Probably benign lesions at breast magnetic resonance imaging: preliminary experience in high-risk women.

BACKGROUND: The purpose of the current study was to determine the frequency of 'probably benign' interpretations at breast magnetic resonance (MR) imaging screening of high-risk women and the frequency of subsequent malignancy in these women.
METHODS: The authors performed a retrospective review of the records of 367 asymptomatic women with normal mammograms who had a high risk of developing breast cancer and whose first breast MR imaging screening examinations occurred during a 2-year period. The number of "probably benign" interpretations was determined, and follow-up data were obtained.
RESULTS: A "probably benign" interpretation was given to 89 (24%) of 367 women. Follow-up MR imaging was performed for 70 (79%) of 89 women, with a median follow-up time of 11 months (range, 1-24 months). Twenty women subsequently underwent biopsy at a median of 9 months (range, 1-18 months) after MR imaging, due to progression on follow-up MR imaging (n=14) or for other reasons (n=6). Malignancies were found in 9 women, constituting 45% of the 20 women who underwent biopsy and 10% of the 89 women with probably benign lesions. In 6 of the 9 women who developed malignant disease (7% of the 89 women in the study), the malignancy was detected by follow-up MR imaging of an area that previously was interpreted as probably benign. Tumor histology was ductal carcinoma in situ (DCIS) in five patients and infiltrating ductal carcinoma in four.
CONCLUSIONS: A "probably benign" interpretation was given to 24% of high-risk women at their first breast MR imaging screening examination. In 7-10% of women with "probably benign" interpretations, malignant disease subsequently developed in an area initially judged to be probably benign; of these malignancies, more than half were DCIS, and more than half were detected by MR imaging only. Copyright 2003 American Cancer Society.