Clinical Report: False-Positive Recall Rates in Malmö Breast Tomosynthesis Screening
Overview
The Malmö Breast Tomosynthesis Screening Trial (MBTST) evaluated false-positive recall rates comparing one-view digital breast tomosynthesis (DBT) with two-view digital mammography (DM). The study found that false-positive recalls were higher with DBT alone, primarily due to stellate distortions, but rates decreased over time, indicating a learning curve. Detailed analysis of radiographic appearances and biopsy outcomes provided insights into the nature of false positives in DBT screening.
Background
Digital breast tomosynthesis (DBT) is emerging as a potential replacement or complement to digital mammography (DM) in breast cancer screening. False-positive recalls are common in breast cancer screening and can cause psychosocial distress and reduced re-attendance. Prior studies have reported mixed results regarding false-positive rates with DBT compared to DM. Understanding the characteristics and outcomes of false-positive recalls is essential before widespread implementation of DBT in screening programs.
Data Highlights
Parameter
DBT Only
DM Only
DBT + DM
False-positive recall rate
Higher initially, decreased over time
Lower compared to DBT initially
Combined recalls
Dominant radiographic appearance
Stellate distortions
Asymmetries and calcifications
Mixed appearances
Work-up procedures
Included DM, ultrasound, biopsies
Similar work-up
Similar work-up
Biopsy outcomes
Benign lesions predominated
Benign lesions predominated
Benign lesions predominated
Key Findings
False-positive recall rates were initially higher with DBT alone compared to DM alone, mainly due to increased recalls of stellate distortions.
False-positive recall rates with DBT decreased over the course of the trial, suggesting a learning curve effect among radiologists.
Radiographic appearances leading to false-positive recalls differed between modalities, with stellate distortions more common in DBT and asymmetries and calcifications more common in DM.
Most false-positive recalls resulted in benign biopsy outcomes, including benign cysts, fibroadenomas, and radial scars.
Work-up for false-positive recalls typically involved additional imaging and biopsies, with fine needle aspiration still used during the trial period.
False-positive recalls were categorized based on the modality that led to recall, allowing detailed comparison between DBT and DM findings.
Clinical Implications
Clinicians should be aware that DBT may initially yield higher false-positive recall rates due to specific radiographic findings such as stellate distortions, but these rates may improve with experience. Understanding the nature of false positives in DBT can help optimize patient counseling and management strategies to minimize psychosocial distress and unnecessary procedures. Continued training and experience are important to reduce false-positive recalls in DBT screening programs.
Conclusion
The MBTST provides valuable insights into false-positive recall characteristics in DBT versus DM screening, highlighting a higher initial false-positive rate with DBT that decreases over time. These findings support the need for ongoing evaluation and training when implementing DBT in breast cancer screening.
References
MBTST Study Group 2024 -- Evaluating False-Positive Recall Rates in the Malmö Breast Tomosynthesis Screening Study
