by Alyssa M. Voss, M.P.H.
In epidemiology, what can be learned from eight cases of an extremely rare tumor observed in the same clinic? Quite a lot, in fact, thanks to an inquisitive mother who asked whether her daughter’s tumor was caused by “the medicine” the mother took during pregnancy and to the astute physicians who decided to follow up on her question.
The medicine that the mother referred to was diethylstilbestrol, or DES, a synthetic estrogen manufactured and marketed as a drug that would prevent miscarriage and other complications of pregnancy. DES became clinically available in the United States during the early 1940s and, for the next 30 years, was prescribed to millions of pregnant women. By the late 1950s, a series of clinical trials had been published that showed no efficacy in preventing the adverse events that DES was supposed to keep from occurring. Nonetheless, the drug continued to be prescribed.
Then, in 1971, Dr. Arthur Herbst, Dr. Howard Ulfelder, and Dr. David Poskanzer published their report of eight clear cell adenocarcinomas (CCA) of the vagina and cervix among young women (aged 14 to 22 years old) in a clinic in Boston and their documented prior exposure to DES. The researchers’ findings were noteworthy because CCA is extremely rare and is usually found in much older women. The seminal paper established the association between in utero exposure to DES and a woman’s increased risk of cancer. The paper also prompted the U.S. Food and Drug Administration to issue a warning to physicians that the drug should not be prescribed to pregnant women.
Various clinics across the country initiated clinical and follow-up studies of the effects of DES in females and males exposed to the drug in utero. However, according to Robert N. Hoover, M.D., Sc.D., Director of DCEG’s Epidemiology and Biostatistics Program (EBP), “Over time, enthusiasm and funding for these efforts progressively waned, so that by the late 1980s, most of the follow-up work had stopped.”
Recognizing the need to continue observing DES-exposed mothers and their offspring, Dr. Hoover reached out to several principal investigators at the various study centers and suggested reviving the cohorts and pooling the data to better answer the many questions that remained. Having the larger cohort would allow sufficient statistical power to find associations for rarer outcomes that otherwise would be missed in the individual cohorts.
In 1992, a combined cohort was assembled and named the “DES Follow-up Study.” It included DES-exposed and non–DES-exposed mothers and their offspring from cohorts that had been enrolled during the 1970s along with clinical and other information from that era. The combined study pooled 5,067 daughters exposed to DES in utero and 2,387 non-exposed females. It also included 2,001 DES-exposed sons and 2,111 non-exposed males. Among the mothers, 5,441 were exposed to DES during pregnancy and 4,036 were not exposed, making it the largest follow-up study of DES in the country. Dr. Hoover and Rebecca Troisi, Sc.D., a staff scientist in EBP, have been DCEG’s coprincipal investigators of the study.
Subjects were re-contacted and invited to participate in the combined cohort by completing the 1994 baseline questionnaire and subsequent periodic questionnaires. Since that time, numerous key findings have emerged. “DES is an endocrine-disrupting chemical,” Dr. Troisi stated. “From the many laboratory studies that have been done, we know these types of chemicals can cause cancer, birth defects, and other developmental abnormalities in the reproductive tract, and the effects of exposure are most severe when it occurs during fetal development.” The study team explored these leads and confirmed that effects of DES were evident not only in DES-exposed mothers but also in both daughters and sons exposed to the drug in utero.
In 2001, the study team found that mothers who had been prescribed DES during pregnancy had a 20 to 30 percent increased risk of breast cancer than non–DES-exposed women in the same age group. A 2009 study of DES-exposed sons found that risks for cryptorchidism, epididymal cyst, and testicular inflammation or infection were twice those of non–DES-exposed males and were three times as great among sons exposed to DES before the 11th week of gestation. The sons’ abnormalities, however, had no overall adverse effect on their fertility or reproductive outcomes.
Using data from the 1994, 1997, and 2001 follow-up questionnaires, the study team, led by Dr. Hoover, published an article in the October 6, 2011, issue of the New England Journal of Medicine (NEJM) that included the most comprehensive analysis of data on women exposed to DES while in utero, illustrating the cumulative effects of this exposure over the course of a woman’s lifetime. Among 4,653 DES-exposed and 1,927 non–DES-exposed women, the authors evaluated 12 adverse health and reproductive outcomes—infertility, spontaneous abortion, ectopic pregnancy, loss of second trimester pregnancy, preterm delivery, preeclampsia, stillbirth, neonatal death, early menopause, cervical intraepithelial neoplasia grade 2+, breast cancer, and CCA—all of which had been linked previously to DES exposure in preliminary reports from the combined study (see Table 1).
|Clear cell adenocarcinoma||40 times higher|
|Neonatal death||8 times higher|
|Preterm delivery||4.7 times higher|
|Loss of second trimester pregnancy||3.8 times higher|
|Ectopic pregnancy||3.7 times higher|
|Stillbirth||2.4 times higher|
|Infertility||2.4 times higher|
|Early menopause||2.4 times higher|
|Cervical intraepithelial neoplasia||2.3 times higher|
|Breast cancer||1.8 times higher|
|First trimester miscarriage||1.6 times higher|
|Preeclampsia||1.4 times higher|
Compared with non–DES-exposed women, women exposed to DES in utero had eight times the risk of neonatal death, almost five times the risk for preterm delivery, and nearly four times the risk of having an ectopic pregnancy. In addition, DES-exposed daughters were twice as likely to experience infertility. Risks of breast cancer or neoplastic cervical lesions were approximately doubled and persisted until age 55. Finally, the data showed that the 40-fold increased risk of CCA, the cancer diagnosed in young women exposed to DES in utero, remained elevated until at least age 40.
The 2011 paper by Dr. Hoover and colleagues also was the first to assess possible dose-response effects between the amount of DES exposure in utero and the risk of all outcomes by using a unique marker called vaginal epithelial changes (VEC), a phenomenon observed particularly in women exposed to high doses of DES early in gestation. These changes appear as red granular patches in the epithelium of the vagina and represent a precursor to CCA. According to the report in NEJM, women with VEC experienced even higher risks for 9 of the 12 outcomes evaluated than DES-exposed women without VEC.
Findings from the studies of DES-exposed women have provided the impetus and the laboratory model for the current widespread interest in the biologic impact of endocrine disruptors, particularly from fetal and other early-life exposures. “The DES story has many valuable lessons for clinical medicine, in particular for an entire generation of clinicians who, for the most part, have been unaware of this signal event in medical history,” Dr. Hoover noted. These lessons include the value of clinical trials; the special vulnerability of the fetus; the importance of being alert to unusual events and searching for possible causes; and, perhaps most important, the need for systematic long-term surveillance for adverse outcomes of drug exposures. Dr. Hoover further added, “It is clear that without the startling occurrence of a cluster of a rare cancer in adolescent girls, we would be unaware of the massive burden of multiple common adverse health outcomes ultimately endured by DES-exposed women.”
The DES Follow-up Study will continue to follow DES-exposed daughters as they move into the menopausal years and beyond (i.e., the ages when cancers more commonly occur) as well as follow DES-exposed sons. In addition, the study recently has expanded to include granddaughters of DES-exposed and non–DES-exposed women, referred to as the “third generation.” Because of the relatively young age of the research participants, these studies will focus on exploring the basis for biological persistence of DES effects across generations.
Read After DES: Tracking the Harms of a Prenatal Drug Exposure in the NCI Cancer Bulletin Archive.
View The DES Story: Lessons Learned, a video produced by the NCI Cancer Bulletin.