EP031/#958 Is the extemporaneous pathological examination reliable for the evaluation of surgical margins during conservative treatment for breast cancer?

Abstract

Objectives For a long time, axillary lymph node dissection (ALND) was routinely performed in patients with an involved sentinel lymph node (SLN). However, in 30 to 50% of cases, the non-sentinel lymph nodes (NSLN) were not involved, and these patients would have suffered the morbidity of ALND excessively. The aim of our study was to identify the risk factors for NSLN involvement in patients with a positive SLN. Methods We included patients with early breast cancer and positive sentinel node who underwent ALND in Salah Azaiez Institute of Oncology between 2005 and 2018. We analyzed retrospectively the clinicopathological data to predict NSLN involvement. Results Among the 77 selected patients, 36% did not have any NSLN involvement during the pathological examination of the ALND product. Univariate analysis using a=0.05 as the significance level, showed that radical surgery(p=0.05), tumor size>30 mm(p=0.01), number of extracted SLN£2 (p=0.02), number of positive SLN>1(p=0.01), ratio positi-veSLN/Extracted SLN>0.5(p=0.05), macrometastasis(p<10 -3 ), SBRIII grade(p=0.007), and Ki67>20%(p=0.04) were predictive of NSLN involvement. In multivariate analysis, the type of surgery, the tumor size, the Ki67 level and the ratio Positive SLN/Extracted SLN were excluded. Only the number of extracted SLN£2 (OR=18.518, CI=1.402-250, p=0.027), the number of positive SLN>1(OR=9.624, CI=1.266-73.172, p=0.029), SBRIII grade (OR=58.82, CI=2.86-1000, p=0.008), and macrometastasis (OR=759.19, CI=10.166-56698.2, p=0.003) were found to be independent risk factors of NSLN involvement. Conclusions Our results prove that there is a correlation between tumors' clinicopathological features and NSLN involvement. Therefore, a careful study of these criteria could avoid unnecessary ALND in patients with positive SLN who do not need it.