Abstract P1-01-21: Is it safe to skip intra-operative frozen section examination of sentinel lymph node for core biopsy-diagnosed DCIS?

Abstract

Abstract Background Our initial experience on sentinel lymph node (SLN) biopsy with on-table frozen section examination has shown to be an accurate alternative to axillary dissection in the staging of invasive breast cancer. Patients with a preoperative diagnosis of ductal carcinoma in-situ (DCIS) by core biopsy follow a similar treatment pathway because the risk of axillary metastasis and histological upstaging in the final excisional specimen is not negligible. Intra-operative frozen section reporting facilitates axillary clearance in a single operation when necessary. However, with the declining role of completion axillary dissection for those with minimal SLN involvement, its significance in DCIS is questioned. Methods A retrospective analysis was performed for patients with core biopsy-diagnosed DCIS undergoing SLN biopsy at the time of definitive surgery. Those with evidence of microinvasion in the core biopsy specimen were excluded. SLN were identified by blue dye, radioactive isotope, or combined mapping technique, and examined by intra-operative frozen section. Full axillary dissection was performed if frozen section was positive for metastasis, whereas patients with failed SLN localization were spared from the procedure. All lymph nodes harvested would undergo further paraffin section with immuno-histochemical staining. The results of SLN biopsy in relation to the final pathology of the primary tumour and the axillary disease status were analysed. Results From March 2002 to March 2013, 297 patients who had a core biopsy diagnosis of DCIS underwent simultaneous SLN biopsy with their definitive excisional operation. SLN localization was successful in 283 patients (95.3%). Metastases in SLN were found in 19 (6.4%) of them. Axillary dissection, either in the same or second operative setting, was completed in 12 (4%) patients. Only four (1.3%) showed additional positive lymph nodes in the rest of axilla. All of these four patients had invasive disease revealed in their primary lesions upon final pathological assessment. Overall, 83 patients (27.9%) had histological upstaging from DCIS to microinvasive or invasive carcinoma. It was more commonly associated with the presence of a palpable (48.5% versus 21.6%, p = 0.0067) or radiological (46.3% versus 19.7%, p = 0.0047) mass lesion. Out of the 214 patients who remained in the category of “pure DCIS”, six (2.8%) had SLN metastases. Three proceeded to axillary dissection and their SLN were confirmed to be the only positive ones in the axilla. Conclusion Although SLN metastases were rare in patients with “pure DCIS”, underestimation of invasive disease was frequent on core biopsy. Performing SLN biopsy during definitive surgery for DCIS allowed correct nodal staging in a single operation. Nevertheless, in contrast to invasive carcinoma, frozen section examination could safely be foregone since the additional yield of completion axillary dissection was shown to be slim in “pure DCIS” even if SLN shows positive results. However, management of SLN in DCIS with high risk features should follow that of invasive breast cancer in view of the considerable chance of histological upstaging. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-01-21.