Introduction: Digital breast tomosynthesis has been shown to increase invasive cancer detection rates at screening compared to full field digital (2D) mammography alone, and some studies have reported a reduction in the screening recall rate. No prospective randomised studies of DBT have previously been published. This study compares recall rates with 2D mammography with and without concurrent DBT in women in their forties with a family history of breast cancer undergoing incident screening.
Materials and methods: Asymptomatic women aged 40 to 49 who had previously undergone mammography for an increased risk of breast cancer were recruited in two screening centres. Participants were randomised to screening with 2D mammography only at the first study screen followed a year later by screening with 2D plus DBT, or vice versa. Recall rates were compared using an intention to treat analysis. Reading performance was analysed for the larger centre.
Results: 1227 women were recruited. 1221 first screens (604 2D, 617 2D+DBT) and 1124 second screens (558 2D+DBT, 566 2D) were analysed. Eleven women had screen-detected cancers: 5 after 2D, 6 after 2D+DBT. The false positive recall rates were 2.4% for 2D and 2.2% for 2D+DBT (p=0.89). There was a significantly greater reduction between rounds in the number of women with abnormal reads who were not recalled after consensus / arbitration with 2D+DBT than 2D (p=0.023).
Conclusion: The addition of DBT to 2D mammography in incident screening did not lead to a significant reduction in recall rate. DBT may increase reader uncertainty until DBT screening experience is acquired.