E1292. Timing is Everything: Optimizing Post-Lumpectomy Breast MRI for Close and Positive Margins
  1. Regan Ferraro; UCLA Medical Center
  2. Tiffany Chan; UCLA Medical Center
  3. Melissa Joines; UCLA Medical Center
Breast MRI is increasingly being used after breast conservation therapy to assess for residual disease in the setting of close or positive margins. MRI can be useful to assess the extent of residual disease after lumpectomy and for possible multifocal disease, which can guide treatment planning. However, the enhancement in and around the surgical bed can be difficult to interpret as post-surgical inflammation or residual tumor, and there is no clear consensus on optimal timing of breast MRI. The purpose of our study was to assess the optimal timing and diagnostic accuracy of breast MRI after surgery with close or positive margins.

Materials and Methods:
All breast MRIs performed between 9/1/2013 and 9/1/2019 to evaluate residual disease following close or positive surgical margins were included in this retrospective study. Examinations performed as follow-ups after clean margins had been obtained or performed remotely (over one year) after close or positive margins were excluded. Additionally, examinations where the patient received interim chemotherapy or radiation treatment were excluded. All cases were correlated with final histopathology when available.

At our institution, 8186 breast MRIs were performed between 9/1/2013 and 9/1/2019. Of those, 52 met our inclusion criteria as the first MRI performed after close or positive margins to assess for residual disease. Of those 52 cases, 45 proceeded to re-excision, and 25 cases were positive for residual disease on final histopathology. Overall accuracy by the radiologist in predicting the re-excision pathology was 48.1%. Radiologists equivocated (favored residual disease or post-surgical inflammation equally) in 26.9% of cases. Higher accuracy (52.6% vs. 45.5%) but more equivocation (36.8% vs. 21.1%) was observed with early MRI (0-14 days after surgery) compared to late MRI (>14 days after surgery).

Our study reviewed the diagnostic accuracy of breast MRI after close or positive margins, showing that radiologists have relatively high accuracy in predicting final histopathology. However, there is also a high rate of equivocation, where the radiologist favored residual disease and post-surgical inflammation equally. Our results suggest that early MRI is preferred, resulting in higher rates of accuracy despite increased equivocation by radiologists.