
Digital breast tomosynthesis (DBT), commonly known as 3D mammography, has recently been the focus of two research articles from Norway. As seen at right, the X-ray tube moves through an arc of about 30 degrees while a series of low-dose images are acquired, and a 3D volume of the breast is then reconstructed via a mathematical algorithm. The synthesized images can be played back as a movie and, with the removal of overlapping tissues, cancerous lesions are more readily identified (click on the image below left).
In a study published in Radiology, Bjørn Østerås showed that DBT found more malignant tumours compared with traditional 2D mammography, especially if the cancer presented as a spiculated mass or architectural distortion. He and his colleagues used data from a screening trial of 24,000 women and established that DBT had both higher true-positive rates and fewer false-negative findings for most breast density and patient age groups. Interestingly, there was no statistical difference in women with extremely dense breasts.
Tron Anders Moger and his team conducted an extensive cost-effectiveness study in which they compared DBT with 2D mammography and published their findings in the European Journal of Health Economics. They wrote, “On the one hand, DBT will have higher costs than mammography due to the additional investment required and longer times needed for screen readings. On the other hand, lower recall rates, less diagnostic workup … and earlier detection will possibly reduce the cost per woman screened.” However, they found that even though DBT costs decreased in response to various factors – such as lower DBT system prices, reduced storage costs, and shorter reading times – it was still more expensive than mammography.As previously highlighted, the British Medical Journal has a history of publishing articles that are sceptical about the benefits of breast screening. One of their editors, Jeanne Lenzer, has just written an opinion piece on the increased adoption of DBT in the United States despite experts still arguing about the benefits of 3D mammography. She highlighted the TMIST project, a randomized controlled trial comparing DBT with 2D mammography, and quoted Daniel Kopans of Harvard who said the superiority of DBT was a “no brainer” and who believed that TMIST was “an enormous and potentially dangerous waste of money.” Strong words indeed!
An editorial that accompanied the paper by Østerås in Radiology asked, “Isn’t it time to take this method to the next level and employ DBT as a stand-alone screening technique, combining it with automated 3D ultrasound to create the screening ‘dream team’?” Now there’s an idea worth exploring!
Obviously, DBT is until now not superior to FFDM regarding the prevented interval carcinomas. That is the only proxy outcome parameter that counts in screening.
DBT is a technology push and interesting management of expectations. It takes many years to prove superiority. Better sensitivity is not per definition better outcome in screening, but this can hardly be explained to lay people.
If comparing DBT to FFDM is a no brainer is the question. As radiologist I would prefer a new contrast to compare to X-ray and that is ultrasound and not pseudo 3D information.