"A drug prescribed to combat brittle bones has been shown to prevent invasive breast cancer", The Times reported. The newspaper said a study has found that raloxifene - a drug used to prevent and treat osteoporosis - reduces the risk of developing invasive breast cancers by more than 50%. The drug works by binding to oestrogen receptors in the body, and by doing this it could be preventing some of the effects of oestrogen "that spur cancer growth".

Previous research has suggested that raloxifene could potentially reduce the occurrence of oestrogen-receptor positive breast cancer, and this large study provides supportive evidence for this. However, the actual role that the drug might play in the prevention of this type of breast cancer is uncertain.

It is necessary to point out that the women taking raloxifene were more likely to suffer blood clots and fatal strokes compared with those taking a placebo. To balance the benefits with any potential harms such as these, it is important to know the absolute number of people who would benefit from treatment. The Daily Telegraph correctly quotes the study as stating that there was a reduction of 1.2 cases of hormone-sensitive breast cancer per thousand women treated for a year. The number of fatal strokes or blood clots is not reported in the journal article.

Where did the story come from?

Dr Deborah Grady and colleagues from the University of California, San Francisco VA Medical Center, Imperial College, London, and other institutions across the US carried out the research. The study was funded by Eli Lilly and Company, Indianapolis. The study was published in the (peer-reviewed ) Journal of the National Cancer Institute.

What kind of scientific study was this?

The original study was a double blind randomised controlled trial designed to investigate whether raloxifene reduced the risk of coronary heart disease in postmenopausal women. This latest report provided additional data from the trial, including the effects of the drug on the risk of breast cancer.

The study enrolled 10,101 women between June 1998 and August 2000, and was carried out in 177 sites in 26 different countries. The women were all postmenopausal and either had documented coronary heart disease (CHD) or where believed to be at an increased risk of CHD due to factors such as age, diabetes, high blood pressure, high cholesterol or smoking. The researchers excluded any women who had suspected breast cancer or prior history of breast cancer. Other reasons for exclusion included if they were at particularly high cardiovascular risk after having had a recent heart attack, heart failure, or bypass graft, or had suffered other serious medical illnesses such as liver or kidney disease. They also excluded women who had recently used any hormone replacement tablets or patches.

Breast cancer risk was established at the start of the study by asking about known risk factors such as family history, number of children, age when periods started, and age at menopause. A breast examination was also performed, and the women were only included if they had results of a mammogram in the year before the study.

The women were randomly assigned to receive either 60mg raloxifene daily (5044 women) or an identical inactive placebo drug (5057). All participants and investigators were unaware which treatment they were receiving. The women were treated and followed for an average period of five and a half years. During this time they were contacted twice a year and asked about their adherence to the drug, any adverse effects of treatment, and any outcomes noted. Breast examinations and mammograms were given every two years.

Any women developing cancer received full care and assessment by an oncologist who was also unaware of which treatment the women had been taking. The oncologist considered cancer type, size, invasiveness and stage of the cancer, and whether it was oestrogen-receptor positive or negative. Time to first breast cancer was the main outcome that the researchers considered in their statistical analysis.

What were the results of the study?

There were no differences between the raloxifene and placebo groups at the start of the study in terms of the women's characteristics or the presence of risk factors for breast cancer. In both groups, 80% of women completed the study, and there were no differences in the uptake of follow-up between the groups in terms of repeat breast examinations or mammograms.

In the placebo group, 76 women were diagnosed with breast cancer (at a rate of 0.29% per year) compared with 52 women in the raloxifene group (at a rate of 0.20% per year). Raloxifene significantly reduced the risk of developing breast cancer by one third compared to placebo. When the researchers split the cases of breast cancer into invasive (86% of cases) and non-invasive, the reduction in risk from taking raloxifene compared to placebo was a significant 44%; however, there was no significant difference seen between the groups in the much smaller proportion of women with non-invasive cancer. Most of the women with invasive breast cancer were oestrogen receptor positive (73%), and in these women, taking raloxifene gave a significant 45% reduced risk of developing invasive breast cancer compared with women taking placebo.

Taking raloxifene gave no benefit to the smaller number of women with oestrogen-receptor negative cancers. Raloxifene was no different from placebo in terms of histological type of tumour, size, stage or grade of tumour, or whether there was lymph node involvement. When the women were split up into different categories of risk factors (e.g. age, number of children, family history) the effect of raloxifene gave variable results, with a general trend towards reduced risk for those taking raloxifene in all groups of women. However, some results were statistically significant and others not.

What interpretations did the researchers draw from these results?

The researchers concluded that raloxifene reduces the risk of invasive oestrogen-receptor positive breast cancer in postmenopausal women regardless of any underlying risk factors that they have.

What does the NHS Knowledge Service make of this study?

This large and well-conducted study corroborates previous reports that raloxifene reduces the risk of invasive oestrogen-receptor positive breast cancer in postmenopausal women. However, the potential role of this treatment in preventing this type of breast cancer in healthy women is at the present time unclear. Several points should be noted:

- The trial was conducted in a particular group of postmenopausal women with CHD or risk factors for CHD. Results may not necessarily be applicable to other women and further research will be needed to confirm the findings.

- Additionally, as the trial was primarily designed to investigate the effects of raloxifene in reducing the risk of cardiovascular events (which was not found), the trial may not have been sufficiently powered to accurately detect differences in breast cancer outcomes between sub-groups of women, for example those with other less common stages or types. There were relatively few cases of breast cancer, and in particular non-invasive cancer, and the number of outcomes may have been too few to find statistically significant differences between the groups.

- Any benefit of taking raloxifene to reduce risk of breast cancer must be balanced against the risk of treatment. Raloxifene is known to increase the risk of venous blood clots, a fact that the authors confirmed was found by this study (data not given in this report). The authors also say that raloxifene increased the risk of fatal stroke. The drug should also not be used in women who fall into the categories that were excluded from the study, such as those with cancer of the uterus, any unexplained uterine bleeding, or those with kidney or liver disease. - Any benefit of taking raloxifene in women who already have oestrogen-receptor positive invasive breast cancer was not investigated here.

Further research is needed to look at the role of raloxifene in the prevention of breast cancer in other groups of women. It should be compared to receiving no treatment, as well as being compared to other treatments that act in a similar way upon the oestrogen receptor, such as tamoxifen. Data on the numbers of women who could be expected to suffer strokes or blood clots when taking raloxifene would help put this study into perspective.

Links to the headlines

Breast cancer risk reduced by brittle bone drug.The Daily Telegraph, June 11 2008
Breast cancer: brittle bone drug can stop disease taking hold.The Times, June 11 2008
Brittle bone drug 'prevents cancer'.Channel 4, June 11 2008

This news comes from NHS Choices