| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Articles |
Beth Israel Medical Center New York, New York 10003
Address correspondence and requests for reprints to: Barnett Zumoff, Division of Endocrinology and Metabolism, Beth Israel Medical Center, 1st Avenue at 16th Street, New York, New York 10003.
 |
Introduction |
|---|
 Top Introduction References |
|---|
To give such assurances, the endocrinologist must be convinced that there is indeed little or no danger from estrogen administration. Unfortunately, the literature has not reached a consensus about estrogen’s dangers. Since 1941, there have been 70 epidemiological studies dealing with estrogen administration and the risk for breast cancer, 7 meta-analyses of this problem, and uncounted reviews, and though I believe that both the weight of opinion and the weight of evidence indicate that the risk is exceedingly small or nonexistent (10, 11, 12, 13, 14, 15), there are respected voices that still say there is indeed a risk, though probably only with prolonged administration (16, 17, 18). This difference of opinion persists despite the fact that new epidemiological studies continue to appear in the literature at a steady pace, in the hope, so far unrealized, that better experimental design will produce the "definitive" study that will resolve the controversy once and for all.
A solution to the dilemma may be at hand. Because the literature had arrived at a consensus that alcohol consumption is a risk factor for breast cancer (19), the large Nurses’ Health Study, which sought to evaluate the role of estrogen administration as a risk for breast cancer, was designed to stratify for alcohol consumption as a possible associated risk. Analysis of the data, as reported by Colditz et al. (20), yielded a startling finding: only women who consumed alcohol manifested an increased risk of breast cancer with estrogen administration; those who did not consume alcohol showed no increase in risk. To quote the authors: "Among women who did not consume alcohol, the risk of breast cancer was not increased by current use of menopausal hormones ... among alcohol consumers, current hormone users were at increased risk of breast cancer." This is at odds with the statement made by Susan Love in The New York Times (21) that "taking hormones for more than 10 years could increase a woman’s risk of developing breast cancer. The Nurse’s Health Study, a definitive 14-year study of 122,000 nurses issued in 1995, estimated that women between ages 60 and 64 who took hormones for at least five years increased their risk of getting breast cancer by 71 percent. They increase their risk of dying of breast cancer by 45 percent." This is a misstatement in two regards: 1) The Nurses’ Health Study showed an overall increase in breast cancer incidence of 41%, not 71%, and said nothing at all about the risk of dying of the cancer—other workers have reported that the death rate from breast cancer discovered during estrogen use is substantially lower than with other breast cancers, as Colditz et al. mention; 2) The Nurses’ Health Study showed that women who did not consume alcohol had no increased risk of developing breast cancer at all. Colditz et al. noted the finding about the role of alcohol consumption in their discussion, but did not place great emphasis on it despite its potentially profound implications; they merely commented that "the apparent difference in risk due to estrogens according to level of alcohol intake is unexpected and intriguing. Further study is needed ..."
One might think that other epidemiologists would hasten to redesign
their studies and reanalyze their data in the light of this important
observation, but I have found only one study that has done so, the
almost equally large Iowa Women’s Health Study, reported by Gapstur
et al. (22). The data from that study are quite unequivocal
about the critical role of alcohol consumption in determining the risk
of breast cancer with estrogen administration: only women who consumed
5 grams of alcohol or more per day manifested an increased risk of
breast cancer with estrogen administration; those who consumed less
alcohol than that or none at all showed no increase in risk (Fig. 1
).
|
Gapstur et al. were also hesitant about placing great
emphasis on their finding about the role of alcohol because they were
concerned that there was no "biologically plausible" mechanism by
which alcohol could exert the observed effect; but recently a
biologically plausible mechanism has been reported: Ginsburg et
al. (23) have found that when a postmenopausal woman receiving
estrogens consumes alcohol her blood level of estradiol rises acutely,
by about 300% (Fig. 2).
|
1) The elevation of blood estradiol by administration of postmenopausal
estrogen replacement therapy is only modest, i.e. from
castrate levels to values characteristic of the follicular phase of the
menstrual cycle (23); these values may be near a threshold value for
breast cancer-promoting effects, so that some women will manifest
increased risk and some will not, depending on their varying genetic
susceptibility factors, including family history of breast cancer;
abnormal BRCA1, BRCA2, or p53 genes; and elevated capacity to
16
-hydroxylate estradiol (24) [increased 16-hydroxylation may be
a risk factor for breast cancer (25, 26)].
2) When alcohol is consumed by a postmenopausal woman receiving estrogens, the blood estradiol is raised to values characteristic of the periovulatory peak in the menstrual cycle (23), which may be above the threshold of breast-cancer-promoting effects for all women regardless of genetic background, resulting in a clearly demonstrable increase in risk in whole populations.
The question of whether estrogen administration favors the development of breast cancer has been the Gordian Knot of endocrinology. Hundreds of workers have attempted to unravel it by epidemiological studies, meta-analyses, and literature reviews, but without success. Alexander the Great opened his Gordian Knot by slashing it apart with his sword; it now seems that we may be able to dissolve our Gordian Knot in alcohol. If we can assure our patients that consuming less than 5 grams of alcohol daily will eliminate any breast-cancer-promoting risk of estrogen administration, we will have made a very important therapeutic advance.
Received March 26, 1997.
Accepted March 27, 1997.
|
References |
|---|
|
TopIntroductionReferences |
|---|
-hydroxylase
activity in Caucasians. J Surg Oncol. 4:255–273.[Medline]
-hydroxylation of estradiol: a possible risk marker for
breast cancer. Ann NY Acad Sci. 464:138–151.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  S. Fan, Q. Meng, B. Gao, J. Grossman, M. Yadegari, I. D. Goldberg, and E. M. Rosen Alcohol Stimulates Estrogen Receptor Signaling in Human Breast Cancer Cell Lines Cancer Res., October 1, 2000; 60(20): 5635 - 5639. [Abstract] [Full Text]
|
|
|
|
 |
|
 B. Zumoff Alcohol, Estrogens, and Breast Cancer J. Clin. Endocrinol. Metab., July 1, 1997; 82(7): 2375 - 2378. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
