NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Robertson C, Arcot Ragupathy SK, Boachie C, et al. The Clinical Effectiveness and Cost-Effectiveness of Different Surveillance Mammography Regimens After the Treatment for Primary Breast Cancer: Systematic Reviews, Registry Database Analyses and Economic Evaluation. Southampton (UK): NIHR Journals Library; 2011 Sep. (Health Technology Assessment, No. 15.34.)

Cover of The Clinical Effectiveness and Cost-Effectiveness of Different Surveillance Mammography Regimens After the Treatment for Primary Breast Cancer: Systematic Reviews, Registry Database Analyses and Economic Evaluation

The Clinical Effectiveness and Cost-Effectiveness of Different Surveillance Mammography Regimens After the Treatment for Primary Breast Cancer: Systematic Reviews, Registry Database Analyses and Economic Evaluation.

Show details

Appendix 28Review of management guidelines for breast cancer


In order to assess the cost-effectiveness of different surveillance mammography regimens after the treatment for primary breast cancer, the clinical effectiveness of standard treatment options of any subsequent cancers were required for the economic model.

The objective of this review is to determine the effect of treatment options on survival in patients who are identified as having breast cancer during surveillance. After searching relevant guidelines, the recent updated NICE guideline Early and Locally Advanced Breast Cancer, published in February 2009, was identified as providing the best available evidence of treatments for early breast cancer relevant to the UK.24 This review, therefore, is mainly based on this guideline and the source data used to inform the guideline: the EBCTCG83 and Adjuvant! Online computer program.84 Using these sources, estimates of survival following various treatment options were prepared using the Adjuvant! Online computer program due to its flexibility.


At the beginning of this review, initial scoping searches were carried out into identify relevant local, national or international guidelines. Eleven guidelines were identified describing various treatments or managements of primary breast cancer.2125,28,111115 The most recent of these, and most applicable to the UK, were the NICE guidelines published in February 2009.24 Few data were available on the effectiveness of treatments for cancers identified following treatment for a primary cancer. As a consequence, it was judged that, in the absence of data in the literature directly relevant to our study question, the best source of information would relate to treatment of primary cancer. It was judged that, of the guidelines available, the best available summary of existing evidence on the clinical effectiveness of treatments of breast cancer, including early, locally advanced and advanced disease, was provided by the NICE guidelines. The literature searches used to inform these guidelines considered papers published up to July 2008.

NICE guideline: Early and Locally Advanced Breast Cancer

The NICE guideline Early and Locally Advanced Breast Cancer,24 published in February 2009, updated and developed guidance from three NICE technology appraisals: 109 (docetaxel), 108 (paclitaxel) and 107 (trastuzumab).99,116,117 The evidence on clinical effectiveness of diagnoses and treatments for early and locally advanced breast cancer is based on the systematic review of relevant clinical literatures and critical appraisal.

Search strategy

Papers that were published or accepted for publication in peer-reviewed journals were considered as relevant. Search filters, such as those to identify systematic reviews and RCTs, were applied to the search strategies when there was a wealth of these types of studies. No language restrictions were applied to the search; however, foreign language papers were not requested or reviewed (unless of particular importance to the question). Any evidence published before July 2008 was included. The following databases were included in the literature search:

  • The Cochrane Library
  • MEDLINE and PREMEDLINE 1950 onwards
  • Excerpta Medica (EMBASE) 1980 onwards
  • Cumulative Index to Nursing and Allied Health Literature (CINAHL) 1982 onwards
  • Allied & Complementary Medicine (AMED) 1985 onwards
  • British Nursing Index (BNI) 1994 onwards
  • PsycINFO 1806 onwards
  • Web of Science 1970 onwards [specifically SCI Expanded and Social Sciences Citation Index (SSCI)]
  • System for Information on Grey Literature In Europe (SIGLE) 1980–2005
  • BioMed Central 1997 onwards
  • NRR
  • CCT.

Types of studies included in the guideline

Relevant guidelines, systematic reviews and RCTs of different treatments for early or locally advanced breast cancer. In the absence of RCT evidence, the observational studies were considered in the review such as cohort, case–controls, etc.

Types of interventions considered in the guideline

Table 77 summarises the treatment interventions included in the NICE guideline. Surgery is considered as the first line of treatment in primary breast cancer, such as mastectomy, BCT and surgery to the axilla. Adjuvant therapies were used in management of breast cancer after the surgery, including hormonal therapy, chemotherapy, biological therapy and radiotherapy.

TABLE 77. Treatments considered in the NICE guideline.


Treatments considered in the NICE guideline.

Critical appraisal of studies included in the NICE guideline

One researcher independently scanned the titles and abstracts of every article. Full texts were obtained for any papers that were considered potentially relevant or where there was insufficient information. The researcher then applied the inclusion/exclusion criteria to determine which studies were relevant. Included papers were critically appraised and data extracted. Quality assessment was based on the SIGN criteria.


The NICE guideline overviewed the best clinical evidence of treatment options derived from the studies that it reviewed and appraised.24 In addition to BCS or mastectomy, adjuvant treatments are used and the selection of adjuvant therapies depends on tumour factors (ER and HER2 status) and patient characteristics (age and menopausal status). However, the NICE guideline did not report the detailed data but rather made a recommendation on the use of adjuvant treatments. The guideline suggested that two sources were more likely to provide reliable data on the effectiveness of adjuvant treatments: (1) a series of overviews and meta-analyses of different treatments in the EBCTCG that provides the effectiveness of adjuvant treatment after surgical therapy for early breast cancer, and (2) Adjuvant! Online, based on a US population, to estimate the efficacy of adjuvant therapies (hormonal or chemotherapy) after initial surgical treatment.

Effect of adjuvant treatments on survival based on overview in EBCTCG

The NICE guideline suggested that overviews of meta-analyses in EBCTCG appear to provide the best evidence for estimating risk for treating breast cancer in the UK. The EBCTCG was established in 1984. The most recent publications from the EBCTCG relate to:

  • adjuvant polychemotherapy in ER-poor breast cancer: meta-analysis of individual patient data from the randomised trials118
  • chemotherapy and hormonal therapy for early breast cancer: effects on recurrence and 15-year survival in an overview of the randomised trials3
  • effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials.2

Some adjuvant treatments tested in the 1980s have clear evidence that they substantially reduce 5-year recurrence rates and also substantially reduce 15-year overall mortality rates (e.g. tamoxifen, polychemotherapy regimens and radiotherapy). Further improvements in long-term survival could be available from newer drugs, or better use of older drugs. The reviews, however, although excellent, do not provide data readily useable in an economic model, as it is not possible to estimate survivals for specific types of cancer. However, for completeness a summary of key findings is presented below.

Adjuvant chemotherapy

The number of randomised trials of chemotherapy increased substantially over the first two decades of the EBCTCG overview, with a shift from trials comparing chemotherapy with no chemotherapy, to trials of different types of chemotherapy. In the first cycle of the overview, 31 randomised trials of no chemotherapy versus chemotherapy using one or more drugs were included. This involved a total of 9000 women, of whom 2900 had died. Subsequent cycles refined this to focus on trials of prolonged multiagent chemotherapy. By the third cycle of the EBCTCG overview, this analysis was based on 18,000 women from 47 trials, and it had risen further to 60 trials (29,000 women and 10,000 deaths) by the fourth cycle. The recent meta-analysis from randomised trials is to assess the long-term effects of adjuvant polychemotherapy regimens in ER-poor breast cancer, and the extent to which these effects are modified by age or tamoxifen use.118 This study analysed 6000 patients with ER-poor breast cancer in 46 trials of polychemotherapy against not, and about 14,000 women with ER-poor breast cancer in 50 trials of tamoxifen versus not.

Adjuvant hormonal therapy

The first EBCTCG review of tamoxifen, in the mid-1980s, included data from a total of 16,500 women in 28 randomised trials, of whom nearly 3800 were known to have died. This grew to 40 trials (30,000 women and 8200 deaths) in the second cycle; and then to 55 trials, involving > 12,000 deaths among 37,000 women. By the 2000–5 cycle of the EBCTCG overview, the growth in the number of trials included had levelled off at 56, but further accrual to some of these trials and five more years of follow-up for many of them meant that the number of women in the analysis had increased to 48,000, with a total of 18,000 deaths.3 This represented 88% of the total number of women randomised into eligible trials of adjuvant tamoxifen versus no immediate tamoxifen worldwide.

Adjuvant radiotherapy

The most recent overview of radiotherapy and differences of surgery published in 20052 was based on the analysis of 42,000 women in 78 randomised treatment comparisons (radiotherapy vs no radiotherapy, 23,500; more surgery vs less surgery, 9300; more surgery vs radiotherapy, 9300). A total of 24 types of local treatment comparison were identified to help relate the effect on local (i.e. locoregional) recurrence to that on breast cancer mortality.

Effect of treatments on survival for breast cancer using the Adjuvant! Online computer program

Adjuvant! Online is a tool of assessment of risk of an individual patient developing recurrent disease and/or dying within 10 years. Adjuvant! Online draws information from mortality statistics in the USA, the SEER database, and meta-analyses and individual clinical trials. Based on well-validated factors, such as age, menopausal status, ER status, tumour size and grade, nodes status, etc., predictions can be made about survival for alternative adjuvant treatment regimens, such as chemotherapy, endocrine, etc. However, survival estimates are derived from the US population. Version 8 of this tool may underestimate the risk of mortality and the benefit of trastuzumab in HER-2-positive patients. Table 78 describes the information used to predict recurrence and mortality.

TABLE 78. Prognostic factors in Adjuvant! Online.


Prognostic factors in Adjuvant! Online.

The NICE guideline summarised the following issues in its critical appraisal of this tool:

  • The predictions made by Adjuvant! Online are based on the published methodology, which has been updated periodically as evidence of treatment effectiveness and data on risk factors become available.
  • Help files and published descriptions of the tool make clear some of the assumptions and limitations that underpin the methodology. The impact of these individual assumptions is difficult to assess. Adjuvant! Online deals with key uncertainties by alerting the user to them at relevant points.
  • Survival estimates are derived from the US population. Quantifying the impact on survival of socioeconomic background and of ethnic differences between US and UK populations is difficult.
  • Adjuvant! Online is already used in the UK and is designed to incorporate the Oxford overview meta-analyses.
  • The strongest evidence of Adjuvant! Online validity for the UK is derived from comparisons between predictions and observed outcomes using a Canadian population. This study found its predications to be reliable for most groups. Further validation is under way using a European population.

Survival estimates for treatment options using the Adjuvant! Online computer program

Adjuvant! Online integrates patient-related information (age and comorbidity) and tumour-related information (nodal status, tumour size, histological grade, ER status and histological subtype) to make estimates of mortality caused by cancer or from other causes. Table 79 reports an example of mortality estimates for a woman with breast cancer depending upon her characteristics and tumour factors when the woman is 40 years old, has perfect health, tumour grade is undefined and oestrogen status is positive.

TABLE 79. Mortality rate when age = 40 years.


Mortality rate when age = 40 years.


Decisions on the treatment for every woman with breast cancer should be based on the best evidence. This requires a combination of information about the patient and tumour along with evidence on the effectiveness of the treatments being considered. This evidence needs to be as reliable as possible. It was judged by the research team that the NICE guideline Early and Locally Advanced Breast Cancer provided the best available summary of evidence of breast cancer treatments. However, the NICE guideline did not report estimates of clinical effectiveness of treatment options. Based on the NICE guideline, there are two reliable sources that produce the estimates of effectiveness of adjuvant treatments: EBCTCG overviews and Adjuvant! Online. Of these Adjuvant! Online was sufficiently flexible to allow relevant data to be estimated for the economic model. Estimates from Adjuvant! Online provide similar/dissimilar estimates to the EBCTCG overviews, which, arguably, are more applicable to a UK population. However, as EBCTCG overview subset analyses are carried out largely as a set of univariate subset analyses it is impossible to tell if two univariate effects are independent.

Adjuvant! Online integrates patient-related information (age and comorbidity) and tumour-related information (nodal status, tumour size, histological grade, ER status and histological subtype) to make estimates of mortality and recurrence. However, survival estimates are derived from the US population.

© 2011, Crown Copyright.

Included under terms of UK Non-commercial Government License.

Bookshelf ID: NBK100086
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (4.4M)

Other titles in this collection

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...