Novel Nomograms-Based Prediction Models for Patients with Primary Undifferentiated Pleomorphic Sarcomas Resections

Abstract

Background: Undifferentiated pleomorphic sarcomas (UPS) were one of the most common soft tissue sarcomas. As UPS had relatively high potentials of recurrence and metastasis, we designed two nomograms to better predict the overall survival (OS) and time to recurrence (TTR) for patients who underwent primary surgery. Methods: The data of UPS patients who underwent primary surgery were extracted from Shanghai Cancer Center, Fudan University. Multivariate analyses were performed using Cox proportional hazards regression to identify independent prognostic factors. Kaplan–Meier analysis was used to compare differences for patients who underwent primary surgery in OS and TTR. Nomograms were designed with the help of R software and validated using calibration curves and receiver operating characteristic curves (ROC). Results: Kaplan–Meier curves showed that patients with older ages (p = 0.0024), deeper locations (p = 0.0422), necrosis (p < 0.0001), G3 French Federation Nationale des Centres de Lutte Contre le Cancer (FNCLCC) classification (p < 0.0001), higher Ki-67 (p < 0.0001), higher mitotic index (p < 0.0001), R1/R2 resections (p = 0.0002) and higher invasive depth (p = 0.0099) had shorter OS than the other patients while patients with older ages (p = 0.0108), necrosis (p = 0.0001), G3 FNCLCC classification (p < 0.0001), higher Ki-67 (p = 0.0006), higher mitotic index (p < 0.0001) and R1/R2 resections (p < 0.0001) had shorter TTR compared with those without. Multivariate analyses demonstrated that mitotic rates and surgical margin were independent factors for TTR while age and invasive depth were independent factors for OS. Three parameters were adopted to build the nomograms for 3- and 5-year OS and TTR. The Area Under Curve (AUC) of this nomogram at 3- and 5-year TTR reached 0.802, 0.814, respectively, while OS reached 0.718, 0.802, respectively. Calibration curves for the prediction of 3- and 5-year OS and TTR showed excellent agreement between the predicted and the actual survival outcomes. Conclusions: Some important parameters could be used to predict the outcome of individual UPS patients such as mitotic age, rates, surgical margin, and invasive depth. We developed two accurate and practicable nomograms that could predict 3- and 5-year OS and TTR for UPS patients, which could be involved in the modern medical decision-making process.