Letters Breast screening

Will ductal carcinoma in situ defeat breast cancer screening?

BMJ 2009; 338 doi: http://dx.doi.org/10.1136/bmj.b957 (Published 09 March 2009) Cite this as: BMJ 2009;338:b957
  1. John R Benson, consultant breast surgeon1
  1. 1Addenbrooke’s Hospital, Cambridge CB2 0QQ
  1. john.benson{at}addenbrookes.nhs.uk

    Gøtzsche and colleagues focus on the inadequacy of patient information leaflets in the NHS breast screening programme.1

    The failure to disclose information on ductal carcinoma in situ has aptly been described by a patient representative as a “closely guarded state secret.” Ductal carcinoma in situ constitutes up to a fifth of screen detected cancers but only 2-5% of symptomatic tumours before national screening.2 Its occurrence in 15-39% of routine autopsy studies suggests that some non-invasive lesions detected mammographically and subsequently excised would have been of no clinical consequence during their lifetime.3 4

    This non-obligate progression (especially of non-high grade lesions) poses a dilemma in clinical management of non-invasive disease.5 Once ductal carcinoma in situ has been detected, treatment must be offered, which mandates mastectomy in about 30% of cases and often radiotherapy in cases amenable to breast conservation. Undoubtedly some patients with low grade lesions undergo surgery for a condition which would never have progressed to life threatening disease. This so called pseudodisease represents overdiagnosis in the screening programme, which is effectively tapping into a reservoir of indolent non-progressive ductal carcinoma in situ.4

    Such overdiagnosis and inconsistent labelling of proliferative lesions discovered on biopsy after recall have undermined the net benefits of breast screening. A degree of psychosocial morbidity and prognostic liability has been generated which has been ignored and excluded from patient information leaflets. It must now be incorporated to satisfy the basic tenets of informed consent and conform with the principle of primum non nocere.


    .Cite this as: BMJ 2009;338:b957


    • Competing interests: None declared.