCs act, including nonmonotonic dose-responses, low-dose effects, and developmental vulnerability, these findings can be much better translated to human health.

Armed with this information, researchers, physicians, and other healthcare providers can guide regulators and policymakers as they make responsible decisions.

Discussion (DES and Fertility-specific)

Diethylstilbestrol and beyond: perhaps the best-studied endocrine-based example is in utero exposure to diethylstilbestrol (DES), a potent synthetic nonsteroidal estrogen taken by pregnant women from the 1940s to 1975 to prevent miscarriage and other complications. DES was prescribed at doses from less than 100 mg (in most cases) upward to 47 000 mg, with a median dose of 3650 to 4000 mg in the United States (IARC 2012). Most women received low doses (ie, 5 mg) and increased their intake (up to 125 mg) as symptoms or pregnancy progressed, translating to doses of about 100 μg/kg to 2 mg/kg DES per day. In 1953, a study proved DES was ineffective. Its use was discontinued when a subset of exposed daughters presented with early-onset vaginal clear-cell adenocarcinoma, with a 40-fold increase in risk compared to unexposed individuals. A highly significant incidence ratio for clear-cell adenocarcinoma was also found in the Dutch DES cohort, a population that may have had lower exposures than US women. It was subsequently determined that exposed offspring of both sexes had increased risk for multiple reproductive disorders, certain cancers, cryptorchidism (boys), and other diseases, although the risk for sons is more controversial. New data are emerging to implicate increased disease risk in grandchildren. Not surprisingly, a plethora of examples is emerging for increased disease susceptibility later in life as a function of developmental exposures to EDCs that include BPA, phthalates, PCBs, pesticides, dioxins, and tributyltin (TBT), among others.

Epigenetics and transgenerational effects of EDCs: EDC-induced epigenetic changes are also influenced by dose of exposure, and they are tissue specific. Thus, it is important to consider both dose of EDC and the tissue before making firm conclusions about the epigenetic effects of EDCs. DNA methylation changes are the best-studied mechanism in this regard. For example, prenatal exposure to DES caused hypermethylation of the Hoxa10 gene in the uterus of mice and was linked to uterine hyperplasia and neoplasia later in life. Beyond the effects of prenatal exposure to DES on the daughters exposed in utero are suggestions that this leads to transgenerational effects of the chemical on the reproductive system, although whether this is linked to DNA methylation changes in humans is unknown.

Little is known about the ability of EDCs to cause histone modifications and whether this leads to transgenerational effects in animals or humans. The herbicides paraquat and dieldrin caused histone modifications in immortalized rat mesencephalic dopaminergic cells, and the insecticide propoxur causes histone modifications in gastric cells in vitro. DES caused histone deacetylation in the promoter region of the cytochrome P450 side chain cleavage (P450scc) gene. Further studies, however, need to be conducted to identify other EDCs causing histone modifications in animals and humans and to determine whether such modifications lead to transgenerational effects.

Female Reproductive Health: in the past 5 years, no new information became available on the effects of DES on the postnatal human ovary. Recent animal studies indicate that DES adversely affected the postnatal ovary. Neonatal exposure to DES inhibited germ cell nest breakdown (408) and caused the formation of polyovular follicles in mice, likely by interfering with the ERβ pathway and inhibiting programmed oocyte death and germ cell loss. It also reduced the primordial follicle pool and increased atresia in prepubertal lambs, and it caused polyovular ovaries in hamsters. Although these previous studies provide solid evidence that DES adversely affects ovarian structure in a variety of species, studies are needed to determine whether other synthetic estrogens adversely affect the ovary.

Effects of EDCs on uterine structure and function: synthetic estrogens are well known disruptors of uterine structure and function in humans and animals. Consistent with previous studies, recent data indicate that neonatal DES exposure caused endometrial hyperplasia/dysplasia in hamsters and increased uterine adenocarcinoma and uterine abnormalities in Donryu rats. Neonatal DES exposure also caused the differential expression of 900 genes in one or both layers of the uterus. Specifically, DES altered multiple factors in the PPARγ pathway that regulate adipogenesis and lipid metabolism, and it perturbed glucose homeostasis, suggesting that DES affects energy metabolism in the uterus. In the mouse uterus, DES altered the expression of chromatin-modifying proteins and Wnt signaling pathway members, caused epigenetic changes in the sine oculis homeobox 1 gene, and decreased the expression of angiogenic factors. DES also altered the expression of genes commonly involved in metabolism or endometrial cancer in mice, and it activated nongenomic signaling in uterine myometrial cells and increased the incidence of cystic glands in rats.

Effects of EDCs on the vagina: only a limited number of studies assessed the effects of EDCs on the vagina, and of these, all but one on phthalates focused on DES. A recent study of women showed an association between in utero exposure to DES and clear cell carcinoma of the vagina, confirming previous findings. Furthermore, DES disrupted the expression of transformation-related protein 63, which makes cell fate decisions of Müllerian duct epithelium and induces adenosis lesions in the cervix and vagina in women.

Studies in mice showed that DES induced vaginal adenosis by down-regulating RUNX1, which inhibits the BMP4/activin A-regulated vaginal cell fate decision; induced epithelial cell proliferation and inhibited stromal cell proliferation (520); and caused persistent down-regulation of basic-helix-loop-helix transcription factor expression (Hes1, Hey1, Heyl) in the vagina, leading to estrogen-independent epithelial cell proliferation. Neonatal exposure to DES caused persistent changes in expression of IGF-1 and its downstream signaling factors in mouse vaginas. It also up-regulated Wnt4, a factor correlated with the stratification of epithelial cells, in mouse vaginas. Interestingly, the simultaneous administration of vitamin D attenuated the ability of DES to cause hyperplasia of the vagina in neonatal mice.

In the one study in the previous 5 years that did not focus on DES, polypropylene and polyethylene terephthalate did not increase vaginal weight in Sprague-Dawley rats. Although a few studies have been conducted during the previous 5 years on the effects of EDCs on the vagina, such studies are very few in number, small in scope, and focused on DES. Thus, future studies are needed in this largely understudied area before we fully appreciate whether other EDCs impair the vagina.

Premature ovarian failure/early menopause: combined data from three studies on DES indicated that in utero exposure was associated with an increased lifetime risk of early menopause in women (602). However, animal studies have not determined whether DES exposure causes premature ovarian failure. Thus, future studies should focus on this issue.

Fibroids: a few recent studies confirmed the known association between DES exposure and fibroids. In the Sister Study, in utero exposure to DES was positively associated with early-onset fibroids. Similarly, in the Nurses’ Health Study II, prenatal DES exposure was associated with uterine fibroids, with the strongest risk being for women exposed to DES in the first trimester. Given the consistency in findings, future studies should be focused on determining the mechanism by which DES exposure increases the risk of fibroids.

Click to download the summary.
Click to download the full study.

More DES DiEthylStilbestrol Resources

Have your say! Share your views