Risk Factors for Non-Sentinel Lymph Node Metastasis in Breast Cancer Patients with Positive Sentinel Lymph Node
H Gobbi, HS Bartels, VFZ Marinho, ACS Porto, SMT Carvalho, CABT Osorio, A Ribeiro-Silva, GE Silva, FA Soares. Federal University Minas Gerais, Belo Horizonte, Brazil; Hospital A.C.Camargo, Sao Paulo, Brazil; University Sao Paulo, Ribeirao Preto, Brazil
Background: The role of complete axillary lymph node dissection (ALND) after identification of metastasis in sentinel lymph node (SLN) biopsy of woman with breast cancer has been questioned. Up to 60% of patients with positive SLN are found to have no other metastasis in non sentinel lymph node (NSLN). The aim of our study was to evaluate risk factors for NSLN metastasis in SLN-positive patients and to propose a mathematical model (nomogram) to predict the likelihood of finding additional positive nodes at ALND.
Design: We reviewed 326 cases of patients with breast cancer and positive-SLN divided into two groups according to the nodal involvement in the ALND: patients with all non-sentinel nodes negative for metastasis and patients with at least one positive NSLN. Pathological features of the primary tumor (tumor size, histological tumor type and grade, mitotic index, nuclear grade, lymphovascular invasion, estrogen and progesterone receptor status), and SLN (number of SLN, detection method of metastasis and size of SLN metastasis) were assessed. Data were submitted to univariate and multivariate logistic regression, followed by construction of a mathematical model (nomogram) to predict the presence of additional disease in the non-SLN of these patients.
Results: The univariate and multivariate analyses identified the following risk factors for involvement of NSLN with the respective p values: length of the largest SLN metastasis (p <0.001, p = 0.002), number of positive SLN (p = 0.006, p = 0.04) and negative SLN (p = 0.010, p = 0.004), and lymphovascular invasion (LVI), (p = 0.075, p = 0.085). The nomogram was created using size of largest SLN metastasis, number of positive and negative SLN, and LVI. The nomogram was discriminating, with an area under the receiver operating characteristic (ROC) curve of 0.70.
Conclusions: Our data showed that the size of largest SLN metastasis, number of positive and negative SLN were predictive risk factors for metastatic involvement of NSLN in patients with positive-SLN. Our nomogram, similar to other models, may represent an additional tool to help physicians and patients who decide whether or not a complete ALND should be performed.
Wednesday, March 24, 2010 9:30 AM
Poster Session V # 46, Wednesday Morning