Breast Excision Specimens Evaluated by Micro-Computed Tomography (Micro-CT) with Histopathological Correlations
Leopoldo J Fernandez, Julliette M Buckley, Owen P Aftreth, Rong Tang, Mansi Saksena, Yukako Yagi, James S Michaelson, Frederick C Koerner, Elena F Brachtel, Barbara L Smith. Massachusetts General Hospital, Boston
Background: Breast conserving surgery is standard of care in the treatment of breast cancer. Re-excision is required in 25-40% to provide negative margins on permanent histopathological evaluation. Improved methods to intraoperatively determine margin involvement by carcinoma would be desirable. In this study, we explore a novel method of imaging excision specimens by micro-CT for rapid visualization of the removed mass and its relationship to the margins.
Design: Fourteen breast excision specimens for breast cancer from 13 consented female patients were evaluated with a table top micro-computed tomography scanner (micro-CT), Skyscan®1173 (Skyscan, Belgium). Scanning took <15 minutes, followed by routine histopathological processing and reporting. Micro-CT images of the excision specimens were evaluated for mass size and relation to margins. Whole slide images of the histologic slides were obtained by Nanozoomer 2.0-HT®(Hamamatsu LTD, Japan).
Results: Average mass size was 1.9cm by micro-CT (range 0.9-3.5cm) and 1.3cm on pathology (range 0-2.2cm). Margins were positive/<0.1cm in 86% (n=12) by micro-CT and 64% by histology (n=9). Margins were free (≥0.2 cm) in 7% by micro-CT (n=1) and 14% (n=2) by histopathology. Additional shaved margins (not evaluated by micro-CT) were negative in 71% (n=10). Tumors consisted of 12 invasive ductal carcinomas (IDC), 1 invasive lobular carcinoma and 1 healing biopsy site with mass-like appearance on micro-CT but no residual carcinoma. 57% showed ductal carcinoma in-situ histologically (n=8).
IDC with mass at margin (→) by micro-CT (A,B), macroscopically (C) and histologically on whole slide image (D).
Conclusions: In this small pilot study, tumor size was slightly overestimated by micro-CT compared to pathology. Specimen margins were more often considered involved by tumor on micro-CT than confirmed histologically. Densely fibrous lesions, even if they do not contain tumor cells histologically, may be difficult to distinguish from tumor on micro-CT. Analysis of a larger cohort is necessary to determine predictive value of micro-CT assessment for breast excisions.
Wednesday, March 21, 2012 1:00 PM
Poster Session VI # 3, Wednesday Afternoon