Related Experiment Video
Updated: Jul 14, 2026
Modeling Brain Metastases Through Intracranial Injection and Magnetic Resonance Imaging
Published on: June 7, 2020
Assessment of treatment response in recurrent medulloblastoma using craniospinal MRI
Juliane Hennenberg1, Leo Hofer1, Bence Nemeth2
1Medical University of Vienna, Department of Biomedical Imaging and Image-guided Therapy, Vienna, Austria.
Background:
To evaluate whether quantitative craniospinal MRI assessment of baseline tumor burden provides prognostic value in patients with recurrent, previously irradiated medulloblastoma undergoing MEMMAT (Metronomic Antiangiogenic) therapy.
Methods:
We analyzed craniospinal MRI of 40 patients with recurrent, previously irradiated medulloblastoma enrolled in the MEMMAT trial (April 1, 2014, and March 31, 2021). Ependymal, leptomeningeal, and local relapse lesions were retrospectively manually segmented on T1-contrast-enhanced (T1CE) and diffusion-weighted imaging (DWI) at baseline and follow-up (best response). Lesion count and volume were quantified. Bland-Altman analyses and intraclass correlation coefficients (ICC) assessed inter-sequence agreement., logistic regression evaluated associations between baseline tumor burden and progressive disease.
Results:
Patients achieving Complete Response (n = 6/40) showed mean monthly decreases of - 12.7% in T1CE lesion count and - 12.9% in volume. Partial Response patients (n = 9/40) showed similar declines (-10.3% and - 13.0%). Stable Disease patients (n = 5/40) demonstrated minimal decreases (-0.8% and - 2.3%). Progressive Disease patients (n = 16/40) showed increases of 33.7% in lesion count and 56.2% in volume. Logistic regression indicated trends toward higher baseline tumor burden predicting PD (volume OR 1.18, p = 0.08; lesion count OR 1.05, p = 0.075). Agreement between T1CE and DWI was limited (ICC 0.35 for lesion count, 0.47 for volume), with 23% of patients misclassified using either modality alone. Survival analyses demonstrated significantly shorter overall survival in patients with baseline lesion volumes ≥ 5.5 ml (log-rank p = 0.003), while a similar association for progression-free survival did not reach statistical significance (p = 0.053).
Conclusions:
Combined intracranial and intraspinal T1CE and intracranial DWI assessment improves response evaluation. Exploratory analyses suggested that higher baseline tumor burden may be associated with progression, although the observed associations were of borderline statistical significance and require validation in larger cohorts.
