Mrs Marisa Jenkins
MNg MScMed (ClinEpi)
School of Public Health
|Telephone||+61 2 9036 7675|
Thesis title: Digital breast tomosynthesis (3D-mammography) in population breast screening.Supervisors: Nehmat HOUSSAMI
Screening for the early detection of breast cancer is currently performed in Australia using digital mammography (DM) technology which is consistent with current practice in developed countries . Screening has been estimated to reduce mortality due to breast cancer in the order of 15% - 25% . Despite this mortality benefit, there is ongoing debate about important limitations of DM. Critics suggest the benefits may be outweighed by harms such as false positives and negatives particularly in younger women , and there are also ongoing concerns around over-diagnosis . Some of these limitations may be ameliorated to some extent with emerging imaging technology such as Digital Breast Tomosynthesis (DBT) also known as three dimensional (3D) mammography  .
Currently, standard DM is based on two views (2D) which can hamper its ability to detect some cancers due to 'normal' tissue overlap masking lesions and the possibility of incorrectly diagnosing normal tissue as suspicious or abnormal for the same reason[3-5] . DBT has the advantage over DM as it produces better visualisation of breast tissue because it is examined as multiple thin slices which can lead to an increase in test accuracy by reducing both false negative and false positive results [3-7]. The ability of DM to accurately detect a cancer when one exists is less than optimal with sensitivity (true positive rate) estimated in one study to be between 41% and 72% and the ratio of false positive to true positive readings between 5:1 and 12:1.
Why is the issue important?
Any improvement in the effectiveness of screening methods from current levels is important in that it has the potential to increase cancer detection capability whilst reducing both the physical and psychological costs to women. Of particular concern is the psychological impact of a false positive reading which, initially for some women, has been reported to be no different to women who actually have breast cancer.  The psychological impact of a false positive reading may also persist for several years . It is important that new technologies are carefully evaluated for patient outcomes to ensure that the benefit to harm trade-off is favourable.
What are some of the knowledge gaps?
Rapidly emerging research into DBT instead of, or as an adjunct to, standard DM has comprised primarily of small retrospective observational studies showing promising results for DBT . More recently, interim results from three, and final results from one, large prospective population studies in the UK, Europe and Scandinavia [4, 7-9] are also consistently showing favourable detection measures for DBT. These more recent studies provide more rigorous evidence and are a platform to guide further research on the subject.
Despite the promising results outlined above, the current evidence is limited because there are no randomised studies of DBT v DM, no studies being conducted in the Australian context and there is a lack of information on screening outcomes (such as impact on interval cancers) and cost effectiveness. Currently, there is insufficient evidence of incremental screening benefit from adding DBT to standard DB to recommend the introduction of DBT into standard screening programs in Australia  or internationally . Also, adding DBT to standard (DB) mammography increases radiation to the breast, which may be unacceptable for population screening programs .
The purpose of this research is to add to the evidence base on the use of DBT particularly focusing on key evidence gaps [1,11] . In order to do this, the following components are planned:
1. A Systematic Review and meta-analysis of breast cancer detection by DBT vs standard mammography in population breast screening.
Given that the literature is growing rapidly in the field of DBT, my research will commence with a comprehensive study-level systematic review of the evidence, updating and extending that reported by Houssami . The evidence review will provide a systematic approach to further defining evidence gaps on DBT screening that are worthy of research and will also be the basis for identifying studies to be invited to contribute to a collaborative Individual Patient Data (IPD) Meta-analysis (see component 3 below).
2. Conduct original, primary research study based on an important evidence gap identified in the systematic review: I will explore, develop and undertake a study of DBT for population screening, envisaged to involve conducting either a:
o NSW-based pilot trial of DBT screening.
o Contributing to a collaborative international study using newer DBT screening technology that supports reconstructed 2D images (hence this technology would potentially reduce the radiation burden to the breast by eliminating the 2D acquisition).
3. Individual Patient Data (IPD) meta-analysis/project: Effectiveness of breast tomosynthesis in population screening.
Performing IPD meta-analysis is useful in overcoming limitations of the study level approach as it provides important covariate information at the individual participant level. Pooling the data from several studies will allow us to report more precise estimates of cancer detection rates, and should enable more precise estimation of interval cancer rates in DBT-screened women.
This work will involve planning and co-ordinating an international collaborative IPD database to conduct IPD meta-analysis addressing key evidence gaps on DBT screening; the research plan includes collecting screen-detection data, cancer characteristics, and potentially interval cancer data, from the prospective trials of DBT screening (identified in component 1 of my research plan). This work involves collating databases into a common database and liaising with several international teams, and can potentially inform global breast screening practice.
1. Lauby-Secretan, B., et al., Breast-Cancer Screening — Viewpoint of the IARC Working Group. NEJM, 2015. 372(24): p. 2353-2358.
2. Glasziou, P. and Houssami, N., The evidence base for breast cancer screening. Prev Med, 2011. 53(3): p. 100-102.
3. Houssami, N. and Skaane, P., Overview of the evidence on digital breast tomosynthesis in breast cancer detection. Breast, 2013. 22(2): p. 101-108.
4. Ciatto, S., et al., Integration of 3D digital mammography with tomosynthesis for population breast-cancer screening (STORM): a prospective comparison study. Lancet Oncol, 2013. 14(7): p. 583-589.
5. Houssami, N., et al., Breast screening using 2D-mammography or integrating digital breast tomosynthesis (3D-mammography) for single-reading or double-reading – Evidence to guide future screening strategies. Eur J Cancer, 2014. 50(10): p. 1799-1807.
6. Brodersen, J. and Siersma, V.D. , Long-Term Psychosocial Consequences of False-Positive Screening Mammography. Ann Fam Med, 2013. 11(2): p. 106-115.
7. Lång, K., et al., Performance of one-view breast tomosynthesis as a stand-alone breast cancer screening modality: results from the Malmö Breast Tomosynthesis Screening Trial, a population-based study. Eur Radiol, 2016. 26(1): p. 184-190.
8. Gilbert, F., et al., 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. Health Technol Assess, 2015. 19(4).
9. Skaane, P., et al., Prospective trial comparing full-field digital mammography (FFDM) versus combined FFDM and tomosynthesis in a population-based screening programme using independent double reading with arbitration. Eur Radiol, 2013. 23(8): p. 2061-2071.
10. Department of Health , Digital Breast Tomosynthesis - Overview of the evidence and issues for its use in screening for breast cancer - April2013 Commonwealth of Australia
11. Houssami, N., Digital breast tomosynthesis (3D-mammography) screening: data and implications for population screening. Expert Rev Med Devices, 2015. 12(4): p. 377-379.