A study of more than 2,500 women has found ultrasound just as good as mammography at detecting breast cancer.

On the downside, ultrasound generated more false positives than mammography. However, the radiation-free modality also spotlighted more invasive and node-negative cancers—and the number of women recalled for additional testing post-ultrasound became comparable on incidence-screening rounds.

The study’s authors, led by Wendie Berg, MD, of the University of Pittsburgh, conclude that their results suggest ultrasound could satisfactorily substitute for mammography in developing countries.

In the U.S., they state in their study discussion, ultrasound might supplement standard mammography for women with dense breasts who don’t meet high-risk criteria for screening MRI, as well as for high-risk women with dense breasts who balk at or refuse MRI.

The research team looked at 2,662 participants who were enrolled at 20 sites in the U.S., Canada and Argentina in the American College of Radiology Imaging Network (ACRIN) 6666 trial and who completed three annual screens (7,473 exams) with ultrasound and film-screen (n = 4,351) or digital (n = 3,122) mammography and had biopsy or 12-month follow-up.

Noting cancer detection, recall and positive predictive values, they found:

• 110 women had 111 breast cancer events; 89 (80.2 percent) were invasive cancers, median size 12mm.
• The number of ultrasound screens to detect one cancer was 129, and for mammography the number of needed screens was 127.
• Cancer detection was comparable for each of ultrasound and mammography at 58 of 111 (52.3 percent) vs. 59 of 111 (53.2 percent), with ultrasound-detected cancers more likely invasive (53 of 58, 91.4 percent, median size 12mm) vs. mammography at 41 of 59 (69.5 percent, median size 13mm).
• Invasive cancers detected by ultrasound were more frequently node-negative, 34 of 53 (64.2 percent) vs. 18 of 41 (43.9 percent) on mammography.

Meanwhile, in 4,814 incidence screens in years two and three, ultrasound had higher recall and biopsy rates and a lower positive predictive value for biopsy than mammography.

“Training would be necessary for any facility planning to offer screening ultrasound, also true for developing countries,” Berg et al. write in their study discussion. “With appropriate training, ultrasound is no more operator dependent than interpreting mammography.”

The authors add that, while they previously showed in ACRIN 6666 that invasive lobular cancer and low-grade invasive ductal carcinoma are overrepresented among cancers seen only on ultrasound, they did not glean detailed molecular subtype results for the cancers in the present study.

The study was published in the December Journal of the National Cancer Institute and is posted in its entirety online.