Hormone therapy for menopausal symptomsBMJ 2012; 344 doi: http://dx.doi.org/10.1136/bmj.e815 (Published 08 February 2012) Cite this as: BMJ 2012;344:e815
- Helen Roberts, associate professor women’s health
A recently published and much publicised paper by Shapiro and colleagues, the last in a series of four, evaluated the effects of hormone therapy on the risk of breast cancer.1 The authors of the four review articles applied epidemiological principles to the findings of two randomised placebo controlled studies from the Women’s Health Initiative (WHI; 27 347 women) and two observational studies—the Collaborative Reanalysis (53 865 women) and the Million Women Study (MWS). Shapiro and colleagues concluded in their fourth paper that the MWS had design defects, that it contained multiple biases, and that its findings were thus not robust enough to show that hormone therapy increased the risk of breast cancer.
All observational studies are inherently biased because subjects are not randomly assigned to treatment or control. Adjustment for confounders and careful design of observational studies help to reduce bias. However, because there is no independent variable, such studies can tell us only about association not causation.
The MWS was published in the Lancet in August 2003,2 and a flurry of letters was published in a print issue later that year, many of which raised the same concerns about bias recently highlighted by Shapiro and colleagues. The authors of the MWS replied,3 and the sequence continued over the years: concerns about the believability4 and worry about uncritical acceptance of the MWS data,5 followed by more responses from the authors of the MWS.6
Although observational studies add to our knowledge, they cannot replace randomised trials. Analysis of data from the WHI found a decreased risk of early stage breast cancer and ductal carcinoma in women randomised to receive oestrogen only.7 Subgroup analysis showed that the reduction in breast cancer was statistically significant only for women who complied with treatment, had not used hormones before study entry, and had started oestrogen more than five years after the menopause. No benefit was seen for women who started oestrogen treatment at the time of the menopause.8 This “gap time concept” of a reduced risk of breast cancer only if oestrogen is started late contrasts with the gap time hypothesis of a potential decrease in cardiovascular disease if oestrogen is started early.9 Shapiro and colleagues’ evaluation of the WHI findings concluded that treatment with oestrogen only does not increase the risk of breast cancer and may even reduce it, although the last possibility is based on statistically borderline evidence.10Fewer breast cancers were diagnosed in the first four years of follow-up in women in the WHI who were randomised to receive combined oestrogen and progestin. This is thought to be the result of hormonally induced increased density of breast tissue, which leads to delayed mammographic diagnosis of cancer. After four years, breast cancer rates were higher in women on combined hormone therapy, and diagnosed cancers were larger and more advanced.11 Women who had used hormone treatment before joining the study were at higher risk of breast cancer than those who were treatment naive, but a significant increasing trend in risk of breast cancer over time was seen for this last group.11
Pretreatment clinical or laboratory characteristics may be discovered that will help identify women who, because of genetic predisposition, are at increased risk of adverse events with hormone treatment. The role of progestogens in the development of breast cancer needs to be clarified. A recent review suggests that women who use progesterone or dydrogesterone instead of progestogens have a lower risk of breast cancer.12 Because treatment with oestrogen alone seems to be associated with lower risk, local delivery of progestogen to the endometrium is a potential option. However, a recent case-control study found increased odds of breast cancer in women with a levonorgestrel intrauterine system.13 Other strategies to deliver combination treatment are being investigated; oestrogen combined with selective oestrogen receptor modulators has the potential to improve symptoms without affecting the breast and has positive effects on lipids and bone.14
How should we advise women while we wait for better treatment solutions? The second article in Shapiro and colleagues’ series concluded that potential biases in the combined hormone treatment arm of WHI reduced the robustness of an association between treatment and breast cancer.15 However, they acknowledged that the use of oestrogen plus a progestogen could possibly increase the risk of breast cancer. This is how most clinicians would frame the issue when discussing the risks of using combined hormone treatment. The increased risk of breast cancer associated with use of combined oestrogen-progestogen (hazard ratio 1.24) is similar to risks conferred by delayed menopause or moderate use of alcohol. Although an increase in risk of this size may be important for public health, individual women may not consider it enough to change their minds about using hormone treatment.
Women should continue with regular breast screening, and those with dense breast tissue may need more frequent screening. The primary aim of the WHI was to see if the use of hormone treatment decreased heart disease, as observational studies had found. The study was not designed to determine the risks of hormone use for symptoms in early menopause. It was not powered for subgroup analysis in the 50-59 year age group, and numbers of adverse events were small. Healthy women have a low absolute risk of adverse events, whether they use hormone treatment during early menopause or not.
Cite this as: BMJ 2012;344:e815
Competing interests: The author has completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; not externally peer reviewed.