At the CCP, Professor Stephen Duffy is continuing his exceptional research track record in cancer epidemiology, screening and policy.
His new publication reports the results of the TOMMY trial that tested the effectiveness of adding a 3D imaging technique to standard mammography.
8 August 2017
In a standard breast screening radiograph (mammogram), overlapping breast tissue may hide some cancers (false negatives) or make normal tissue appear abnormal (false positives).
A newer technique called digital breast tomosynthesis (DBT) takes a series of low-dose radiographs of the breast that are processed by a computer to construct a 3-dimensional DBT image. This allows abnormalities in the breast be seen more clearly and could make it easier to see small cancers and decrease the risk of ‘false alarms’.
DBT images are usually read together with a standard two-dimensional (2D) mammogram. However, this undesirable double radiation exposure could be avoided if an equivalent 2D mammogram could be created from the individual DBT images.
The study compared the accuracy of reading the following screening combination techniques to identify breast cancer:
(1) a 2D mammogram
(2) a 2D mammogram together with a DBT and
(3) a synthetic 2D mammogram with a DBT.
Data from 7061 cases were analysed: 6021 cases from women (47–73 years) recalled after routine screening for further tests, and 1040 cases from women (40–49 years) with a family history of breast cancer attending their annual breast screenings.
This project was funded by the NIHR Health Technology Assessment programme
The TOMMY trial: a comparison of TOMosynthesis with digital MammographY in the UK NHS Breast Screening Programme--a multicentre retrospective reading study comparing the diagnostic performance of digital breast tomosynthesis and digital mammography with digital mammography alone.
Gilbert FJ, Tucker L, Gillan MG, Willsher P, Cooke J, Duncan KA, Michell MJ, Dobson HM6, Lim YY, Purushothaman H, Strudley C, Astley SM, Morrish O, Young KC, Duffy SW. Health Technol Assess. (2015) (4):i-xxv, 1-136. PMID: 25599513