Development and validation of nomograms for predicting survival in patients with de novo metastatic triple-negative breast cancer

Abstract

AbstractMetastatic triple-negative breast cancer (mTNBC) is a heterogeneous disease with a poor prognosis. Individualized survival prediction tool is useful for this population. We constructed the predicted nomograms for breast cancer-specific survival (BCSS) and overall survival (OS) using the data identified from the Surveillance, Epidemiology, and End Results database. The Concordance index (C-index), the area under the time-dependent receiver operating characteristic curve (AUC) and the calibration curves were used for the discrimination and calibration of the nomograms in the training and validation cohorts, respectively. 1962 mTNBC patients with a median follow-up was 13 months (interquartile range, 6–22 months), 1639 (83.54%) cases died of any cause, and 1469 (74.87%) died of breast cancer. Nine and ten independent prognostic factors for BCSS and OS were identified and integrated to construct the nomograms, respectively. The C-indexes of the nomogram for BCSS and OS were 0.694 (95% CI 0.676–0.712) and 0.699 (95% CI 0.679–0.715) in the training cohort, and 0.699 (95% CI 0.686–0.712) and 0.697 (95% CI 0.679–0.715) in the validation cohort, respectively. The AUC values of the nomograms to predict 1-, 2-, and 3-year BCSS and OS indicated good specificity and sensitivity in internal and external validation. The calibration curves showed a favorable consistency between the actual and the predicted survival in the training and validation cohorts. These nomograms based on clinicopathological factors and treatment could reliably predict the survival of mTNBC patient. This may be a useful tool for individualized healthcare decision-making.