Return To Abstract Listing

Scientific Session 07 — SS07: Breast Imaging - MRI

Tuesday, May 7, 2019

Abstracts 1913-2417

2406. Frequency and Outcomes of New Suspicious Lesions on Breast MRI in the Setting of Neoadjuvant Therapy

Eckstein D*,  Price E,  Hayward J,  Joe B,  Lee A. University of California - San Francisco, San Francisco, CA

Address correspondence to D. Eckstein (

Objective: In women undergoing neoadjuvant chemotherapy for locally advanced breast cancer, breast MRI is commonly performed to monitor treatment response prior to surgery. Occasionally, new suspicious breast MRI findings distinct from the site of known malignancy arise during this surveillance period. Although it would be unusual for a new site of cancer to arise during therapy, there are no studies on the outcomes of these findings to guide management. We aim to determine the frequency and outcomes of new suspicious findings detected on breast MRI in the setting of neoadjuvant chemotherapy.

Materials and Methods: An IRB-approved retrospective database review identified all breast MRI examinations performed to assess response to neoadjuvant chemotherapy between 2010 and 2018. BI-RADS assessments were extracted from the imaging reports, and those assessed as BI-RADS 4 or 5 for suspicious findings were identified. We included all cases with new suspicious lesions found after initiation of neoadjuvant treatment. Examinations were excluded if there was no baseline pretreatment MRI for comparison (7 cases) or if the suspicious finding was present on the baseline pretreatment MRI and was unsampled on follow-up MRI (44 cases). Lesion characteristics were obtained from the radiology reports. Malignant outcomes were determined by pathology. Benign outcomes were determined by pathology or follow-up imaging. Descriptive statistics were performed.

Results: A total of 419 breast MRI examinations were performed in 297 women to assess response to neoadjuvant chemotherapy. After exclusions, 28 MRI examinations (7%) with new suspicious findings constituted the final study cohort. All study cases were assessed as BI-RADS 4. Sixteen lesions (57%) were contralateral to the known malignancy, 11 (39%) ipsilateral, and 1 (4%) bilateral. Lesions included mass (18, 64%), nonmass enhancement (8, 29%), and focus (2, 7%). Twenty-five cases had either tissue diagnosis or sufficient imaging follow-up (>2 years) to establish a diagnosis. In 3 cases, the outcomes were unknown (e.g., insufficient follow-up). In cases with known outcomes, none (0/25) of the new suspicious findings were malignant; 44% (11/25) were proven benign at percutaneous biopsy, 20% (5/25) at surgical pathology, and 36% (9/25) presumed benign by follow-up imaging, demonstrating resolution of the suspicious lesion or stability without malignancy for >2 years.

Conclusion: New suspicious findings on breast MRI can occur when monitoring response to neoadjuvant therapy, present in 7% of cases in our study. However, these findings are highly unlikely to represent a new site of malignancy. In this cohort, the cancer yield for new suspicious lesions was 0% for all 25 cases that had tissue diagnosis or sufficient imaging follow-up. Our findings suggest that new suspicious findings on MRI arising during neoadjuvant therapy are almost certainly benign, although larger studies across facilities are needed to confirm whether biopsy may be safely averted in these scenarios.