-
DBT doesn’t improve screening metrics in breast cancer survivors
It is time to stop comparing DBT to FFDM in different populations. In this paper by Nia et al comparing FFDM to DBT in patients who have been treated for breast cancer they, apparently, compared women who had only FFDM to those who had DBT for follow-up screening. This creates a major potential for selection bias. There is no way of knowing if the populations in each group were comparable in their prior probability of recurrent or new disease.
Since every DBT study should include FFDM in the MLO and CC projections, it is, literally, impossible for FFDM to be better than DBT! Certainly, anyone who uses DBT for screening the general population knows that there are cancers that are evident on the planar DBT images that are not evident on the 2D FFDM images, so it is impossible for FFDM, by itself, to be better than (or even equivalent) to DBT!
Investigators should study the same women and randomly read the planar images first (before adding the 2D images) alternating with the FFDM images (2D) first on the same case if they still wish to compare the value of adding the planar images, but it is time to stop comparing FFDM and DBT in differing patient populations since this is, likely, grossly inaccurate.
-
1 Reply
-
.
