Objective: This study aimed to develop and validate a nomogram and risk stratification system for the overall survival of pediatric patients with medulloblastoma after surgical repair.
Patients and methods: In this multicenter, retrospective study, consecutive patients who underwent surgery for medulloblastoma at Shanghai Children's Medical Center and the First Affiliated Hospital of Fujian Medical University from 2010 to 2022 formed the training and external validation datasets, respectively. Univariable and multivariable Cox regression analyses were performed to identify variables associated with mortality in the training dataset. A nomogram prediction model was developed based on independent variables in the multivariable Cox regression analysis to predict the 1-, 3-, and 5-year overall survival. The area under receiver operating characteristic curve (AUC) and calibration curve were used to evaluate the discrimination and calibration of the nomogram. A risk stratification system based on the median risk score was also established to divide patients into two risk groups.
Results: In the training dataset, Cox regression analyses identified tumor size, brainstem involvement and chemotherapy as independent predictors for overall survival. The AUC of the nomogram was 0.75 at 1 year, 0. 75 at 3 years, 0.77 at 5 years in the training dataset, 0.74 at 1 year, 0.70 at 3 years, and 0.70 at 5 years in the validation dataset. The calibration curve for the probability of 1-, 3-, and 5-year survival showed good agreement between the nomogram prediction and actual observation in the training and validation datasets. The risk stratification system could perfectly classify patients into two risk groups, and the overall survival in the two groups had a good division.
Conclusions: This low-cost, convenient, and noninvasive nomogram can be translated into clinical practice as a tool for risk stratification and individualized prognosis prediction for children with medulloblastoma.