| Background and ObjectiveUndifferentiated pleomorphic sarcoma(UPS)is the most common subtype of soft tissue sarcoma.UPS is more common in deep soft tissues,which makes it difficult to be diagnosed in early clinical practice.At the same time,UPS is prone to local recurrence and metastasis after operation,and the prognosis is poor.Current studies have shown that UPS prognosis is associated with some clinical features.However,there is no report on the survival prediction models for overall survival(OS)and cancer-specific survival(CSS)of UPS patients based on the same large sample data.In this study,based on the advantages of large samples in the Surveillance,Epidemiology,and End Results(SEER)database,the clinical information of UPS was retrieved and analyzed,and then the independent influencing factors affecting the prognosis were determined.On this basis,the 3-year and 5-year CSS and OS prognostic prediction models of UPS patients were constructed and validated,to provide the reference for clinical treatment decisions of UPS.MethodAccording to inclusion and exclusion criteria,a total of 1079 patients were screened from the SEER database and randomly divided into training cohort(n=755)and validation cohort(n=324).Firstly,the demographic and clinicopathological characteristics of the patients were described,and the correlation between the training cohort and the validation cohort was compared by the chi-square test.Kaplan-Meier method and univariate Cox regression analysis were used to analyze the possible prognostic factors included.Multivariate Cox regression analysis was performed to further analyze the prognostic factors with statistically significant differences and determine the independent prognostic factors.The method was stepwise backward regression.Finally,based on the identified independent prognostic factors,the R language was used to establish nomograms of CSS and OS for patients with UPS.The concordance index(C-index),the area under curve(AUC)of the receiver operating characteristic(ROC)curve,and the calibration curve were used to evaluate the discrimination and calibration of the nomogram.Decision curve analysis(DCA)was used to evaluate the clinical utility of the model.Result1.Survival analysis:In this study,both the median overall survival and 60%cancer-specific survival of UPS patients were 83 months,and no cancer-specific median survival was observed.The 3-year and 5-year CSS were 71.4%and 64.0%respectively,and the 3-year and 5-year OS were 65.5%and 55.6%respectively.2.Independent prognostic factors:By combining the results from the K-M method and the Cox regression analysis,age,tumor site,tumor size,tumor depth,N stage,M stage,tumor grade,surgery,and radiotherapy were identified as independent prognostic factors for CSS and OS in UPS patients.Surgery and radiotherapy could improve the prognosis of patients to some extent.3.Construction of nomogram:Nomograms of 3-year and 5-year CSS and OS were established based on the above independent prognostic factors.The results showed that the top three factors that had great influence on CSS were M stage,tumor size and surgery,and the top three factors that had great influence on OS were M stage,age and tumor size.4.Validation of nomogram:(1)C-index and AUC:The C-indexes of nomograms of the CSS and OS in the training cohort were 0.776 and 0.756 respectively,and the C-indexes of those in the validation cohort were 0.770 and 0.757,respectively.Both results were higher than those in the TNM stage system(0.717 and 0.673).A similar conclusion could be observed in AUC.In the training cohort,the AUC of nomogram of 3-year and 5-year CSS were 0.805 and 0.814,and the AUC of nomogram of 3-year and 5-year OS were 0.785 and 0.795,respectively.In the validation cohort,both AUC of nomogram of 3year and 5-year CSS were 0.821,and the AUC of nomogram of 3-year and 5-year OS were 0.814 and 0.812,respectively.For the TNM staging system,the AUC of 3-year and 5-year were 0.733 and 0.750(CSS),and 0.703 and 0.712(OS),respectively,all of which were lower than the data demonstrated in the nomogram.(2)Calibration curve:The calibration curves of the nomogram and the TNM staging system fitted well with the 45° line,showing the consistency of prediction probability and actual probability.(3)Decision curve analysis:The results of decision curve analysis showed that both the nomogram and the TNM staging system had clinical net benefit within a wide range of threshold probability,and the nomogram had higher clinical net benefit than the TNM staging system on the whole.Conclusion1.Age,tumor site,tumor size,tumor depth,N stage,M stage,tumor grade,surgery,and radiotherapy were independent prognostic factors for undifferentiated pleomorphic sarcoma.Surgery and radiotherapy could improve the prognosis of patients to some extent.Whereas,the effect of gender,race,and chemotherapy on prognosis was not found.2.The nomogram established in this study could help predict 3-year and 5-year CSS and OS of UPS patients individually,with good discrimination,calibration,and clinical practicability.This could provide reference for clinicians and patients to make better clinical decisions. |
PDF Full Text Request