A Systematic Review of Evidence on the Clinical Effectiveness of Surveillance Imaging in Children With Medulloblastoma and Ependymoma

Surveillance imaging aims to detect tumour relapse before symptoms develop, but it's unclear whether earlier detection of relapse leads to better outcomes in children and young people (CYP) with medulloblastoma and ependymoma. This systematic review aims to identify relevant literature to determine the efficacy of surveillance magnetic resonance imaging (MRI) for CYP with medulloblastoma and ependymoma compared to symptomatic detection. 11 databases and 2 trial registries were searched in March 2025. Studies evaluating MRI surveillance imaging in CYP with medulloblastoma and ependymoma were included. The primary outcome of interest was overall survival (OS) from diagnosis. Studies were screened independently. Data extraction/quality assessment (using QUIPs) were conducted by one reviewer and checked by a second. Narrative synthesis and post-hoc meta-analyses of the proportion of relapses detected by surveillance imaging were conducted. Of 9,575 records screened, seven studies including 196 CYP with medulloblastoma and 309 with ependymoma were eligible. All were deemed moderate/high risk of bias in at least one domain. Single-proportion meta-analysis showed most relapses were detected by surveillance imaging in medulloblastoma (66.7%; 95% CI:60.1-73.2%) and ependymoma (72.6%; 95% CI:67.6-77.7%). Data on OS from diagnosis by method of relapse detection was reported in two studies: neither provide conclusive evidence that earlier detection improves survival. We conclude that while surveillance imaging detects relapses more frequently than symptomatic detection, there is limited high-quality evidence that earlier detection improves survival. Future prospective research should be conducted and should provide more granular reporting of patient characteristics and survival outcomes from diagnosis/end of treatment.