Low BI-RADS 3 lesion cancer rate suggests rethinking follow-up

Ultrasound-detected lesions assessed as Breast Imaging-Reporting and Data System (BI-RADS) category 3—meaning probably benign—have a total malignancy rate of 0.8 percent, according to the results of a multicenter imaging trial published online Aug. 20 in Radiology

Based on the low malignancy rate, study authors Richard G. Barr, MD, PhD, of Northeast Ohio Medical University in Youngstown, and colleagues suggested extending the follow-up imaging interval from the current guideline-recommended six months to a full year.

“Yearly follow-up for US [ultrasound] BI-RADS category 3 lesions would substantially decrease the number of follow-up examinations, and thereby improve the cost-benefit ratio of US screening,” they wrote, adding that unnecessary biopsies and imaging can also cause added anxiety for patients.

Using ultrasound as a supplement to screening mammography is becoming more common due to ultrasound’s ability to detect lesions in dense breast that might not be visible with mammography. However, with an increase in cancer detection there’s also an increase in false-positive findings, and with a substantial number of lesions classified as “probably benign,” a sound follow-up strategy is necessary.

To determine how many of these likely benign BI-RADS 3 lesions are actually malignant—and in turn inform how aggressive follow-up exams should be—Barr and colleagues analyzed information from the American College of Radiology Imaging Network trial 6666 database, a study involving mammography and ultrasound screenings for women with dense breasts at 21 sites around the U.S. 

There were 745 BI-RADS 3 lesions among the more than 2,600 trial participants, and these lesions had a median size of 7mm. The proportion of lesions classified as BI-RADS 3 was stable at around one-quarter even as the total number of BI-RADS 3 lesions fell after the second and third year of screening.

Six lesions were found to be malignant, for a total malignancy rate of 0.8 percent, reported the authors, with an average size of 10mm for the detected cancers. After six-month follow-up, only one lesion exhibited suspicious changes. A second and third lesion had suspicious changes at 12-month and 24-month follow-up imaging, respectively.

“The cancers detected due to changes at one year had not spread beyond the breast, with similar prognosis if the lesion was biopsied on initial detection,” said Barr in a press release.

Under current recommendations, approximately 20 percent of patients who undergo three annual rounds of screening ultrasound will be recommended for diagnostic follow-up. Since there are low rates of malignancy and suspicious changes in BI-RADS 3 lesions, extended follow-up interval may be appropriate and could reduce the number of biopsies and follow-up imaging performed, according to the authors.