Several statistical models have been developed to predict the likelihood of finding additional axillary nodal disease when the sentinel lymph node is found to harbor breast cancer. Hessman and colleagues¹ have presented an analysis of the accuracy of nomograms from Memorial Sloan-Kettering Cancer Center (MSKCC) and Stanford in predicting nonsentinel lymph node metastasis in patients with breast cancer treated at the Oregon Health & Science University from October 1, 1999, through January 31, 2008, that had positive sentinel node biopsies and subsequent axillary lymph node dissection.

Unlike many previously published reports validating the statistical accuracy of various nomograms for predicting nonsentinel lymph node metastases, Hessman and colleagues have attempted to demonstrate the clinical utility of these complex mathematical models. For patients with high risk of disease, how often is the tool right? The positive predictive value for patients with at least 80% risk of additional nodal disease was 91% for MSKCC and 62% for Stanford. For patients with low risk of disease, how often is the tool wrong? The false-negative rate for patients who had 10% or less risk of additional nodal disease was 4% for MSKCC and 8% for Stanford. Except for a small subset of patients with minimal nodal disease (N0i聽+聽or N1mi), the MSKCC nomogram outperformed the Stanford nomogram.