Glioblastoma (GBM) effects on quantitative MRI of contralateral normal appearing white matter

Hatef Mehrabian; Wilfred W Lam; Sten Myrehaug; Arjun Sahgal; Greg J Stanisz

doi:10.1007/s11060-018-2846-0

Glioblastoma (GBM) effects on quantitative MRI of contralateral normal appearing white matter

J Neurooncol. 2018 Aug;139(1):97-106. doi: 10.1007/s11060-018-2846-0. Epub 2018 Mar 28.

Authors

Hatef Mehrabian^{1

2}, Wilfred W Lam³, Sten Myrehaug^{4

5}, Arjun Sahgal^{6

4

5}, Greg J Stanisz^{3

6

7}

Affiliations

¹ Medical Biophysics, University of Toronto, Toronto, ON, Canada. [email protected].
² Radiology and Biomedical Imaging, University of California, San Francisco (UCSF), 1700 - 4th St., Suite BH 201, San Francisco, CA, 94158, USA. [email protected].
³ Medical Biophysics, University of Toronto, Toronto, ON, Canada.
⁴ Radiation Oncology, Sunnybrook Health Sciences Centre, Toronto, ON, Canada.
⁵ Department of Radiation Oncology, University of Toronto, Toronto, ON, Canada.
⁶ Physical Sciences, Sunnybrook Research Institute, Toronto, ON, Canada.
⁷ Department of Neurosurgery and Pediatric Neurosurgery, Medical University, Lublin, Poland.

PMID: 29594656
DOI: 10.1007/s11060-018-2846-0

Abstract

Purpose: The objective was to investigate (with quantitative MRI) whether the normal appearing white matter (NAWM) of glioblastoma (GBM) patients on the contralateral side (cNAWM) was different from NAWM of healthy controls.

Methods: Thirteen patients with newly diagnosed GBM and nine healthy age-matched controls were MRI-scanned with quantitative magnetization transfer (qMT), chemical exchange saturation transfer (CEST), and transverse relaxation time (T₂)-mapping. MRI scans were performed after surgery and before chemo-radiation treatment. Comprehensive qMT, CEST, T₂ data were acquired. A two-pool MT model was fit to qMT data in transient state, to calculate MT model parameters [Formula: see text]. CEST signal was isolated by removing the contributions from the MT and direct water saturation, and CEST signal was calculated for Amide (CEST_Amide), Amine (CEST_Amine) and nuclear overhauser effect, NOE (CEST_NOE).

Results: There was no difference between GBM patients and normal controls in the qMT properties of the macromolecular pool [Formula: see text]. However, their free water pool spectrum was different (1/R_aT_2a_,_patient = 28.1 ± 3.9, 1/R_aT_2a_,_control = 25.0 ± 1.1, p = 0.03). This difference could be attributed to the difference in their T₂ time ([Formula: see text] = 83 ± 4, [Formula: see text] = 88 ± 1, p = 0.004). CEST signals were statistically significantly different with the CEST_Amide having the largest difference between the two cohorts (CEST_{Amide,patient} = 2.8 ± 0.4, CEST_{Amide,control} = 3.4 ± 0.5, p = 0.009).

Conclusions: CEST in cNAWM of GBM patients was lower than healthy controls which could be caused by modified brain metabolism due to tumor cell infiltration. There was no difference in MT properties of the patients and controls, however, the differences in free water pool properties were mainly due to reduced T₂ in cNAWM of the patients (resulting from structural changes and increased cellularity).

Keywords: Chemical exchange saturation transfer (CEST); Glioblastoma (GBM); Normal appearing white matter (NAWM); Quantitative magnetization transfer (qMT).

MeSH terms

Brain / diagnostic imaging*
Brain Neoplasms / diagnostic imaging*
Brain Neoplasms / therapy
Female
Functional Laterality
Glioblastoma / diagnostic imaging*
Glioblastoma / therapy
Humans
Image Processing, Computer-Assisted
Magnetic Resonance Imaging*
Male
Middle Aged
White Matter / diagnostic imaging*

Abstract

MeSH terms

Grants and funding