Freddy T. Nguyen, Adam M. Zysk, Eric J. Chaney, Jan G. Kotynek, Uretz J. Oliphant, Frank J. Bellafiore, Kendrith M. Rowland, Patricia A. Johnson, Stephen A. Boppart. Cancer Research 2009-01-01. Full Text
Background: The decrease in the number of breast cancer deaths has largely been attributed to increased awareness, earlier detection, and improved treatment options. However, as the number of breast-conserving surgeries rose over the years, the need for negative margins and little or no residual disease has become critical to help reduce the chances of local recurrence. OCT is a high resolution imaging modality that has been used to image tumor margins in an NMU-carcinogen-induced rat mammary tumor model. Due to the location of breast lesions, the use of needle-based imaging probes may be used to further extend the reach of the OCT imaging beam by incorporating an optical fiber into biopsy needle tips, providing real-time information to guide biopsies or to place localization wires.
Material & Methods: A clinical spectral domain OCT system was developed with a super luminescent diode light source centered at 1310 nm with a bandwidth of 92 nm yielding an axial resolution of 8.3 µm. The beam delivery sample arm uses a 60 mm achromatic lens to focus 4.75 mW of light to a 35.0 µm spot size (transverse resolution) with a confocal parameter of 1.47 mm. The patients included in this study had primary breast tumors diagnosed by needle-biopsy and were in need of surgical resection, as determined by their physicians. At Carle Foundation Hospital, the OCT system was placed inside the operating room during breast conserving surgical procedures to image the tissue specimens. The OCT images were evaluated by a single operator allowing for consistent classification based on the level of scattering intensity and heterogeneity, scattering profile, and physical extent of the highly scattering area.
Results: An initial training data set of OCT images from 17 patients was used to establish standard imaging protocols and standard evaluation criteria of the surgical margins. Of the 20 additional tissue specimen imaged for the feasibility study, 11 were identified as having a positive or close surgical margin and nine as a negative margin under OCT. In comparing to the H&E histology, there were 9 true positives, 9 true negatives, 2 false positives, and 0 false negatives yielding a sensitivity of 82% and specificity of 100%.
Discussion: With an imaging penetration depth of 2-3 mm, equivalent to that used for histological assessment, OCT provides unique real-time cellular-level imaging to identify positive and close margins. In these studies, areas of higher scattering tissue with an irregular or heterogeneous pattern were identified, differentiating them from the abundant adipose tissue found in normal breast tissue. The small nucleus to cytoplasm (N/C) ratio is observed with low-scattering adipocytes compared with the larger N/C ratio found from highly-scattering tumor cells. These intraoperative imaging studies have demonstrated the ability for OCT to identify positive surgical margins.