Validation of online calculators to predict the non-sentinel lymph node status in sentinel lymph node-positive breast cancer patients

Purpose This study evaluated and compared the predictive values of the Memorial Sloan-Kettering Cancer Center (MSKCC) nomogram and the Stanford nomogram for predicting non-sentinel lymph node (SLN) metastasis in patients with SLN metastasis, which were the only nomograms available online, and verified their usefulness in the macrometastasis or micrometastasis/isolated tumor cell (ITC) subgroups. Methods Eighty-nine patients with a positive SLN biopsy who underwent axillary lymph node dissection were analyzed. The predicted probability of non-SLN metastasis was calculated using a computerized model from the websites for each nomogram. The results were compared using the area under the curve (AUC) of the receiver operating characteristics curve for each model. The falsenegative and false-positive rates were also calculated. Results The AUC for the entire population was 0.701 with the MSKCC nomogram and 0.756 with the Stanford nomogram. The AUCs of macrometastasis and micrometastasis/ITC groups were 0.680 and 0.469 with the MSKCC nomogram, and were 0.676 and 0.574 with the Stanford nomogram, respectively. Although false-negative cases were not identified, the false-positive rates were high in both subgroups when using these nomograms. Conclusions This independent comparison found no significant difference between the two nomograms. In this study, these nomograms could not reliably predict positive non-SLN in patients with SLN micrometastasis/ITC.