select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to study looking at HRT use and risk of breast cancer

Research, published in The Lancet, reports an association between Hormone Replacement Therapy (HRT) for the treatment of menopause and risk of breast cancer.

This Round Up accompanies an SMC Briefing.

A Before The Headlines also accompanies this Round Up


Prof Janice Rymer, Consultant Gynaecologist and Vice President of the Royal College of Obstetricians and Gynaecologists, said:

“Women and doctors should be reassured that the findings of this study do not add anything new in terms of the effects of hormone replacement therapy. Research shows that, for most women, HRT helps to manage menopausal symptoms and is safe.

“Women must be informed of the small increase in risk of breast cancer so they can weigh these up against the benefits that they may have from taking HRT. Every woman experiences the menopause differently and symptoms vary. These can be extremely debilitating and have a significant impact on a woman’s physical and psychological health, career, social life and relationships. Unfortunately, many women are still suffering in silence and are reluctant to seek advice and support due to concerns around the risks of breast cancer associated with HRT.

“The findings from this research should be helpful to both women and doctors, particularly around when considering whether to start hormone therapy, for how long and which preparation they could take – whether it includes oestrogen and progestagen combined, or oestrogen alone. These findings should not put women off taking HRT if the benefits – such as protection of bones and decrease in cardiovascular risk – outweigh the risks.

“To put the risk into context, a woman has greater risk of developing breast cancer if she is overweight or obese compared to taking HRT. Women must be aware of the effect of obesity and alcohol which increase the risk of breast cancer and modifies the additional risk of HRT.”


Mr Haitham Hamoda, Consultant Gynaecologist and Chair of the British Menopause Society, said:

“We welcome this further data on the incidence of breast cancer which will help us counsel our patients and women in general better. This paper provides further data on the impact of estrogen and progesterone combined and estrogen that adds more detail to that we have already gathered from overall assessment of the literature. It also includes some new information on different types of progestogen that surprisingly were found to not vary as much as had been thought.

“Of particular interest though is the impact of estrogen, and different regimens of combined HRT on obese women, where the former is found to have little effect but the increase with the latter is greatest with continuous combined HRT. However, in practice this must be weighed against the rapidly rising incidence of endometrial cancer which is significantly decreased by the continuous combined preparations.

“The overall findings from this study are in keeping with the NICE menopause guideline recommendations which show a small increase in risk of breast cancer with HRT. Women must be informed of the data on breast cancer risk with HRT to help them make an informed decision. This should also be considered in comparison to the risk of breast cancer with other lifestyle factors such as alcohol intake and obesity which have been shown to be associated with a higher risk compared to that with HRT. This should also be taken in the context of the overall benefits obtained from using HRT including symptom control and improving quality of life as well as considering the bone and cardiovascular benefits associated with HRT use.’’


Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is, in my view, a very careful, thorough, excellent piece of research. That’s hardly surprising, given the worldwide extent of the collaboration that produced it, and the many years that several of the leading researchers have spent examining these issues. The overall finding, that using menopausal hormone therapy (MHT), except vaginal oestrogens, is associated with an increased risk of breast cancer, isn’t new, but what is new is the extensive detail on many aspects of this association, in quantifying how any increase in risk depends on the type of MHT, the length of time for which a woman used MHT, and the length of time since she stopped. 

“The main findings come from a considerable number of observational studies. Because the studies are observational, there must always remain some doubt about how far the association between MHT use and breast cancer risk is one of cause and effect. That’s because there are differences between post-menopausal women who use or do not use MHT, or between women with different patterns of use, apart from the differences in their MHT use, and these other differences may be the actual cause, or part of the cause, of any increased risk. However, the researchers give several reasons why they consider that the associations are probably largely causal, and my own feeling is that there is enough evidence from several sources to be reasonably sure that MHT is an important part of the cause.

“However, it’s worth mentioning that there are some issues with the evidence. First, the researchers decided to base their main conclusions only on data from prospective studies, where women are followed up for long periods of time, with their MHT use and any breast cancer cases being recorded. There have also been many retrospective studies of these associations, where women are investigated in later life when they will, or will not, have already had a breast cancer diagnosis, and then their past MHT use is investigated. Not using the data from retrospective studies might sound like throwing away potentially useful data. However, the detailed figures on both prospective and retrospective studies are available in the research reports (including the appendix) anyway. Details of the strength of the association are different between retrospective and prospective studies, with the MHT use generally being less strongly associated with breast cancer in retrospective than prospective studies, particularly in retrospective studies carried out in North America. But, generally speaking, retrospective studies are more likely to be subject to several forms of bias than are prospective studies. The researchers give their reasons why these biases may be a particular problem on retrospective studies of this association, and I find these convincing, so I am happy with their decision to base the main results on the prospective studies only.

“Then, there are results from a limited number of women who took part in randomised clinical trials of MHT, mostly 10 or more years ago. Though the numbers involved are small relative to the observational studies, randomised trials generally avoid biases due to confounders (differences between women other than in MHT use). On combined oestrogen and progestagen therapy, the randomised studies produced results on breast cancer risk that are pretty similar to the results from prospective studies anyway, but on oestrogen-only therapy, the randomised trials in fact found evidence of a slight protective effect. The researchers on this new study give reasons for this difference, and state why they think they may be misleading in relation to the usual patterns of MHT use, but some may not agree with this reasoning. What this does demonstrate is that these issues are complicated, and we don’t know everything yet.

“It’s good to see that the research paper and the press release give some clear indications of how much the risk of breast cancer will be increased by different types of MHT, assuming that the associations are indeed largely causal and that the risk estimates have not been much affected by remaining biases in the observational data. I’ve nothing to add to these numbers.  The increased risks are not by any means huge anyway, but, in my view, very much worth taking into consideration. There’s a balance to be struck; MHT relieves many menopausal symptoms and may reduce the risk of other conditions such as osteoporosis. I’m no clinician, and I’m not a woman so have no direct experience of this, but I’d definitely recommend that anyone who has concerns should talk things through with health professionals before changing any MRT they are using.”


Prof Stephen Evans, Professor of Pharmacoepidemiology, London School of Hygiene & Tropical Medicine (LSHTM), said:

“This is a report of a mammoth task to bring together the individual data from 58 studies involving 568859 women from many different countries.

“It shows some findings that are generally accepted now and some that are new. The accepted findings are that MHT can cause an increased occurrence of breast cancer in users who take it for more than 5 years and there is a higher risk in those who take it for more than 10 years. There is already good evidence that, while oestrogen combined with a progestagen reduces the risk of endometrial cancer compared with oestrogen-only MHT, the combination has a higher risk than oestrogen only MHT for breast cancer which outweighs the benefit of endometrial cancer risk reduction.

“The new findings are principally 1) that even in years 1 to 4 of use, MHT confers a risk of breast cancer though this risk is smaller than that with longer use; 2) the increased risk that is strongly dependent on duration of use has a “carry-over” effect which also depends on duration of use and lasting possibly 10 or 15 years after stopping; 3) there are clear differences between the different types of MHT, with topical vaginal oestrogens having the lowest risk and possibly conferring no extra risk; 4) the overall implications of both the evidence of risk prior to five years of use and the extent of the “carry-over” after use has ceased suggests that the overall role of MHT in breast cancer incidence is greater than thought previously.

“All these remarks assume that it is MHT that is the causal factor. The authors have used data on individuals to try and ensure that users and non-users were as similar as possible given their data. Similarity would only be guaranteed using randomised allocation to MHT. Their data are compatible with the randomised data which is much smaller in extent and has only been able to examine a very much smaller number of women starting MHT at the time of the menopause. The authors essentially note limitations of observational data; that the design of the studies cannot be as sure of the results as they could be if randomised. The great consistency shown in the findings do make it much more likely than not that these effects are causal.

“Some of the data will have related to products no longer in current use, but the evidence is consistent about what differences make a difference (mainly addition of a progestagen in combined MHT, and longer duration of use).

“The authors do a very good job in translating their results into numbers of practical use. Their interpretation is cautious but clear. These results should not be used to cause alarm among women, but there is no doubt that they should follow the advice given by the MHRA in the UK and the FDA in the US that MHT be used for the shortest time that it is needed. MHT does offer real benefits for menopausal symptoms, but its use beyond one year seems to confer a steadily increasing risk of breast cancer with increasing years of use.

“Breast cancer is in many cases a very treatable disease, but the accompanying research letter, based only on data from the UK Million Women Study, shows that the same pattern of risk – increases with increasing duration of use and addition of a progestogen- also applies to deaths from breast cancer which are rarer than the incidence of breast cancer.

“This is a “tour de force” in what has been done and the way it has been done – the findings cannot be dismissed.”


Prof Paul Pharoah, Professor of Cancer epidemiology, University of Cambridge, said:

“This is a very large study combining data from a large number of different studies.  It has been carefully conducted and analysed and clearly reported.  There are no major, truly new findings in that we have known for many years that hormone replacement therapy (HRT) after the menopause is associated with an increased risk of breast cancer.  What this study has done is to refine some of the estimates of risk associated with different types of HRT as well as the risks associated with using HRT for different lengths of time and the risks after stopping HRT.

“There is little doubt that the reported associations are causal – although these data are from observational studies that are by their nature correlative, there is good evidence from clinical trials that the association is causal.

The IF behind these results is that noted by the authors themselves in the summary. The reason for that caveat is that while we can be sure that the association is a causal one, we cannot be sure that the sizes of the risks reported have not been affected by confounding.  A consequence of this is that the absolute numbers may be slightly over-estimated. However, it seems unlikely that any over-estimation is very large.

“While the authors do provide some estimates of absolute risks it is worth considering the implications for a typical woman considering hormone replacement.  Assume a 50 year old woman in the UK who has started having symptoms of the menopause is considering taking combined estrogen and progestogen HRT for five years and no more.  If 1000 such women take HRT for five years, 66 would be expected to get breast cancer by the age of 70.  If these women did not take HRT then 58 would be expected to get breast cancer.  A difference of 8 in 1000.  Most of that increase happens in the first five years when the woman is taking HRT.”


Type and timing of menopausal hormone therapy and breast cancer risk: individual participant meta-analysis of the worldwide epidemiological evidenceby Beral et al. was published in The Lancet at 23:30 UK time on Thursday 29 August. 

DOI: 10.1016/S0140-6736(19)31709-X


Declared interests

Prof Kevin McConway: “Prof McConway is a member of the SMC Advisory Committee, but his quote above is in his capacity as a professional statistician.”

Prof Stephen Evans: “I was not in any way involved in the study, but I was an independent expert present when the results of this study were presented to the UK MHRA very recently. I was formally an independent expert member of the European Medicines Safety Committee (2012-2018).”

Prof Paul Pharoah: I have previously done consultation for Kay Scholer as an expert in cancer epidemiology in relation to trials about hormone replacement therapy and cancer.

None others received.

in this section

filter RoundUps by year

search by tag