DBT Better Detects Cancer in Dense Breasts

Digital breast tomosynthesis (DBT) provides overall greater cancer detection than digital mammography, especially among women with dense breast tissue, according to a presentation at the American Roentgen Ray Society (ARRS) Annual Meeting in Toronto, Canada.

Researchers from the Ibrahim N. Einstein Medical Center, Philadelphia, PA, carried out a retrospective review of 29,377 screening mammograms (11,221 digital mammography and 18,156 DBT) performed at their institution from July 1, 2011, to December 31, 2013.

"There are a lot of data showing that screening with DBT increases cancer detection, but much less is known about the effect of density and lesion type on detection rates," coauthor Caroline Ling said in a release.

The researchers reviewed each BI-RADS category 0 examination, as well as breast density and lesions for which patients were called back. The breast density was separated into two categories: nondense (entirely fatty or scattered fibroglandular) or dense (heterogeneously dense or extremely dense), and the lesion types were categorized as mass, asymmetry, distortion, or calcification.

All pathologically proven cancers found on screening mammography were reviewed during this time period and up to three months after the last screening mammogram.

The results showed that the overall cancer detection rate was 3.1% in the DBT group and 2.2% in the digital mammography group, resulting in an overall cancer detection rate that was 38% greater in the DBT group (not statistically distinguishable).

Patients with dense breasts were 67% more likely to be diagnosed with cancer in the DBT group than the digital mammography group:

Patients with nondense breasts were 20% more likely to be diagnosed with cancer in the DBT group compared with the digital mammography group.

“The cancer detection among callbacks for mass and asymmetry was also higher in the DBT group for patients with dense breasts,” the researchers wrote in the abstract. “Patients with dense breasts called back for mass and asymmetry were 0.5 and 3.2 times more likely, respectively, to have cancer in the DBT group. Conversely, the cancer detection among callbacks was similar between the DBT and digital mammography groups in nondense breasts for mass and asymmetry (p = 0.7, 1.0).”

The researchers concluded that overall cancer detection was greater in the DBT group relative to the digital mammography group. The most striking increase was cancer detection among women with dense breasts called back for mass and asymmetry relative to nondense breasts. They suggest that a larger study is needed to determine statistical significance.