Digital breast tomosynthesis (DBT) and digital mammography (DM) detect breast cancers with comparable tumor characteristics, according to a new study published in Radiology.
“Before the broad implementation of DBT in screening, it is important to assess whether the detection of additional cancers leads to an actual screening benefit (ie, reduced breast cancer mortality) or whether mainly low-risk tumors are detected, potentially resulting in overdiagnosis,” wrote Kristin Johnson, Lund University in Malmö, Sweden, and colleagues. “Because data from randomized DBT trials with breast cancer mortality as the outcome are likely to be limited, studies on cancer biology as a surrogate can elucidate this issue.”
The researchers explored data from the Malmö Breast Tomosynthesis Screening Trial, a population-based study that compared breast cancer screening using DBT and DM. More than 14,000 women participated from Jan. 27, 2010, to Feb. 13, 2015, undergoing both one-view DBT and two-view DM during a single visit. A team of seven radiologists read the examinations.
Overall, 139 breast cancers were detected in 137 women. Forty-two of those cancers were only detected by the DBT reading arm, eight were only detected by the DM reading arm and 89 were detected by both the DBT and the DM reading arms. The proportion of cancers detected and proportion of women recalled after consensus were both higher in the DBT reading arm.
Of the 139 breast cancers, 118 were invasive, and the other 21 were ductal carcinomas in situ. And of those 118 invasive cancers, 37 were additional cancers detected only with one-view DBT. For both reading arms, the majority of additional cancers detected were early stage luminal A-like cancers.
Meanwhile, no differences were observed between the cancers detected by DBT and DM when it came to whether the tumor size was 20 mm or smaller, node-negative status or luminal A-like subtype.
“Our results suggest that DBT screening will not necessarily change the panorama of biologic profiles of screen-detected cancers, but rather that more of the same type of cancers would be detected,” Johnson and colleagues wrote.
The authors also noted that the Malmö Breast Tomosynthesis Screening Trial was not originally created to focus on the issue of overdiagnosis, so it’s hard to make any concrete observations from these findings.
“Because DBT alone depicted one additional low-grade ductal carcinoma in situ, its contribution to possible overdiagnosis of this type of cancer does not appear to be a major impediment for the use of DBT in screening,” the authors wrote. “However, because the predominant additional cancers detected at DBT consisted of the luminal A–like subtype, we cannot eliminate the possibility that DBT would add to the number of overdiagnosed cancers.”