| Background: High-grade endometrial mesenchymal sarcoma(HGESS)is a malignant tumor occurring mainly in the body of the uterus,for which no specific incidence statistics are available,with a high degree of malignancy,easy metastasis and recurrence in early stages,and low survival rates.It is highly susceptible to occur in middle-aged and elderly women,and patients have a high proportion of competing risks during follow-up,posing new challenges to the establishment of cancer prediction models.Therefore,we need to further understand the causes of patients’ death,analyze the competing factors of their death,explore the prognostic influences of high-grade endometrial mesenchymal sarcoma-specific death,and develop HGESS intervention,monitoring strategies to improve patients’ prognosis and survival quality.Objective: To analyze the factors influencing the prognosis of patients with high-grade endometrial mesenchymal sarcoma by various methods such as constructing competing risk models and drawing nomogram plots,and to provide new ideas for the prognosis analysis of patients with HGESS.Results: In this study,the overall survival time of all patients ranged from 0 to 192 months,with a median survival time of 9 years;the 3-year cancer-specific mortality rate in the training set was 45.6% and the 5-year cancer-specific mortality rate was 48.9%,with a mean follow-up time of 4.394 years;the 3-year cancer-specific mortality rate in the validation set was 33.4% and the 5-year cancer-specific mortality rate was 44.9%,with a mean The follow-up time was 4.778 years.The results of univariate survival analysis suggested that patients’ age,marital status,tumor stage,chemotherapy status,radiotherapy status,surgery status,tumor size and lymph node detection status were possible predictors of death in patients with high-grade endometrial mesenchymal sarcoma and were associated with patients’ survival time,after excluding variables withabsolute coefficients equal to 0 using Lasso regression.chemotherapy status,surgery status,tumor size and lymph node detection status could be used as predictors affecting death in patients with high-grade endometrial mesenchymal sarcoma.The consistency index of the training set of line plots in this study was 0.740(95% CI:0.684-0.796),the area under the ROC curve for 3-year survival was 0.709,and the area under the ROC curve for 5-year survival was The internal validation of the prediction model constructed in this study was performed by 10-fold cross-validation,and the test results showed that the fit of the calibration curves for the training and validation sets at 3 and5 years was 0.765 and 0.763,respectively.5-year calibration curves fit well,indicating good agreement between the actual observed and model predicted values.Conclusion: High-grade endometrial mesenchymal sarcoma is an extremely rare type of cancer,and research on its etiology,diagnosis,and prognosis-related factors is still in progress,but there is no doubt that,as a malignant tumor with a high mortality rate,when there is a competing risk,we should make a judgment based on the actual situation of the patient,use the correct method for data processing and survival analysis,and construct a more accurate survival prognostic model.In this study,we analyzed the data in the SEER database,used Lasso regression and univariate and multifactor competing risk model regression analysis to screen for predictors of mortality specific to high-grade endometrial mesenchymal sarcoma,and after excluding competing events,we successfully constructed a 3-year and 5-year competing risk model for patients with HGESS,and used it to draw a The Nomogram graph,with the consistency index,area under the ROC curve and correction curve,proved that this line graph has good discriminatory ability,high reliability and accuracy,and can provide reference for future clinical decisions. |
PDF Full Text Request