Surveillance guidelines
Due to the lack of evidence-based data on the most appropriate protocols for breast surveillance of BC patients, published practice guidelines give somewhat conflicting information. No consensus exists about how frequently patients should undergo mammography and clinical breast examination (if any), for how long they should be seen at short intervals, and whether or when it is appropriate for them to return to normal screening [
2,
10,
21,
22].
For example, the United Kingdom NHS National Institute for Health and Clinical Excellence guidelines of 2009 recommend that patients with early BC, including ductal in situ carcinoma, be offered annual mammography until they enter the national screening programme. Patients who are already eligible for screening should have annual mammography for 5 years. For patients who reach the screening age as well as those who complete 5 years of annual mammographic surveillance, it is recommended that the screening frequency be stratified according to the patient risk category [
20].
The British Association of Surgical Oncology recommends that patients be followed-up for 5 years, although this period may vary with local protocols. As far as the frequency of mammography is concerned, the Association refers to the guidelines from the Royal College of Radiologists which suggest 1- to 2-year intervals for up to 10 years [
23].
The 1998 clinical practice guidelines for the care of BC from the Canadian Steering Committee [
24], as well as the 2005 guideline update [
25], recommend that mammography and clinical breast examination be more frequent in the first few years and be performed indefinitely at approximately 1-year intervals, although the frequency may be adjusted according to individual patient's needs.
The 2006 update of the American Society of Clinical Oncology guidelines recommend that all patients have a careful history and physical examination every 3 to 6 months for the first 3 years, every 6 to 12 months for years 4 and 5, and annually thereafter, and that a mammogram be obtained at least 6 months after completion of radiation therapy and every year thereafter [
26]. It is of note that no advice is given regarding patient discharge to primary care, whereas previous guidelines of 1999 suggested that this should occur after 10 years [
12].
Finally, the European guidelines for mammography screening recommend that screen-detected BC patients have annual mammography [
27] or periodical (not otherwise specified) physical examination and mammography [
28]. In the section dedicated to annual surgical follow-up [
29], it is stated that bilateral mammograms should be obtained and should have a technical quality equivalent to those provided in the organised screening programme. One could suggest that follow-up mammograms of exactly the same quality as screening mammograms can easily be found in screening centres. Unfortunately, the text does not address where follow-up mammography should be performed nor for how long. However, in the section that deals with the definition of the population eligible for screening [
30], it reads that patients with previous BC (probably including both screen-detected and prevalent BC patients)
" ... may ..." be excluded from the programme. In other words, and contrary to a widespread belief, the exclusion of BC survivors from organised screening programmes is presented as an option, not a necessity.
A Table in the Additional file
1 gives a summary of the above surveillance guidelines.
One key point should be noted here. Most guidelines place emphasis on providing frequent mammography and clinical examination in the first 3 to 5 years after diagnosis, and suggest that follow-up may be less frequent or even terminated thereafter [
12,
18]. This is not in accordance with the temporal pattern of the risk of relapse. The pooled incidence of true local recurrence, new BC in other areas of the ipsilateral breast, and new contralateral BC is constant for at least 15 years [
11,
12]. By implication, if any breast surveillance is to be offered, there is no obvious reason why it should be made less frequent or even discontinued after a few years [
18]. Current guidelines seem to be based on the assumption that there exists an acute high-risk phase of follow-up. In fact, this is only the case for the detection of scar recurrences and distant metastases. It clearly appears that most current guidelines are influenced, if not dictated, by surgical and oncological priorities. By implication, they have limited or no relevance in terms of long-term surveillance for the early detection of new ispilateral and contralateral diseases.
This explains why most guidelines pay so much attention to the frequency of follow-up mammography and so little to the complementary question of which health setting is most appropriate. The option of providing mammographic surveillance through organised screening programmes is simply not taken into consideration.
Surveillance practices in organised screening programmes
The inconsistency of practice guidelines, their emphasis on the early post-treatment phase, and their ambiguity and omissions are among the major reasons why BC survivors are permanently or temporarily or partially excluded from many - but not all - organised screening programmes worldwide. It is important to note that there is also a lack of consistency and clarity in policies and procedures within the same countries.
In 1999, a survey of organised screening programmes and mammography registries in 25 member countries of the International Breast Cancer Screening Network was performed. The objective was to determine what specific data relating to screening monitoring were collected. Information on whether women attending screening had a history of BC was available from 15 countries. Most of the national representatives of these countries stated that women with a history of BC were specifically not invited to screening [
31].
In Sweden, for example, women who ask to be excluded and women with previous BC are not invited, and this is also the case for women with newly diagnosed, screen-detected cancer [
32]. In Poland [
33] and Hungary [
34], non-eligible women are defined as those who have undergone surgical treatment for BC. In Switzerland, gynaecologists and radiologists have agreed that BC survivors need closer medical follow-up outside organised screening programmes [
35]. In Canada, the target population at the national level is defined as asymptomatic women aged 50-69 years with no prior diagnosis of BC [
36]. Recently, some provincial programmes have started to include BC survivors. These women, however, are excluded from routine statistical reports for the first 5 years after diagnosis [
37].
The Italian Group for Mammography Screening guidelines do not establish a management standard for women with a history of BC. However, they state that only patients who are known with certainty to be followed-up in the clinical setting can be excluded from an organised screening programme, and that the exclusion criteria should be clearly defined during the planning phase [
38]. As the planning and implementation of screening are decentralized to the health care district level, approaches differ in many respects [
39‐
41]. In general, BC survivors are indefinitely excluded. In situations where identifying patients is impossible or impractical, the invitation letter asks them to inform the screening service of their status, with exclusion occurring post-hoc. In the health care district of Rome, those patients who are being followed-up clinically are asked to send a signed declaration to the screening service in which they state their willingness to be excluded [
42].
There are important organised programmes that use different, if not opposite, approaches. For example, it is not the policy of the United Kingdom NHS Breast Screening Programme to exclude BC survivors. The only women who are discontinued from the programme are those with bilateral mastectomy, those who are considered by general practitioners to be medically unsuitable due to terminal illness, and those who have explicitly asked not to be invited. In all other circumstances, women have the opportunity of making an informed choice about whether or not to accept a screening invitation on each and every occasion it is made [
43]. The United Kingdom Society of Radiographers has suggested that patients with a recent history might be followed-up outside the programme [
44].
In Denmark, all women aged between 50 and 69 years at the start of an invitation round are invited to screening unless they have specifically asked to be excluded [
45].
In Australia, the national mammography screening programme selects women on the basis of age alone (50-69 years) because "...
The basis on which women are eligible for screening must be clear and unequivocal". On the website of the BreastScreen Australia programme, it also reads that "
Generally there is no dispute about a women's age..." and that "
Any policy is only effective if it is able to be implemented" [
46]. This is obvious, but so often forgotten.
The screening management of BC survivors is not a binary option,
i.e. exclusion or inclusion. It must be borne in mind that organised screening programmes no longer have a static design. It may happen that BC survivors are systematically invited and managed with non-standard procedures. One of these is the so-called early or short-interval rescreen, an opportunity that may be extended to inadvertently invited BC survivors (when their status is unknown to the service). Early rescreen may be used for several other conditions including poor technical quality of mammograms, high levels of mammographic breast density, hormone replacement therapy use, and family history of BC [
47‐
49]. Where and with what frequency early rescreen is practiced, however, is hard to establish. Information is sparse. In Australia, the reasons for offering screening more often than every 2 years include a previous diagnosis of in situ carcinoma [
50]. In northern Italy, BC survivors inadvertently invited are among the 0.6% of total screening participants who are placed on early rescreen [
49]. A survey of NHS mammography screening centres [
47] showed that 1.1% of total screening participants underwent early rescreen and that 21% of these were made up of BC survivors. In Canada, the total estimated rate of early rescreen is above 10% [
37].
Another hybrid model of BC survivor management, which combines screening with a clinical service, has been developed in the health care district of Rome. Apart from some exceptions, all women receive a personal invitation to participate in the organised screening programme. BC patients who are not followed-up elsewhere are asked to contact the screening centre to arrange a follow-up visit that is equivalent to an assessment for a positive mammography result, after which they are referred for specialist oncology consultation [
42].
These are only a few examples of the solutions that have been adopted for the provision of mammographic surveillance of BC survivors. Given the fact that organised screening activities are highly decentralised in most countries, it can be assumed that the way in which BC survivors are managed varies to a far greater extent than is suggested by the information available.
Surveillance practices in the clinical setting
Variability in approaches to breast surveillance of BC survivors is not confined to organised screening, but is also observed in the clinical setting [
20]. A survey of a random sample of symptomatic breast imaging units in England documented an impressive disparity of protocols [
51]. The same patient, depending on her place of residence, can be discharged to the national screening programme immediately after treatment or have an annual mammography for 10 years, although most units recommend a 5-year duration of follow-up. In certain units, the schedule for patients treated with breast conserving surgery is based on a different frequency of mammography for the ipsilateral and contralateral breast. Such a scheme has a high risk of being misunderstood by patients, as suggested by the fact that they are more aware of the risk of ipsilateral relapse [
18], and of being mismanaged by the follow-up service.
In England and elsewhere, the degree of within- and between-centre variation in follow-up strategies has probably increased over time because of a qualitative change in the service workload. In all populations targeted by screening, the growth in the number of BC patients to be followed-up is associated with a broadening range of clinical situations due to the fact that the presence of screen-detected cancers amplifies the heterogeneity of biological characteristics of the disease [
52]. This, in turn, may widen the spectrum of proposed follow-up schedules. Some breast imaging units use one or more tumour-related risk factors for local recurrence to modulate the duration and frequency of examinations [
51].