[224] Utility of Frozen Section for Intra-Operative Evaluation of Sentinel Lymph Nodes in Breast Cancer

UR Kundu, KK Hunt, S Krishnamurthy. MD Anderson Cancer Center, Houston, TX

Background: Intraoperative evaluation of axillary sentinel lymph nodes (SLN) allows the surgeon to complete axillary dissection in one setting at the time of primary breast surgery if diagnosed as positive for metastatic tumor. There is currently no consensus regarding the optimal method for intra-operative evaluation of SLNs. This study was initiated to evaluate the utility of frozen section (FS) for intraoperative evaluation of SLNs in patients with and without pre-operative neoadjuvant chemotherapy and their ability to detect macro and micrometastasis.
Design: SLN biopsies performed in 2007 with FS analysis for immediate assessment were included. SLNs were sliced at 2 mm intervals along the short axis and submitted entirely for FS which were stained by hematoxylin and eosin method. Final pathologic examination (FP) of formalin fixed and paraffin embedded tissue was regarded as the gold standard. FP included one H&E stain of the first section and cytokeratin immunostain of the third section. The overall sensitivity and specificity of FS for detecting metastatic tumor in patients with and without chemotherapy for detection of micro(>0.2mm and<2mm) and macro metastasis(>2mm) was calculated.
Results: The study included 358 patients; 73 received neoadjuvant chemotherapy (NC). FP examination revealed metastatic carcinoma in 11/73 (15.1%) patients who received NC and in 82/285 (28.8%) patients who did not receive NC (non-NC). The average size of micro-metastasis was 0.75 mm and the average size of macro-metastasis was 5.4 mm. In the NC group, FS detected 5/11 patients with metastasis to SLN; 2 with micro-metastasis and 3 with macro-metastasis. In the non-NC group, FS detected 50/82 patients with metastasis; 15 with micro-metastasis and 35 with macro-metastasis. The sensitivity of FS for detecting micro-metastasis was 33% in the NC group compared to 43% in the non-NC group. The sensitivity of FS for detecting macro-metastasis was 60% in the NC group compared to 76 % in the non-NC group. The specificity was 100% for both groups. The false negative rate in the NC group vs non-NC group was 66.7% vs 60% for micrometastasis and 40% vs 23.9% for macro-metastasis.
Conclusions: 1) The overall sensitivity of FS for intraoperative evaluation for detection of micrometastasis (33%vs43%) and macrometastases (60%vs76%) was lower in patients who received pre-operative chemotherapy in comparison to those who did not receive any such therapy. 2)The false negative rate for detection of macro-metastasis was lower than that of micro-metastasis in both groups.
Category: Breast

Monday, March 9, 2009 9:30 AM

Poster Session I Stowell-Orbison/Autopsy Award # 36, Monday Morning