Prediction score model for non-sentinel and four or more nodal metastases using a combined method of one-step nucleic acid amplification and histology in sentinel node-positive breast cancer patients.

BACKGROUND: ALN dissection (ALND) is the only way to obtain information on ALN metastasis status accurately when sentinel lymph node (SLN) metastasis is present. In this study, we established a model for intraoperatively predicting non-SLN metastasis and the presence of four or more ALNs (pN2), based on the combined use of one-step nucleic acid amplification (OSNA) and histological examination following SLN biopsy.

MATERIALS AND METHODS: Subjects comprised 318 consecutive breast cancer patients (cTis-3, N0) who underwent SLN biopsy with a combination of OSNA and histological examination, and who were found to have SLN metastasis and were treated by ALND. We allotted points to each patient according to their SLN metastasis status as defined by both OSNA and histology, then defined the "National Cancer Center sentinel lymph node metastatic score" (NCS score) based on the total points. Correlations between the NCS score and both non-SLN metastasis and pN2 status were analyzed by logistic regression analysis. The accuracy of this score was evaluated using receiver operating characteristic (ROC) analysis.

RESULTS: The NCS score was significantly correlated with both non-SLN metastasis and pN2 status (adjusted odds ratio: 1.26 for non-SLN metastasis, 1.56 for pN2 status). The area under the ROC curve (AUC) of the NCS score demonstrated 0.74 for non-SLN metastasis, 0.91 for pN2 status.

CONCLUSIONS: The NCS score was a strong independent predictor of non-SLN metastasis and pN2 status. Use of this score will facilitate the selection of optimal adjuvant therapies without requiring unnecessary ALND.