318P - Margin Assessment in Breast Cancer Lumpectomy Specimens with Infrared Diffuse Reflectance Spectroscopy(Drs)

Abstract

ABSTRACT Aim: Breast conserving surgery (BCS) is the primary choice of treatment for early stage breast cancer. The number of positive resection margins is still reported 10-20%, emphasizing the need for a tool to discriminate malignant tissue during surgery. Here we investigate whether optical imaging by means of DRS is able to discriminate normal breast tissue from tumor tissue during surgery. Methods: Specimens form thirteen patients undergoing BCS were analyzed. All specimens were processed by the standard protocol of the pathologist. After slicing the lumpectomy specimen, DRS measurements of healthy tissue, malignant tissue and tissue near the border of the tumor were obtained. All measurement locations were photographed to be able to link histopathology with DRS measurements. In addition, two mastectomy specimens were intensively analyzed by performing measurements of the entire surface of the sliced specimen to determine the accuracy of DRS to identify tumor borders. A custom made grid with holes for the probe was laid over the cassette containing the specimen to obtain spectra in a more controlled setting. For each measurement location fat and water content was derived from the spectra with an analytical fitmodel. Results: From the lumpectomy specimens spectra were obtained from 170 locations (34 benign, 27 malignant & 109 border). In healthy tissue a high fat content and low water content was measured whereas in malignant tissue the opposite was seen. The ratio (Fat/Water) calculated from these parameters allowed to discriminate tumor tissue from healthy tissue with a sensitivity of 97% and a specificity of 92%. The ratio calculated for the border locations was in the range between the values found in malignant tissue and healthy tissue. Some border locations resembled tumor tissue whereas others were more similar to healthy tissue. These findings were confirmed with our grid measurements. Conclusions: DRS is able to distinguish healthy tissue from malignant tissue in breast cancer resection specimens. Furthermore, DRS can detect the gradual change of healthy tissue to tumor tissue, and is therefore a promising technique to detect tumor borders during surgery. Especially, since DRS can be applied real time during the surgical procedure. Disclosure: B.H.W. Hendriks and T.M. Bydlon:The author who is affiliated with Philips Research and Philips Healthcare only has financial interests in the subject matter, materials, and equipment, in the sense that he is an employee of Philips. All other authors have declared no conflicts of interest.