BackgroundThis study aimed to construct a nomogram for Breast sarcoma (BS) to predict the prognosis of patients with BS accurately and provide a theoretical basis for individualized treatment.MethodsPatients selected from the Surveillance, Epidemiology and End Results (SEER) database from 2000 to 2018 were assigned to a training group (TG, n = 696) and an internal validation group (IVG, n = 299) at a 7:3 ratio. Cox regression analysis was performed on the TG, and statistically significant factors were used to establish a nomogram to predict 3-, 5-, and 10-year overall survival (OS). The nomogram’s predictive power was validated using data from patients who attended our institution as the external validation group (EVG, n =79).ResultsCox regression analysis identified five factors, which were used to construct the nomogram. Good prediction accuracy was demonstrated using calibration curves. The concordance (C) indices for TG = 0.804 (95% confidence interval (CI) 0.777–0.831) and IVG = 0.761 (0.716–0.806) were higher than those based on 8th American Joint Committee on Cancer (AJCC8) stage: TG = 0.695 (0.660–0.730), IVG = 0.637 (0.584–0.690). The EVG also had a high C-index: 0.844 (0.768–0.920). Decision curve analysis showed that nomogram has larger net benefits than the AJCC8. The Kaplan–Meier curves of the nomogram-based risk groups showed significant differences (p < 0.001).ConclusionsThe nomogram could accurately predict 3-, 5-, and 10-year OS and provided nomogram-based risk stratification, which could help physicians to personalize treatment plans for patients with BS.
Read full abstract