Quantitative and Multi-Parametric 3D Magnetic Resonance Imaging of Primary Brain Tumors

In traditional IGRT the radiation treatment is planned and simulated on computers using x-ray computed tomography (CT) images alone or a combination of CT and MRI. The CT mostly provides information about attenuation of radiation beams needed for the dose simulations because most tumours are more readily identified and contoured with MRI. However, fusion of CT and MR images is prone to error and is a time-consuming process that cannot be automated reliably. Previous research (Stanescu et al.) has shown that the attenuation information can be obtained from MRI which, unlike CT, does not use ionizing radiation to create images. Eliminating the CT scan is therefore possible and beneficial to both the patient, who avoids an additional dose of diagnostic x rays, and to the health care system which saves resources that can be used elsewhere.

Our new MRI acquisition and processing techniques (performed at 3 tesla i.e. 3T) enable:

1. the extraction of 4 different quantitative parameters (hence "multi-parametric MRI", quantitative MRI or relaxometry) that are normally not accessible in traditional MRI, and
2. the automatic classification of tissues (e.g. bone, air, adipose, soft tissue, etc.) which is needed for dose computation in IGRT planning

In the first part of the study (dosimetry) these MRI methods will be used to generate a pseudo-CT to replace the traditional CT data. The hypothesis is that dosimetry can be accurately calculated for primary brain cancer patients using the pseudo-CT, thus allowing radiation treatment planning using MRI only. The IGRT treatments planned using qMRI will be compared to those planned conventionally.

The second part of the study (treatment monitoring) aims to evaluate the ability of qMRI techniques to provide clinical information such as distinguishing between progression and pseudo-progression, assessing treatment effectiveness or prognosis. The hypothesis is that qMRI can provide increased sensitivity to biological changes in tumors associated with disease progression over conventional (T1- or T2-weighted) MRI. The reasoning is that by providing quantitative, rather than weighted, images, direct numerical comparisons can be made between images acquired at different time points or at different centres. With traditional MRI, only limited, qualitative comparisons of tumor morphology or relative intensity within the same image can be made. Therefore, 3 or more follow-up MRI scans will also be acquired 3, 6 and 12 months after treatment and at recurrence, transformation, or pseudoprogression to monitor the effectiveness of the treatment.