ReviewScreening women at high risk of breast cancer on the basis of evidence
Introduction
Following a long media and journal debate on its merits and effectiveness, screening for breast cancer has become a part of the regular way of life in women from the age of 50 years in most prosperous countries. This screening is based on trial evidence, not only of overall effectiveness [1], [2], [3], [4], [5], [6], but also of the best way to conduct the screening [7], [8], [9], [10]. Film-screen mammography is the method employed, and for 20 years has not been surpassed by newer methods of imaging [11] or of blood or tissue analysis [12], despite substantial research effort. Within the last 5 years the genes that render women in some families unusually prone to breast cancer have been identified [13], [14], [15], [16]. Women with a strong family history of breast cancer can be tested to see if they carry a faulty copy of one of these genes [17]. Many more women have family histories that are not sufficiently strong to merit gene testing, but they are nevertheless at increased risk, and geneticists can predict the lifetime probability of their developing breast cancer [18].
Women at all levels of increased risk wish for a strategy to handle their susceptibility [19], [20]. Several approaches are on offer at present — prevention, for example with tamoxifen [21], [22], [23], [24]; early diagnosis by screening; surgery viz. prophylactic mastectomy with reconstruction to remove the tissue at risk of developing cancer [25], [26], [27]; or lifestyle alterations [28]. The risks and benefits of all of these approaches are the subject of research effort and this paper looks at the evidence that is relevant to the screening of women with an increased risk of breast cancer.
Screening aims to advance the diagnosis of such cancers in order that treatment can be offered when their prognostic factors favour survival [29]. This paper maps out the evidence needed to create a screening strategy specifically designed for women identified as being at increased risk, who commonly develop cancer at a younger age than the general population and who are, therefore, likely to have anxieties from their twenties, before routine screening is offered, and throughout their adult life [19].
Section snippets
Mammography
Although hitherto the best method available [19], mammography has some important disadvantages as a screening modality:
- •
it uses ionising radiation, which itself is a potential carcinogen [30], [31], this limits the age and frequency with which it can be used [32];
- •
the breast must be compressed in order to obtain high quality images [33];
- •
some cancers are not seen on mammography, particularly in radiographically dense breasts [34], [35], [36];
- •
interpretation is highly skilled [37], and has not yet
Women 40–49
Recent debate in the USA was generated by a National Institutes of Health Consensus Development Conference on breast cancer screening for women ages 40–49 [68]. It led to extensive discussion about the evidence from randomised controlled trials on the effectiveness of mammographic screening of the normal population in this age range [69]. All the trials that were evaluated had originally addressed the effectiveness of screening women in a much wider age band, and so the evaluation of women aged
The nature of the cancers in carriers of breast cancer susceptibility gene mutations
A large series of cancers in women carrying defects of the BRCA 1 and 2 genes have been reviewed by a panel of expert pathologists [102], [103]. It is now known from these studies that the pathological features of these cancers are different, and certain histological types are found to be more frequently associated with the gene mutations than in sporadic cancers. The relationship of these differences to mammographic morphological features and density is not known. It is possible that there are
What is known to date of mammography in women at high risk?
Mammography is being undertaken in women at higher risk than the normal population in an informal manner. It has not been evaluated by any trials. Information is available from clinics where outcomes are documented. These studies may carry biases that could have been avoided in a properly designed prospective study [104], [105]. The Manchester Cancer Family History Clinic has published its first figures [106]. Their high risk clinic offers mammography to women who have a one-in-three lifetime
Alternative imaging techniques that might be useful
Radiation risk and lower sensitivity as a result of breast density have been the impetus to promote various other breast imaging techniques for women at high risk, including ultrasound and magnetic resonance imaging [87]. The greater incidence justifies the greater expense and effort involved, although these techniques would not be suitable for screening the normal population.
Risk analysis and balance of risks
The impact and value of all these interventions will be assessed from statistical analysis and health economics. The ultimate recommendations for screening will depend heavily on the incidence of cancer in the particular population under consideration [104], [106]. In general, the false-positive cases are likely to be as many whatever the level of cancer risk. The importance of the problem of benign queries will, therefore, be determined by the cancer incidence in the population. For a
Research studies required
It is the duty of those who care for these women to offer interventions based on evidence so that women are not made anxious unnecessarily, falsely reassured or put at worthless risk. The discussion of the topics above shows how weak the evidence is for much of the screening that is or may be offered to women at high risk. More information about dose and the effectiveness of screening by mammography is needed stratified by age. Any new screening intervention should be subject to randomised
Discussion
With greater medical knowledge of the genetic predisposition to breast cancer and research on-going to expand knowledge further, there is public demand for screening strategies. From the evidence presented it can be seen that the problem of risk from radiation is more important at a younger age, and in those with genetic susceptibility to radiation damage. Since the risk arising from a family history of breast cancer is known from a young age, and may be important in the 30s and 40s, it is
Conclusion
The screening that is offered to women at high risk is based throughout on deficient evidence. Those who care for these women, and the governments who provide health services, must furnish the funded research to create a strategy properly founded on evidence. More information is needed on the radiation susceptibility of women at high risk, stratified for age, taking into account different gene mutations. The radiological features of the carcinomas found in those at high risk should be studied,
Acknowledgements
I am grateful to the following colleagues who have kindly contributed by editing the text of this review: Dr Peter Britton, Consultant Radiologist and Dr David Goodman, Senior Physicist, Addenbrooke's Hospital, Cambridge, Professor Michael Peters, Professor of Nuclear Medicine, Dr Lynda Bobrow, Lecturer in Histopathology, Dr Evis Sala, Research Associate in Epidemiology, Dr James Mackay, Lecturer in Clinical Genetics, Cambridge University, Dr Ros Eeles, Clinical Senior Lecturer and Honorary
References (133)
- et al.
14 Years of follow-up from the Edinburgh randomised trial of breast-cancer screening
Lancet
(1999) Role and evaluation of mammography and other imaging methods for breast cancer detection, diagnosis, and staging
Semin. Nucl. Med.
(1999)- et al.
Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial
Lancet
(1998) - et al.
Prevention of breast cancer with tamoxifen: preliminary findings from the Italian randomised trial among hysterectomised women. Italian Tamoxifen Prevention Study
Lancet
(1998) - et al.
Pain coping and the pain experience during mammography: a preliminary study
Pain
(1997) - et al.
Just how valuable is double reporting in screening mammography?
Clin. Radiol.
(1997) - et al.
Estimating mean glandular dose using proprietary mammography phantoms
Appl. Radiat. Isot.
(1999) - et al.
Early-onset breast cancer in A-bomb survivors
Lancet
(1993) - et al.
‘At risk’ for breast cancer
Lancet
(1997) - et al.
Breast cancer and radiotherapy in ataxia-telangiectais heterozygote
Lancet
(1996)