Outcome and follow-up of diethylstilbestrol (DES) exposed individuals

INTRODUCTION — Diethylstilbestrol (DES) is a nonsteroidal estrogen first synthesized in 1938. It was initially used for postpartum lactation suppression and treatment of postmenopausal symptoms but quickly became an accepted intervention for prevention of miscarriage, preterm delivery, and other obstetric problems. DES exposure is associated with increased risks of cervical and vaginal cancer, congenital anomalies of the female and male genitourinary tracts, and adverse effects on fertility and reproductive function.

DES remains the only firmly established transplacental carcinogen in humans. Because of this risk, it is important to continue to monitor exposed offspring and their mothers for long-term health effects. Patients who took DES during pregnancy should inform their health care providers and their offspring. Offspring should also inform their health care providers of in utero exposure. We generally agree with the American Cancer Society and National Cancer Institute (NCI) recommendations for follow-up of DES-exposed individuals [1,2]. These recommendations reflect expert opinion based on the data described above. Where the recommendations are vague, we have provided our approach to these patients.

The outcomes and care of DES-exposed individuals are reviewed here. Other issues related to female reproductive tract anomalies or vaginal or cervical cancer are discussed separately. (See "Congenital uterine anomalies: Clinical manifestations and diagnosis" and "Benign cervical lesions and congenital anomalies of the cervix" and "Invasive cervical cancer: Epidemiology, risk factors, clinical manifestations, and diagnosis" and "Vaginal cancer".)

IDENTIFYING THE EXPOSED POPULATION — It is estimated that five million pregnant patients worldwide received DES to improve pregnancy outcome. Use of DES in pregnant patients to prevent miscarriage began in approximately 1938, but use did not become common until the late 1940s, and it was approved by the US Food and Drug Administration (FDA) for this indication in 1947 [3]. Although clinical trials in the 1950s demonstrated that it was ineffective for prevention of adverse pregnancy outcomes [4], it remained in use until 1971 in the United States and until the early 1980s in some other countries (eg, France stopped use in 1977 [5]). The peak years of DES use were approximately 1952 to 1954, and the range was approximately 1945 to 1971. In 1971, in utero exposure to DES was linked to the occurrence of vaginal clear cell adenocarcinoma (CCA) in female offspring [6], which prompted the FDA to advise against its use in pregnancy [7]. DES is no longer commercially available for human use in North America.

PATHOPHYSIOLOGY OF DES EXPOSURE — DES readily crosses the placenta. Exposure to DES during a critical period of organogenesis disorganizes the developing uterine muscle layers; causes maldevelopment of the uterotubal junction; and prevents stratification of the vaginal epithelium and resorption of vaginal glands, resulting in vaginal adenosis.

In murine studies, the mechanism for DES-induced adenosis is blocked expression of Trp63 in müllerian duct epithelium at epithelial estrogen receptor alpha [8]. DES has also been shown to alter normal programming of the gene families Hox and Wnt, which affect differentiation of the reproductive tract [9-13].

DES is also important as the model agent for endocrine-disrupting chemicals and their effects on fetal development and adult health [14,15]. Aberrant DNA methylation has been proposed as a molecular mechanism whereby endocrine disruptors such as DES regulate developmental programming [16].

PATIENTS WHO RECEIVED DES DURING PREGNANCY (DES MOTHERS) — Most patients who received DES were of higher socioeconomic status and thus very few very young (<20 years old) patients would have been exposed. Therefore, in 2020, the most common age range for DES mothers would be 87 to 107 years (many are deceased), with the very youngest mothers being approximately 69 years of age.

Clinical follow-up for DES mothers — DES mothers should have regular preventive care.

Patients who took DES in pregnancy should inform their children of in utero exposure, so they can receive appropriate health care and follow-up.

●As in all patients, recommendations for initiation, method, and frequency of breast cancer screening should be based on patient-specific composite risk factors, including DES exposure. In the absence of any other risk factors for breast cancer, the lifetime risk of breast cancer in DES mothers is estimated to be approximately 17 percent, which is considered mildly to moderately elevated (where low risk is <15 percent and high risk is >20 to 25 percent lifetime risk). A detailed discussion of initiation, frequency, and modalities for breast cancer screening based on patient-specific factors can be found separately.

●DES mothers have no increased risk for cervical, vaginal, or other cancers. They should be screened for cervical cancer according to standard guidelines. (See "Screening for cervical cancer in resource-rich settings".)

Breast cancer risk — There appears to be a slightly increased incidence of breast cancer among patients who were prescribed DES during pregnancy (DES mothers) [17-22]. The largest and most comprehensive study on this topic compared long-term cancer risk in a database of 3844 patients prescribed DES during pregnancy with an unexposed comparison group of 3716 patients [18]. Patients who took DES had a small increased risk of breast cancer (relative risk [RR] 1.27, 95% CI 1.07-1.52); looked at in another way, their lifetime risk of breast cancer increased from one in eight (approximately 12 percent) to one in six (approximately 17 percent). There was no evidence of an interaction between the use of postmenopausal hormone therapy and DES in relation to breast cancer risk, and the risk did not increase over time. The higher risk did not appear to be due to inherent characteristics of the patients who were given DES (ie, patients with pregnancy problems) [20].

Other cancers — Patients who were prescribed DES during pregnancy are not at increased risk of vaginal or cervical cancer. They also do not appear to be at an increased risk of ovarian or endometrial cancer [17-22].

Mortality — A study of 3857 DES-exposed and 3818 unexposed patients reported no substantial increase in mortality rates among patients prescribed DES compared with unexposed patients with the exception of breast cancer mortality [23]:

●All-cause mortality (RR 1.06, 95% CI 0.98-1.16)

●Mortality from cerebrovascular disease (RR 1.08, 95% CI 0.78-1.47)

●Mortality from cardiovascular disease (RR 0.98, 95% CI 0.83-1.16)

●Mortality from breast cancer (RR 1.27, 95% CI 0.96-1.69)

●Cancer mortality (all types; RR 1.07, 95% CI 0.94-1.23)

FEMALES EXPOSED TO DES IN UTERO (DES DAUGHTERS) — With peak use between 1952 to 1954, the average age of DES daughters in 2021 would be approximately 67 years of age, with the youngest being approximately 50 in 2021. Thus, most DES daughters are peri- or postmenopausal, and fertility and pregnancy impacts are typically no longer part of clinical care.

In utero exposure to DES has been associated with several potential risks in female offspring (DES daughters):

●Cervicovaginal clear cell adenocarcinoma (CCA)

●Cervical intraepithelial neoplasia (CIN)

●Congenital anomalies and epithelial changes of the reproductive tract

●Subfertility and adverse pregnancy outcomes

●Earlier age at menopause

●Breast cancer

Clinical follow-up for DES daughters — Clinical follow-up for patients with in utero DES exposure includes:

●Screening for cervical and vaginal cancer – Historically, DES-exposed daughters were advised to have an annual gynecologic examination with screening for cervical and/or vaginal cancer, and no stopping age has been determined. However, most guidelines from major expert groups (eg, American College of Obstetricians and Gynecologists [ACOG], American Society for Colposcopy and Cervical Pathology [ASCCP]) do not specifically address cervical and vaginal cancer screening in such patients.

In our practice, we perform the following:

•Annual pelvic examination.

•During speculum examination, we rotate the speculum to allow inspection of the entire epithelial surface of the vagina and cervix. Adenosis may appear red and granular, while squamous metaplasia may be indistinguishable from normal squamous epithelium without iodine staining. Clear cell carcinoma of the cervix or vagina usually presents as a polypoid mass, but findings may vary and any raised, hard, or fleshy lesions should be evaluated further. Vaginal neoplasms most often occur on the anterior wall.

•Cervical and vaginal cytology specimens are obtained. Specimens are obtained from all four quadrants (circumference) of the upper vagina and the cervix (if present). All the specimens can be placed on one slide/container.

If screening for CCA is the only indication for screening, we do not perform human papillomavirus (HPV) testing, as oncogenic HPV is probably not a factor in development of CCA.

•Colposcopy is not performed routinely as a screening tool and is used as in the general population to further evaluate an abnormality found on gross visualization or palpation or to evaluate abnormal cytology. Routine colposcopy adds unnecessary cost and has not proven to be essential for the detection of clear cell carcinoma, likely due to the lack of specific vascular changes associated with the disease and the fact that some tumors may be intramural and difficult to visualize on colposcopy [24].

When performed, the entire vagina and cervix should be examined with staining with Lugol iodine. Any abnormalities are biopsied; we do not perform random biopsies. (See "Colposcopy".)

•Bimanual pelvic and rectal examinations include palpation of the entire length of the vagina, including the fornices, and a thorough digital examination of the cervix. This is essential for detection of CCA and may be the only way to identify these lesions; cytology will be negative if tumor cells have not penetrated the vaginal epithelium. Areas of thickening or induration or nodules should raise suspicion for CCA and should be biopsied.

•We continue annual screening as long as the patient is a candidate for intervention if cancer is diagnosed. However, the age at which to discontinue screening is unknown. No studies have evaluated when screening can be discontinued in such patients, and there are no recommendations from expert groups regarding an upper age limit. While there is a theoretical concern that there may be a second peak in related cancers as the cohort ages [25,26], as DES daughters become older, the relative benefits and risks (eg, false positives) of annual screening becomes more unclear [1].

●Breast cancer screening – Breast cancer screening of DES-exposed daughters is similar to that of non-DES exposed individuals and is discussed in detail separately. (See "Screening for breast cancer: Strategies and recommendations".)

●Hormonal medications – DES-exposed patients may be concerned about using hormonal therapies, but there is no evidence postmenopausal estrogen therapy confers a higher risk of adverse effects in DES daughters than in unexposed patients [18,27].

Neoplastic disease — DES daughters have a greatly increased risk of vaginal or cervical CCA, but the absolute number of cases of this disease remains low. DES exposure in utero is also associated with a higher risk of squamous cell cervical cancer or precancers. It is uncertain whether there is an increased risk of breast cancer.

Vaginal or cervical clear cell adenocarcinoma — DES daughters have a 40-fold increase in risk of CCA of the vagina and cervix compared with unexposed patients [28,29]. However, the cumulative lifetime incidence of CCA is very low, estimated to be 1 case per 1000 to 2000 DES-exposed daughters. A registry established to track cases of cervical and vaginal CCA worldwide included 775 cases as of 2015; approximately two-thirds of these cases were patients exposed to DES in utero [25,30-32].

In patients in the registry, overall survival at five years was approximately 80 percent, but slightly higher in DES daughters than other patients (86 versus 81 percent), and at 20 years was approximately 70 percent [32].

Most cases of DES-associated CCA have been diagnosed among females in their late teens and 20s, with a range of 7 to 48 years [25]. By comparison, most cases of CCA among nonexposed patients occur postmenopausally, although early and late age peaks have been reported in nonexposed patients [33].

There are no data to allow clear conclusions about the upper age limit for the diagnosis of DES-associated CCA, especially regarding late risk in DES daughters after menopause. Although there is no firm evidence, a possible increase in incidence of DES-associated CCA at age ≥40 years has been suggested [25,26]. For this reason, DES daughters should be screened for CCA throughout their lives. (See 'Clinical follow-up for DES daughters' above.)

Oncogenic HPVs are unlikely to be a cofactor in the development of CCA of the vagina or cervix; however, if there is an association, it is less than that reported for squamous or non-CCAs [34].

Ninety percent of cases of CCA related to DES have been diagnosed at stage I or II [35]. Survival rates at these stages are good (80 to 90 percent), but recurrences as late as 20 years following diagnosis have been documented. (See "Vaginal cancer", section on 'Adenocarcinoma'.)

Vaginal or cervical adenosis — Adenosis of the vagina or cervix is defined as the development of columnar epithelium of the ectocervix and vagina rather than normal squamous epithelium. The columnar epithelium may cover the surface or form glands within the stromal wall. As a result, the squamocolumnar junction, which is normally near the external os or in the endocervical canal, can be in the vagina. DES is one cause but not the only cause of adenosis; it may be related to other hormones or environmental chemicals (endocrine disrupters) [30].

In a large cohort study of DES-exposed patients (the National Cooperative Diethylstilbestrol Adenosis [DESAD] project), one-third of these patients had vaginal and cervical adenosis [36]. The prevalence of adenosis in DES daughters has been reported as high as 91 percent in some studies compared with 4 percent in non-DES-exposed controls [30].

Adenosis often regresses over time; in the DESAD cohort, the majority of patients with vaginal adenosis underwent vaginal epithelial changes, and the lesions appeared to regress [37]. Thus, treatment may be unnecessary except in rare patients who are symptomatic. For those patients who are symptomatic, simple excision or fulguration with CO₂ laser or unipolar cautery can be considered [38,39]. Although benign, adenosis is considered a precursor of CCA; the factors that promote malignant transformation of adenosis to CCA are not known, and no clear transformation from adenosis to CCA has been documented [40]. In addition, the treatment of vaginal adenosis did not significantly reduce the rate of new dysplasia of in utero DES-exposed patients [38].

Squamous cervical cancer — The risk of squamous cervical precancer appears to be increased in DES daughters. The excess risk may be due to an increase in the size of the cervical transformation zone, leading to greater susceptibility to HPV infection.

The risk of high-grade CIN was increased almost twofold in DES-exposed daughters (cumulative risk to age 55 years: CIN2 or higher [CIN2+] was 5.3 versus 2.6 percent in controls; hazard ratio [HR] 2.10, 95% CI 1.41-3.13) [41]. The risk was higher up to age 45 (age <45: HR 2.47, 95% CI 1.55-3.94) but was not apparent in patients age 45 and above, based on an imprecise estimate. Risks were higher in patients exposed earlier in gestation and in those with documented vaginal epithelial changes.

A study from the Netherlands reported an increased risk of CIN1 among approximately 12,000 DES-exposed daughters compared with expected rates from the general population, but no overall increase in CIN2, 3, or invasive cervical cancer was found. An increased risk of CIN2 was found among DES-exposed daughters with vaginal/cervical epithelial changes, but the authors speculated that this increase, as well as the increase in CIN1, was likely due to increased screening in these daughters. DES exposure was originally ascertained by self-report as opposed to medical records [42], which was a limitation of this study.

Breast cancer — The available data on risk of breast cancer in female patients exposed to DES in utero are conflicting [26,43,44]; there is no evidence of an increased risk of breast cancer before age 40 years [45]. Breast cancer in males is extremely rare, and to date there is no evidence to suggest that the risk of breast cancer is elevated among males exposed to DES in utero.

An increased risk in female patients ≥40 years of age exposed to DES in utero was observed in a prospective cohort study from the United States (3.9 versus 2.2 percent, HR 1.82, 95% CI 1.04-3.18) [43]. However, a subsequent study suggested that the increased risk in exposed female patients has attenuated as the cohort has aged, with an overall HR for prenatal DES exposure compared with no exposure of 1.07 for all ages combined, and a standardized incidence ratio (SIR) of 1.17 (95% CI 1.01-1.36) [46], comparing DES-exposed with population-based cancer incidence rates.

A similar but larger study from the Netherlands also found little association between prenatal DES exposure and breast cancer [26]. Continued follow-up is warranted to determine whether the increase in breast cancer observed in the earlier United States study was a transient finding that will disappear as the patients age.

Other cancers — There is no established risk of an increase in other types of cancer in DES daughters.

One study found no overall increase in risk of all cancers combined, or individual cancers, except for breast cancer and CCA, when comparing rates in the exposed daughters with expected rates based on the Surveillance, Epidemiology, and End Results (SEER) population (SIR 1.01, 95% CI 0.94-1.8) [28]. A subsequent report from this study also found no overall increase in all cancers combined (SIR 1.02, 95% CI 0.91-1.14) but did report an increase in pancreatic risk (SIR 2.43, 95% CI 1.21-4.34) based on 11 observed cases in the exposed. A small increase in non-Hodgkin lymphoma was also observed (SIR 1.72, 95% CI 1.08-2.61).

Another study found an increase in melanoma among exposed daughters before, but not after, age 40 years (before age 40 years: SIR 1.59, 95% CI 1.08-2.26) [26]. A limitation of this study is that DES exposure was not documented by medical records.

Reproductive tract abnormalities — Congenital reproductive tract anomalies associated with in utero DES exposure include:

●Uterus – Hypoplastic uterus, T-shaped uterine cavity (figure 1A-C and image 1), constrictions and adhesions of the endometrial cavity

●Cervix – Hypoplasia, collar, hood, polyps, ectropion (adenosis of the ectocervix)

●Vagina – Ridge, transverse septa, vaginal adenosis

Uterus — Uterine changes (hypoplasia, abnormalities of the endometrial cavity) in nonpregnant patients are likely to be asymptomatic, although one study found a decrease in the duration and amount of menstrual bleeding, which may have been related to the uterine changes [47]. Cycle length was not affected.

An expert group suggests making the diagnosis of a T-shaped uterus if all of the following criteria are present on three-dimensional ultrasound (image 2) [48]:

●Lateral internal indentation depth ≥7 mm

●Lateral indentation angle ≤130 degrees

●T-angle ≤40 degrees

They consider the diagnosis borderline if two criteria are present and do not make the diagnosis if ≤1 criterion is present.

Cervix and vagina — Gross cervical abnormalities (hypoplasia, collar, hood, polyps) are visualized in approximately 20 percent of exposed patients. In one study, approximately 75 percent of DES-exposed patients had an abnormal hysterosalpingogram; however, this estimate is likely to be high since it was derived from a population undergoing evaluation for infertility [49]. Vaginal abnormalities (ridge, septa) are less common. Reproductive tract anomalies may be associated with a higher risk of pregnancy complications. (See 'Reproductive function issues' below.)

Vulva — No changes in the vulva associated with DES exposure have been reported.

Benign gynecologic tumors — The risk of ovarian cysts does not appear to be increased in DES-exposed offspring, but an increase in paraovarian cysts has been reported [50]. There may be a slightly increased risk of fibroids [51].

Reproductive function issues — The US Food and Drug Administration (FDA) advised against DES use in pregnant patients in 1971. Exposure may have continued in some United States patients and in other countries through the 1980s. At some point, all of these patients will be postmenopausal, and reproductive impacts will no longer be part of clinical care, as noted above.

Infertility — The majority of patients exposed to DES in utero do not have difficulty becoming pregnant, but DES exposure did appear to confer a higher risk of subfertility. The largest comprehensive study of 3769 DES-exposed and 1654 unexposed patients found that DES-exposed patients were more than twice as likely to experience infertility (attempting but failing to conceive over a period of ≥12 months) compared with unexposed patients (cumulative risk to age 45 years: 33.3 versus 15.5 percent; HR 2.37, 95% CI 2.05-2.75) [43]. The infertility risk was highest for patients with vaginal epithelial changes at a young age, a marker of high DES dose and exposure early in gestation.

Infertility does not appear to be due to an increased prevalence of ovulatory or hormonal problems in DES-exposed patients but appears related to uterine and tubal abnormalities [52]. Endometriosis may also be a factor. A large prospective cohort study of nurses found that the risk of endometriosis was increased by 80 percent among patients who reported that they had been exposed to DES prenatally [53].

Ectopic pregnancy — DES daughters have an increased risk of ectopic pregnancy. This was illustrated by a prospective cohort study of 2692 DES-exposed and 1293 unexposed patients who had ever been pregnant [43]. After adjustment for number of pregnancies, DES-exposed patients had an 11.7 percent excess risk of ectopic pregnancy (cumulative risk to age 45 years: 14.6 versus 2.9 percent; HR 3.72, 95% CI 2.58-5.38). The higher risk of ectopic pregnancy is most likely related to tubal and uterine abnormalities [54].

DES daughters should be evaluated for ectopic pregnancy early in the first trimester. A definitive diagnosis of intrauterine pregnancy can be made 4.5 to 6 weeks after the last menstrual period and when the serum human chorionic gonadotropin concentration is at least 1000 to 1500 international units/L. (See "Ectopic pregnancy: Clinical manifestations and diagnosis".)

Pregnancy loss and preterm birth — Patients exposed to DES in utero also experienced higher risks of miscarriage and premature birth, possibly related to the structural abnormalities of the cervix/uterus seen in DES daughters. The possibility of cervical insufficiency leading to midtrimester pregnancy loss or preterm birth was not eliminated by either a normal cervical appearance or a prior term delivery in the DES-exposed patients.

A prospective cohort study compared the pregnancy outcome of 2692 patients exposed to DES in utero with 1293 unexposed controls who had ever been pregnant [43]. DES exposure was associated with increased cumulative risk to age 45 years of spontaneous abortion (50.3 versus 38.6 percent; HR 1.64, 95% CI 1.42-1.88), preterm delivery (53.3 versus 17.8 percent; HR 4.68, 95% CI 3.74-5.86), and loss of second-trimester pregnancy (16.4 versus 1.7 percent; HR 3.77, 95% CI 2.56-5.54). Although the risk of preterm birth was increased, most DES daughters delivered at term in their first pregnancy (64 versus 85 percent of unexposed patients) [55]. The risk of pregnancy complications appears to be higher in patients with a reproductive tract abnormality [43,56,57].

Secondary sex ratio — A study that evaluated the secondary sex ratio of offspring born to DES-exposed patients did not find a significant effect in exposed patients (odds ratio [OR] 1.05, 95% CI 0.95-1.17) [58]. However, when the results were evaluated according to dose and timing of first exposure to DES, there was an increase in the proportion of male births among patients exposed early in gestation to higher doses of DES (OR 1.24, 95% CI 1.04-1.48 for first trimester exposure to 5 or more grams of DES) compared with no exposure.

Preeclampsia — Early in utero DES exposure appears to be associated with an increase in risk of preeclampsia (cumulative risk to age 45 years: 26.4 versus 13.7 percent; HR 1.42, 95% CI 1.07-1.89 adjusted for number of births) [43].

Stillbirth — In utero DES exposure appears to be associated with an increase in risk of stillbirth (cumulative risk to age 45 years: 8.9 versus 2.6 percent; HR 2.45, 95% CI 1.33-4.54) [43]. However, there were only 70 stillbirths in the entire cohort of DES-exposed and unexposed patients.

Earlier menopause — A longitudinal study of patients exposed to DES in utero found that they reached natural menopause slightly earlier than unexposed patients (51.5 versus 52.2 years) and were 1.5 times more likely to experience natural menopause at any given age compared with unexposed patients of the same age [59]. In a subset of the study cohort that had complete information on the cumulative dose of DES given during pregnancy, there were significant dose-response effects (eg, a twofold higher risk of menopause in patients exposed to the highest dose level [>10 grams]). In the same study population, DES-exposed patients (n = 3393) were more than twice as likely to have an early age at natural menopause (<45 years of age) as unexposed patients (n = 1682; 5.1 versus 1.7 percent; HR 2.35, 95% CI 1.67-3.31) [43]. Animal studies have reported reductions in ovarian follicles among DES-exposed mice, which support the biologic plausibility of these findings.

Postmenopausal hormone levels — A small pilot study of 40 prenatally DES-exposed and 20 unexposed patients evaluated total estrogens as well as parent estrogens and estrogen metabolites [60]. All patients in the study were postmenopausal, and patients who had ever used hormone therapy or who had had cancer were excluded. In general, DES-exposed patients had slightly higher concentrations of estrogens, after controlling for time since menopause, body mass index, parity, and alcohol use. In exposed compared with unexposed patients, concentrations were increased of total estrogens (15 percent higher) and parent estrogens (27.1 percent higher). Ratios of estrogen metabolites were also lower in DES-exposed patients, including ratios of 2-hydroxylation pathway metabolites (path2):parent estrogens (36.5 percent lower) and path2:16-hydroxylation pathway (path16; 28.8 percent lower).

Autoimmune diseases — The weight of evidence in humans suggests that autoimmune diseases are not more common among DES-exposed offspring [61,62]. One cohort study reported an increased risk of confirmed cases of rheumatoid arthritis in patients under 45 years, but this was counterbalanced by a reduced risk among exposed patients 45 years and older [63]. Long-term follow-up and further research as the DES-exposed population ages are needed, as animal studies have indicated that DES can affect immune function [64].

Psychological disorders — Among studies of psychosexual development in DES-exposed offspring compared with an unexposed group, some reported an increased prevalence of depression, anxiety, and anorexia nervosa; these studies were mostly based upon small numbers and had methodologic flaws [65]. One large series including 2684 males and 5686 females reported no association between in utero DES exposure and self-reported mental illness [66]. A follow-up study of this cohort reported no association between prenatal DES exposure and depression in either females or males [67].

Obesity — Although animal studies have reported increased body fat among female mice exposed to DES prenatally [68,69], human data have been equivocal [70]. Interestingly, patients exposed to lower DES doses had an increased risk of obesity compared with patients exposed to high doses of DES. More studies are needed to determine the risk relationship between DES exposure and obesity in humans.

Other

●Diabetes and cardiovascular disease – A preliminary study reported there may be a small increase in diabetes and heart disease in DES-exposed patients.

In a questionnaire study of approximately 3500 DES-exposed and 1500 unexposed patients, the risk for diabetes among DES-exposed patients was increased (adjusted HR 1.49, 95% CI 1.04-2.12) [71]. Self-reported values for hypertension and high cholesterol were also slightly elevated, and there was a trend toward increased cardiovascular disease (CVD). In another study, researchers updated the data on CVD through 2013 and validated self-reports of CVD by medical records. Results based on reported conditions were consistent with the earlier study. Medical records were obtained for approximately 60 percent of reported CVD outcomes and were validated in 78 percent. When the analysis was restricted to verified outcomes, including CVD deaths from linkage with the National Death Index, the association between prenatal DES exposure and CVD persisted, with a 74 percent (HR 1.74, 95% CI 0.82-3.68) increase in coronary artery disease risk and 163 percent (HR 2.63, 95% CI 0.90-7.74) increase in myocardial infarction. DES was not related to an increase in stroke. Adjustment for possible mediating conditions (hypertension, high cholesterol, and diabetes) had little effect on the results, suggesting that other mechanisms, such as changes in hormones or epigenetic markers, may be involved. On the other hand, overdiagnosis among DES-exposed individuals is a possible explanation.

●Sexual orientation and gender identity – Data on sexual orientation and gender identity are limited. In a cohort study including 5154 patients (64 percent females), females with prenatal DES exposure were less likely to identify themselves as having a lesbian or bisexual orientation, while DES-exposed males were slightly more likely to identify themselves as having a gay or bisexual orientation, but estimates were imprecise [72].

MALES EXPOSED TO DES IN UTERO (DES SONS) — Health effects of in utero DES exposure among male offspring (DES sons) have not been studied as extensively as in females.

Clinical follow-up for DES sons — There are no recommendations for special screenings or tests.

In the general population, there is no evidence that routine screening for testicular cancer through periodic clinical examinations or self-examinations improves health outcomes. Symptomatic males should report their symptoms (eg, painful or tender mass, testicular firmness, scrotal heaviness, hematuria) to their health care provider. (See "Screening for testicular cancer" and "Clinical manifestations, diagnosis, and staging of testicular germ cell tumors".)

Genitourinary abnormalities — The largest study of genitourinary abnormalities in DES sons included 1157 males prenatally exposed to DES [73]. Compared with 1038 unexposed males, DES-exposed males were at increased risk of epididymal cysts (risk ratio [RR] 2.4, 95% CI 1.5-4.3), cryptorchidism (RR 1.9, 95% CI 1.1-3.4), and testicular inflammation/infection (RR 2.5, 95% CI 1.5-4.4). Exposure prior to 11 weeks of gestation was associated with the highest risk. There was no increase in other genitourinary abnormalities (varicocele, structural abnormalities of the penis, urethral stenosis, benign prostatic hypertrophy, or inflammation/infection of the prostate, urethra, or epididymis).

A long-term follow-up study of male offspring exposed to large doses of DES during a randomized trial at the University of Chicago in the 1950s found that 15 percent of DES sons reported a genitourinary abnormality compared with only 5 percent of unexposed sons [74]. As above, abnormalities occurred more frequently among males who were exposed before 11 weeks of gestation. DES sons had no impairment of sexual function, as measured by frequency of intercourse or reported episodes of decreased libido.

However, others have not observed an increased risk of reproductive abnormalities in DES sons [75,76].

Infertility — Most studies report no impairment of fertility in males with in utero DES exposure [74,75,77-80]. As an example, in the large placebo-controlled study discussed above, DES sons (with or without genital malformations) had no impairment of fertility when assessed in terms of whether they had ever impregnated a person, age at the birth of their first child, average number of children, medical diagnosis of a fertility problem, or length of time to conception in the most recent pregnancy of the female partner [74].

Testicular and prostate cancers — The role of DES in the development of testicular cancer is a controversial subject. Cryptorchidism (a risk factor for testicular cancer) has been reported in the sons of patients exposed to DES or oral estrogens during pregnancy, and some studies reported an association between DES or other exogenous hormone exposure during pregnancy and testicular cancer [81-84]. More rigorous studies have failed to prove, but have not definitely excluded, an epidemiologic link between DES exposure and testicular cancer [75,85-88]. Assessment of DES exposure in most of these studies was based on maternal self-report many years after the pregnancy and may have been inaccurate. The number of cases analyzed was small, further limiting the power of the studies. In the only cohort study with medical record confirmation of exposure status, the testicular cancer incidence rate in exposed sons was approximately three times that of the unexposed but was based on only six exposed and two unexposed cases [87].

While studies in mice have demonstrated an association between DES exposure and an increased rate of rete testis cancer and prostate cancer [89], there is no evidence of an association between DES exposure in humans and prostate cancer or other cancers. In a prospective study including over 2900 male patients (mean age 62 years), patients with prenatal DES exposure compared with no DES exposure had similar rates of overall cancer and prostate cancer; an unexpected risk reduction in urinary system cancers was observed in DES-exposed patients (hazard ratio [HR] 0.48, 95% CI 0.23-1) [90].

THIRD-GENERATION EFFECTS — The terms third-generation DES daughter or son refer to a child of a DES daughter (ie, a female exposed to DES in utero). Few human studies on the transgenerational effects of DES have been published. Adverse effects are suggested by some data, but further study is needed. Epigenetic alterations are the proposed mechanism of third-generation effects [91].

●A prospective cohort study assessed menstrual and reproductive outcomes with baseline and follow-up questionnaires. A total of 796 females with a mother exposed to DES in utero and 469 females with an unexposed mother were included in the baseline survey [91]. Based on follow-up data from a total of 381 exposed and 280 unexposed included in this cohort, third-generation DES daughters had a slightly increased prevalence of irregular menses (prevalence ratio [PR] 1.32, 95% CI 1.1-1.6, 25 versus 19 percent), amenorrhea (PR 1.26, 95% CI 1.06-1.49, 18 versus 15 percent), and preterm delivery (PR 1.54, 95% CI 1.35-1.75, 12 versus 8 percent). There was little difference in age at menarche. An analysis within a subset of females whose mothers had vaginal epithelial changes found that the prevalence of amenorrhea was slightly higher among third-generation DES-exposed compared with unexposed.

A previous study of a subset of the population in the study cited above [91] reported the mean age at menarche was 12.6 years in both DES-exposed and unexposed third-generation daughters, but daughters of the exposed patients attained menstrual regularization later (mean age of 16.2 versus 15.8 years) and were more likely to report irregular menstrual periods (odds ratio [OR] 1.54, 95% CI 1.02-2.32) [92].

●A study examined 28 DES third-generation daughters in whom 61 percent of their mothers had well-documented genital tract changes associated with DES [93]. No lower genital tract DES-associated abnormalities were found in these daughters, who ranged in age from 15 to 28 years.

●A cohort study from the Netherlands reported a higher prevalence of hypospadias in third-generation DES sons [94]. These results were based on a small number of cases but were confirmed in a subsequent retrospective study from France [95]. By contrast, a subsequent larger study that documented in utero DES exposure did not find a statistically significant association between DES and hypospadias in the third generation (OR 1.7, 95% CI 0.4-6.8) [96].

●A case-control study from the Netherlands observed an association between esophageal atresia/tracheoesophageal fistula in third-generation offspring and maternal exposure to DES in utero [97]. In addition, a survey of patients participating in a large, multicenter study of prenatal DES exposure in the United States reported an increase in congenital anomalies among third-generation sons (OR 1.53, 95% CI 1.04-2.23) and daughters (OR 2.35, 95% CI 1.44-3.82) compared with unexposed controls [98]. Third-generation daughters appeared to have an excess of congenital heart disease. Reporting bias may have played a role in these findings.

●No clear increase in cancer risk in third-generation males and females has been demonstrated. A possible increase in ovarian tumors in third-generation females was observed; however, this was based upon only three cases in the exposed group and is thus uncertain [99].

●A study nested within the Nurses' Health Study reported that third-generation exposure to DES may be associated with an increased risk of attention deficit hyperactivity disorder (ADHD) in exposed grandchildren [100]. A total of 1.8 percent of patients in the cohort (n = 861 of 47,540 patients) reported in utero exposure to DES. ADHD among offspring was reported by 8 percent of exposed versus 5 percent of unexposed mothers (OR 1.36, 95% CI 1.1-1.67). When the analysis was stratified by trimester of DES exposure, the association was highest for offspring of mothers who had been exposed to DES during the first trimester (OR 1.63, 95% CI 1.18-2.25). There was little evidence for effect modification by sex of the offspring, although ADHD was more common among male offspring, as expected.

Animal studies have found increased rates of tumors, including uterine, ovarian, and rete testis tumors, among third-generation mice exposed to DES [101,102]. The incidence of tumors was low, and the tumors tended to occur in older animals.

RESOURCES

●Registry for Research on Hormonal Transplacental Carcinogenesis (Clear Cell Cancer Registry)

University of Chicago, Department of Obstetrics and Gynecology

●Resources that provide information and support for individuals who were exposed to DES:

•DES Action USA (https://desaction.org/)

•American Cancer Society (American Cancer Society)

•National cancer institute (https://dceg.cancer.gov/research/what-we-study/des-study)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Vulvar cancer and vaginal cancer".)

SUMMARY AND RECOMMENDATIONS

●Patients who took DES during pregnancy (DES mothers)

•Patients who took diethylstilbestrol (DES) during pregnancy (DES mothers) should inform their health care providers and their offspring. Offspring should also inform their health care providers of in utero exposure. (See 'Introduction' above.)

•DES mothers have a slightly increased risk of developing breast cancer. Such patients follow current guidelines for initiation, method, and frequency of breast cancer screening and prevention for their age and risk group. (See 'Breast cancer risk' above and 'Clinical follow-up for DES mothers' above.)

●Females exposed to DES in uterus (DES daughters)

•Females exposed to DES in utero (DES daughters) appear to be at increased risk for:

-Cervicovaginal clear cell adenocarcinoma (CCA) (see 'Vaginal or cervical clear cell adenocarcinoma' above)

-Cervical intraepithelial neoplasia (CIN) (see 'Squamous cervical cancer' above)

-Congenital anomalies and epithelial changes of the reproductive tract (figure 1A and figure 1B and figure 1C) (see 'Reproductive tract abnormalities' above)

-Infertility, ectopic pregnancy, pregnancy loss, preterm birth (see 'Infertility' above and 'Ectopic pregnancy' above and 'Pregnancy loss and preterm birth' above)

-Earlier menopause (see 'Earlier menopause' above)

•Data on increased risk of breast cancer in females exposed to DES in utero are conflicting. (See 'Breast cancer' above.)

•For DES-exposed daughters, we suggest performing yearly cervical and vaginal cytology and continuing this screening as long as the patient is a candidate for intervention if cancer is diagnosed rather than following cervical screening recommendations in the average risk population (Grade 2C). When collecting the cytology specimens, all four quadrants of the upper vagina should be sampled and sent for analysis, in addition to the cervical specimen (if present). (See 'Clinical follow-up for DES daughters' above.)

●Males exposed to DES in utero (DES sons) have a higher prevalence of benign epididymal cysts, cryptorchidism, and testicular inflammation/infection. It is unclear whether they have any increased risk of testicular/prostate cancer. They do not appear to have an increased frequency of infertility. No special screening or tests are recommended. (See 'Males exposed to DES in utero (DES sons)' above.)

●Third generation – Few human studies on the transgenerational effects of DES have been published. Adverse effects are suggested by some data, but further study is needed. (See 'Third-generation effects' above.)