Validation and limitations of use of a breast cancer nomogram predicting the likelihood of non-sentinel node involvement after positive sentinel node biopsy

Background: Axillary lymph node dissection (ALND) for patients with positive sentinel lymph nodes (SLNs) is currently under discussion in the literature. The breast cancer nomogram (BCN), an online tool developed by the Memorial Sloan-Kettering Cancer Center (MSKCC), aims to predict the risk of positive non-SLN in SLN-positive patients. The purpose of this study was to test the accuracy of the nomogram on patients with macrometastatic and micrometastatic SLN-positive biopsy findings. Methods: Patient characteristics, tumor pathology, and positive SLN characteristics were collected on 588 consecutive patients who underwent completion ALND. The MSKCC BCN tool was used to calculate risk of metastases for all 588 cases that included a subgroup of the 213 patients with SLN micrometastases. The BCN was performed for positive SLN biopsy findings regardless of the method of metastasis detection. Evaluation of the BCN was performed by the area under the curve method. Results: The BCN applied to all 588 patients achieved an area under the receiver operating characteristic curve (ROC) of .724 (range, .677-.771) compared with .76 in the MSKCC study. When the tool was applied solely to micrometastases found by hematoxylin and eosin staining and metastases found by immunohistochemistry, the area under the ROC was .538 (range, .423-.653). Conclusions: The MSKCC nomogram has been validated for all the patients having a metastatic SLN at the Institut Curie. However, this model was not reliably predictive for positive non-SLN in cases with micrometastic positive SLN.