Getting perspective on MHT safety

7 minute read

MHT has a long history of expert disagreement around risk and safety. Here's a guide to what experts on both sides of the debate have to say on the latest controversy

In 2002, the now infamous Women’s Health Initiative (WHI) trial including over 16,000 postmenopausal women resulted in a global media storm.1

In the furore, the study results were often misunderstood and misrepresented. The end result was that MHT was subsequently demonised as an intolerable breast cancer risk and, worldwide rejected it as an option to treat menopausal symptoms. Its usage plummeted and has never returned to the level seen before the trial.2

Until recently, developments in therapy, a lot of slow release education, and the publishing of more representative longitudinal data on risk, have been responsible for a substantive recovery in confidence in the role and usage of MHT, even among GPs, many of whom, like their patients been scared off recommending MHT by the WHI paper.

Last month, a new study published in The Lancet,3 again around the risk of breast cancer,   re-ignited the debate around MHT and risk. But unlike 2002, the Lancet paper was not picked up in a big way by the consumer media, GPs were not stampeded by panicked patients, and following the paper’s publication experts have been critiquing and criticising the methodology and interpretation of these latest results. The debate is being played out without the hysteria of before in the specialist medical media.

The following summarises the key arguments of both sides, and the opinions of some experts on how the paper should or could be interpreted.

Methodology and key findings

The Lancet paper is based on a meta-analysis of data from 58 observational studies involving more than 600,000 postmenopausal women, of whom 108, 000  developed breast cancer between 1992 and 2018. More than half of these women who developed breast cancer(51%) had taken MHT, of which developed breast cancer.

The meta-analysis found that  there was a significantly heightened risk of developing breast cancer among postmenopausal women who had taken any form of  MHT other than topical vaginal oestrogen. The risk varied with time on MHT but can be summarised as all types of MHT, except vaginal oestrogen, increased the risk, including tibolone, with a relative risk of 1.57.

In women, the lifetime risk of developing breast cancer is generally accepted as being approximately one in eight. With the estimate of a 1.57 increase in relative risk with MHT, the additional number of women who will to develop this malignancy because of this treatment is likely to be in the hundreds of thousands, the Lancet paper suggests.

To provide more detailed perspective, Dr Stephen Birrell, from the Department of Medicine, University of Adelaide, summarised the results of the Lancet paper in a recent MJA article. He estimated that the absolute increase in risk for women taking MHT were:

Five years usage:

  • Continuous cMHT — 1 in 50 develop breast cancer (BC)
  • Intermittent cMHT — 1 in 70 develop BC
  • Oestrogen only MHT — 1 in 200 develop BC

     Ten years usage:

  • Continuous cMHT — 1 in 25 develop BC
  • Intermittent cMHT — 1 in 35 develop BC
  • Oestrogen only MHT — 1 in 100 develop  BC.

However, some experts have urged caution accepting these findings. They point out :

  • Observational studies, especially pooled ones, can create poor data outcomes because they will always include a degree of  “unquantifiable confounding effects”, which can’t be easily corrected for. Even though large randomised controlled studies take a long time and are expensive,they usually produce a much higher standard of interpretable data
  • That the study is unidimensional, that is, it only looked at one aspect of the data, that being breast cancer risk, rather than considering other potentially important risks or mitigating factors, such as the effect of MHT on bone density and fracture risk, reduced CVD and alleviation of what can sometimes be very debilitating menopausal symptoms.
  • The study’s median diagnosis year is 1999, in which case, it is not taking into account important changes to therapy formulations and prescribing practices that have occurred in the last few years.
  • There is selective reporting of data in some parts of the paper. For example, the paper suggests that postmenopausal women under the age of 45 who take MHT increase their risk of breast cancer, which is true, but the increase in risk is the same as if the women had developed MHT in her 50s.
  • The study only contemplates risk of breast cancer, not actual mortality rates, which can vary significantly.

Supporters of the paper, acknowledge that the study is based on observational data, and that it is indeed a unidimensional study, just on breast cancer risk. But they say critics should not underestimate  the potential importance observational data  or what could be interpreted as “real world evidence”.

In his MJA reply to a piece written by endocrinologist Professor Sue Davis, from Monash University, Dr Birrell says “The data presented in the Lancet article are extremely valuable and concur with what most researchers in hormone-dependent breast cancer thought was happening. We have not listened to epidemiologists in the past, much to our chagrin.” 5

Dr Birrell’s reference to epidemiologists underlines the fact that the Lancet paper data has been collated after the fact from various population studies, published and unpublished, rather than derived from randomised controlled studies, which is generally acknowledged as the gold standard of research.

Supporters of the paper go on to argue that the consistency of the findings for nearly all forms of MHT argue for the conclusions to be valid. They summarise the absolute increase in risk for five and ten year usage for continuous combined MHT, intermittent combined MHT and oestrogen only MHT. At the extreme end of combined MHT conducted for 10 years they say that MHT can lead to a one in 25 relative change of developing breast cancer.

The Lancet article concludes that: “In Western countries there have been about 20 million breast cancers diagnosed since 1990, of which about one million would have been caused by MHT use.”3

Weighing up both sides of debate

In his MJA piece, Dr Birrell does not address a key concern with the Lancet study findingswhich ProfessorDavis raises.  She points out that the median diagnosis of the study period is 1999, 20 years ago, and therefore the study does not take into consideration change that have occurred in terms of MHT delivery methods, formulations and treatment schedules which are significant. Professor Davis says that randomised controlled studies are needed before the sort of conclusions reached in this paper could be justified.5

Professor Davis also says that the paper also does not consider a number of other important aspects related to the health of post-menopausal women,  such as the relative risk of breast cancer associated with MHT compared with the risk of morbidity associated with osteoporosis or CVD which will be lower in those women taking MHT.

She points out as well that findings like those in the Lancet study must always be placed in the context of “the whole women”.

As an example she says that “we know that for postmenopausal women aged less than 55 years, 75% have vasomotor systems, which are severe for nearly 30% of women. The negative impact of vasomotor symptoms on wellbeing is almost as severe as housing insecurity.”

The supporters and detractors of Lancet paper tend to line up as population health and breast cancer researchers versus women’s health and menopause clinicians and researchers.

The latter say that one paper based on observational data and focussed on just one aspect of MHT,  is too narrowly defined to be practical at the treatment level. At least until better data on the most recent therapies can be obtained.

Professor Rod Baber, of the University of Sydney Department of Medicine, and a past  president of the International Menopause Society told The Medical Republic recently that his advice to his patients, if they had come across this paper was to not look at the findings in isolation.6

“It is most likely that that breast cancer risks are exaggerated in this paper and that modern regimens using natural hormones as a part of MHT do not have the same risks,” he said.

“But of course, each women should discuss her situation and her individual risks with her own doctor”.


  1. JAMA ,2002;288(3):321-333
  2. Langer, RD, Climacteric, 2017, 20 (2)
  3. Lancet, 2019;394:1159-68
  4. Davis, S MJA, Insight Sept 9, 2019
  5. Birrell, S, MJA Insight, Sept 19, 2019
  6. The Medical Republic, Sept 16, 2

End of content

No more pages to load

Log In Register ×