Prevention of breast cancer

• Modifiable lifestyle factors may reduce the risk of developing breast cancer. Obesity is associated particularly with post-menopausal breast cancer. Diet is important, and exercise equivalent to running for up to 8 hours each week reduces the risk of breast cancer, both in its own right and through reducing obesity.
• Alcohol consumption may be responsible for 5.8% of breast cancers in Australia and it is recommended to reduce this to two standard drinks per day. Drinking alcohol and smoking increases the risk for breast cancer and, therefore, it is important to quit tobacco smoking.
• Prolonged use of combined oestrogen and progesterone hormone replacement therapy and oral contraceptives may increase breast cancer risk and this must be factored into individual decisions about their use.
• Ionising radiation, either from diagnostic or therapeutic radiation or through occupational exposure, is associated with a high incidence of breast cancer and exposure may be reduced in some cases.
• Tamoxifen chemoprevention may reduce the incidence of oestrogen receptor positive cancer in 51% of women with high risk of breast cancer. Uncommon but serious side effects include thromboembolism and uterine cancer. Raloxifene, which can also reduce osteoporosis, can be used in post-menopausal women and is not associated with the development of uterine cancer.
• Surgical prophylaxis with bilateral mastectomy and salpingo-oophorectomy can reduce the risk of breast cancer in patients carrying BRCA1 or BRCA2 mutations. For preventive treatments, mammographic screening can identify other women at high risk.

Breast cancer prevention includes strategies that range from a change in lifestyle behaviours to alter modifiable risk factors, to the prophylactic use of chemoprevention or surgery — based on genetic predisposition or other factors that identify patients at high risk — and the treatment of pre-invasive lesions found by population mammographic screening.

The strategy for this narrative review was to search PubMed and the grey literature available on the internet for the most recent reviews and meta-analyses and for pivotal articles describing each preventive action, and to supplement those articles with others from the references of the reviews.

Many of the major risk factors for breast cancer cannot be changed. Breast cancer is more common in women than men by 100-fold.1 In women, reproductive factors, such as early menarche and late menopause — which increase the exposure of breast tissue to oestrogen and progesterone — also increase the risk of developing cancer.2 The incidence of breast cancer also increases with age, with 77% of breast cancer in Australian women occurring over the age of 50 years.1 A family history of first degree relatives (mother, sister or daughter), although associated with only one in nine breast cancers, almost doubles the risk compared with women who have unaffected relatives.3,4 An increase in breast density,5 which is associated with increased cancer risk, can also be a heritable trait, while inheriting mutated genes also increases this risk. For example, inheriting either BRCA1 and BRCA2 mutations, although they are only associated with less than 10% of breast cancers, increases the lifetime risk (to 75 years old) of breast cancer by 47% and 32% respectively.6

Lifestyle risk factors

The lifestyle factors that can be changed include diet, exercise, alcohol and tobacco consumption, the use of combination hormonal therapy, and the exposure of the breast to radiation. These have been estimated as being responsible for 40% of breast cancers. Also, an early age for a first full term pregnancy and breastfeeding have been associated with decreasing the risk of breast cancer. A recent meta-analysis has shown that these reproductive behaviours change the incidence of specific subtypes of breast cancer.7 Older age at first birth was associated with an increase in the luminal subtype of breast cancer, while having ever breastfed a baby reduced the risk of luminal and triple negative breast cancer, that is, breast cancer that does not have the receptors for oestrogen, progesterone or human epidermal growth factor receptor 2 (HER2).

Obesity

After reviewing the available studies, the World Cancer Research Fund (WCRF) has stated that there is convincing evidence for a link between obesity and being overweight, and post-menopausal breast cancer.8 In Australia, the proportion of breast cancers attributable to being overweight or obese is estimated to be 8%.9 A recent analysis of the observational study from the Women’s Health Initiative has shown that the duration of being overweight is associated with an increasing incidence of obesity-related cancers. Every 10-year increase in adult obesity duration is related to a 5% increase in the risk of post-menopausal breast cancer.10

Exercise has been found to decrease the risk of post-menopausal breast cancer independent of its impact on obesity.8 In pre-menopausal women, the average risk reduction is 30–40% and may be greatest in those of normal or low bodyweight.11

The risk of breast cancer decreases with increasing levels of physical activity.12 The question of how much exercise reduces this risk was recently reviewed by analysing 174 studies reported between 1980 and 2016, 35 of which were related to breast cancer. The findings were that it may require the equivalent of 15–20 hours of brisk walking each week or 6–8 hours of running to have an impact on reducing cancer, which is up to five times the previous recommendation of the World Health Organization.13 Compared with insufficiently active women, the risk of breast cancer decreased by 3% for low activity, 6% for moderate activity and 14% for high activity levels.13

The role of exercise in preventing breast cancer may be that exercise lowers hormone levels in pre-menopausal women and levels of insulin and insulin-like growth factor. It also improves the immune response and assists with weight maintenance.14

General dietary advice for a healthy lifestyle recommends limiting energy dense foods, red meat, salt and sugary drinks while eating more fresh fruit and vegetables. The evidence for specific dietary factors, such as fat intake relating to developing breast cancer is not definitive.15

Alcohol

The International Agency for Research on Cancer lists alcohol as a group 1 carcinogen, that is, a known cause of cancer. The WCRF reports that the evidence associating breast cancer with the consumption of alcohol is convincing.15,16 The cumulative amount of alcohol consumed is associated with an increased risk of breast cancer, which is not linear as the dose response curve becomes steeper at higher alcohol intakes, indicating greater risk with heavier alcohol intake.17

The National Health and Medical Research Council recommends that a low risk of alcohol consumption for healthy adults is two standard drinks (equivalent of 10 g or 12.5 ml pure alcohol per drink). At that level, the lifetime risk of dying from an alcohol-related disease is less than 0.4 per 100 people.18,19 In Australia, the proportion of breast cancers attributable to alcohol has been estimated to be 5.8%.19

The main mechanism for the carcinogenicity of alcohol is that its main metabolite, acetaldehyde, is carcinogenic.20

Hormone use

Although individual women may benefit from hormone replacement therapy (HRT) to manage symptoms of menopause, population studies show that the use of combined hormones with both oestrogen and progesterone is associated with an increased risk of breast cancer.21 The Women’s Health Study quantified the relative risk of breast cancer with HRT use as 1.37, while the Million Women Study found the relative risk to be 2.14.22,23 This risk increases with the duration of HRT (usually after the first 3 years), but seems limited to the period in which the hormones are taken, and it returns to the levels of those who have not taken HRT by 5 years after ceasing treatment.24 The risk is also higher in women who start HRT close to the onset of menopause rather than later.25 When HRT use declined in some countries, because of publicity about the link with breast cancer, the breast cancer incidence fell.26 If oestrogens are used alone, most studies report no increased risk of breast cancer.27

There is some evidence from a review of 50 studies in 1996 that longer term users of oral contraceptives have a small increase in the risk of breast cancer, which returns to baseline by 10 years after ceasing them.28

Although this should no longer occur, in utero exposure to diethylstilboestrol has an increased breast cancer risk.29

Tobacco

There has been no consistent association between tobacco smoking and breast cancer. The results are confounded by concomitant alcohol use. A recent meta-analysis showed that there was a 24% increase in the likelihood of developing cancer only in women who also currently or previously drank alcohol if they smoked as compared with non-smokers. The greatest risk was in those who commenced smoking before their first menstrual period, who had a 61% increased risk. Those who started to smoke tobacco at least 11 years before their initial full term pregnancy had a 45% higher risk of developing breast cancer.30

Ionising radiation

Ionising radiation is radiation with sufficient energy to break electrons off atoms and disrupt the DNA in cells. Exposure can be to large doses, such as was observed after nuclear explosions or accidents, or when receiving radiation therapy. Small doses of radiation, such as may be experienced with multiple computed tomography (CT) scans or mammograms, where the exposure is cumulative may be problematic, but in some situations have been found to stimulate an immune response.31 Although the overall risk of breast cancer following exposure to ionising radiation is low, it is greater if the exposure occurs at a younger age. It is also greater if women already have gene mutations that increase their risk of cancer and it may be additive to the impact of exogenous oestrogens on the risk of developing breast cancer.32,33 The implication of these findings is that benefits of diagnostic imaging by using x-rays or CT scans should be carefully assessed against this risk, particularly in younger patients, and this may include consideration of alternative tests such as ultrasound or magnetic resonance imaging (MRI) scans. Reassessing the indications for therapeutic radiation and, where possible, the better targeting of therapeutic radiation therapy to reduce the exposure of the breast, will also help reduce this risk.

Occupation

Studies examining exposure to ionising radiation among radiological technologists demonstrated an elevated risk of mortality from breast cancer for those who operated equipment before 1950 and if they started working at younger ages.34,35 In a recent review, airline workers have been associated with an increased risk of breast cancer, which may be partly due to cosmic radiation exposure.36 Nightshift work has also been associated with an increased mortality from breast cancer as reported in a recent meta-analysis of prospective cohort studies.37 Other occupational exposures, such as to pesticides, solvents and heavy metals, have insufficient evidence to characterise their association with breast cancer.38

Chemoprevention

For women at high risk of breast cancer, taking medication may reduce their chances of developing breast cancer, but there is no survival advantage yet documented.

Selective oestrogen receptor modulators

Tamoxifen is one of a series of selective oestrogen receptor modulators (SERMs) which reduce the impact of oestrogen on the breast. High risk includes women with a strong family history of breast cancer or who have a biopsy that shows pre-invasive cancer pathology, such as lobular carcinoma in situ (LCIS), atypical ductal hyperplasia or atypical lobular hyperplasia. Those who will benefit from tamoxifen are women over 35 years with a risk of breast cancer of more than 0.66% on the Gail model.39 The Gail model is a risk assessment tool which scores demographic risk factors, such as age and ethnicity, along with histological features, such as positive ductal carcinoma in situ (DCIS) and LCIS, family history and menstrual history.39 The side effects of tamoxifen include hot flushes, night sweats, vaginal dryness and discharge, and cataracts. Rarer but more serious toxicities are thromboembolic events which cause strokes, deep venous thrombosis and pulmonary emboli, and the development of endometrial and uterine cancer.

Raloxifene is a second generation SERM that is not quite as effective as tamoxifen, but does not increase the risk of uterine cancer and also helps prevent osteoporosis. It is often used for post-menopausal women at high risk.

A meta-analysis of 10-year data from nine randomised trials of SERMs as chemopreventive agents in women at high risk of breast cancer showed a 38% reduction in overall breast cancer incidence and a 51% reduction for oestrogen receptor positive cancers.40 This preventive effect of tamoxifen can last for 20 years.41 Raloxifene’s impact on prevention does not last as long and has about 76% the efficacy of tamoxifen.

Third generation SERMs, such as lasofoxifene and arzoxifine, which also treat reduced bone mineral density, have been trialled with reduction in the development of invasive breast cancer, but these drugs are not yet approved for use.

Despite the impressive results of chemoprevention of SERMs in women at high risk, the uptake of this preventive strategy has been poor. A meta-analysis of 36 studies of participation in chemoprevention showed the uptake to be 16.3%, higher in trial settings (25.2%) than in non-trial settings (8.7%).42 A higher uptake was associated with older age, having an abnormal biopsy result, higher risk, fewer concerns about side effects and having a clinician recommend it.42

Aromatase inhibitors

Aromatase inhibitors, such as anastrazole, letrozole and exemestane, are used to treat post-menopausal women who have oestrogen receptor positive breast cancer. Although they may have side effects, such as hot flushes, vaginal dryness, joint and muscle pains and fatigue, and may increase the risk of osteoporosis, they are not associated with thromboembolic phenomena or uterine cancer like the SERMs.42 They have been tested as chemopreventive agents for oestrogen receptor positive tumours in post-menopausal women and had sufficient activity to be tested against tamoxifen.43,44

Anti-human epidermal growth factor receptor 2 drugs

Small trials have been mounted to test trastuzumab in patients with human epidermal growth factor receptor 2 (HER2)-positive DCIS. Lapatanib is also being tested pre-operatively in patients with HER2-positive DCIS to determine whether its use prevents invasive cancer.42

Aspirin

Aspirin has been associated with chemoprevention in patients with high risk colorectal cancer. To explore the association between aspirin use and post-menopausal breast cancer, women with no history of breast cancer were followed and divided into those who used aspirin and those who did not. The use of aspirin decreased the breast cancer incidence in women with a family history of breast cancer or a history of benign breast disease.45 Further studies are required.

Mammographic screening

In Australia, population mammographic screening for early detection of breast cancer is provided for the target group of women aged between 50 to 74 years. It has been shown to reduce mortality from breast cancer. However, because breasts can develop pre-invasive lesions before developing invasive cancer, a mammogram is important for preventing invasive breast cancer in addition to early diagnosis of invasive disease when it can still be cured, thereby reducing the mortality from breast cancer. As an example, in 2010 in the United Kingdom, mammographic screening diagnosed 125.7 cancers per 100 000 woman years, whereas the pre-invasive lesions of DCIS and LCIS were detected at a rate of 18 per 100 000.46

Screening by mammography or MRI may be recommended in younger women with a high risk of developing breast cancer, such as women with BRCA1 or BRCA2 mutations or multiple close relatives with breast cancer, particularly if they developed cancer at a younger age.

The other feature that mammography can assess is breast density. Differences in density relate to the different proportions of the dense fibro-glandular tissue to the less dense adipose tissue, which causes a different appearance on mammography measured by less or more radiolucency. Increased breast density has been shown to be a risk factor for non-familial breast cancer. Although higher density may make cancers more difficult to see, and increased breast density has been related to other risk factors — such as family history, not having had children and exogenous hormone use — breast density is nonetheless an independent risk factor. A higher proportion of dense tissue has been associated with a four to five times higher risk of developing breast cancer.5 Moreover, the cancers diagnosed in women with increased breast density are often larger, higher grade and with more lymphatic invasion and lymph nodes and higher stage, but this does not increase the risk of death when accounting for these factors.47 The importance of assessing such risk factors is that it may identify candidates for chemoprevention to reduce the chances of developing breast cancer.

Genetic testing

When the genes BRCA1 and BRCA2, which code for tumour suppressor proteins, are mutated, the lifetime risk of breast, ovary and other cancers is increased. This is the case in up to 25% of inherited breast cancers and 10% of breast cancers overall.48 It is estimated that BRCA1 carriers have a 55–65% chance of developing breast cancer by the age of 70 years and BRCA2 carriers have a 45% chance of developing breast cancer by 70 years, depending on their other risk factors.49

Genetic counsellors usually determine who needs to be tested based on aspects of their family history, such as first degree relatives with breast cancer diagnosed before 50 years, multiple breast primaries of bilateral cancers, male breast cancer, ovarian cancer, or Ashkenazi Jewish ethnicity.

There are other genes associated with an increased risk of breast cancer.48 These are found in inherited disorders such as Li–Fraumeni syndrome and Fanconi anaemia. The PALB2 gene associated with Fanconi anaemia codes for a tumour suppressor gene which produces a protein that interacts with the proteins produced by BRCA1 and BRCA2 to help with DNA strand break repairs.50

Combining modifiable risk factors with non-modifiable factors, including single nucleotide polymorphisms and family history, is being explored to develop a more accurate model for predicting the absolute risk of women developing breast cancer.51 This will enable the identification of people who will benefit from risk reduction strategies to prevent breast cancer. These may include screening at an earlier age, which can introduce other imaging techniques such as MRI. Chemoprevention can also be considered in these higher risk groups, and prophylactic surgery which removes as much of the breast tissue as possible becomes a further option.

Surgical prevention of breast cancer

For BRCA1 and BRCA2 mutations, which carry an increased risk of breast and ovarian cancer, bilateral salpingo-oophorectomy reduces the risk of ovarian and fallopian tube cancer and halves the risk of breast cancer. The mutation carriers will also benefit from bilateral prophylactic mastectomy, which reduces their breast cancer risk by up to 95%, more so than non-mutation carriers who may also be offered the procedure because of high risk.51,52 A more controversial surgical area is contralateral mastectomy after the diagnosis of cancer in the other breast. This has not been associated with increased survival.53 Reconstructive surgeries can restore the symmetry of the breasts as an alternative to contralateral mastectomy if that is the motivation for requesting it.54

As with chemoprevention, uptake of these surgical preventive strategies is lower or later than recommended.55 The risks and benefits of these strategies must be assessed for each individual.

Conclusions

Preventing breast cancer begins with identifying those at higher risk because of their pattern of genetic mutations; demographic features, such as age, sex and ethnicity; reproductive and hormonal factors; family history of breast cancer; past history of benign breast disease; or lifestyle behaviours. Mammographic screening can detect pre-invasive lesions that can be treated. Modifiable lifestyle behaviours include increasing exercise and dietary modification to prevent obesity, moderating alcohol intake, not smoking, balancing the risks when deciding on taking exogenous hormones, and reducing exposure to radiation and other occupational exposures. In addition, patients at high risk may benefit from chemoprevention with tamoxifen and other hormones, prophylactic bilateral surgical mastectomies, or salpingo-oophorectomies.