Vascular habitat analysis using dynamic susceptibility contrast-enhanced perfusion MRI for differentiating glioblastoma from solitary brain metastases.
- 1Department of Radiology, Daping Hospital, Army Medical University, Chongqing, China.
- 2Chongqing Clinical Research Center of Imaging and Nuclear Medicine, Chongqing, China.
- 3Department of Pathology, Daping Hospital, Army Medical University, Chongqing, China.
Abstract
Objective
To investigate tumor vascular habitat parameters based on dynamic susceptibility contrast-enhanced perfusion-weighted imaging (DSC-PWI) for differentiating glioblastoma (GBM) from solitary brain metastasis (SBM).Methods
In this single-center retrospective study, imaging data from 141 patients with pathologically or clinically confirmed malignant brain tumors (74 GBM, 67 SBM) were analyzed. The hemodynamic tissue signature (HTS) method was used to segment the lesions and construct four tumor vascular habitats: high angiogenic tumor (HAT), low angiogenic tumor (LAT), potentially infiltrated peripheral edema (IPE), and vasogenic peripheral edema (VPE). Pearson correlation analysis was used to quantify the correlations among habitat hemodynamic parameters. Habitat characteristic parameters were calculated, and ROC were plotted to evaluate the diagnostic performance. A nomogram model was constructed using logistic regression.Results
GBM exhibited significantly higher median relative cerebral blood volume (rCBVmedian) and median relative cerebral blood flow (rCBFmedian) than SBM across all four habitat regions (all p < 0.001). In terms of volumetric parameters, the ratio of IPE volume to peritumoral volume (V (IPE)/V (P)) was significantly higher in the GBM group (p < 0.001). Among individual habitat parameters, LAT rCBFmedian showed the best diagnostic performance with an AUC of 0.786 (95% CI: 0.709-0.850). In the GBM group, the perfusion parameters of the IPE and VPE were significantly positively correlated with those of the HAT, whereas no such correlation was found in the SBM group. A nomogram incorporating lesion location, enhancement pattern, and LAT-rCBFmedian yielded an AUC of 0.857 (95% CI: 0.79-0.92) for distinguishing GBM from SBM, with good calibration.Conclusion
Tumor vascular habitat parameters based on DSC-PWI demonstrate significant differences between newly diagnosed GBM and SBM, and the nomogram shows certain value in differentiating between the two tumor types.
Related
- Quantitative Pharmacokinetic Analysis of Dynamic Contrast-Enhanced Perfusion in Multiparametric Prostate MRI: An Exploratory Intra-Patient Study Differentiating Malignant Lesions from Normal Tissue
- Qualitative Dynamic Contrast-Enhanced MRI Time–Intensity Curve Analysis for Differentiating Non-Progressive Treatment-Related Changes from Tumor Progression After Stereotactic Radiotherapy for Brain Metastases
- Efficacy of arterial spin labeling magnetic resonance perfusion in differentiating recurrent or residual high-grade glioma from post-treatment changes: A retrospective hospital-based study.
- Three-dimensional multimodal fractal dynamics integrating spatial metabolism, cytoarchitecture, and perfusion for preoperative prediction of microvascular invasion in colorectal liver metastases
- Explainable DSC Perfusion Radiomics for Post-Treatment Glioblastoma Assessment Using a Soft-Voting Machine-Learning Ensemble
BibTeX
@article{li2026vascular,
title = {{Vascular habitat analysis using dynamic susceptibility contrast-enhanced perfusion MRI for differentiating glioblastoma from solitary brain metastases.}},
author = {Xuan Li and Hanwei Wang and Linlan Zeng and Mimi Zhao and Ling Feng and Huan Xie and Qisheng Ran and Qiang Ma and Shunan Wang},
journal = {Frontiers in oncology},
year = {2026},
month = oct,
doi = {10.3389/fonc.2026.1904498},
url = {https://europepmc.org/article/MED/42908618},
note = {CC BY},
}
RIS
TY - JOUR
TI - Vascular habitat analysis using dynamic susceptibility contrast-enhanced perfusion MRI for differentiating glioblastoma from solitary brain metastases.
AU - Xuan Li
AU - Hanwei Wang
AU - Linlan Zeng
AU - Mimi Zhao
AU - Ling Feng
AU - Huan Xie
AU - Qisheng Ran
AU - Qiang Ma
AU - Shunan Wang
PY - 2026
DA - 2026/10/07
JO - Frontiers in oncology
DO - 10.3389/fonc.2026.1904498
UR - https://europepmc.org/article/MED/42908618
AB - Objective To investigate tumor vascular habitat parameters based on dynamic susceptibility contrast-enhanced perfusion-weighted imaging (DSC-PWI) for differentiating glioblastoma (GBM) from solitary brain metastasis (SBM).Methods In this single-center retrospective study, imaging data from 141 patients with pathologically or clinically confirmed malignant brain tumors (74 GBM, 67 SBM) were analyzed. The hemodynamic tissue signature (HTS) method was used to segment the lesions and construct four tumor vascular habitats: high angiogenic tumor (HAT), low angiogenic tumor (LAT), potentially infiltrated peripheral edema (IPE), and vasogenic peripheral edema (VPE). Pearson correlation analysis was used to quantify the correlations among habitat hemodynamic parameters. Habitat characteristic parameters were calculated, and ROC were plotted to evaluate the diagnostic performance. A nomogram model was constructed using logistic regression.Results GBM exhibited significantly higher median relative cerebral blood volume (rCBVmedian) and median relative cerebral blood flow (rCBFmedian) than SBM across all four habitat regions (all p < 0.001). In terms of volumetric parameters, the ratio of IPE volume to peritumoral volume (V (IPE)/V (P)) was significantly higher in the GBM group (p < 0.001). Among individual habitat parameters, LAT rCBFmedian showed the best diagnostic performance with an AUC of 0.786 (95% CI: 0.709-0.850). In the GBM group, the perfusion parameters of the IPE and VPE were significantly positively correlated with those of the HAT, whereas no such correlation was found in the SBM group. A nomogram incorporating lesion location, enhancement pattern, and LAT-rCBFmedian yielded an AUC of 0.857 (95% CI: 0.79-0.92) for distinguishing GBM from SBM, with good calibration.Conclusion Tumor vascular habitat parameters based on DSC-PWI demonstrate significant differences between newly diagnosed GBM and SBM, and the nomogram shows certain value in differentiating between the two tumor types.
SN - 2234-943X
ER -