Learning from a Mistake? The History of DES (Diethylstilbestrol)

Chris Kahlenborn
(chapter ten)
Breast Cancer: Its link to Abortion
and the Birth Control Pill
Reproduced with Permission

DES is the abbreviation for a synthetic estrogen called diethvlstilbestrol, which was given to pregnant diabetic mothers because certain researchers (de, White and Hunt in 1943) reported that it would decrease a pregnant diabetic woman's risk of having a miscarriage. It was later used for other indications such as pre-eclampsia, premature labor, etc. It was usually given between the 8th and the 15th week of pregnancy and continued until well into the third trimester. The initial trials started in the late 1940s but women took it regularly from the 1950s until the 1960s. In the mid 1950s, it was noted that DES did not decrease a woman's rate of miscarriage and over time it was discovered that the daughters of the women who had taken DES experienced a higher incidence of both vaginal and cervical cancer1. But what about the mothers who took DES? Did they need to worry about getting any particular cancer? After all, DES is an artificial estrogen and it was given during the first trimester of pregnancy. Either of these two facts should give a woman cause to worry.

Researchers continued to monitor women for any observation of cancer, especially breast cancer. For almost 25 years, researchers claimed that there was no link between DES and breast cancer risk in the women who had used it. Finally, in 1978, Bibbo et al published a study which showed a 47% increased risk of developing breast cancer, but the authors under-emphasized the findings because they were not statistically significant: "We detected no statistically significant differences between the two groups in the frequency and types of uterine, ovarian, breast and other reproductive-tract abnormalities." [2, p.766] Studies in the 1980s confirmed what many had feared: DES causes a statistically significant increase in breast cancer.

How much does DES elevate the risk of breast cancer? Colton et al published a classic article3 which provided a detailed account of this. Two tables which contain information from that article serve to illustrate:

Table 10A:
Risks of DES Found in Various Studies
Author Percent Increase Year of Study Number
of Subjects
Relative Risk
Bibbo et al [2] 46% increase* 1978 693 1.46 (not significant)
Beral & Colwell [1] infinite risk 1980 79 infinite risk CI's
not given
Vessey et al [see 3] not reported 1983 319 not reported
et al [see 3]
37% increase* 1984 1531 1.37 (0.83-2.28)
Colton et al [3] 35% increase 1993 2864 1.35 (1.05-1.74)

* This result reflects a trend toward an increased risk but does not attain statistical significance.

Table 10A shows the progression of data from the various studies since 1978. Every study points to an increased risk of breast cancer from DES, except the Vessey study, which did not report any results. Perhaps the most significant result is found in the study by Colton et al, especially given the size of the study and the fact that it had the longest latent period. Note that it calculated a statistically significant 35% risk overall. Data and results from the Colton study [3] give us more important information as noted in Table 10B, which follows:

Table 10B:
Breast Cancer Risk To Women Who Took DES
Time Since Exposure (years) Percent Increase Relative Risk
0-9 13% increase* 1.13 (0.38-3.37)
10-19 13% increase* 1.13 (0.70-1.81)
20-29 36% increase* 1.36 (0.93-1.99)
30 or more 33% increase* 1.33 (0.95-1.87)
less than 40 22% increase* 1.22 (0.48-3.09)
40-49 18% increase* 1.18 (0.74-1.87)
50-59 18% increase* 1.18 (0.82-1.71)
60 or over 47% increase 1.47 (1.02-2.13)

* This result reflects a trend toward an increased or decreased risk but does not attain statistical significance.

What lesson should we learn from Table 10B? First, it should be noted from the upper half of the table, that it took more than 20 years before the risk of DES and breast cancer began to show a trend toward an increased risk. Second, we see that even after 30 years, the trend toward an increased risk of DES remained high -- over 30%. Third, the bottom half of the table shows us an extremely important finding: The risk of DES appears to be greatest in women older than 60.

Q-10A: Why does the present author bring up the example of DES?

DES is extremely important because it provides us with a historical lesson of how a synthetic estrogen, when given during pregnancy, caused a significant increase in breast cancer 20-30 years after its first use. DES provides us with a possible medical corollary to both the oral contraceptive pill (OCP) and/or abortion. What is that correlation? DES contained a synthetic estrogen; today's OCPs contain synthetic estrogens. Although the estrogens are certainly different (de, DES was a non-steroidal whereas today's OCPs contain steroidal estrogens) and are given at different times and doses, one fact stands out: DES, a synthetic estrogen, causes breast cancer despite studies which found no risk for almost 25 years. Just as important, DES was noted to have its greatest effect in women after the age of 60. This same phenomenon may be happening in regard to early OCP use. It was already noted in Chapters 8 and 9 that virtually every study since 1980 to date, which looked at women who took oral contraceptives before their first full-term pregnancy (FFTP) and were under the age of 45 on or before 1995, showed an increased risk in breast cancer. In the past, some researchers have imprudently reassured women "not to worry" because they predict that this increased risk will not be sustained as women approach their 50s and 60s. This sort of statement must be evaluated in light of the historical example of DES and its increasing risk in women 60 years of age and older. Better for the laity to make this decision than the "experts."

Q-10B: Does the historical example of DES shed any light on the question of the link between an abortion performed early in a woman's reproductive life and breast cancer?

It may. As was noted, DES was given to pregnant mothers in an attempt to prevent miscarriages. Thus, DES and induced abortion have one thing in common: they both are a "hormonal blow" to the body at a very sensitive time in the development of the breast. When DES was given, especially between the 8th and 15th weeks of pregnancy, it introduced a foreign synthetic estrogen into a woman's body at a critical stage -- when she was pregnant and her breast cells were dividing rapidly. In an analogous way, an early induced abortion results in an abrupt decrease in a woman's hormone levels during this very same critical stage. Thus, although different, each one is a "hormonal blow" to the body at a time when the breast may be especially sensitive -- during early pregnancy. If we recall that it took about 25 years to note the true dangers of DES and the increased risk of breast cancer in mothers who took it, similarly, we may just be beginning to see the real increase in breast cancer in women who had abortions performed early in their reproductive lives as data from the late 1990s and early 21st century come in. In addition, if induced abortion follows a similar pattern as DES, we may notice the biggest increases in breast cancer in women after the age of 60.

Q-10C: Is there any similarity between the fertility drug Clomid and DES?

There may be. Clomid®* (clomiphene), made by Hoechst Marion Rousel, is a fertility drug which acts upon the pituitary gland, so that it increases the hormones involved in ovulation (called LH and FSH), which results in a tremendous increase in ovarian activity. Clomid® "is capable of interacting with estrogen receptors."4 It is therefore possible that it may affect a woman's risk of developing breast cancer years after it is first used.

*Clomid and the other commonly used fertility drugs have been noted to markedly increase the risk of ovarian cancer. In a large collaborative study which pooled the results of several studies, Whittemore et al5 found that women who used fertility drugs and who did not get pregnant had a 2600% increased risk in obtaining ovarian cancer compared to women who never took fertility drugs. Spiritas recalculated the data and noted that Whittemore's data actually resulted in an 1100% increased risk6.

Q-10D: It was noted that daughters of women who took DES experienced a higher incidence of both vaginal and cervical cancer [1]. Could the daughters of women who took Clomid® or OCPs also experience such effects?

The PDR (Physicians' Desk Reference) states that "Newborn rats, injected (with Clomid®) during the first few days of life, also developed metaplastic changes in uterine and vaginal mucosa. . ." (PDR, 1997). Thus, theoretically the daughters who are born to women who have recently taken Clomid~ could also experience an increased risk of cervical and vaginal cancer.

The same question could be asked of women who become pregnant while taking OCPs. In 1996 Potter7 found that the pregnancy rate for "typical use" in women using OCPs was 7%. This means that tens of thousands of women have and will continue to become pregnant while taking OCPs. In these incidences DES and the OCP are both artificial hormones that would have been taken at some point in pregnancy, so it is not unreasonable to ask: Will the daughters of women who take OCPs have higher incidences of breast, cervical, vaginal or other cancers? What about the effects on the sons of these women? For example the rate of testicular cancer has risen by 50% over the past 25 years. Is it possible that these men were conceived while their mothers were taking oral contraceptives? The answers to these questions may have catastrophic consequences and yet to the best of this author's knowledge, the questions have never been asked. A formal study of these areas is urgently needed.


1 Beral V, Colwell L. Randomized trial of high doses of stilboestrol and ethisterone in pregnancy: long-term follow-up of mothers. Br Med J. 1980; 281: 1098-1101. [Back]

2 Bibbo M, Haenszel W, et al. A twenty-five-year follow-up study of women exposed to diethylstilbestrol during pregnancy. N Engl J Med. 1978; 298; 763-767. [Back]

3 Colton T, Greenberg ER, et al. Breast cancer in mothers prescribed diethylstilbestrol in pregnancy. JAMA, 1993; 269: 2096-3000. [Back]

4 Physicians' Desk Reference (1997). Description of Clomid (R). [Back]

5 Whittemore AS, et al. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 U.S. case-control studies. II. Invasive epithelial ovarian cancers in white women. Collaborative Ovarian Cancer Group. Am J Epidemiol. 1992; 136: 1184-1203. [Back]

6 Spiritas R, et al. Fertility drugs and ovarian cancer: Red alert or red herring. Fertility and Sterility. 1993; 59: 291. [Back]

7 Potter LA. How effective are contraceptives? The determination and measurement of pregnancy rates. Obstet Gynecol. 1996; 88: 13S-23S. [Back]