Walker1 reported an increased frequency of development of uterine adenocarcinoma and ovarian cystadeno‐carcinoma in diethylstilbestrol (DES) gestationally exposed mice. Subsequently, Walker and Kerth2 reported a multigenerational effect of DES in mice noting an increased frequency of uterine and cervical adenocarcinoma in third‐generation mice (offspring of the mice exposed in utero to DES). More recently, Newbold et al3 reported an increased occurrence of uterine adenocarcinoma in the female offspring of mice who had been exposed to DES at varying periods of time during and immediately after the completion of gestation. They observed an increased prevalence of these tumors with age. These studies have raised considerable concern regarding the potential for the development of reproductive tract abnormalities in the female offspring of mothers who were exposed in utero to DES. The current report is of the findings observed in 28 female offspring of mothers who were exposed in utero to DES.
MATERIALS AND METHODS
Baylor College of Medicine was one of the original participants in the National Diethylstilbestrol Adenosis (DESAD) project, which was funded by the National Cancer Institute in 1974 to investigate the prevalence and natural history of genital tract anomalies and clear cell adenocarcinoma in women exposed to DES in utero (second‐generation exposed women).4 Enrollment in this study continued through January 1, 1980. Patients were recruited from three sources. The first group was comprised of women identified as having been exposed to DES through a review of hospital or physician office prenatal records. The second group of women with documentation of in utero DES exposure were referred to the screening clinic by physicians. The third group was comprised of women who were self‐referred to the clinic and who brought with them written evidence of in utero DES exposure. The second‐generation women enrolled in the study were followed yearly with clinical examinations through 1980 and were mailed questionnaires from 1984 through 1989. Further follow‐up of these patients continued with participants being mailed a detailed questionnaire on cancer risk factors and health history in 1994 and again in 1997.
In 1997, a letter was mailed to the in utero exposed second‐generation participants in the Baylor study who had given birth to a female offspring. The letter inquired as to the willingness of these third‐generation offspring to come into the clinic for a free examination. A total of 105 of the mothers still living in Texas responded affirmatively. Of these, 70 were believed to be “eligible” for examination (over age 16). To date, a total of 28 third‐generation offspring have been examined, three of whom were delivered from one mother. Informed consent was obtained from each of the women to perform the examination. Several attempts have been made to encourage the remainder to participate, but many of the third‐generation women have refused for a variety of reasons, including being away at college or because they had already seen a gynecologist and been told they had no abnormalities.
Each third‐generation participant was given the same questionnaire that had been given to her mother at the time of her enrollment in the original DESAD project. In addition, the examination carried out in the clinic was identical to the examination carried out on her mother. This included breast and abdominal examination, bi‐manual pelvic examination, inspection of the cervix grossly, and colposcopy of the cervix and vagina. A cervical‐vaginal smear was obtained on each individual with a separate sample taken from the upper vagina as well as from the cervix, and iodine staining of the vagina and cervix was performed. All the examinations were performed by a single gynecologist who had extensive experience in the examination of DES‐exposed women. This same physician had also examined the majority of the mothers.
The records of the 26 mothers of these 28 patients were reviewed, and findings recorded at the time of their initial DESAD examination were noted.5,6 The findings included the presence of vaginal epithelial changes and/or structural change in 28% of record‐review participants and 74% of the walk‐in and referral groups.5 These findings were quite similar to those observed in the entire cohort of women participating in the DESAD project.5,6
For statistical evaluation, Fisher exact test was used for differences in DES‐related changes between in utero exposed mothers and their third‐generation daughters. This test was applied to 26 mothers and 26 daughters because of the assumption that daughters from the same family would not be statistically independent. The difference in findings between the mothers and their daughters was highly significant (P > .001). The study was reviewed and approved by the Institutional Review Board of Baylor College of Medicine.
All of the exposed mothers were white, middle‐class women, similar to the population of women originally enrolled in the DESAD project.4 The mean age of the mothers when they entered the DESAD study was 23.2 (range 14–32, standard deviation [SD] 4.0), and the mean age of the third‐generation daughters at the time of pelvic examination was 20.1 (range 15–28, SD 4.2) (Table 1). The mothers currently are an average of 44.2 years (range 35–49, SD 3.0). The average age of menarche of the mothers was 12.9 (range 11–16, SD 1.2) and that of the daughters 12.8 (range 11–16, SD 1.5). Fourteen of the daughters were sexually active, and two were parous. Two of the 26 mothers exposed to DES in utero demonstrated structural changes of the cervix and/or upper vagina, ten demonstrated structural changes of the cervix and or vagina as well as vaginal epithelial changes, and four had vaginal epithelial changes alone. The structural changes of the cervix included cockscomb, cervical collar, and/or pseudopolyp. The vaginal epithelial changes consisted of the presence of adenosis, nonstaining of the vaginal epithelium with Lugol's solution, and acetowhite change of the vagina after the application of 4% acetic acid to the cervix and vagina. The findings were considered negative in ten of the women (Table 2). Thus, 16 (61.5%) of the mothers exposed to DES demonstrated cervical and/or vaginal changes characteristic of DES exposure.
The findings observed in the 26 mothers exposed in utero to DES were related to the time of exposure to DES during their pregnancies. There was no apparent difference in the frequency of these changes between those exposed in the first trimester and those exposed during the second trimester. Findings in all 28 of the third‐generation offspring of these women were completely normal (P < .01 compared with mothers). There were no cases of endometrial, ovarian, cervical, vaginal carcinoma, or of intraepithelial neoplasia of the cervix or vagina found in any of the third‐generation participants.
This is a preliminary report of an ongoing study investigating changes seen in female offspring of women exposed in utero to DES. This study was stimulated by reports,1–3 which have generated significant concern among DES‐exposed women with female offspring. The cumulative risk of DES‐related clear cell adenocarcinoma of the vagina and cervix is estimated to range between 0.14 and 1.4 per 1000.7 Therefore, the number of women in this report is too small to have any meaning as regards the risk for adenocarcinoma. Furthermore, the adenocarcinoma of the endometrium reported by Newbold et al3 tended to develop as the mice aged, and the mean age of the third‐generation women reported here was 20.1 (range 15–28). Thus, it is possible that, as these women become older, adenocarcinoma of the endometrium could begin to occur with increasing frequency. However, the report by Hatch et al8 was reassuring in that with an average age of 38, no increase in the occurrence of adenocarcinoma of the endometrium in women exposed in utero to DES was observed. What is significant is the fact that in a group of mothers exposed in utero to DES who were found to have either structural changes of the cervix and/or vaginal epithelial changes with a high frequency (61.5%), none of the daughters were found to have similar findings. If structural changes were to occur in the third‐generation offspring, they would have presumably developed by the age at examination. The average age at which the third‐generation daughters were examined was similar to the average age of their mothers when they enrolled in the project. These negative findings are reassuring and suggest that it is likely that any abnormalities found in DES‐exposed women will occur at a markedly reduced rate in their daughters or may not be seen at all. However, the third‐generation daughters may still be different from the general female population and the number of examined women at the present time may have been too small to exclude this possibility.
Vaginal epithelial changes were found in 29% of record‐review exposed mothers and 63% of the referral and walk‐in patients. These findings are comparable to those noted on enrollment in the DESAD study where 34% of the record‐review patients and 62% of the referral and walk‐in patients demonstrated these changes. Twenty‐five percent of the record‐review women in the DESAD project demonstrated structural changes of the cervix and 45% of the walk‐in and referral patients demonstrated these findings as well. In the present study, structural changes were found in 14% of the record‐review mothers and 58% of the walk‐in and referral mothers. It would appear that the 26 mothers whose daughters were examined had findings similar to those observed overall in the DESAD population of women. Thus, these 26 mothers appear to be representative of this cohort.
Of some concern is the fact that only 28 of the 70 “eligible” third‐generation daughters were examined. This raises the possibility of some selection bias. However, in the overwhelming majority of cases, those women not seen for examination were either away at college or had already been examined by a gynecologist and had been told that no abnormalities were present. An additional factor that could lend some bias to our findings is the fact that 61.5% of the mothers were known to have either structural changes of the cervix or vaginal epithelial changes, suggesting that those mothers knowing that some abnormality was present were more likely to insist that their daughters be seen in our clinic. However, rather than detracting from our results, this probably lends further strength to our findings. If all the mothers had been “normal,” one would question whether the lack of abnormal findings in their daughters had any significance. The fact that so many of the mothers demonstrated DES‐associated changes would make it more likely that their daughters would demonstrate similar findings if there was in fact a third‐generation carryover effect.
A variety of theories have been proposed for the mechanism by which a multigenerational effect of DES could occur. In mice, the carcinogenicity of DES can apparently be transmitted from the prenatally‐exposed offspring to the next generation. Germ cell mutation has been implicated as the mode of transmission, and the study of Walker and Kerth2 would certainly suggest a genotoxic effect of DES upon the germ cells. Another possibility is that the multigenerational cancer effect was transmitted by imprinting. In contrast to the findings by Walker,1 Walker and Kerth,2 and Newbold et al,3 Forsberg and Halling9 were unable to detect a third‐generation effect in female mice after the mothers' neonatal exposure to DES.
Although the numbers are small, the striking difference in findings of lower genital tract abnormalities between the mothers and their offspring suggests that the multigenerational effect of DES observed in mice is likely to be dramatically lower in humans, if present at all. The only recommendation we can make on the basis of our study is that third‐generation daughters should be carefully examined for the presence of DES‐associated changes seen in many of the DES‐exposed women. However, they can be reassured that, on the basis of preliminary findings, it is probably unlikely that any significant abnormalities will be seen in third‐generation daughters.
1. Walker BE. Tumors of female offspring of mice exposed prenatally to diethylstilbestrol. JNCI 1984;73:133–40.
2. Walker BC, Kerth LA. Multi-generational carcinogenesis from diethylstilbestrol investigated by blastocyst transfers in mice. Int J Cancer 1995;61:249–52.
3. Newbold RR, Hanson RB, Jefferson WN, Bullock BC, Haseman J, McLachlan JA. Increased tumors but uncompromised fertility in the female decendents in mice exposed developmentally to diethylstilbestrol. Carcinogenesis 1998; 19:1655–63.
4. LaBarthe D, Adam E, Noller KL, O'Brien PC, Robboy S, Tilley BC, et al. Design and preliminary observation of the National Cooperative Diethylstilbestrol Adenosis (DESAD) project. Obstet Gynecol 1978;51:453–8.
5. O'Brien PC, Noller KL, Robboy SJ, Barnes AB, Kaufman RH, Tilley BC, et al. Vaginal epithelial changes in young women enrolled in the National Cooperative Diethylstilbestrol Adenosis (DESAD) project. Obstet Gynecol 1979;53:300–8.
6. Jeffries JA, Robboy SJ, O'Brien PC, Bergstralh EJ, LaBarthe DR, Barnes AB, et al. Structural anomalies of the cervix and vagina in women enrolled in the Diethylstilbestrol Adenosis (DESAD) project. Am J Obstet Gynecol 1984;148:59–66.
7. Melnick S, Cole P, Anderson D, Herbst AL. Rates and risks of diethylstilbestrol — Related clear cell adenocarcinoma of the vagina and cervix. N Engl J Med 1987;316:514–6.
8. Hatch EE, Palmer JR, Titus-Ernstoff L, Noller KL, Kaufman RH, Mittendorf R, et al. Cancer risk in women exposed to diethylstilbestrol in utero. JAMA 1998;280:630–4.
© 2002 The American College of Obstetricians and Gynecologists
9. Forsberg JG, Halling A. Failure to detect a second-generation effect in female mice after neonatal treatment with estrogen (diethylstilbestrol). Acta Anat 1992;114:103–6.