The establishment and evaluation of a new model for the prediction of Children B-ALL based on TARGET: A SQUIRE-compliant study.

摘要: B lymphocytic leukemia (B-ALL) is a hematopoietic malignant disease characterized by an accumulation of early cells. This study aimed to construct children B-ALL Nomogram prediction model based on Therapeutically Applicable Research Generate Effective Treatments database, so as further guide clinical diagnose and treatment.Clinical data related were collected from the TARGET among which, stage II used model, while I utilized external verification model. The factors analyzed through Lasso regression analysis screen risk for construction In addition, capacity accuracy verified internally externally using ROC curve, C-index calibration respectively.A total 1316 enrolled in this study. revealed that, Age, Gender, WBC, CNSL, MRD29, BMR, CNS R, BCR-ABL1, BMA29, DS, DI important prognostic factors. values internal models 0.870 0.827, respectively, revealing ideal discriminating capacity. Besides, curve had high contact ratio, which suggested favorable consistency between incidence predicted actual incidence. Moreover, AUC 0.858, 0.787, 0.898, 0.867, indicating predicting 3- 5-year survival rates with B-ALL.The plotted exhibits practicability rate children, contributes patients screening intervention.