In the DBTUST study reported in the Journal of Clinical Oncology, Berg et al found that the addition of screening ultrasound to digital breast tomosynthesis (DBT) improved detection of cancer in patients with dense breasts—but also increased the rate of false-positive findings.
Study Details
In the prospective study, conducted between December 2015 and June 2021, 6,179 patients at three centers in western Pennsylvania consented to undergo three rounds of annual screening with DBT and technologist-performed handheld ultrasound, interpreted by two radiologist observers. The initial screening ultrasound occurred within 8 weeks of initial DBT; ultrasound was performed concurrently with DBT at 12 months and 24 months after initial ultrasound, with additional clinical follow-up at 36 months.
Key Findings
Among a total of 17,552 screens, 126 cancers were detected in 125 patients, yielding a rate of 7.2 cancers detected per 1,000 screens.
In year 1, the rates of cancer detection per 1,000 screens were 5.0 with DBT alone and 6.3 with DBT plus ultrasound (difference = 1.3/1,000, P = .005). In years 2 and 3 combined, rates were 4.9/1,000 with DBT alone and 5.9/1,000 for DBT plus ultrasound (difference = 1.0/1,000, P < .001).
In year 1, rates of false-positive findings were 7% with DBT alone and 11.5% with DBT plus ultrasound (P < .001). In years 2 and 3, rates of false-positive findings were 5.9% with DBT alone and 9.7% with DBT plus ultrasound (P < .001).
Although it was associated with reduced specificity, the addition of ultrasound to DBT resulted in a significant improvement in receiver operating characteristic area under the curve values, from 0.83 to 0.92 in year 1 (P = .01); smaller, numeric improvements were observed in year 2 (0.90 vs 0.93, P = .190) and year 3 (0.82 vs 0.86, P = .160).
Across 3 years among the 6,179 patients, 1,007 (16.3%) had a false-positive recall based on DBT and an additional 761 (12.3%) had a false-positive recall based on ultrasound (P < .001).
Across 3 years among the 6,179 patients, 172 (2.8%) had a benign biopsy associated with DBT and 230 (3.7%) had a benign biopsy associated with ultrasound (P < .001).
The investigators concluded, “Overall added cancer detection rate of ultrasound screening after DBT was modest… but potentially overcomes substantial increases in false-positive recalls and benign biopsies.”
Wendie A. Berg, MD, PhD, of the Department of Radiology, University of Pittsburgh School of Medicine, Magee-Womens Hospital of UPMC, is the corresponding author for the Journal of Clinical Oncology article.
Disclosure: The study was supported by the National Cancer Institute and Pennsylvania Breast Cancer Coalition. For full disclosures of the study authors, visit ascopubs.org.
