Spectrally encoded confocal microscopy for diagnosing breast cancer in excision and margin specimens.

Elena F Brachtel, Nicole B Johnson, Amelia E Huck, Travis L Rice-Stitt, Mark G Vangel, Barbara L Smith, Guillermo J Tearney, Dongkyun Kang,

Laboratory investigation; a journal of technical methods and pathology, January 18, 2016

A large percentage of breast cancer patients treated with breast conserving surgery need to undergo multiple surgeries due to positive margins found during post-operative margin assessment. Carcinomas could be removed completely during the initial surgery and additional surgery avoided if positive margins can be determined intraoperatively. Spectrally encoded confocal microscopy (SECM) is a high-speed reflectance confocal microscopy technology that has a potential to rapidly image the entire surgical margin at subcellular resolution and accurately determine margin status intraoperatively. In this study, in order to test the feasibility of using SECM for intraoperative margin assessment, we have evaluated the diagnostic accuracy of SECM for detecting various types of breast cancers. Forty-six surgically removed breast specimens were imaged with an SECM system. Side-by-side comparison between SECM and histologic images showed that SECM images can visualize key histomorphologic patterns of normal/benign and malignant breast tissues. Small (500 μm × 500 μm) spatially registered SECM and histologic images (n=124 for each) were diagnosed independently by three pathologists with expertise in breast pathology. Diagnostic accuracy of SECM for determining malignant tissues was high, average sensitivity of 0.91, specificity of 0.93, positive predictive value of 0.95, and negative predictive value of 0.87. Intra-observer agreement and inter-observer agreement for SECM were also high, 0.87 and 0.84, respectively. Results from this study suggest that SECM may be developed into an intraoperative margin assessment tool for guiding breast cancer excisions.Laboratory Investigation advance online publication, 18 January 2016; doi:10.1038/labinvest.2015.158.

Pubmed Link: 26779830

DOI: 10.1038/labinvest.2015.158