A Novel Nomogram for Estimating a High-Risk Result in the EndoPredict® Test for Estrogen Receptor-Positive/Human Epidermal Growth Factor Receptor 2 (HER2)-Negative Breast Carcinoma.

Background/Objectives: The EndoPredict^® assay has been widely used in recent years to estimate the risk of distant recurrence and the absolute chemotherapy benefit for patients with estrogen (ER)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer. However, there are no well-defined criteria for selecting patients who may benefit from the test. The aim of this study was to develop a novel nomogram to estimate the probability of obtaining a high-risk EndoPredict^® result in clinical practice. Methods: The study cohort comprised 348 cases of T1-3/N0-1a/M0 ER-positive/HER2-negative breast carcinoma. A multivariate analysis was conducted using a training cohort (n = 270) based on clinicopathological features that demonstrated a statistically significant correlation with the EndoPredict^® result in a univariate analysis. The predictive model was subsequently represented as a nomogram to estimate the probability of obtaining a high-risk result in the EndoPredict^® assay. The predictive model was then validated using a separate validation cohort (n = 78). Results: The clinicopathological features incorporated into the nomogram included tumor size, tumor grade, sentinel lymph node status, pN stage, and Ki67. The internal validation of the model yielded an area under the curve (AUC) of 0.803 (95% CI = 0.751, 0.855) in the receiver operating characteristic (ROC) curve for the training cohort, with an optimal sensitivity and specificity at a threshold of 0.536. The external validation yielded an AUC of 0.789 (95% CI = 0.689, 0.890) in its ROC curve, with optimal sensitivity and specificity achieved at a threshold of 0.393. Conclusions: This study presents, for the first time, the development of a clinically accessible nomogram designed to estimate the probability of obtaining a high-risk result in the EndoPredict^® assay. The use of easily available clinicopathological features allows for the optimization of patient selection for the EndoPredict^® assay, ensuring that those who would most benefit from undergoing the test are identified.