
It was five years ago that we published a key article in Diagnostics, an open-access journal, in which we described our dual-modality device that combines automated breast ultrasound (ABUS) and full-field digital mammography (FFDM) in a single platform. The same journal has just published an article by a Romanian group (click here) that compared ABUS and digital breast tomosynthesis (DBT) with FFDM in women with dense breast tissue. Their findings add further evidence in the search for the ideal screening approach for early detection of breast cancer.
The purpose of the study by Ioana Boca and her colleagues was to evaluate the success of FFDM + ABUS in the diagnosis of breast cancer compared to FFDM + DBT. This retrospective study included 50 patients with dense connective tissue – BI-RADS level C (heterogeneously dense) or BI-RADS level D (extremely dense). Equipment used for FFDM and DBT to acquire cranio-caudal (CC) and mediolateral oblique (MLO) views was a Senographe Essential mammography unit made by GE Healthcare. Equipment used to acquire ABUS images of patients lying supine was an Invenia system, also manufactured by GE.
The figure at left (© MDPI) highlights the finding for one patient who had an indication for further investigation. Skin thickening is visible in the DBT image in the MLO view of the right breast (red arrow) and confirmed in the upper ABUS image (white star). Normal skin thickness of the left breast is shown in the lower ABUS image (white dot). It turned out the skin thickening, and subcutaneous swelling were due to previous radiotherapy.
The diagnostic performance of the imaging techniques was based on five parameters: sensitivity; specificity; positive predictive value (PPV); negative predictive value (NPV); and accuracy. As seen in the table below right, FFDM + ABUS had significantly higher values for specificity, PPV and accuracy compared to FFDM + DBT. The latter combination had significantly higher sensitivity, while there was essentially no difference in NPV.
The authors concluded: “ABUS and DBT are suitable as additional diagnostic imaging techniques to FFDM in women at intermediate risk of developing breast cancer through the presence of dense breast tissue. In this study, DBT reduced the false negative results, while the use of ABUS resulted in an increase in specificity.” Their results provide further evidence to support a screening approach that combines DBT followed by ABUS or, better still, employing a system that combines DBT and ABUS in a single platform.
The major flaw in this study is the fact that DBT should always include FFDM MLO and CC projections. It is part of the DBT study. These are needed to facilitate comparison with previous studies and to help detect clustered calcifications. When “paging through” DBT planes it is possible to overlook a cluster of calcifications even though calcifications are ultimately seen as well and even better on the planar images (Kopans DB, Gavenonis S, Moore RH, Halpern E. Calcifications In The Breast And Digital Breast Tomosynthesis. The Breast Journal 2011 Nov-Dec;17(6):638-44).
There are still some cancers that are not seen on DBT (because they do not distort the architecture and have no interface with fat, or associated microcalcifications) that will be seen on US. A device that combines DBT and US would greatly facilitate the detection of these cancers by having DBT and US perfectly registered to facilitate review.
Thanks you for this feedback, Dan. Your comments are well made. I agree that this study had some flaws which the authors did, in fact, acknowledge in their discussion, including the lack of comparison with previous images.
I think it would have been useful for them to have compared FFDM alone (the gold standard) with DBT + ABUS. I suspect that comparison would have been revealing.