There is a known association between the use of menopausal hormone therapy (MHT) and breast cancer. An article by Beral et al for the Collaborative Group on Hormonal Factors in Breast Cancer (Lancet 2019 doi: 10.1016/S0140-6736(19)31709-X) attempts to quantify the risk for women in developed countries for the various forms of therapy (estrogen alone and in combination with progesterone) for different durations of use and the timing of commencement in relation to each woman’s menopause transition. The mode of hormone delivery and the woman’s weight also need to be taken into account because being over-weight or obese is a risk factor for breast cancer, probably due to the endogenous estrogen she produces.
Given these variables, the following data are published for women initiating MHT soon after their menopause and calculating their risk of developing breast cancer over the next 20 years:-
|MHT use for 5 years||Breast cancer risk|
|Estrogen only use||6.8%|
|Intermittent progesterone use||7.7%|
|Continuous combined use||8.3%|
Other points made in the article about breast cancer risk in relation to MHT risk are:
- “For 10 years of use, the 20-year increases in incidence would be about twice as great as for 5 years of use.”
- The risk was greater for estrogen receptor positive than estrogen receptor negative disease
- Oral or transdermal administration risks were not substantially different
- Progestagenic constituents had no different effects
- There was little excess risk after one year of use
- Topical genital estrogen application appeared to be risk free
Editorial opinion – This is an important article published in one of the world’s foremost journals and presents information about MHT and the risks of breast cancer. As such it deserves the attention of primary care and specialist healthcare providers as many women will seek guidance about the findings and may ask for interpretation of the conclusions.
In my opinion there are questions about how this paper should be interpreted.
Type of research – the data analysed was epidemiological evidence that brought together associations between MHT and breast cancer. It was not a review of randomised trials and the authors made use of published and unpublished findings. The approaches used do not delegitimise the research but do place limitations on the outcomes concluded.
The study design is complex. The authors worked from 58 studies and had to take into account the woman’s age and menopausal status at commencement of MHT; any exclusion criteria which may have biased selection; the type of MHT and its mode of delivery; personal factors (for example BMI) and duration of use. To quote the editorial commenting on the article “The complexity of the study design makes it difficult to appraise the results and most of us will take the results on face value.” (Kotsopoulos Lancet 2019.doi: 10.1016/S0140-6736(19)31901-4).
I have no problem believing the calculations but I do have queries about the interpretations of the findings in the overall context of a menopausal woman’s health, wellbeing and longevity.
Interpretations of results – It is unusual for a paper to start the interpretation of the results with the words “If these associations are largely causal, then …….”. The authors do not quote a line of reasoning whereby a causal association is posited between MHT and breast cancer risk. There may be a “causal association”, and I suspect there is one but it may be far more complex than the intervention/outcome calculation they present. I can imagine that genetic susceptibility will need to be taken into account to explain the association which they postulate as largely causal.
Another unusual presentation is the way the findings are framed, particularly with the absolute increase in risk from 6.3% (no use) to 8.3% (combined MHT use over 5 years).
The authors choose to describe this as a 1 in 50 additional women will develop breast cancer if they use MHT.
There are different ways to present the risk – for example:
The 6.3% risk of breast cancer with no MHT is a 1 in 16 chance over 20 years.
The 8.3% risk of breast cancer with combined MHT is a 1 in 12 chance over 20 years.
Benefits and harms balance. This article describes the link between MHT and breast cancer from the statistical viewpoint. As clinicians, we need to assess the patient holistically from her personal and family perspective, plus her hopes and fears at the time of her menopause transition.
The data presented here need assimilation into the “big picture” which includes her expectations, lifestyle and quality of life. Other considerations are the effects of MHT on cardiovascular health and disease, bone and skeletal integrity, mental well-being from symptom relief, sexual problems averted and cutaneous improvements which need to be weighed against the negative effects of breast cancer and thrombotic risks. Ultimately mortality rates need to be taken into account – a topic not addressed in this paper.
The same authors did subsequently publish data on MHT and breast cancer mortality (Beral et al Lancet 2019 doi: 10.1016/S0140-6736(19)32033-1). From the Million Woman Study it is stated that 0.8% women died from breast cancer (never users 0.74% and combined users 0.85%).
If not placed in context, these mortality rates could give the impression that MHT increases a woman’s chance of dying, but this would again fail to take into account the “big picture.” Although MHT does increase mortality rates from breast cancer, the actual figures are low, and the difference between never users and current combined MHT users is 0.1%. Past users with less than 5 years of prior use did not have excess mortality.
However, breast cancer is only one factor in the array of MHT benefits or harms – and on balance overall mortality rates are no different if MHT is used or MHT is not used.
(Rossouw et al JAMA 2007;297:1465-77; Manson et al JAMA 2017;318:927-38; Bae & Yoon J Meno Med 2018;24:139-42; Holm et al BJOG 2019;126:55-63).
- The data presented are based on epidemiological evidence not randomised trials.
- Much of the data comes from the “Million Women Study” which is not accepted in its entirety by all experts in the field and the WHI research which itself represented a particular group of mainly older women using products that would not be prescribed today.
- They do not give a composite picture of MHT benefits and harms, or of the influence of MHT on overall mortality rates.
- The figures have been collected over decades and are not materially different to those published at the turn of the century in the WHI study.
- It is also a summation of the data at the present time, meanwhile breast cancer rates are falling and survival rates are rising – rapidly, with the risk of dying from the disease in developed countries declining by nearly a half in the last 3 decades.
- The next set of data on MHT may give a very different picture.
- In the opinion of this reviewer, the statistics drawn together in this paper do not provide sufficient evidence for any change in MHT prescribing policies.
Finally there are questions which remain unanswered:
- How does the addition of a progestagen to MHT raise the risk of breast cancer?
- Obese women produce considerable amounts of endogenous estradiol, and are at higher risk of breast cancer than women of normal BMI. Why do they not increase their risk if they are treated with exogenous hormones?
- The WHI data for estrogen-only MHT in hysterectomised women showed a decreased risk of breast cancer. The statistics presented here show an increased risk for estrogen-only MHT. Is there a timing hypothesis for breast cancer susceptibility or how else can these anomalies be explained?
The RCOG and the British Menopause Society put out a statement in response to this paper which can be found at: https://thebms.org.uk/2019/08/joint-rcog-bms-statement-in-response-to-the-lancet-study-on-hrt-use-and-breast-cancer-risk/
All rights reserved © September 2019
This article comes with the compliments of the editor of JASS (Athol Kent) who wishes to draw your attention to a special offer that is soon to be offered to RANZCOG members concerning subscriptions to JASS.
The offer is a reduced subscription cost from US$100 per annum to US$80 for RANZCOG members for the first year. Should you, as an AMS member, wish to avail yourself of this offer straight away, you are welcome to access the JASS website www.getjass.com and apply by using the RANZCOG link.