Research Paper Volume 16, Issue 2 pp 1536—1554

Figure 5. Using GSE17118 as a validation dataset for the risk model. (A) Patients who died during follow-up had a higher risk score than those who survived. (B) Risk score was found to be a useful tool to assess the prognosis of sarcoma patients based on principal component analysis. (C) Patients with higher risk scores had a worse overall survival according to Kaplan-Meier analysis. (D) As shown in the ROC curve, risk scores had good predictive value for survival in sarcoma patients (AUC > 0.72). (E) Risk score was found to be an independent risk factor for OS in both univariate and multivariate Cox regression analyses. (F) The risk score’s predictive power was significantly higher than that of other clinical characteristics, according to the C-index analysis. Abbreviations: HR: hazard ratio; ROC: receiver operating characteristic. Statistical significance is indicated by an asterisk: ^***p < 0.001.