Objective: Medulloblastoma is the most common malignant intracranial tumor affecting the pediatric population. Despite advancements in multimodal treatment over the past 2 decades yielding a 5-year survival rate > 75%, children who survive often have substantial neurological and cognitive sequelae. The authors aimed to identify risk factors and develop a clinically friendly online calculator for prognostic estimation in pediatric patients with medulloblastoma.
Methods: Pediatric patients with a histopathologically confirmed medulloblastoma were extracted from the Surveillance, Epidemiology, and End Results database (2000-2018) and split into training and validation cohorts in an 80:20 ratio. The Cox proportional hazards model was used to identify the univariate and multivariate survival predictors. Subsequently, a calculator with those factors was developed to predict 2-, 5-, and 10-year overall survival as well as median survival months for pediatric patients with medulloblastoma. The performance of the calculator was determined by discrimination and calibration.
Results: One thousand seven hundred fifty-nine pediatric patients with medulloblastoma met the prespecified inclusion criteria. Age, sex, race, ethnicity, median household income, county attribute, laterality, anatomical location, tumor grade, tumor size, surgery status, radiotherapy, and chemotherapy were variables included in the calculator (https://spine.shinyapps.io/Peds_medullo/). The concordance index was 0.769 in the training cohort and 0.755 in the validation cohort, denoting clinically useful predictive accuracy. Good agreement between the predicted and observed outcomes was demonstrated by the calibration plots.
Conclusions: An easy-to-use prognostic calculator for a large cohort of pediatric patients with medulloblastoma was established. Future efforts should focus on improving granularity of population-based registries and externally validating the proposed calculator.
Keywords: artificial intelligence; machine learning; medulloblastoma; oncology; predictive analytics; survival; tumor.