Medication to prevent breast cancer — too much to swallow?

Using medication to prevent breast cancer in women at moderate and high risk is a cost-effective1 and immediately implementable strategy for reducing the burden of breast cancer in Australia. There is Level 1 (strong) evidence that selective oestrogen receptor modulators (SERMs) such as tamoxifen and raloxifene reduce breast cancer risk by up to 40% in these women.2 The absolute risk reduction depends on an individual’s risk factor profile.

These agents are rarely prescribed in Australia for breast cancer prevention (for complex but potentially modifiable reasons3) even in women at high risk, although they are now endorsed in an Australian guide and in the American Society of Clinical Oncology practice guideline.2,4 We summarise the evidence and suggest ways to implement this intervention in Australia. Women at high risk of breast cancer have the most favourable risk–benefit ratio, and hence are the focus of this article, but consideration of use of SERMs for women at moderately increased risk is also appropriate.

The average lifetime risk of breast cancer for an Australian woman is 1 in 9 or 11%. A very small proportion (< 1%) of Australian women have a mutation in a major breast cancer predisposition gene, such as BRCA1, BRCA2, p53 or PTEN, resulting in a high lifetime risk for breast cancer of up to 80%. A larger group has a strong family history of breast cancer but no documented gene mutation. These women have a lifetime risk of at least 30%. A history of some types of benign breast disease and higher mammographic density also increase a woman’s risk. Importantly, an individual’s risk changes with age.5 To aid clinicians, the National Breast and Ovarian Cancer Centre (NBOCC) has developed a risk assessment guide and the online Familial Risk Assessment — Breast and Ovarian Cancer (FRA-BOC) risk assessment tool (one of several available6), (http://canceraustralia.nbocc. org.au/fraboc/). “High risk” throughout this article includes category 3 of the NBOCC guide (Box 1).

Several large randomised controlled trials (RCTs), which included over 25 000 women with at least moderately increased risk, have demonstrated the efficacy of SERMs.7-10 These trials showed that 5 years of tamoxifen use at 20 mg daily reduces breast cancer risk by 40%.2 The absolute benefit is thus between 12% and 32% for a woman at high risk (as defined in the previous paragraph), depending on her underlying absolute risk of breast cancer. Raloxifene is about 75% as effective as tamoxifen in postmenopausal women,11 with an arguably better side effect profile. Raloxifene has not been tested in premenopausal women. Both medications reduce the incidence of hormone receptor-positive cancers. SERMs also reduce breast density, which may facilitate earlier mammographic diagnosis of hormone receptor negative cancers.12 The reduction in breast cancer incidence is maintained for at least 5 years after a patient discontinues tamoxifen.2,7 Although none of the RCTs was powered for a mortality end point, reduced breast cancer incidence translates to reduced physical and psychosocial morbidity and treatment cost savings,1 and is a worthwhile and clinically meaningful end point.

The commonest side effects of SERMs are hot flushes, sweats and altered vaginal secretions. In the RCTs, 10%–12% more women using tamoxifen had vasomotor and vaginal symptoms than in the placebo group; however, no difference was found in the incidence of depression, weight gain or overall quality of life.13 For the minority of women who experience significant problems, the SERM can be discontinued.

SERMs double the risk for deep vein thrombosis and thromboembolism during use. The absolute incidence is around 4 per 1000 women-years of use (0.4% per year).11 For premenopausal women, with their low background risk of deep vein thrombosis, the absolute risk is even smaller; around 0.1% per year10 (comparable to the excess risk of thrombotic events in women who take third-generation oral contraceptives).14 SERMs did not significantly increase the risk for coronary heart disease or stroke in prevention trials.15 Women who smoke or those with a previous history of thrombosis should probably not be offered SERMs for risk reduction. Other contra-indications are listed in Box 2.

Tamoxifen is associated with an excess risk of endometrial carcinoma of 0.25%–0.4% per year during use in postmenopausal women. The risk is not significantly increased in premenopausal women.10 Raloxifene does not increase the risk of endometrial cancer.11 Tamoxifen is a potential teratogen, and therefore is not appropriate for women planning pregnancy in the treatment period or women not using reliable contraception. There is a modest increase in the risk of cataracts (relative risk, 1.14; 95% CI, 1.01–1.29) in women taking tamoxifen.10 Bone density and fracture rates are improved by 5 years of SERM use after menopause,15 and raloxifene is currently prescribed for this indication.

Before September 2010, the Australian NBOCC guide suggested consideration of prevention trials but made no recommendations about off-trial SERM use. Current use of SERMs preventively in Australia is very low. Of 3788 women at high risk of breast cancer enrolled in a national cohort study (kConFab), fewer than 3% have used SERMs for prevention, and only 0.3% have done so while not enrolled in a clinical trial.16

The main alternatives to SERMs are screening and, for women at high risk, risk-reducing surgery (Box 3). Only 5% of women at high risk in the kConFab cohort have undergone risk-reducing surgery, and the vast majority are not using any preventive strategy.16,17 Annual breast imaging for early detection is suggested but does not reduce risk and is not proven to decrease mortality in premenopausal women at high risk.18 Aromatase inhibitors are under investigation for breast cancer prevention, with early results showing a benefit for exemestane,19 but longer term follow-up is needed, and use outside a clinical trial is not recommended in any guide. The International Breast Cancer Intervention Study (IBIS II), an RCT of anastrozole versus placebo, is currently recruiting participants in Australia.

There are multiple potentially reversible barriers to recommendation of SERMs by clinicians and acceptance by women (Box 4). Some of these have recently been addressed, notably by revision of the NBOCC guide, which now suggests consideration of SERMs for women with moderate or high risk of breast cancer (Box 3).

Referring to preventive SERMS as “chemoprevention” creates confusion with chemotherapy and results in negative attitudes towards this strategy.20 The consumer group Breast Cancer Network Australia recommends use of the term “risk-reducing medication” (Gerda Evans, Advocate, Breast Cancer Network Australia, personal communication, 24 August 2011). Tamoxifen is often imprecisely referred to as “hormone therapy”, potentially reducing its acceptance in a community aware of the risks of cancer with long-term hormone replacement therapy.21

Concern about side effects is common. Descriptions of relative rather than absolute risks of side effects can result in overestimation of the risk of rare side effects. This limits patient acceptance, as does lack of precision in estimating the magnitude of absolute benefit, which depends on multiple individual factors.

Several contributors to the reluctance of doctors to prescribe SERMs are documented, and include a previous lack of endorsement of SERMs for risk reduction in guidelines, deficiencies in knowledge of risk assessment or patient selection, and time constraints.3 The new NBOCC guide and FRA-BOC risk assessment tool do much to address these issues. Australian doctors identify the lack of a Pharmaceutical Benefits Scheme subsidy as a barrier to prescribing, but tamoxifen and raloxifene are relatively inexpensive (less than $1 and $2.50 per day respectively) and likely to be within the means of most patients.

Raloxifene is registered with the Therapeutic Goods Administration for the indication of breast cancer prevention in women at high risk, but registration for this indication has never been sought for tamoxifen and, as an off-patent drug, there is little incentive for pharmaceutical companies to champion its listing. This lack of registration for breast cancer prevention is cited as another barrier to prescription of tamoxifen. Nevertheless, off-label prescribing of tamoxifen is an appropriate approach because high-quality evidence exists to support its use.22 Tamoxifen and raloxifene are approved by the Food and Drug Administration for primary prevention of breast cancer in the United States.

A minority of Australian women at increased risk of breast cancer regularly attend a risk management clinic where SERMs can be prescribed and monitored,23 but most require an alternative model of care. General practitioners and breast surgeons are ideally placed to identify women at increased risk and discuss, prescribe and monitor risk-reducing medication. GPs have a central role in prescribing for prevention of other conditions such as cardiovascular disease, and are a trusted source of prevention advice and management. Information is lacking on why Australian GPs and breast surgeons are not currently prescribing SERMs for prevention, but it is likely that they would benefit from further education and better tools to identify women at high risk. Tools such as the FRA-BOC risk assessor and the IBIS risk calculator (http://www.ems-trials.org/riskevaluator/) allow rapid assessment of breast cancer risk, but it remains difficult to quantitatively assess an individual woman’s absolute risk of side effects due to SERMs. Cancer Australia has developed a fact sheet which provides helpful information about contraindications and practicalities of prescribing. An evidence-based, online decision aid that enables personalised risk–benefit assessment is urgently needed.

Australian women at moderate and high risk for breast cancer are not routinely offered SERMs, a proven preventive strategy. We hope that the recent update of the NBOCC guide including a suggestion for consideration of SERM use, along with the easy availability of online risk assessment tools, will increase preventive SERM use. Improved education and decision support for clinicians may help reduce the controversy surrounding the use of SERMs for breast cancer prevention and ultimately reduce breast cancer incidence.