Digital Breast Tomosynthesis May Reduce Development of Advanced Cancers

Digital breast tomosynthesis (DBT) increased breast cancer detection rates and showed invasive cancers with a lower rate of advanced cancers compared with digital mammography, according to results from a retrospective study.

In the review of 272,938 screenings for breast cancer, DBT had a higher cancer detection rate than digital mammography (5.3 vs 4.0 cancers per 1,000 screening exams, P=0.001), as well as a lower recall rate (7.2% vs 10.6%, P

Furthermore, DBT “depicted invasive cancers with a lower rate of advanced cancers compared with digital mammography, with further improvement observed at incident rounds of screening,” they observed in Radiology.

Specifically, the proportion of advanced cancers was lower with DBT than digital mammography (32.6% vs 43.6%, respectively, P=0.04), suggesting DBT is detecting cancers at an earlier stage.

Broken down further into prevalent cases (in which the patient had no previous exam performed with the same modality) and incident cases (in which the patient had a previous exam performed with the respective modality), the advanced cancer rates were similar among prevalent cases (39.1% with DBT vs 42.9% with digital mammography, P=0.98). However, among incident cases, DBT showed a significantly lower advanced cancer rate than digital mammography (29.1% vs 43.7%, respectively, P=0.009).

“These results are the most striking; they suggest that DBT screening may reduce advanced cancer development,” said Soo-Yeon Kim, MD, PhD, and Ok Hee Woo, MD, PhD, both of the Korea University College of Medicine in Seoul, in an accompanying editorial.

Kim and Woo pointed out that a large ongoing randomized study — the TMIST trial — is designed to test whether DBT outperforms digital mammography in reducing the development of breast cancer.

“According to the TMIST hypothesis, the reduction in advanced cancers could result in cancers that are more easily treatable, leading to improved quality of life and better prognosis,” they wrote. “The present study indirectly supports the TMIST hypothesis using retrospective data from a single institution.”

In their retrospective study, Philpotts and colleagues analyzed consecutive cancer cases detected by screening mammography from August 2008 through July 2021 at Yale University/Yale-New Haven Health. The data included 10 years of mammograms performed with DBT, which became the standard at Yale facilities beginning in 2011, and mammograms performed with digital mammography between 2008 and 2011.

The study cohort included 272,938 screening mammograms, of which 35,544 were performed with digital mammography and 237,394 performed with DBT. A total of 1,407 cancers were detected, 142 by digital mammography and 1,265 by DBT.

The mean age in the digital mammography group was slightly younger than the DBT group (59 vs 62 years), and there was a higher percentage of white patients in the DBT group (73.8% vs 56.3%).

The ratio of invasive cancers to ductal carcinomas in situ was comparable between the DBT (76.5%:23.5%) and digital mammography (71.1%:28.9%) groups, and the mean invasive cancer size was also similar (1.44 cm vs 1.36 cm, respectively).

DBT and digital mammography had similar rates of invasive cancer subtypes — low grade (26.5% vs 29%, respectively), moderate grade (57.2% vs 51%), and high grade (16.1% vs 20%).

There was no significant difference in interval cancer rates between groups (0.2 per 1,000 with DBT vs 0.14 per 1,000 with digital mammography).

In their editorial, Kim and Woo noted that the study had several limitations, including the fact that it was a retrospective study from a single center that used a mammography machine from a single vendor, and also conducted supplemental ultrasounds in women with dense breasts.

“These specific conditions may not be generalizable to other institutions’ data and population-based screening,” they wrote.

Furthermore, there was a wide discrepancy in the length of the study period between the two groups, they added. “Some staff radiologists might have remained consistent during the study period, and their cumulative experience in interpreting mammography may have had a favorable impact on DBT outcomes.”