Women classified with BI-RADS 3 lesions should undergo follow-up surveillance imaging at six months due to a small risk that such lesions could in fact be malignant.

Women with BI-RAD 3 lesions should seek six month follow-up imaging, says study

May 26, 2020
by John R. Fischer, Senior Reporter
A new study says women who receive benign mammograms for breast lesions should seek follow-up surveillance imaging at six months due to a small risk that such lesions could, in fact, be malignant.

Radiologists at UPMC Magee-Womens Hospital and the University of Pittsburgh based their study on the Breast Imaging and Reporting System (BI-RADS) established by the American College of Radiology. They specifically based it on the BI-RADS 3 category, a group in which most findings are probably benign, except for a few that have less than a 2% chance of being cancer.

“The majority of cancers were diagnosed at or right after the six-month follow-up, so it actually is important to get these patients back in that six-month time frame,” said lead study author Dr. Wendie A. Berg, a breast cancer researcher at the University of Pittsburgh School of Medicine and professor of radiology at UPMC Magee-Womens Hospital, in a statement.

The BI-RADS system classifies findings based on the risk of breast cancer, with BI-RADS 2 lesions considered benign and those in BI-RADS 6 confirmed by biopsy to be malignant. The risk associated with the BI-RADS 3 category is known to make it confusing for physicians and cause anxiety in patients.

Berg and her colleagues analyzed outcomes from six-, 12-, and 24-month follow-ups of findings that were probably benign and were initially identified on recall from screening mammography in the National Mammography Database (NMD). They also evaluated results for women who were recalled from screening mammography with BI-RADS 3 assessment at additional evaluation over a period of almost 10 years. Participants in the study group had no personal history of breast cancer and underwent either a biopsy or two-year imaging follow-up.

Of the 43,628 women with a BI-RADS 3 assessment after screening mammography recall, 810 were diagnosed with cancer. Malignancy rates of this group substantially exceeded those of women downgraded to BI-RADS 1 or 2 at each follow up.

About a third developed ductal carcinoma in situ (DCIS), an early-stage, noninvasive form of breast cancer. Twelve percent of invasive cancers, however, diagnosed within six months with node staging spread to the lymph nodes.

The findings according to Bergman illustrate the need for short-interval follow-up imaging surveillance, a source of contention among breast imagers, as earlier research suggests that the rarity of malignancies among BI-RADS 3 patients means that such exams can be safely pushed back from six months to one year. A majority of that research, however, was conducted before the creation of the NMD.

“I thought that we should really look at this again, now that we’ve got this large database we can work with and try to find out what happened with these patients,” said Berg, who adds that “the important thing about this paper is that these data come from a wide number of facilities across the United States, so it really brings to bear that, yes, this is the appropriate practice and yes, you still need to see these patients in six months.”

A similar study presented at the ARRS 2019 Annual Meeting in Hawaii this month deemed short-interval follow-up MR as an efficient imaging method for identifying early-stage breast cancer among findings that are deemed probably to be benign. Using it could help to avoid unnecessary biopsies, according to its authors.

“Many patients and providers question the utility of BI-RADS 3 in MR, particularly as some insurers do not cover the costs of the short interval follow-up MR,” said co-author Christine Edmonds, of the department of radiology at Massachusetts General Hospital in a statement. “This study clarifies that short interval follow-up MR is a valuable method of identifying early-stage breast cancer while avoiding unnecessary biopsies.”

The study by Berg and her colleagues was published in Radiology.