Advanced MR Imaging Protocol for Glioblastoma

Introduction:
An MRI protocol has been developed and evaluated on a 3T Siemens scanner in the University of Michigan and led by Dr. Yue Cao. This advanced MRI protocol is intended for identifying the most aggressive components of glioblastoma (GBM) for radiation boosting and for reliable therapy response assessment at multi-institutes.

Background:
It is well known that post-Gd T1 weighted and FLAIR images underestimate and overestimate the tumor volume of GBM, respectively, and cannot assess therapy response reliably. In the last decade, physiological and metabolic imaging biomarkers (including both MRI and PET) have been developed for therapy assessment, and to a lesser extent for definition of treatment target of GBM.(1-16) Considering the wide availability, we focus on MRI techniques. Several MRI techniques have shown the predictive values for OS and PSF, including cerebral blood volume (CBV) (6-10), choline to N-acetylaspartate ratio (Cho/NAA) from 1H-MR spectroscopy imaging (1H-MRSI) (11-13), and functional diffusion map (fDM) derived from two time points of conventional diffusion-weighted (DW) imaging (with b-value < 1000 s/mm2) (4, 5). Recently, we developed an imaging technique for detection of hypercellular components of GBM using high b-value DW imaging by suppressing MR signals from edema and normal tissue.(3) We found that 1) the hypercellular volume (HCV) was a negative predictor for PSF, 2) the non-enhanced HCV could be treated inadequately due to poorer detection by conventional MRI, and 3) the mis-treated HCV resulted in rapid progression. These findings suggest that the HCV is an aggressive component of GBM, and could be targeted by intensified radiation doses or surgery to improve outcome. Also, this technique is easily deployed in a large scale clinical trial. Furthermore, biological heterogeneity of GBM indicates single imaging modality may not be sufficient to detect different image-phonotypes and assess heterogeneity response in GBM. Based upon these evidences, we develop the following MRI protocol for GBM.

MRI Protocol series:

1. Localizer
2. 3D T1-weighted images pre-contrast
3. 2D FLAIR images
4. 2D multiple b-value diffusion weighted images
5. T1/B1 mapping
6. Dryrun for DCE
7. DCE series with contrast
8. 3D T1-weighted images post-contrast
9. DTI series

Details of the scan protocols are given in the attached documents.