Diffusion tensor invasive phenotypes can predict progression-free survival in glioblastomas.

Area of Science:

• Neuro-oncology
• Medical imaging
• Radiology

Background:

• Glioblastoma multiforme (GBM) is a highly aggressive brain tumor with poor prognosis.
• Tumor invasion into surrounding brain tissue limits treatment efficacy.
• Previous studies indicate variable GBM invasion patterns, ranging from <1 cm to >3 cm from the tumor core.

Purpose of the Study:

• To utilize Diffusion Tensor Imaging (DTI) to characterize GBM invasive patterns.
• To correlate identified invasion patterns with progression-free survival (PFS).

Main Methods:

• Retrospective analysis of 25 GBM patients treated per EORTC/NCIC protocol.
• Post-operative imaging using 1.5 T MRI with standard anatomical and DTI sequences.
• Classification of patients into diffuse, localized, or minimally invasive patterns based on DTI-derived p and q maps.

Main Results:

• Median PFS differed significantly across invasion patterns: 278 days (diffuse), 605 days (localized), and 820 days (minimally invasive).
• A minimally invasive pattern was associated with a significantly longer PFS.
• Three-quarters of patients with a minimally invasive phenotype remained progression-free at 24 months (p=0.010).

Conclusions:

• Diffusion tensor imaging can differentiate three distinct invasive phenotypes in GBM.
• These phenotypes exhibit significantly different progression-free survival outcomes.
• A minimally invasive GBM phenotype, observed in approximately 20% of patients, predicts a delayed progression.