A scoring system to predict nonsentinel lymph node status in breast cancer patients with metastatic sentinel lymph nodes: A comparison with other scoring systems

Background: The majority of breast cancer patients with metastatic sentinel lymph node (SLN) do not harbor additional metastasis in non-SLN. It is unclear which patients with metastatic SLN require axillary lymph node dissection (ALND). The aim of this study was to identify predictive factors of non-SLN metastasis and to develop a scoring system. Methods: The training dataset consisted of 184 breast cancer patients. The independent validation dataset consisted of 82 breast cancer patients. The receiver operating characteristic (ROC) curve was drawn and the area under the ROC curve (AUC) was calculated to assess the discriminative power of the scoring systems. Results: Multivariate analysis revealed that non-SLN status was predicted by preoperative ultrasonographic findings of the axilla, lymphovascular invasion, increasing tumor size, increasing number of metastatic SLN, and decreasing number of nonmetastatic SLN. Based on multivariate logistic regression, we developed a scoring system for predicting non-SLN metastasis. The AUC for our scoring system was superior to other published scoring systems when identical validation data were applied. Conclusion: The likelihood of metastatic non-SLN correlated with preoperative ultrasonographic findings of the axilla, increasing pathologic size of the primary tumor, presence of lymphovascular invasion, increasing number of metastatic SLN, and decreasing number of nonmetastatic SLN. Our scoring system appears to be effective and accurate for selecting patients for whom ALND can be avoided.