Authors
Cunha GR, Forsberg JG, Golden R, Haney A, Iguchi T, Newbold R, Swan S, Welshons W.

Title
New approaches for estimating risk from exposure to diethylstilbestrol.

Journal
Environmental Health Perspectives 107 Suppl 4:625-30. 1999.

In utero exposure to diethylstilbestrol (DES), a potent estrogen, once prescribed to pregnant women to prevent miscarriages, produced gross abnormalities of the reproductive tract in addition to a rare vaginal adenocarcinoma in affected female offspring. Amidst growing concern of the potential adverse effects of environmental estrogens, the National Institute of Environmental Health Sciences workshop reviewed the DES incident and the lessons learned regarding the identification of estrogenic compounds, the interpretation of animal studies and the extrapolation of these results to humans. DES and estrogens were known to induce breast cancer in post-natal rats and mice and to perturb urogenital development in fetal rodents. Could proper interpretation of these studies have prevented the clinical use of DES in pregnant women and the tragedy that unfolded?

Assessment of the risks posed by environmental estrogens requires a fundamental understanding of early embryo and fetal development and the identification of the critical periods of development that are highly sensitive to endocrine modulation. This information can be obtained using animal models, classic embryology studies and cell lines. Dose-response studies must also be conducted to ascertain the dose ranges that produce urogenital abnormalities and cancer. Many models designed to investigate endocrine disruptors use doses above the range of human exposure. The development of low dose models will help identify the quantitative and qualitative effects of low dose estrogen exposure. The mechanism of action of environmental estrogens, including DES, is presumed to be via the estrogen receptor (ER). Characterization of ER expression, tissue distribution, isoforms and receptor affinities during development is essential to best interpret the effects of estrogenic compounds on development.

The human data available following in utero DES exposure is fairly abundant, however, there are still unanswered questions. Developmentally abnormalities would certainly be sensitive to DES during critical periods of differentiation. Thus, the exact dose range and timing of DES exposure that induces urogenital abnormalities should be determined. Many DES-affected offspring are only now reaching 50 years of age, the age where reproductive cancers typically occur. Continued follow-up of DES survivors is essential to further our understanding of the long-term effects of gestational exposure.

Additional exposure studies using DES must now use animal models. To best extrapolate the results from animal studies to humans requires the selection of a suitable animal model, the selection of an appropriate dose range to best approximate human exposure and studies designed to examine the effects of DES exposure during fetal development. The mouse, rat and hamster have all been used to some extent, with the DES-mouse model perhaps the best animal model. Results from perinatally-treated DES mice are very similar to the urogenital abnormalities seen in humans. Using animal models, sensitive biological endpoints can be used to identify selective estrogen response modulators (SERMs) at the low dose range of the compound (same range as endogenous estrogen). There are many biological markers (HOX genes, lactoferrin) that may be used to quantify estrogenic action. These markers, if conserved, would be particularly useful to extrapolate results from animal models to humans. The use of transgenic mice lacking one or both of the ER isoforms would enable characterization of the mechanism of action of DES and other estrogenic compounds. This would confirm that the adverse effects induced by estrogenic compounds are mediated by ER, and may also indicate separate receptor-mediated pathways resulting in either teratogenicity and/or carcinogenicity.

The DES experience has provided most of what we know today about the adverse effects of estrogenic compounds and remains perhaps, the most clear-cut example of estrogenic toxicity. Lessons from the DES experience are vital to the prevention of a similar situation and enable a greater understanding of the actions of estrogens during fetal development. While we are far more cognizant of the exquisite sensitivity of fetal development to all pharmaceuticals (DES, thalidomide), this does not preclude effects of dietary phytoestrogens, stored lipophilic pesticides in addition to accidental or industrial exposures to estrogenic compounds. For these reasons, continued investigation into the mechanisms of action of these chemicals, their relative potencies and the sensitivity of each stage of development to estrogenic exposure is essential.

>Go to next summary in this workshop