Postmenopausal Hormone Therapy and Breast Cancer: A Danish Cohort Study of Mortality
Abstract & Commentary
By Leon Speroff, MD, Editor, Professor of Obstetrics and Gynecology, Oregon Health and Science University, Portland, is Editor for OB/GYN Clinical Alert
Synopsis: Breast cancer mortality is increased in hormone users because of an increase in incidence despite a better prognosis because of more favorable tumors.
Source: Stahlberg C, et al. Breast cancer incidence, case-fatality and breast cancer mortality in Danish women using hormone replacement therapy—a prospective observational study. Int J Epidemiol. 2005;34:931-935.
The Danish Nurse Cohort was established in 1993. A total 10,874 women were available for follow-up after excluding hysterectomized women and cases of breast cancer identified at baseline. There were 2,726 current users (25.1%) of hormone therapy, 1,582 past users (14.5%), and 6,566 never users at baseline in 1993. Two hundred forty-four women developed invasive breast cancer through 1999 (74.8% ductal carcinomas and 15.6% lobular carcinomas). Cases were followed to mid-April, 2004. Mortality was assessed in 2 ways: case-fatality analysis for breast cancer death was based on 71 deaths in the 244 cases; breast cancer mortality for death from breast cancer was based on 71 deaths in the entire population of 10,874 women. Hazard ratios (HR) compared users to never users. The results reflect a variety of hormone products and regimens (see Table 1).
The calculations in Table 1 are adjusted for age only. The authors then calculated all-cause mortality based on 1,225 deaths, adjusting for smoking, alcohol use, and physical activity—finding no effect of hormone therapy on all-cause mortality (see Table 2).
Stahlberg and associates concluded that breast cancer mortality is increased in hormone users because of an increase in incidence despite a better prognosis because of more favorable tumors.
Commentary
It is worth repeating the key message of this report because it is a little confusing. Stated as simply as possible: hormone users have a greater mortality rate from breast cancer because they have more breast cancers. Even though the case fatality rate is better (more favorable prognosis), the increased incidence produces a net effect of an increase in mortality. This means that an entire population of current users of hormone therapy would have an increased risk of dying of breast cancer, even though the risk of dying from breast cancer once the disease is diagnosed in current users is lower (the case fatality rate). This is consistent with the 3-fold increase in estrogen receptor positive breast cancers in this study.
Danish case-control and cohort studies are noted for their ability to accurately obtain information from the national registries. Nevertheless, there are important limitations to this study. The most glaring problem is that not all causes of death could be verified and a death in a woman with a diagnosis of breast cancer was assumed to be a breast cancer death. (Stahlberg et al state that this problem would be balanced by similar numbers in the user and nonuser groups, but this is an assumption.) In this population of older women, you would expect deaths from other causes to outnumber deaths from breast cancer. This single assumption by the epidemiologists could have skewed the results.
In addition, the case fatality and breast cancer mortality calculations were adjusted only for age, the authors stating that a relatively small number of deaths precluded multiple adjusting. But there are so many factors that influence the risk of breast cancer, including age of menarche, age at menopause, age at first full-term pregnancy, parity, and age at diagnosis, use of mammography, presence of benign breast disease (specifically with atypical hyperplasia), body size, and alcohol intake. How can we know that the results did not reflect an imbalance in some of these factors? The results with all-cause mortality disagree with studies in the literature that indicate a reduction in overall mortality. These are reasons to be cautious regarding the authors’ conclusions.
But, this was a clever analysis and the results are disturbing. The conclusion is totally dependent on the finding of an increase in relative risk for breast cancer in the current users. It remains a puzzle why nothing shows up in the follow-up of past users. Does this indicate hormonal effects on pre-existing tumors? The data in this study cannot help us to answer this important question.
The Danish Nurse Cohort was established in 1993. A total 10,874 women were available for follow-up after excluding hysterectomized women and cases of breast cancer identified at baseline.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content