Development and validation of nomograms integrating immune‐related genomic signatures with clinicopathologic features to improve prognosis and predictive value of triple‐negative breast cancer: A gene expression‐based retrospective study

Abstract Purpose Accumulating evidence indicated that triple‐negative breast cancer (TNBC) can stimulate stronger immune responses than other subtypes of breast cancer. We hypothesized that integrating immune‐related genomic signatures with clinicopathologic factors may yield a predictive accuracy exceeding that of the currently available system. Methods Ten signatures that reflect specific immunogenic or immune microenvironmental features of TNBC were identified and re‐analyzed using bioinformatic methods. Then, clinically annotated TNBC (n = 711) with the corresponding expression profiles, which predicted a patient's probability of disease‐free survival (DFS) and overall survival (OS), was pooled to evaluate their prognostic values and establish a clinicopathologic‐genomic nomogram. Three and two immune features were, respectively, selected out of 10 immune features to construct nomogram for DFS and OS prediction based on multivariate backward stepwise Cox regression analyses. Results By integrating the above immune expression signatures with prognostic clinicopathologic features, clinicopathologic‐genomic nomograms were cautiously constructed, which showed reasonable prediction accuracies (DFS: HR, 1.79; 95% CI, 1.46‐2.18, P