This site is intended for health professionals only

At the heart of general practice since 1960

HRT and cancer what are the risks?

Dr Richard Sullivan sets out the results of the latest research

Over the past year the public has been bombarded about the risks of HRT. Risks and benefits in relation to cancer, heart disease and osteoporosis have become blurred in the minds of even the most seasoned follower of this field. The result has been confusion in the minds of most of the public and many doctors. One thing is certain – the days of HRT as a panacea are over. But it still retains a role on the basis that a properly informed discussion about risks and benefits takes place between the woman and her doctor.

Breast cancer is one of the areas of greatest concern to women. In the UK some 41,300 new cases are diagnosed every year and, despite excellent improvements in five-year survival and a declining mortality rate, almost 13,000 women die of the disease every year.

We also know women consistently overestimate their lifetime risk of breast cancer, but that individual risk perceptions vary greatly depending on cultural background, prior experience of the disease and other factors.

Essentially there are two groups where risk-benefit assessments apply – healthy women with no history of breast cancer and women who have been treated for breast cancer.

No breast cancer history

For women without a history of the disease, the Million Women Study1 is the source of data on the risks of cancer with HRT. This large, well-designed observational study has allowed estimates of the number of extra cases of breast cancer occurring after five and 10 years of HRT use.

It is important to note these are estimates for additional cases: for example, the baseline incidence for breast and endometrial cancers for women who do not take HRT is 32 per 1,000 women (breast) and five per 1,000 women (endometrial). This is important as women need to understand the context of the additional risk.

No data exists for the number of additional cases of breast cancer for tibolone, but it is thought to lie somewhere between oestrogen-only and combined HRT.

Balanced against this is an estimate from another study, the Women's Health Initiative2, of a relatively much smaller reduction in the number of bowel cancers from 16 to 10 cases per 10,000 women (not per 1,000). It also reported no increase in risk of breast cancer for oestrogen-only HRT during the seven years of the research, but in contrast to the Million Women Study it was insufficiently powerful to detect the small number of additional cancers.

The Committee on Safety of Medicines has published clear guidance on EPINET for helping women make an informed choice about starting or continuing HRT with their doctors. Useful information can also be found on this site regarding cardiovascular risks.

History of breast cancer

The situation for women who have had

breast cancer is much more complex and uncertain. Around 60 per cent of premenopausal women who are treated for early breast cancer suffer premature ovarian failure. For many, HRT can provide substantial symptomatic relief.

But it is clear that exogenous hormones are strong initiators and/or promoters of breast cancer. To quantify the potential risk of HRT in causing breast cancer recurrences in these women a number of clinical trials were set up. Recently both Scandinavian (HABITS and Stockholm) and UK trials were suspended

on the advice of the Data Monitoring and Safety Committees after one of these trials (HABITS) began to show a statistically significant increase in the risk of recurrence of breast cancer with HRT.

But because these trials were terminated early the numbers recruited to date were small and in two of them no significant increase was found. In other words, it is a much more uncertain risk.

In essence these results tell us the following. The five-year risk of recurrence of treated early breast cancer is 20 in 100 without HRT; HRT may increase this to 30 in a 100. But the results are so uncertain that the actual risk may either be none – that is 20 in 100 – or as high as 50 in 100.

A further factor that needs to be considered is how this risk changes if women are taking tamoxifen or one of the new aromatase inhibitors (for example anastrozole). There is simply no data on this, although one might speculate that the risk might be lower. This second scenario is not one that many GPs will have to face, but it makes the point that in any discussions about risk with patients it is essential to communicate the uncertainty associated with numbers.


Balancing the risks of HRT compared with other causes is an inexact art but it helps to put the figures into some context (see box 2).

A sobering fact is that around 20 per cent of all deaths from all categories are due to smoking. Something like 60 per cent of cancer deaths could be avoided by eradicating smoking.

Like all medicines, HRT has risks associated with its use. Careful explanation of these risks, the certainty with which the research community hold the supporting data and comparisons with other avoidable causes of cancer are necessary to allow women to make an informed choice.

Richard Sullivan is head of clinical programmes, Cancer Research UK

2. HRT risks in context

Deaths in UK women aged 35-69

(% by cause)*

Cancer 47

Circulatory disease 24

Ischaemic heart disease 12

Respiratory system 8

Digestive system 7

Cerebrovascular disease 6

Pneumonia and influenza 4

Diseases of the nervous system 3

Accidents and assaults 3

Infectious disease 1

Potentially from HRT (cancer only) 0.8

*A single cause of death is not always specified

Source: Cancer Research UK


1 Million Women Study Collaborators. Breast cancer and hormone replacement therapy in the Million Women Study. Lancet 2003;362:419-27

2 Writing Group for the Women's Health Initiative Investigators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women.

JAMA 2002;288:321-33

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say