Ultrasound turned in a better incremental breast cancer detection rate than tomosynthesis in mammography-negative dense breasts, and a similar false-positive rate, according to interim findings from the ASTOUND trial.

Among more than 3,000 mammography-negative screening participants with dense breasts, 24 additional breast cancers (BC) were detected, of which 23 were invasive. All 23 were detected with ultrasound for an incremental cancer detection rate (CDR) of 7.1 per 1,000 screens (95% CI 4.2-10.0, P=0.006) versus 13 found with tomosynthesis (incremental CDR 4.0 per 1,000 screens, 95% CI 1.8-6.2), reported Nehmat Houssami, MBBS, MPH, PhD, of the University of Sydney, and colleagues.

The false-positive (FP) recall rates for any testing (P=0.26) or for biopsy (P=0.86) did not differ between the two modalities, they wrote in the Journal of Clinical Oncology.

While the analysis "shows that ultrasound has better incremental BC detection than tomosynthesis in mammography-negative dense breasts at a similar FP-recall rate ... future application of adjunct screening should consider that tomosynthesis detected more than 50% of the additional BCs in these women and could potentially be the primary screening modality," the authors cautioned.

Therese B. Bevers, MD, of the MD Anderson Cancer Center in Houston, agreed with the authors.

"It's a great suggestion," she told MedPage Today. "While there is slightly greater cancer detection with ultrasound, there are some system barriers of time, cost, and manpower to performing all those ultrasounds."

Bevers, who was not involved in the study, noted that cost is also an issue. While not all insurance plans cover tomosynthesis, "most places I've heard of have priced it exceedingly competitively." MD Anderson charges patients a standard $60 fee for screening tomosynthesis.

In addition, tomosynthesis studies require only a few minutes more in the same setting as standard mammography, so "[screening tomosynthesis] would keep things moving along in screening large populations," she said.

Study Details
During 2012-2015, the multicenter Adjunct Screening With Tomosynthesis or Ultrasound in Women With Mammography-Negative Dense Breasts trial screened 3,231 dense-breasted asymptomatic women (median age 51) with negative mammograms using both tomosynthesis and physician-performed ultrasonography.



Cross-blinded breast radiologists interpreted the images from each modality separately. Outcome measures were CDR, number of false-positive recalls, and incremental cancer detection rates for each imaging technique.

Invasive lesions detected were 18 invasive ductal carcinomas, four invasive lobular carcinomas, one mixed invasive, and one ductal carcinoma in situ. Mean tumor size was 15.2 mm for tomosynthesis and 15.1 mm for ultrasound.

The incremental CDR for tomosynthesis and ultrasound differed by a CDR of 3.1 per 1,000 screens (95% CI 1.2-3.1) in favor of ultrasound, a yield consistent with previous trials of these modalities, the authors reported.

The incremental CDR for adjunct screening with tomosynthesis and ultrasound where either was positive was 7.4 per 1,000 screens (95% CI 4.4-10.4).

Incremental false-positive recalls occurred in 107 cases (3.33%, 95% CI 2.72%-3.96%). False-positive recalls involving any testing did not differ between the two modalities: 53 for tomosynthesis (1.7%) and 65 (2.0%) for ultrasound. Nor did false-positive recalls for biopsy differ: 22 for tomosynthesis (0.7%) and 24 (0.7%) for ultrasound (P=0.86).

Findings referred for short-term imaging review came from tomosynthesis in 150 screens (4.7%), and from ultrasound in 57 screens (1.8%).

"Our results could be taken to suggest that tomosynthesis is detecting [breast caners] that would have been otherwise masked (on 2D mammography) by overlapping breast parenchyma, but seems less capable than ultrasound at finding cancers that are entirely masked by mammography-dense tissue," Houssami's group wrote. "We assume that some cancers are visible to only one of the physical principles of imaging modalities (x-ray for tomosynthesis vs ultrasound)."

The majority of ultrasound-detected cancers undetected by tomosynthesis were masses, whereas the single malignancy detected by tomosynthesis but missed by ultrasound was an architectural distortion.

The authors pointed out that these results are interim only, and relate to a self-referring population of women with dense breasts and negative mammograms.

Ultrasound screening for dense breasts is resource-intensive and increases false-positive recalls and costs, the authors noted. So with adjunct tomosynthesis detecting 50% of additional cancers, policy makers might consider whether "tomosynthesis could potentially be the primary imaging modality (without any adjunct imaging)," they wrote.

A study limitation was that the authors did not have health information on the participants beyond age and breast density. "Risk-related data were not routinely collected at participating imaging centers," they stated.

Used Together?
One of the historical barriers to implementing ultrasound (US) breast cancer screening in practice has been the high rate of false-positives, explained Wendie Berg, MD, PhD, of Magee-Womens Hospital of University of Pittsburgh Medical Center, in an accompanying editorial.

"Importantly, in preliminary results from the ASTOUND trial, false-positive recalls (2.0%) and biopsies (0.7%) were acceptably low," she noted.

However, she pointed out that "these low rates likely reflect that most of the US screens in ASTOUND were incident screens (with prior examinations available); further, recommendations for short interval follow up (Breast Imaging-Reporting and Data System density categories three) were not considered test positive."

She also acknowledged that ultrasound screening implementation is hampered by lack of standardized training for technologists and potential out-of-pocket costs for patients.

In comparison, "digital breast tomosynthesis, which is essentially 3D mammography, is much easier to implement. Once the equipment is in place, a technologist positions the patient exactly the same way as for standard mammography and simply pushes a button to obtain tomosynthesis instead of (or in combination with) a standard digital mammogram," she wrote.

But Berg suggested that the two modalities can work together. "On the basis of the results from ASTOUND, tomosynthesis still misses a substantial number of invasive cancers in women with dense breasts: supplemental US after tomosynthesis would still be reasonable, although further study is warranted," she wrote.