Model selection for DCE-T1 studies in glioblastoma

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Glioblastoma is an aggressive brain tumor.
• Accurate characterization of tumor vascularity is crucial for treatment planning.

Purpose of the Study:

• To evaluate the utility of dynamic contrast-enhanced T1-weighted MRI with gadopentetate dimeglumine (Gd-DTPA) in characterizing glioblastoma.
• To assess the effectiveness of nested pharmacokinetic models in quantifying tumor vascular parameters.

Main Methods:

• Dynamic contrast-enhanced T1-weighted MRI was performed on 10 glioblastoma patients.
• Nested models (1, 2, and 3) estimating plasma volume (v(p)), K(trans), and k(ep) were applied.
• Model 1 represented non-leaky tissue, while models 2 and 3 indicated increasing contrast leakage.

Main Results:

• Model 3 predominated in aggressive tumor regions, while Model 1 was found in normal brain tissue.
• Tumor v(p) was approximately four times that of white matter.
• K(trans) values showed a tenfold difference between model 2 and model 3 regions.
• The mean interstitial space in model 3 regions was 5.5% with excellent curve fits (R(2) = 0.99).

Conclusions:

• The three-parameter standard model adequately fits dynamic contrast-enhanced T1 data in glioblastoma.
• These models can differentiate between aggressive tumor areas and normal brain tissue based on contrast leakage patterns.
• The findings support the use of these pharmacokinetic models for glioblastoma assessment.