Letter to the Editor;
New England Journal of Medicine
Winnifred B. Cutler, Ph.D., Elizabeth Genovese, M.D.
The New England Journal of Medicine, Volume 322, Number 3, pp. 202-203., Jan. 18, 1990.
© Copyright 1990.
To the Editor:
We are concerned about three specific issues that were not mentioned ore addressed in the paper by Bergkvist et al. Because of the wide spread adverse publicity about hormone-replacement therapy that this paper generated and the tremendous benefit of such therapy to women concerned about bone, cardiovascular, and sexual health and emotional well-being, it is particularly important that studies purporting to show an increased risk for hormone-replacement therapy be well balanced.
The possibility of detection bias was not mentioned, and it may be at work in this study. The women for whom hormone-replacement therapy was prescribed were very likely to have had breast examination. In contrast, there is no reason to expect that the control population was undergoing breast examination. A recent study by Brownson et al.1 of breast cancer screening by mammography among 16,000 women revealed that 88 percent of the prevalent cases were identified at first screening.
For this reason, the finding of an increased incidence of breast cancer in a group of women undergoing breast evaluation would be expected when they were compared with a group of women not undergoing breast evaluation. Particularly in the light of the slow growth of many breast tumors, the possibility of detection bias should be considered in this study. The visits to physicians rather than the coincident use of hormone-replacement therapy may account for the increased incidence of breast cancer that was discovered.
Another concern, particularly with the finding of a possible influence of progestin in association with the increased incidence of breast cancer, is fibrocystic breast disease. We do not know the data for Sweden and that genetic group, but data for France and North America suggest that perhaps one in three women has fibrocycstic breast disease. One third of them, according to the recent work of Dupont and Page2, can be expected to have proliferative disease and are thus at increased risk of breast cancer.
An accepted hormonal treatment for fibrocystic breast disease includes progestin therapy, which has been shown to be effective in reducing pain and the size of the cysts.3 Therefore, unless the authors can reevaluate their data in the light of the presence of fibrocystic breast disease in these women, we cannot tell whether the progestin users are precisely the women one would have expected to be at increased risk of breast cancer. If this was true, the study would be confounded by this uncontrolled but critical component.
The validity of the authors' conclusion is therefore questionable in that many women who choose hormone-replacement therapy or have it prescribed for them may have a different risk of breast cancer and may also be more likely to be screened for breast cancer, with a subsequent higher rate of detection than among peers not receiving hormone-replacement therapy or undergoing evaluation for fibrocystic breast disease.
Winnifred B. Cutler, Ph.D.
Women's Wellness Research (Athena Institute)
Haverford, PA 19041
Elizabeth Genovese, M.D.
1. Brownson RC, Blackwell CW, Pearson DK, Reynolds RJ, Richens JW Jr., Papermaster BW. Risk of breast cancer in relation to cigarette smoking. Arch Intern Med 1988; 148:140-4.
2. Dupont WD, Page DL. Risk factors for breast cancer in women with proliferative breast disease. N. Engl J Med 1985; 312:146-51.
3. Sitruk-Ware LR, Sterkers N, Mowszowicz I, Mauvais-Jarvis P. Inadequate corpus luteal function in women with benign breast disease. J Clin Endocrinol Metab 1977; 44:771-4.