Molecular subtypes, prognostic factors, and treatment optimization in pediatric medulloblastoma: a real-world study from China

Abstract

Background: Medulloblastoma (MB) is the most common malignant pediatric brain tumor, yet systematic studies on molecular characteristics and treatment efficacy in Chinese pediatric MB remain scarce. This study evaluates recurrence and mortality risk factors and the feasibility of intensified chemotherapy.

Methods: A retrospective analysis of 381 MB patients (WNT: 63, SHH: 106, Group 3: 27, Group 4: 185) was conducted. Kaplan-Meier analysis estimated survival rates, and Cox regression identified independent risk factors for recurrence and mortality.

Results: With a median follow-up of 4.8 years, 5-year PFS and OS were 69.9% ± 2.4% and 80.6% ± 2.1%, respectively. WNT-MB had the best prognosis, while Group 3-MB had the worst. Independent recurrence risk factors included high-risk status (HR=2.931, p<0.001), TP53 mutation (HR=1.873, p<0.001), MYCN amplification (HR=1.52, p=0.001), chromosome 12p amplification, and 9q deletion. Mortality was associated with LC/A pathology (HR=1.573, p=0.007), TP53 mutation (HR=2.049, p<0.001), and high-risk status (HR=3.966, p<0.001). TP53 mutations influenced WNT-MB recurrence, and Group 3-MB showed a high recurrence risk even without MYC amplification or metastasis. No treatment-related fatalities were observed.

Conclusion: This study identified key clinical and molecular risk factors associated with recurrence and mortality in pediatric medulloblastoma. The findings underscore the prognostic relevance of TP53 mutations, MYCN amplification, and specific chromosomal alterations, particularly in non-metastatic subgroups. These insights may help guide risk-adapted and personalized treatment strategies in future studies.

Keywords: large cohort; maintain chemotherapy by risk stratification; medulloblastoma; real-world study; risk factors for relapse and mortality.

Figure 1

Study flowchart. This flowchart provides a comprehensive overview of patient enrollment, exclusion criteria, and molecular classification. Additionally, it illustrates the distribution of patients into average- and high-risk categories and presents the chemotherapy completion rates. WNT, Wingless activated. SHH, Sonic Hedgehog activated.

Figure 2

Genomic landscape of pediatric patients with medulloblastoma. This figure illustrates the distribution of significant gene mutations and chromosomal variations among the four molecular subtypes of MB. Key findings include the high prevalence of CTNNB1 mutations and chromosome 6 deletions in the WNT subtype; TP53 mutations and chromosome 7 and 17 amplifications in the SHH subtype; MYC and MYCN amplifications in the Group 3 subtype; MYC, SMARCA4, and MYCN mutations along with chromosome 7 and 17 amplifications in the Group 4 subtype. Genomic alterations were identified via next-generation sequencing or combined with RNA-seq. CMB, classical MB; DMB, desmoplastic nodular MB; LC/A, large cell or anaplastic; SHH, Sonic Hedgehog activated; WNT, Wingless activated. SNV, Single Nucleotide Variant; M0, no evidence for metastatic; M+, metastatic spread of disease.

Figure 3

Survival Analysis by Clinical and Molecular Subgroups: (A) Kaplan-Meier curves for 5-year PFS and OS of all patients with MB. (B) PFS by histological subtype (CMB, DMB, LC/A). (C) PFS by molecular subtype (WNT, SHH, Group 3, Group 4). (D) PFS by risk stratification (average-risk vs. high-risk groups). (E) OS by histological subtype. (F) OS by molecular subtype. (G) OS by risk stratification. OS, overall survival; PFS, progression-free survival.AR, average risk; HR, high risk.CMB, classical MB; DMB, desmoplastic nodular MB; LC/A, large cell or anaplastic; SHH, Sonic Hedgehog activated; WNT, Wingless activated.

Figure 4

Risk factor modeling and survival analysis by new risk classification (A) Forest plot displaying HR and 95% CI for recurrence-associated factors in SHH M0 MB, estimated using univariate Cox regression analysis. (B) Kaplan-Meier survival curves for PFS by the proposed new risk classification in SHH M0 MB. (C) Forest plot showing HR and 95% CI for recurrence-associated factors in the combined cohort of Group 3 and Group 4 M0 MB, derived from univariate Cox regression analysis. (D) Kaplan-Meier PFS estimates stratified by the new risk classification in the combined cohort of Group 3 and Group 4 M0 MB. (E) PFS in the WNT subgroup stratified by TP53 mutation status; (F) PFS in the WNT subgroup stratified by M stage; (G) PFS in the no-MYC amplification subgroup by molecular; (H) PFS in the no-MYC amplification and M0 stage subgroup by molecular. HR, hazard ratio; CI, confidence intervals; MB, medulloblastoma; PFS, progression free survival; WNT, Wingless activated.