Individualized Prediction of Survival Benefit From Locoregional Surgical Treatment for Patients With Metastatic Breast Cancer

Objective: Recently, performing locoregional surgical treatment still remains debatable in patients with metastatic breast cancer (MBC). Current study aimed to develop prognostic nomograms for predicting the long-term survival in MBC patients with or without surgical intervention, thereby assisting clinicians in making individualized choice. Methods: The training set included 5173 patients who were diagnosed with MBC in 2010-2013 from the Surveillance, Epidemiology, and End Results Program, while the validation set comprised 2924 patients diagnosed in 2014-2015. Multivariant Cox hazard model was applied to determine the independent risk factors for overall survival (OS) and breast cancer specific survival (BCSS). Then, individualized pre- and postoperative nomograms for predicting 1- or 3-year survival probabilities were constructed accordingly. Internal and external validations were conducted to determine the accuracy of these nomograms by calculating concordance index (C-index) and plotting calibration curves. Results: The survival analysis indicated that surgical management conferred improved OS and BCSS in patients with metastatic breast cancer. Age, T stage, grade, distant metastatic site, ER, PR and HER2 status, radiation, and chemotherapy were independent risk factors for OS and BCSS both in surgery and non-surgery group. All these factors were subsequently incorporated into the nomogram which showed acceptable predictive capabilities with C-index range of 0.65-0.80 both in training set and external validation set. In addition, a preoperative nomogram incorporating variables capable of being determined before surgery was also built with C-index above 0.70 both in training and validation set. Conclusion: Surgical management in patients with metastatic breast cancer suggests a potential survival advantage. In addition, these well-validated pre- and postoperative nomograms may provide a useful tool to assist clinicians in treatment decision-making and in evaluating patients’ long term prognosis.