Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

HRT and breast cancer

Study
Results
Comment

Bandolier 24 looked at breast cancer risk from the US Nurses' study involving about 120,000 women. That concluded that increasing risk of breast cancer was associated with current use of HRT, longer use of HRT, and perhaps was higher with some HRT preparations than others. Use for more than five years was associated with a relative risk of about 1.5. We now have information from over a million women in the UK, which gives much harder evidence on which to base judgements [1].

Study

Over five years to early 2001 NHS breast screening centres participating included a study questionnaire with a letter of invitation sent two to six weeks before routine mammography. The questionnaire contained questions on various factors, including use of HRT (what, when, for how long) and menstrual history.

Participants were flagged on NHS registers so that cancer registrations and deaths could be identified. Women previously on the cancer register were excluded, except for non-melanoma skin cancer.

The relative risk of breast cancer and breast cancer death was calculated, stratified by age, time since menopause, parity, age at first birth, family history, BMI, geographical area and deprivation index.

Results

There were 1,084,100 women in the study. At recruitment the average age was 56 years. Follow up was 2.6 years of for breast cancer diagnosis and 4.1 years for breast cancer mortality. There were 9,364 diagnoses of invasive breast cancer and 637 deaths from breast cancer. Most women were postmenopausal, and the main analysis was on 830,000 postmenopausal women.

Table 1 shows the main results, in terms of the number of diagnoses of breast cancer or breast cancer deaths per 1,000 women, the relative risk, and the number of users of HRT needed to generate an extra case of breast cancer diagnosis or breast cancer death.


Table 1: HRT use and breast cancer incidence


Use of HRT at baseline
Number per 1,000 women
Relative risk
Number of users for each extra case
Breast cancer diagnosis
Never users
7.4
1.0
Past users
7.0
1.0
not applicable
All current users
11.2
1.7
260
Current use of
Oestrogen only
8.6
1.3
1600
Oestrogen-progestagen
13.5
2.0
160
Tibolone
10.1
1.5
360
Other
9.7
1.4
420
Oestrogen use for
less than 1 year
5.6
0.8
not applicable
1-4 years
8.5
1.3
900
5-9 years
8.8
1.3
700
10 years or more
8.7
1.4
750
Oestrogen-progestagen use for
less than 1 year
9.9
1.5
390
1-4 years
11.8
1.7
220
5-9 years
14.9
2.2
130
10 years or more
15.3
2.3
125
Breast cancer deaths
Never users
0.61
1.0
Past users
0.59
1.1
not applicable
Current users
0.67
1.2
16,000
The relative risk was adjusted for age, time since menopause, parity, age at first birth, family history, BMI, geographical area and deprivation index.

Current use of HRT, but not past use, was associated with increased numbers of cases and deaths. Past use, even for less than a year, was associated with no increased risk. Over 2.6 years there was one extra diagnosis for every 260 women currently using HRT. Over 4.1 years there was one extra breast cancer death for every 16,000 women currently using HRT.

HRT products containing both oestrogen and progestagen were associated with the highest risk of breast cancer diagnosis, particularly over a longer time. Only 160 women needed to use oestrogen and progestagen HRT for there to be one extra case. Oestrogen-only HRT preparations, and tibolone, and other preparations, were also associated with higher risk of breast cancer diagnosis, but at a lower level than with oestrogen and progestagen (Table 1).

We are also given calculations, of use to individual women and from the societal perspective. For the individual woman, the calculation for use of oestrogen only or oestrogen-progestagen for five or 10 years from age 50 results in about one to 19 additional cases of breast cancer per 1,000 women, depending on preparation and duration of use (Table 2).


Table 2: Excess risk by age 65 for women using HRT around the menopause - by type and duration


HRT use
Years of HRT use from age 50
Total cases of breast cancer by age 65, per 1000 women
Excess cases of breast cancer by age 65, per 1000 women
None
0
50
0
Oestrogen only
5
51
1
Oestrogen only
10
55
5
Oestrogen-progestagen
5
56
6
Oestrogen-progestagen
10
69
19

For society, the calculation is that on the assumption that 25% of women aged 50-65 years were current users of HRT during the past decade, then that would have resulted in 20,000 additional cancers because of HRT use. Three-quarters would have come from use of oestrogen-progestagen preparations.

Comment

This is a marvellous study. It gives all of us information on which we can base advice or decisions. Brief use of some HRT preparations at the menopause results in very low increased risk, but some women will find even that additional low risk more than negligible. Other preparations used for longer will give a higher risk, but that increased risk will be acceptable to some women because of benefits from HRT use.

Reference:

  1. V Beral and many others. Breast cancer and hormone-replacement therapy in the Million Woman Study. Lancet 2003 362: 419-427.

previous or next story