The Critical Role of Alcohol Consumption in Determining the Risk of Breast Cancer with Postmenopausal Estrogen Administration

Barnett Zumoff

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
 
Discussing with a postmenopausal woman whether she should take hormone replacement is one of the most important tasks of the clinical endocrinologist. Because of the reported major health benefits to a postmenopausal woman from taking estrogens, including decreased osteoporosis (1, 2, 3), decreased coronary disease (4, 5, 6), possibly decreased Alzheimer’s disease (7), and decreased overall mortality (8, 9), the endocrinologist can probably do such a woman more good by prescribing estrogens and encouraging their continued use than by any other treatment. However, many women resist taking estrogens because they are afraid it might cause breast cancer. Convincing them that there is little or no danger of that is what an endocrinologist has to do to secure compliance with his or her advice. Telling a patient that the benefits of estrogen administration outweigh the danger doesn’t work in many cases; the patient is looking for assurances, not sophisticated explanations of benefit/risk ratios.

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. 1Go).



View larger version (31K):
[in this window]
[in a new window]
 
Figure 1. Data from Gapstur et al. (22). The relative risk (RR) of breast cancer in women who do not consume alcohol and have never used estrogens is set at 1.0. Increasing amounts of alcohol consumption have no significant effect on RR in these never-users of estrogens. Women who have used estrogen but consume little or no alcohol show no significant increase in breast-cancer risk—RRs do not differ significantly from 1.0; in contrast, women who have used estrogen and consume 5 grams of alcohol per day or more (5 grams corresponds to 2/3 ounce of whiskey or 3 ounces of wine) show a highly significant increase in risk (RR of slightly over 1.8).

 
It should be emphasized that no one has published data that disagree with the findings of these two studies concerning the role of alcohol consumption.

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. 2Go).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 2. Data from Ginsburg et al. (23). Postmenopausal women who are not using estrogen replacement show castrate levels of serum estradiol, and alcohol consumption does not increase these levels. This corresponds to the absence of any effect of alcohol on increasing breast-cancer risk in women who have never used estrogens (see Fig. 1Go). Postmenopausal women who are on estrogens have basal serum estradiol levels corresponding to those of the follicular phase of the menstrual cycle; consumption of alcohol results in a rise of serum estradiol to values corresponding to those of the periovulatory peak in the menstrual cycle; the effect lasts for more than 6 h.

 
In light of the studies of Colditz et al., Gapstur et al., and Ginsburg et al., I propose the following hypothesis to account for the great variability of the results of even well-designed studies intended to assess whether estrogen administration is a risk for breast cancer and the apparently critical role of alcohol consumption:

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{alpha}-hydroxylate estradiol (24) [increased 16{alpha}-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
 Top
 Introduction
 References
 

  1. Hutchinson TA, Polansky SM, Feinstein AD. 1979 Postmenopausal estrogens protect against fractures of the hip and distal radius. Lancet. 2:705–709.[Medline]
  2. Weiss NS, Ure CL, Ballard JH, Williams AR, Daling JR. 1980 Decreased risk of fractures of the hip and lower forearm with postmenopausal use of estrogens. N Engl J Med. 303:1195–1198.[Abstract]
  3. Paganini-Hill A, Ross RK, Gerkins VR, Henderson BE, Arthur M, Mack TM. 1981 Menopausal estrogen therapy and hip fractures. Ann Int Med. 95:28–31.[Medline]
  4. Ross RK, Paganini-Hill A, Mack TM, Arthur M, Henderson BE. 1981 Menopausal estrogen therapy and protection from death from ischemic heart disease. Lancet. 1:858–860.[CrossRef][Medline]
  5. Szklo M, Tonascia J, Gordis L, Bloom I. 1984 Estrogen use and myocardial risk: a case-control study. Prev Med. 13:510–516.[CrossRef][Medline]
  6. Stampfer MJ, Willet WC, Colditz GA, Rosner B, Speizer FE, Hennekens CH. 1985 A prospective study of postmenopausal estrogen therapy and coronary heart disease. N Engl J Med. 313:1044–1049.[Abstract]
  7. Tang M-X, Jacobs D, Stern Y, et al. 1996 Effect of oestrogen during menopause on risk and age at onset of Alzheimer’s disease. Lancet. 348:429–432.[CrossRef][Medline]
  8. Bush TL, Cowan CD, Barrett-Connor E, et al. 1983 Estrogen use and all-cause mortality: preliminary results from the Lipid Research Clinics Programs Follow-Up Study. Circulation. 249:903–906.
  9. Criqui MH, Suarez L, Barrett-Connor E, McPhilips J, Wingard DL, Garland C. 1988 Post-menopausal estrogen use and mortality. Results from a prospective study in a defined, homogeneous community. Am J Epidemiol. 128:606–614.[Abstract/Free Full Text]
  10. Boston Collaborative Drug Surveillance Group. 1974 Surgically confirmed gall bladder disease, venous thromboembolism, and breast tumors in relation to postmenopausal estrogen therapy. N Engl J Med. 290:15–19.[Medline]
  11. Casagrande J, Gerkins V, Henderson BE, Mack TM, Pike MC. 1976 Exogenous estrogens and breast cancer in women with natural menopause. J Natl Cancer Inst. 56:839–841.[Medline]
  12. Kaufman DW, Miller DR, Rosenberg L, et al. 1984 Noncontraceptive estrogen use and the risk of breast cancer. JAMA. 252:63–67.[Abstract]
  13. Dupont WD, Page DL. 1991 Menopausal estrogen replacement therapy and breast cancer. Arch Int Med. 151:67–72.[Abstract]
  14. Newcomb PA, Longnecker MP, Storer BE, et al. 1995 Long-term hormone replacement therapy and risk of breast cancer in postmenopausal women. Am J Epidemiol. 142:788–795.[Abstract/Free Full Text]
  15. Stanford JL, Weiss NS, Voigt LF, Daling JR, Habel LA, Rossing MA. 1995 Combined estrogen and progestin hormone replacement therapy in relation to risk of breast cancer in middle-aged women. JAMA. 274:137–142.[Abstract]
  16. Steinberg KK, Thacker SB, Smith SJ, et al. 1991 A meta-analysis of the effect of estrogen replacement therapy on the risk of breast cancer. JAMA. 265:1985–1990.[Abstract]
  17. Hulka BS, Brinton LA. 1995 Hormones and breast and endometrial cancers: preventive strategies and future research. Environ Health Perspect. 103[Suppl 8]:185–189.
  18. Adami H-O, Persson I. 1995 Hormone replacement and breast cancer. A remaining controversy? JAMA. 274:178–179.[CrossRef][Medline]
  19. Longnecker MP, Berlin JA, Orza MJ, Chalmers TC. 1988 A meta-analysis of alcohol consumption in relation to risk of breast cancer. JAMA. 260:652–656.[Abstract]
  20. Colditz GA, Stampfer MJ, Willett WC, Hennekens CH, Rosner B, Speizer EE. 1990 Prospective study of estrogen replacement therapy and risk of breast cancer in postmenopausal women. JAMA. 264:2648–2653.[Abstract]
  21. Love S. 1997 Sometimes Mother Nature Knows Best. The New York Times. Op-Ed Page, Thursday, March 20.
  22. Gapstur SM, Potter JD, Sellers TA, Folsom AR. 1992 Increased risk of breast cancer with alcohol consumption in postmenopausal women. Am J Epidemiol. 136:1221–1231.[Abstract/Free Full Text]
  23. Ginsburg EL, Mello NK, Mendelson JH, et al. 1996 Effects of alcohol ingestion on estrogens in postmenopausal women. JAMA. 276:1747–1751.[Abstract]
  24. Lemon HM. 1972 Genetic predisposition to carcinoma of the breast: multiple human genotypes for estrogen 16{alpha}-hydroxylase activity in Caucasians. J Surg Oncol. 4:255–273.[Medline]
  25. Schneider J, Kinne D, Fracchia A, et al. 1982 Abnormal oxidative metabolism of estradiol in women with breast cancer. Proc Natl Acad Sci 79:3047–3051.
  26. Bradlow HL, Hershcopf R, Martucci C, Fishman J. 1986 16{alpha}-hydroxylation of estradiol: a possible risk marker for breast cancer. Ann NY Acad Sci. 464:138–151.[Abstract]