Characterization of Serial Magnetic Resonance Spectroscopy Imaging in Patients With Malignant Glioma Undergoing Radiotherapy

Overview

Malignant glioma is the most common primary brain tumor in adults. Despite aggressive therapy, less than 40% of these patients are expected to live beyond 5 years. The radiologic imaging of these tumors relies on computed tomography (CT) and magnetic resonance imaging (MRI) - these studies provide good anatomical information about the size and location of the tumor, but are unable to evaluate whether the tumor is still viable or contains metabolic activity, after surgery and, in particular, radiotherapy (RT). This complicates accurate understanding of the status of the tumor during a patient's follow-up. This study proposes to add magnetic resonance spectroscopy, a non-invasive imaging method which can monitor metabolic changes in the tumor, to regular imaging. Understanding the changes that occur in a tumor over the course of radiotherapy could help predict how well a treatment might work, and could also be useful in distinguishing a return of the tumor in an area of radiation damage before it would be obvious on regular imaging.

Background: Malignant glioma is the most common primary brain tumor in adults. Despite aggressive therapy, less than 40% of these patients are expected to live beyond 5 years. External beam radiotherapy following maximal surgical resection is the mainstay of treatment for this group of patients. Recent intensification of local therapy with focused radiotherapy planning has resulted in successful escalation of dose. Further improvement in the therapeutic index of therapy is desirable. Radiologic characterization of glial tumors relies predominantly on CT and MRI images; these studies provide good anatomic information regarding the size and location of the tumor, as well as surrounding structures, but are unable to evaluate viability or proliferative activity of tumors. Thus, the enhancing lesion on CT or MRI may not correspond precisely to areas of viable tumor, especially after surgery and radiotherapy. Also, because contrast enhanced MRI relies on regions of blood brain barrier (BBB) breakdown, it is not tumor specific, thus non-neoplastic processes may lead to findings similar to disease progression. This phenomenon can make conventional radiologic follow-up difficult in patients who have received radiotherapy, as such imaging techniques are limited in their ability to discern radiotherapy effect and necrosis from recurrence and progression. Finally, because they do not discriminate viable tumor, CT and MRI are of limited usefulness in assessing response to therapy, and are unable to effectively predict outcome. Magnetic Resonance Spectroscopy (MRS) is a relatively new technology which may be able to address these issues. The objective of the current study is to investigate the changes that occur in tumor related magnetic resonance (MR) spectra over the course of radical radiotherapy for malignant glioma. The primary endpoint for the study will be to identify characteristic evolving metabolite patterns on MRSI, before, during, and after radiotherapy that correlate with overall survival, and progression-free survival in high grade gliomas. Secondary endpoints will involve correlation of MRSI metabolite patterns with time to progression and Karnofsky performance status. Eligibility Criteria: * Patients must be older than 18 years of age. * Patients must have histologically proven malignant glioma of the brain. * Patients must have bi-directionally measurable enhancing residual disease by T1 weighted image. * Patients must be willing to undergo high dose radiotherapy to the brain for the treatment of their glial tumor. * Patients must be willing and able to comply with all study requirements. * The patient or legally authorized representative must fully understand all elements of informed consent, and sign the consent document. Ineligibility Criteria: Ineligibility criteria include: * History of previous RT to the head and neck region * History of lupus, scleroderma or RT hypersensitivity * Co-existing medical condition precluding radiotherapy * Psychiatric conditions precluding informed consent. * Medical or psychiatric conditions precluding MR studies (eg. pacemaker, aneurysm clips, neuro stimulator, cochlear implant, severe claustrophobia/anxiety) Patients will be approached for study participation at the time of their initial radiation oncology consultation in the outpatient department of the Cross Cancer Institute (CCI). Patients who wish to participate, and satisfy the eligibility and exclusion criteria, will be required to review and sign the consent form at that time. Patients will then undergo regular staging investigations, construction of an immobilizing shell, treatment planning MRI, and CT simulation. These studies are typically completed 2 weeks after the initial consult. Radiotherapy will commence approximately 3-4 weeks after the initial consult. At week 0 of RT, prior to beginning therapy, the patient will undergo the baseline MRS. The mid-RT MRS study will be performed during week 4 of RT. The post therapy scan will take place 2 months post-therapy. From then on, patients will be seen in clinic every 2-4 months for follow-up, and will undergo MRI and MRS scans with each visit for 1 year. Data Collection and Statistical Analysis: The height of each MRS metabolite peak will be measured from voxels within the enhancing MRI lesion and from voxels in normal brain for each patient. Relative metabolite values (normalized to the value in normal brain) will then be generated, as well as relative metabolite ratios (eg. relative choline/relative NAA) for each time point (week 0, week 4, and post-RT, at follow-up). For each patient, the relative metabolite levels (and ratios) will be plotted over time. Patients will then be partitioned into groups, based on similar evolving MRS pattern. For each of the groups, curves of survival and disease free progression will be generated by the Kaplan-Meier method. The curves will be analyzed for statistical significance by the log-rank method. The investigators plan to accrue 30 patients for the present study. They are confident 30 patients will be sufficient to generate statistically significant results. In a study of the effects of brain tumor radiotherapy on normal brain as imaged by NMR spectroscopy, Urtasun et al were able to find statistically significant metabolite changes on proton MRS images with only 10 patients. In addition, the data to be utilized in the retrospective aspect of the study contains information on approximately 30 patients and their MRS scans. The trends found in this data will be used to guide data analysis for the prospective study. Finally, given the relative distribution and frequency of histologies seen in the new patient CNS clinic at the CCI, the investigators feel the target accrual of 30 patients is feasible within the time restraints of the project.