Elsevier

The Lancet

Volume 368, Issue 9552, 9–15 December 2006, Pages 2053-2060
The Lancet

Articles
Effect of mammographic screening from age 40 years on breast cancer mortality at 10 years' follow-up: a randomised controlled trial

https://doi.org/10.1016/S0140-6736(06)69834-6Get rights and content

Summary

Background

The efficacy of screening by mammography has been shown in randomised controlled trials in women aged 50 years and older, but is less clear in younger women. A meta-analysis of all previous trials showed a 15% mortality reduction in invited women aged 40–49 years at study entry, but this finding could be due in part to screening of women after age 50 years. The Age trial was designed to study the effect on mortality of inviting women for annual mammography from age 40 years.

Methods

160 921 women aged 39–41 years were randomly assigned in the ratio 1:2 to an intervention group of annual mammography to age 48 years or to a control group of usual medical care. The trial was undertaken in 23 NHS breast-screening units in England, Wales, and Scotland. The primary analysis was based on the intention-to-treat principle and compared mortality rates in the two groups at 10 years' follow-up. This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN24647151.

Findings

At a mean follow-up of 10·7 years there was a reduction in breast-cancer mortality in the intervention group, in relative and absolute terms, which did not reach statistical significance (relative risk 0·83 [95% CI 0·66–1·04], p=0·11; absolute risk reduction 0·40 per 1000 women invited to screening [95% CI −0·07 to 0·87]). Mortality reduction adjusted for non-compliance in women actually screened was estimated as 24% (RR 0·76, 95% CI 0·51–1·01).

Interpretation

Although the reduction in breast-cancer mortality observed in this trial is not significant, it is consistent with results of other trials of mammography alone in this age-group. Future decisions on screening policy should be informed by further follow-up from this trial and should take account of possible costs and harms as well as benefits.

Introduction

Screening women from age 50 years by mammography has been shown in randomised controlled trials to reduce mortality from breast cancer by around 25% in those offered screening.1 Although efficacy of mammography in women younger than 50 years is less certain, evidence from long-term follow-up of some randomised controlled trials has increasingly suggested a benefit of screening in this age-group. The Malmo Mammographic Screening Trial, which included two cohorts of women aged 45–49 years and 43–49 years at entry, invited for screening by mammography at 18–24 month intervals, showed a significant 36% reduction in breast-cancer mortality in the combined intervention groups at an average follow-up of 15·5 years and 10 years for the two cohorts in this age-group;2 however, when the cohorts were analysed separately with a different model the results were non-significant.1 The Gothenburg trial,3 which invited women aged 39–49 years at entry for mammography at an 18 month interval, showed a significant 44% reduction in the intervention group at 14 years of follow-up. Table 11, 3, 4, 5, 6, 7 shows the details of previous randomised trials. Various meta-analyses of these trials have been undertaken;8, 9, 10, 11 one that included trials of screening by mammography alone estimated a 19% reduction in breast-cancer mortality in women aged 40–49 years at entry (relative risk 0·81, 95% CI 0·65–1·01),12 similar to the findings of an updated Swedish overview, which did not include the Kopparberg group of the Two County study.1 A meta-analysis13 of all trials showed a significant 15% reduction in breast-cancer mortality (0·85, 0·73–0·98) in women aged 40–49 years at entry who were invited to screening. Many countries have now introduced population-based screening, and whereas most programmes include women from age 50 years, several include younger women too.

Most previous trials have not been designed specifically to study the effect of screening in younger women, and where women younger than 50 years at study entry have been included, to what extent any benefit in these women was due to screening after they reached age 50 years is unclear. A trial in Canada, in which women aged 40–49 years were randomly assigned either annual mammography and physical examination or usual care after an initial physical examination, with all women being taught self examination of their breasts, showed no effect of mammography at 13 years of follow-up, although the confidence intervals were wide7 and concerns have been expressed about the quality of mammography in this trial and the use of a volunteer population.14

The Age trial was designed specifically to overcome these issues by studying the effect of annual invitation to mammography starting at age 40 years, compared with an uninvited control group. The women in the control group will receive their first invitation between the ages of 50 years and 52 years, as is policy in the NHS breast-screening programme. In 2005, we published results on the predicted reduction in breast-cancer mortality in the intervention group based on surrogate outcomes measures15 using the pathological characteristics of cancers in both groups to calculate three prognostic indices: the Nottingham Prognostic Index (NPI)16 and indices developed from the Swedish Two County Study17 and the Edinburgh randomised trial of breast cancer screening.18 These indices were used to calculate the predicted number of women dying from breast cancer within 10 years of date of entry in each group of the trial. We present here the first results for observed breast-cancer mortality.

Section snippets

Patients and procedures

The design of the study has been described in detail elsewhere.19 Briefly, 160 921 women aged 39–41 years were randomly assigned between 1991 and 1997 to either an intervention group or a control group in a ratio of 1:2. The trial was undertaken in 23 NHS breast-screening units in England, Wales, and Scotland. Women were identified from lists of patients of general (family) practitioners (GPs) held on local Health Authority databases, and individual randomisation was carried out stratified by

Results

A total of 160 921 women were randomised into the trial; more than 99·9% of these have been successfully flagged at NHSCR. The outcomes of screening in the first 10 years of the trial have been described in detail elsewhere.20 Uptake of screening was 68% at the first (prevalent) screen and 69–70% in those re-invited. Overall, 81% of women attended at least one routine screen; the mean number of screens per woman was 4·5, or 5·6 for those attending at least one screen. Detection rates of breast

Discussion

This trial did not find a significant reduction in breast cancer mortality in women offered annual screening between the ages of 40 years and 48 years. The trial was designed specifically to look at the effect on breast-cancer mortality of inviting women to screening from age 40 years compared with invitation from age 50 years as in the current NHS breast-screening programme. This approach was used to avoid the problem present in previous trials in which some women reached age 50 years shortly

References (37)

  • N Bjurstam et al.

    The Gothenburg Breast Screening Trial

    Am Can Soc

    (2003)
  • S Shapiro et al.

    Periodic screening for breast cancer

    (1988)
  • AB Miller et al.

    The Canadian National Breast Screening Study 1: breast cancer mortality after 11 to 16 years of follow-up

    Ann Intern Med

    (2002)
  • RE Hendrick et al.

    Benefit of screening mammography in women aged 40–49: a new meta-analysis of randomized controlled trials

    Monogr Natl Cancer Inst

    (1997)
  • CR Smart et al.

    Benefit of mammography screening in women ages 40 to 49 years: current evidence from randomized controlled trials

    Cancer

    (1995)
  • K Kerlikowske et al.

    Efficacy of screening mammography: a meta-analysis

    JAMA

    (1995)
  • Breast cancer screening

    (2002)
  • DB Kopans et al.

    The Canadian National Breast Screening Study: a critical review

    Am J Radiol

    (1993)
  • Cited by (419)

    View all citing articles on Scopus

    Members of the Trial Management Group are listed at the end of the report

    View full text