Critically Appraised Topic

Appraiser: Eric W. Terman

Date: October 28, 1997

Question: Is the risk of breast cancer increased in women taking hormone replacement therapy (HRT)?

Clinical Bottom Lines:

1. There was a slightly higher risk of having breast cancer in the HRT group.

2. The relative risk of having breast cancer is 1.023 per year if the patient takes HRT. Over ten years this is 1.25. If you accept the papers findings.

3. Although there may be increase in risk, its impact on death is not known, particularly since the cancers were more likely to be caught at an early stage.

4. Unless there are large contra-indications, this paper would not influence me to prescribe less HRT.

5 I’ll wait for the women’s health initiative trial results, which is an RCT, before changing my practice.

 

Methods/Evidence:

1. A meta-analysis which included not only RCTs, but also retrospective trials

2. The study used patients with cancer as the subjects of the study, with those without cancer as a control..

Comments:

1. This is a large and fairly well constructed meta-analysis.

2. There are two discordant populations of papers: the RCTs and the retrospective. The prospective trials do not show the same risk. The cumulative data is less convincing in this light.

3. There is no way to know if the supposed increase in risk has any impact on life expectancy.

4. We think that HRT is good for preventing osteoporosis and maybe CAD. Also it can alleviate the symptoms of menopause. For these reasons it is not a therapy that should be given up easily. It’s a question of balancing harms.

5. There is potential for lead time bias in this study in that the cancers found in the HRT group were a lower stage suggesting that they were found earlier, which could be a function of better medical attention than of HRT.

6. With as many patients included in the study as they had it is difficult not to find significance.

Reference:

Beral V, and the Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer, Lancet. 1997 350(October 11):1047-1059.