Advanced imaging reveals enhanced malignancy in glioblastomas involving the subventricular zone: evidence of increased infiltrative growth and perfusion

Michael Griessmair, Severin Schramm, Julian Ziegenfeuter, Julian Canisius, Kirsten Jung, Claire Delbridge, Friederike Schmidt‐Graf, Meike Mitsdoerffer, Claus Zimmer, Bernhard Meyer, Marie‐Christin Metz, Benedikt WiestlerJournal of Neuro-Oncology 2024 · DOI

TL;DR

This study demonstrates that glioblastomas infiltrating the subventricular zone (SVZ)—a known neural stem cell niche—exhibit distinct imaging signatures of increased malignancy. Using advanced MRI techniques, researchers found that SVZ involvement correlates with higher tumor perfusion and lower cell density, providing radiographic evidence for the more aggressive, infiltrative growth patterns associated with poor patient outcomes.

What Problem Does This Paper Address?

Glioblastoma is notorious for its "diffuse infiltrative growth," meaning the tumor doesn't just sit in one place; it weaves into healthy brain tissue, making complete surgical removal nearly impossible. Clinical data has long shown that tumors in contact with the subventricular zone (SVZ) result in significantly worse prognoses.

The SVZ is one of the few areas in the adult brain that harbors neural stem cells. The prevailing hypothesis is that tumors originating near or infiltrating this "nest" may acquire stem-cell-like properties, leading to more aggressive behavior. However, standard MRI often fails to capture the nuances of this biology. This paper seeks to bridge that gap by using advanced imaging—specifically perfusion and diffusion metrics—to see if the biological "aggressiveness" of SVZ-involved tumors can be quantified radiographically.

Methods

The researchers conducted a monocentric retrospective study using a highly automated pipeline to minimize human bias:

  • Automated Segmentation: They used the BraTS Toolkit and ANTs (Advanced Normalization Tools) to automatically segment tumor regions (necrosis, enhancing tumor, edema) and neurogenic zones (SVZ and the dentate gyrus) based on standardized brain atlases.
  • Perfusion Imaging: They calculated Cerebral Blood Volume (CBV) from Dynamic Susceptibility Contrast (DSC) MRI. CBV is a proxy for angiogenesis (the growth of new blood vessels), a hallmark of high-grade malignancy.
  • Advanced Diffusion Imaging: Standard Diffusion Tensor Imaging (DTI) can be "polluted" by extracellular fluid (free water). The authors used an Artificial Neural Network (ANN) to perform free-water correction. This allowed them to extract two key metrics:
  • FA-FWE: Free-water corrected fractional anisotropy, which measures the structural integrity of white matter tracts.
  • TVM: Tissue volume mask, which essentially estimates the fraction of a voxel occupied by solid tissue versus fluid, serving as a surrogate for cell density.

Datasets

The study analyzed a cohort of 137 therapy-naive patients diagnosed with IDH-wildtype, grade 4 glioblastoma.

  • Reference Atlases: The SRI24 atlas was used for spatial normalization, and the Julich brain atlas was used specifically for segmenting the dentate gyrus (DG).
  • Validation: While not a new benchmark dataset, the study utilized a well-characterized monocentric glioma registry with detailed neuropathological and molecular diagnostics (including 850k methylation analysis) to ensure diagnostic accuracy.

Key Insights and Ideas

  • The SVZ as a Malignancy Multiplier: The most striking insight is that the extent of SVZ infiltration, rather than just proximity, is the primary driver of these imaging changes.
  • Higher Perfusion (The "Leaky" Tumor): SVZ-infiltrating tumors showed significantly higher CBV. This suggests these tumors are more vascularized, likely driven by higher expression of growth factors like VEGF, which helps the tumor fuel its rapid expansion.
  • The Infiltration Paradox: The study found a negative correlation between SVZ infiltration and TVM (tissue volume). In plain English: as the tumor moves into the SVZ, the cell density per voxel decreases. This isn't because the tumor is "weaker," but because it is more infiltrative. Instead of growing as a dense, solid mass, the cells are spreading out and invading healthy tissue more effectively.
  • Specific to the SVZ: Interestingly, these correlations were not found for the dentate gyrus (another neurogenic zone). This suggests that the SVZ has a unique biological relationship with glioblastoma evolution that other stem cell niches might not share.

Key Results

  • Perfusion: Significant positive correlations were found across all CBV percentiles (e.g., CBV P50: $r=0.26, p=0.001$). High SVZ infiltration cases showed markedly higher blood volume than low infiltration cases.
  • Cell Density: TVM showed a significant negative correlation with SVZ infiltration (e.g., TVM P50: $r=-0.20, p=0.019$), supporting the "more infiltrative, less dense" growth hypothesis.
  • Structural Integrity: FA-FWE (fractional anisotropy) was positively associated with SVZ infiltration ($r=0.18$ to $0.24$ for higher percentiles), which may reflect the tumor's tendency to track along white matter pathways as it invades.
  • Tumor Volume: There was a significant negative correlation between tumor volume and distance to the SVZ ($r=-0.30, p=0.000107$), meaning tumors closer to the SVZ tend to be much larger at the time of diagnosis.

Limitations and Future Work

The authors acknowledge several hurdles:

  • Monocentric Design: The data comes from a single institution, which may limit how well these results generalize to other hospitals or different MRI hardware.
  • Exploratory Nature: Because this was an exploratory study, they did not perform multi-testing corrections (like Bonferroni), meaning some of the weaker correlations require further validation in larger, prospective cohorts.
  • Biological "Why": While the imaging shows that these tumors are different, the exact molecular mechanism—why SVZ cells make the tumor more vascularized—remains a subject for future biological research.

In conclusion, this research provides a "radiographic fingerprint" for the aggressive biology of SVZ-involved glioblastomas, potentially allowing clinicians to identify the most high-risk patients earlier using advanced MRI sequences.

Summary updated

Abstract

Glioblastomas with SVZ infiltration exhibit distinct imaging characteristics, including higher perfusion and lower cell density per voxel, indicating a more infiltrative growth and higher vascularization. Stem cell-like characteristics in SVZ-infiltrating cells could explain the increased infiltration and aggressive behavior. Understanding these imaging and biological correlations could enhance the understanding of glioblastoma evolution.
BibTeX
@article{griessmair2024advanced,
  author = {Michael Griessmair and Severin Schramm and Julian Ziegenfeuter and Julian Canisius and Kirsten Jung and Claire Delbridge and Friederike Schmidt‐Graf and Meike Mitsdoerffer and Claus Zimmer and Bernhard Meyer and Marie‐Christin Metz and Benedikt Wiestler},
  title = {{Advanced imaging reveals enhanced malignancy in glioblastomas involving the subventricular zone: evidence of increased infiltrative growth and perfusion}},
  journal = {Journal of Neuro-Oncology},
  year = {2024},
  doi = {10.1007/s11060-024-04849-2},
}
Figure