Breast cancer screening: what does the future hold?BMJ 2013; 346 doi: https://doi.org/10.1136/bmj.f87 (Published 23 January 2013) Cite this as: BMJ 2013;346:f87
- Cliona C Kirwan, National Institute for Health Research clinician scientist in surgical oncology
- 1Institute of Cancer, University of Manchester, Department of Academic Surgery, Education and Research Centre, University Hospital of South Manchester, Manchester M23 9LT, UK
The role of national breast screening programmes and the quality and transparency of information given to participating women are increasingly the subject of heated debate. In the past 12 months alone, the BMJ, the Lancet, and the New England Journal of Medicine have published 24 articles or communications debating the value of breast cancer screening. After calls for an impartial review of the value of breast screening in the United Kingdom, the findings of an independent panel of experts, led by Professor Marmot, were published in November 2012.1
Currently in the UK, women aged 50-70 years are invited for screening every three years; 2.3 million women were invited during 2010-11. The rate of uptake currently stands at 73.4%, having steadily increased in the past decade.2
The primary aim of screening is to reduce mortality from breast cancer. Reduced breast cancer related mortality is balanced against the cost of screening in terms of physical and psychological harm to women and the financial impact on health services.
Much recent debate has concerned overdiagnosis—that is, diagnosis of a condition that would never cause symptoms or death during a patient’s lifetime. Although over-investigation can cause harm (pain and anxiety from mammography and biopsies), this is usually transitory. The harm of overdiagnosis (anxiety associated with the “cancer” label) and subsequent overtreatment lasts a lifetime. Surgery is associated with anaesthetic risks, surgical complications, and reduced cosmesis. Short term side effects of adjuvant treatments include alopecia, neutropenic sepsis, hot flushes, vaginal dryness, and increased risk of fracture from endocrine treatment. Long term consequences include cardiovascular and respiratory complications (associated with radiotherapy), reduced quality of life,3 and increased treatment induced secondary cancers.
It is difficult to quantify mortality benefit and overdiagnosis. Analyses are based on data from historical randomised controlled trials that used outmoded screening and treatment techniques,4 5 6 7 or estimates are made by extrapolating population data.8 9 Both these methods have limitations. Meta-analyses of trials are complicated by the heterogeneous methods used, such as different age ranges of women screened, screening intervals, and length of follow-up. Data extrapolation makes assumptions about baseline risk of cancer, the proportion of women who participate in screening, and length of lead time (the time between screening and diagnosis).
Marmot and colleagues acknowledged the limitations of their review, which depended on randomised controlled trials performed 20-50 years ago and observational studies with potential biases.1 Nonetheless, on the basis of their estimate of one death from breast cancer avoided for every 235 women invited to screening, they concluded that the UK breast screening programme should continue. However, they also reported that “for every breast cancer death prevented, approximately three overdiagnosed cases will be identified and treated.” Although the review made no comment on all cause mortality, the authors highlighted the need for clear communication of both harms and benefits to women.
The Marmot review provides independent evidence of a 20% relative reduction in death from breast cancer in women invited for breast cancer screening, a finding consistent with other meta-analyses.4 5 6 7 Mortality from breast cancer is decreasing—in the UK, European age standardised mortality rates have dropped from 40.1 per 100 000 in 1990 to 24.4 per 100 000 in 2010.10 Although this reduction is partly due to screening, improvements in adjuvant treatments such as radiotherapy, endocrine therapy, chemotherapy, and trastuzumab have also had a major part to play. The potential for screening to affect mortality is therefore reducing.
Marmot estimated that almost a fifth of breast cancers discovered through screening are overdiagnoses. Previous estimates ranged from 1.7% to 54%.11 This large range reflects the lack of data and the complexity of this type of analysis. Improved screening techniques, such as tomosynthesis, may lead to an increase in identification of smaller inconsequential cancers in the future. However, increased availability of oncoplastic surgery and more accurate radiotherapy should reduce the harms associated with treatment. In addition, improved understanding of the heterogeneity of breast cancer biology may permit identification of those cancers most likely to remain asymptomatic in a patient’s lifetime, thus allowing more conservative treatment for lower risk cancers. A proposed “Low Risk Ductal Carcinoma In Situ Trial” may identify in situ cancers that can be managed conservatively. Oncotype DX and MammaPrint are multigene assays used to identify very low risk cancers for which chemotherapy is unlikely to confer any additional benefit. Trials of partial breast radiotherapy may further reduce toxicity in low risk patients.
The Marmot report currently provides the best evidence on the benefits and harms of breast screening. Given the limitations of the available evidence, the authors correctly conclude that breast screening is worth while. However, it highlights a higher rate of overdiagnosis than has hitherto been communicated to women. The level of risk that is acceptable will vary for the individual woman, and the decision on whether to participate in screening is a personal one. As professionals, we need to help women weigh the benefits and harms of screening by providing them with up to date and transparent information, including information on the uncertainties.
Ongoing analysis of the efficacy of screening must depend on meticulous collection of population data and subsequent modelling. Long term evaluations of randomised controlled trials take decades, so outcome data reflect outmoded treatments. Such trials have a role only in assessing new screening approaches. The impact of overdiagnosis and overtreatment on morbidity, overall mortality, and quality of life in the context of current treatment regimens needs quantitative assessment. With increasing identification of risk factors for developing breast cancer, such as mammographic density,12 the frequency of screening could be adjusted according to each woman’s risk category.
In light of the rate of overdiagnosis associated with breast screening, and the survival improvements in symptomatic cancers, financial review of breast screening programmes is warranted. However, the unquantifiable effects of screening, including increased breast awareness, must be taken into account. Perhaps the greatest priority for now is to minimise the impact of overdiagnosis by improving individualised treatment to reduce overtreatment.
Cite this as: BMJ 2013;346:f87
Competing interests: The author has completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Provenance and peer review: Commissioned; externally peer reviewed.