An FDA advisory panel unanimously voted in favor of an automated ultrasound device for breast cancer screening after a negative mammogram in women with dense breasts.

But in Wednesday’s 13-0 vote, the Radiological Devices Advisory Panel recommended that the expanded indication include a restriction that the device be used only for such women who have not had a prior breast intervention, such as needle biopsy or breast cancer treatment.

The somo-v Automated Breast Ultrasound System (ABUS) is currently FDA approved as an adjunct to mammography in diagnosing breast cancer but not for screening.

The automated aspect of the technology would be helpful in the clinic for screening, Kristin Byrne, MD, chief of breast imaging at Lenox Hill Hospital in New York City, commented in an email to ABC News and MedPage Today.

“The automated image acquisition can help produce consistent results, allowing comparison between prior ultrasound exams in order to detect an interval change,” she explained. “Handheld ultrasound is technologist-dependent, and images vary between different technologists making comparison more difficult.”

Roughly 40% of women have dense breasts, which is associated with a four- to six-fold greater breast cancer risk, according to device manufacturer U-Systems in Sunnyvale, Calif.

The company and FDA reviewers emphasized that the automated breast ultrasound device is not intended to be used as a replacement for diagnostic mammography or diagnostic handheld ultrasound.

The advisory committee’s decisions revolved around results from a pivotal study comparing how many breast cancers interpreting physicians found with an ultrasound scan added as a second screen after a negative mammogram versus mammogram alone in women with more than 50% parenchymal density (BI-RADS composition density 3 or 4).

The 200 asymptomatic cases had been drawn from a prospective multicenter registry in which women had both exams as part of their annual routine screening and cancers found were confirmed by pathology.

In that study, the addition of automated breast ultrasound screening boosted cancer detection with an area under the curve of 0.75 compared with 0.60 for mammography alone (P<0.001).

Accuracy improved as well, with sensitivity rising from 38.5% with mammography alone to 62.4% with the addition of breast ultrasound (P<0.001). Specificity did not significantly change and was 76.2% with the combination.

Adding automated breast ultrasound screening boosted the true positive rate by about 30% and the false positive rate by about 4%. The positive predictive value and negative predictive value were significantly greater with the combination as well.

For dense-breasted women at elevated cancer risk, a separate study by the American College of Radiology Imaging Network (ACRIN) showed that adding ultrasound to mammography in annual screening boosted breast cancer detection but also boosted the number of false positives.

Recently-reported results from that trial showed the same was true on incidence screens after the initial round.

In that higher-risk population with dense breasts, the likelihood of biopsy for any detected suspicious abnormality is much higher than the normal population, with a biopsy rate of 12 lesions for every cancer identified in the ACRIN study, commented R. James Brenner, MD, chief of breast imaging at the University of California San Francisco.

"In a normal population, even with dense breasts, this percentage will be worse, and perhaps unacceptable," he wrote in an email to ABC News and MedPage Today.

The threshold for biopsy is somewhat lower for women with dense breasts, which may contribute to false positives, Byrne added.

However, the advisory panel also voted unanimously that the benefits of using the device under the proposed indication would outweigh its risks for the patients meeting the specified criteria.

For patients who didn't meet those criteria, though, generalizability was raised as an issue by FDA reviewers in documents prepared in advance of the advisory committee hearing.

The study excluded women with prior breast interventions, which represents a substantial proportion of the general screening population.

About 40% of women in the screening population have had a prior biopsy or aspiration at some point; another 15% have had diagnosis or treatment of breast cancer or breast augmentation.

In real-world use, "it is expected that patient records will be available to physicians, and patient management that includes an ABUS scan may be considered for patients who have had prior clinical breast intervention," the FDA summary document released before the panel meeting noted.

Whether prior clinical breast intervention should be specified in the indication until further clinical data is collected or whether the pivotal study results could be generalized was under discussion, but the committee decided to leave the condition on.

"It's typical for radiological or imaging devices to have such kinds of exclusions in their reader studies," U-Systems president Ron Ho, explained to MedPage Today after the panel meeting.

However, it's possible that with further documentation or analyses this exclusion might not make it onto the final indication if approved, he noted.

"The FDA has expressed willingness to work with us on that," Ho said.

The agency doesn't have to follow the advice of its advisory committees, but it usually does. The device is already approved as an adjunct to mammography for screening asymptomatic women in Canada and Europe.

###

By Crystal Phend, Senior Staff Writer, MedPage Today