Approximately one-in-nine women in Ireland will develop breast cancer at some point in their lives, while the incidence rate in men is approximately one-in-1,000. Breast cancer incidence rates are increasing worldwide, from 550,000 in 1975 to over two million in 2018. This increase is partly because people are living longer and their risk increases with age, and partly due to changing lifestyle trends over time. Lifestyle factors that increase breast cancer risk include alcohol consumption, later age at first birth, use of HRT, and dietary changes leading to increased levels of obesity; whereas physical activity and increased duration of breastfeeding reduce that risk. With increased incidence of breast cancer comes a need for better risk assessment within the general population, so that women who are at higher risk of developing breast cancer can be offered more accurate screening techniques, or can be offered interventions to reduce their risk.
Current approaches to risk assessment include genetics clinics, where women are referred for genetic assessment if they have a family history of breast cancer. Risk assessments may also be carried out by GPs, but there is often a lack of time or knowledge to provide an in-depth assessment. Many women will also undergo some form of risk assessment as part of their mammographic breast screening. However, the approach is not standardised and most experts will agree that we need improved risk assessment guidelines to more accurately identify those women at the highest risk of developing breast cancer.
This topic was on the agenda at a research symposium, ‘Current Approaches to Risk Assessment and Management in Breast Cancer’, held on 12 February 2019 in University College Dublin, and organised by BREAST-PREDICT, an Irish Cancer Society-funded Collaborative Cancer Research Centre in the area of breast cancer research (www.breastpredict.com). The keynote speaker, Prof Jack Cuzick, shared his views on the most effective approaches to breast cancer prevention. Prof Cuzick is Director of the Wolfson Institute of Preventive Medicine in London, Head of the Centre for Cancer Prevention, and holds the post of John Snow Professor of Epidemiology at Queen Mary University of London. He has worked extensively in cancer epidemiology and clinical trials, with special interest in prevention and screening.
According to Prof Cuzick, there are two approaches that can be taken to cancer prevention: the first is to advocate for interventions for the population as a whole, aimed at reducing overall risk. Examples of this approach would be initiatives that tackle rising obesity, increase physical activity, or reduce alcohol consumption. Research has demonstrated that exercise and weight loss greatly reduce a woman’s risk of developing breast cancer. The second, more focused approach would be to firstly identify those women who are at the highest risk of developing breast cancer, by improving risk assessment techniques. These women could then be offered increased monitoring to catch any developing cancer at an early stage, or, if appropriate, suitable chemopreventative agents, such as aspirin or tamoxifen, to reduce their risk of developing breast cancer. The key to this focused approach is the use of an accurate risk assessment model to guide selection for increased surveillance and/or chemoprevention, according to the level of risk. This must then be combined with the use of effective, minimally toxic preventative agents in those women requiring this intervention.
Risk management in Ireland
In Ireland, risk assessment in the general population is generally based on family history, whereby unaffected women with a family history of breast cancer are referred either to the Department of Clinical Genetics at Children’s Health Ireland in Crumlin or to the Cancer Genetics Service run by Prof David Gallagher at Trinity St James’ Cancer Institute at St James’s Hospital, Dublin.
Consultant Clinical Geneticist in the Department of Clinical Genetics at Children’s Health Ireland, Prof Andrew Green, also spoke at the research symposium on 12 February. He outlined that, for unaffected women with a family history of breast cancer referred to the service, the screening method is an assessment based on the Manchester criteria for predicting the probability of there being a BRCA1 or BRCA2 mutation in the family. If the unaffected woman has a relative with breast cancer with a confirmed pathogenic variant in either BRCA1 or BRCA2, genetic counselling and a predictive genetic test are offered. Approximately 5-to-10 per cent of breast cancer cases are hereditary, linked to mutations in specific genes such as the DNA repair genes BRCA1 or BRCA2. The cumulative lifetime risk for those carrying BRCA1 mutations is 72 per cent; for those with BRCA2 mutations, the risk is 69 per cent; while for mutations in other genes, the risk is lower: 44 per cent for PALB2, 32 per cent for CHEK2 and 30 per cent for ATM. This compares to the population lifetime risk of approximately 12 per cent.
According to Prof Green, there needs to be more awareness and education in the area of genetic testing both for healthcare professionals and the general public.
Chemoprevention in high-risk women
Back in the early 1980s, Prof Cuzick was working on a trial testing the effectiveness of a new drug at the time, tamoxifen, on treatment of breast cancers, when he observed that tamoxifen not only reduced breast cancer recurrence but also reduced the incidence of ‘new’ breast cancers in the contralateral breast. This was the first indication of the chemopreventative potential of tamoxifen. He subsequently led one of the first large-scale clinical trials to examine the use of tamoxifen as a chemopreventative agent in reducing breast cancer incidence in high-risk individuals. This trial, the International Breast Cancer Intervention Study (IBIS-1), which began in 1992, recruited over 7,000 high-risk women, who were randomly assigned to receive oral tamoxifen treatment or placebo for five years, and followed up for an average of 16 years. Results showed a long-term 30 per cent reduction in breast cancer in the group assigned to receive tamoxifen compared to the placebo group. There were some significant side effects associated with tamoxifen therapy, such as an increased risk of endometrial cancer and thromboembolic complications; however, given the long-term reduction in breast cancer risk, the benefit-to-harm ratio was still favourable. Subsequent trials, including IBIS-2, examined the use of aromatase inhibitors such as anastrozole as preventative agents for high-risk women and found a >50 per cent long-term reduction in breast cancer incidence in women receiving these agents, in addition to reduced toxicity in comparison to tamoxifen.
Despite the demonstrated efficacy of drugs like tamoxifen or aromatase inhibitors in reducing breast cancer incidence in high-risk populations, studies in the UK have shown that many GPs are unaware that the drug can be used in this way, or are aware and reluctant to prescribe it. In 2013, the National Institute for Health and Care Excellence (NICE) recommended that women deemed to be at moderate or high risk of breast cancer should be offered chemoprevention drugs, including tamoxifen. According to Prof Green, there is no policy in Ireland on the use of tamoxifen as a risk reduction measure in healthy women with a family history of breast cancer.
Surgical approaches to risk reduction
In terms of risk reduction, prophylactic surgery remains the most effective modality for reducing breast cancer risk and mortality for individuals with a strong genetic predisposition. Mr Damian McCartan, an Oncoplastic Breast Surgeon at St Vincent’s University Hospital, Dublin, also addressed the symposium in February to outline surgical approaches to risk reduction. The uptake of the surgical approach is likely to increase in the future, as improved awareness and genetic testing options, in conjunction with lower costs, means that more women will face decisions about risk-reducing mastectomy. This is a difficult decision to make, and women must weigh up the possibility of developing a potentially curable disease with the potential complications of surgery. The decision-making process should be supported with genetic counselling, education, clinical psychology and surgical consultations. The timing of surgery is also key, and is influenced by a series of personal factors, such as breastfeeding, career stage, availability of support during recovery, etc. Surgical techniques have advanced over the years, from a simple mastectomy with no reconstruction, to skin-sparing or nipple-sparing techniques with reconstruction. There is a risk of complications with the procedure, including implant loss, which occurs in approximately 10 per cent of patients, with smokers and those with high BMI at increased risk. Post-operative pain is also reported by 15-to-20 per cent of patients. In some cases, later aesthetic or implant-related complications may require further surgery. However, there is no doubt that this approach is effective: studies have shown a 93 per cent risk reduction overall in BRCA carriers, with high levels of satisfaction with the decision to undergo the procedure.
Improved risk assessment tools
While working on the IBIS trials, Prof Cuzick also developed a now widely-used risk evaluator software tool called the Tyrer-Cuzick Model (see figure 1), which incorporates various risk factors including family history, hormonal factors, benign disease, age, body mass index, and genetic factors into a single statistical model to predict breast cancer risk. When the risk factor information is fed into the model, the programme calculates the percentage lifetime risk of breast cancer for the individual in comparison with the general population risk.
While alternative models exist, including GAIL and BOADICEA, the Tyrer-Cuzick model has been shown in independent studies to be the most consistently accurate when compared with other available risk assessment models. It is commonly used, particularly in the US, to determine eligibility for additional MRI screening on top of the standard mammographic screening programme. The most recent version of the Tyrer-Cuzick model has incorporated additional risk factors into the evaluation of risk, including polygenic single nucleotide polymorphisms (SNP) scores, and one relatively lesser-known risk factor – breast density.
The importance of breast density
Most women are aware of the common risk factors when it comes to breast cancer. However, many are not aware of the importance of breast density, which has been described as one of the most significant risk factors of all. Women with dense breasts have a higher proportion of fibroglandular and supportive tissue, whereas those with less dense breasts have a higher proportion of fat. Breast density cannot be measured by look or feel, only by mammography, where dense breast tissue appears white and fatty breast tissue appears dark. Breast density can be categorised into ‘mostly fatty’ (~10 per cent), ‘scattered density’ (40 per cent), ‘heterogeneously dense’ (~40 per cent), and ‘extremely dense’ (~10 per cent). Breast density is one of the strongest independent risk-factors for developing breast cancer, stronger even than age or family history, with women with ‘extremely dense’ breast tissue being four-to-six times more likely to develop breast cancer compared to women with ‘mostly fatty’ breasts. High breast density can also have the effect of ‘masking’ breast cancers, as they also show up as white on a mammogram, and are thus more difficult to see.
On 10 June 2019, BREAST-PREDICT teamed up with a patient advocacy organisation, Being Dense, to host a public seminar in the RCSI entitled ‘Mammographic Breast Density: What women need to know’. National and international experts were invited to discuss the evidence basis for including breast density information in breast cancer screening programmes internationally, and in the context of Irish healthcare and the national breast cancer screening programme, BreastCheck. Speakers included Dr Paula Gordon from the University of British Columbia, a breast radiologist who has been a powerful advocate for notifying women about their breast density; Prof Fidelma Flanagan, Consultant Radiologist at the Mater Hospital, Dublin, and Clinical Director of BreastCheck Eccles Unit, who has played an active role in developing and modernising the BreastCheck screening programme; and Research Fellow Dr Maeve Mullooly from the RCSI, whose research focuses on understanding the biological relationship between mammographic breast density and breast cancer development. The discussion panel was chaired by Prof William Gallagher, Professor of Cancer Biology at University College Dublin and Director of BREAST-PREDICT, and Siobhán Freeney, Breast Density Patient Advocate and Founder of Being Dense.
One of the key messages from the seminar is that mammography saves lives, and that women should participate in a screening programme where it is available to them. In Ireland, the BreastCheck screening programme provided mammograms to almost 140,000 women in 2016, and detected 975 cancers. Translating cancers detected into mortality benefits is not straightforward, but one study presented showed that women aged 40-69 years who chose to participate in an organised screening programme in Sweden had a 60 per cent lower risk of dying from breast cancer within the 10 years after diagnosis. However, having dense breasts impairs the detection of cancer on mammograms, with both appearing white. This makes a delayed or missed diagnosis more likely. According to Dr Gordon, mammograms detect 98 per cent of cancers in women with fatty breasts, but only 48 per cent of cancers in women with dense breasts. Interval cancers are also more common in women with dense breasts, which are cancers diagnosed in the period between two screening mammograms. These cancers are more likely to be larger and lymph node-positive at diagnosis, and therefore require more aggressive treatment, which comes with attendant side effects. Dr Gordon recommends supplementary screening methods, in addition to mammography, for women with dense breasts, such as MRI or ultrasound. Although these additional screening methods can increase the rate of false alarms, she suggests that: “When provided with the facts, most women are willing to put up with the short-lived stress associated with a recall or even a needle biopsy, in exchange for avoiding a missed cancer.”
In the US, 36 states have introduced laws that mandate compulsory density notification to patients when they have a mammogram, with advice for patients with dense breasts to seek further breast screening such as MRI or ultrasound. The US Food and Drug Administration (FDA) was recently tasked with drawing up national guidelines to establish a national minimum standard for including breast density information on mammogram reports. The Canadian province of British Columbia recently adopted the policy of breast density notification to all women who attend public breast cancer screening, and are now providing government-funded supplementary ultrasound for women with high density. Six more Canadian provinces have committed to policy change to be implemented by the end of 2019. These changes in legislation have been brought about primarily due to pressure from grassroots campaigns led by women who have had breast cancers missed because of the masking effect of breast density.
Screening in Ireland
In Ireland, the breast cancer screening programme, BreastCheck, has been running for almost 20 years, and currently offers mammograms to women aged between 50-69 years on a two-year cycle. In that 20 years, the programme has been at the cutting edge of screening, becoming the first mammography screening programme worldwide to go fully digital, and contributing to an improved survival rate for Irish women diagnosed with breast cancer. However, as Prof Flanagan outlined at the breast density seminar in June 2019, mammography is not a perfect tool, particularly in women with dense breasts. Not all women benefit from mammography equally, and women with dense breasts tend to have lower sensitivity in mammogram screening, higher recall rates, higher false positive rates, higher interval cancer rates, and more advanced tumours at diagnosis. Currently, women in Ireland going through the BreastCheck mammographic screening programme are not routinely informed about their breast density when receiving their results, and many are unaware of this important risk-factor.
While it is clear that the benefits of mammography as a screening tool are reduced in women with dense breasts, there are some challenges involved in introducing density notification to the screening system in Ireland. Breast density is currently assessed by a radiologist by eye. This is not an exact science and can lead to inconsistencies in density evaluation. Researchers in Ireland and worldwide are currently developing automated systems for scoring breast density, which may provide opportunities in the future for standardised assessment. The introduction of density notification in the Irish screening system would also necessitate a huge upgrade both to the system infrastructure and the resources required – if women are notified that they have dense breasts, they would ideally be offered supplementary screening by MRI or ultrasound, putting further pressure on the screening programme.
Patient advocate Siobhán Freeney set up the breast density awareness group Being Dense (www.beingdense.com) in 2016, having read an article by the late Dr Nancy Cappello. Dr Capello began advocating for breast density notification in the US in 2004 following her own breast cancer diagnosis. Being Dense promotes awareness about breast density and the relevance of supplementary screening by MRI or ultrasound for women with dense breasts.
References on request