UCSF

Diffusion weighted Magnetic Resonance Imaging has been shown to be very useful for the clinical evaluation of patients with stroke. Although the acquisition of such data is a common feature on most MR scanners, the implication of the parameters that it uses are often not appreciated by the oncology community, especially for the application to patients with gliomas. The potential for providing quantitative information that describes the biological properties of tumor and surrounding brain parenchyma has therefore not been fully explored. The goal of this thesis was to characterize the variations in diffusion imaging parameters and determine how they may be used for the management of patients with brain tumors throughout the treatment of their disease. This thesis utilized MR spectroscopic imaging (MRSI) techniques to provide adjunct metabolic information.

The relationship between the apparent diffusion coefficient (ADC) from DTI and levels of choline containing compounds from MRSI, both of which have been suggested to correlate to cell density, was assessed by grade and subtype within the anatomically heterogeneous tumor regions. There was evidence of a relationship between ADC and choline in some regions of grade IV gliomas but within no regions of grade II gliomas. This suggested independent information to be acquired from both DTI and MRSI. A method to assess deviation from mono-exponential decay in a clinically relevant scan time was developed as another possible method of assessing abnormality in grade IV glioma.

Previously observed variations in grade II glioma subtypes was further assessed. A difference in ADC was evident and based on these findings, a method of visualizing this variation was developed using RGB color maps. This technique was applied on a new cohort of grade II gliomas and histopathology was assessed by image-guided biopsies to examine the histopathological variations.

Diffusion parameters during treatment were evaluated for early non-invasive biomarkers. The ADC changes from mid- to post-treatment suggest such a possible early non-invasive biomarker. The results of this dissertation suggest that diffusion parameters play an important role in assessing gliomas. These are very important steps towards increasing the utilization of imaging in the management of patients with brain tumors.