Development and validation of nomograms for predicting axillary non-SLN metastases in breast cancer patients with 1–2 positive sentinel lymph node macro-metastases: a retrospective analysis of two independent cohorts

Abstract Background It is reported that appropriately 50% of early breast cancer patients with 1–2 positive sentinel lymph node (SLN) micro-metastases could not benefit from axillary dissection (ALND) or breast-conserving surgery whole irradiation. However, whether SLN macro-metastases ALND remains unknown. The aim our study was to develop and validate nomograms for assessing non-SLN metastases in macro-metastases, using their pathological features alone combination STMs. Methods We retrospectively reviewed STMs 1150 two independent cohorts. Best subset regression used feature selection signature building. risk score calculated each patient as a linear selected predictors were weighted by respective coefficients. Results pathology-based nomogram possessed strong discrimination ability metastases, an area under the receiver operating characteristic (ROC) curve (AUC) 0.727 (95% CI: 0.682–0.771) primary cohort 0.722 0.653–0.792) validation cohort. addition CA 15–3 CEA can significantly improve performance (AUC: 0.773 (0.732–0.815) vs. (0.682–0.771), P < 0.001) (0.777 (0.713–0.840) (0.653–0.792), 0.001). Decision analysis demonstrated clinically useful. Conclusion based on be identify SLN. In addition, more accurately than characteristics alone.