Single-scan, multi-contrast imaging of glucose metabolism in patients with glioblastoma

Summary - Noninvasive imaging of disease progression is a critical component in the management and treatment plan of brain tumors. To evaluate tumor growth and treatment response anatomical magnetic resonance imaging (MRI) scans are acquired on regular intervals to detect changes in lesion size. However, the need to acquire multiple MRI scans over several months to be certain about tumor growth can correspond to a waste of time, particularly for patients with glioblastoma (GBM), the most aggressive and most common primary brain tumor, who face an overall survival of less than 2 years. An imaging method that can detect active tumor much faster would be a significant improvement for patients and help clinicians make treatment decisions. In tumors outside of the brain metabolic imaging with 18F-fluoro-deoxy-glucose positron emission tomography (18FDG-PET) can quickly characterize disease progression by detecting increased glucose uptake. However, 18FDG-PET is rarely used for brain tumor imaging because healthy brain also takes up high amounts of glucose which can negate image contrast and make it problematic to distinguish tumor from normal brain. As an alternative we propose to extend our work on the MR-based deuterium metabolic imaging (DMI) combined with simple, oral intake of non-radioactive deuterated glucose, to characterize brain tumors based on the altered metabolic fate of glucose. While both healthy brain and tumors can have high glucose uptake, aggressive tumors differ by metabolically converting a high fraction of glucose to lactate, a phenomenon known as the Warburg effect. DMI has an advantage over 18FDG-PET because it can directly detect various metabolites, including lactate, and generate multiple types of metabolism-based image contrast from a single scan. Preliminary data show high image contrast based on glucose metabolism in GBM at diagnosis and at recurrence after treatment, detected with DMI. Furthermore, DMI is highly robust, safe, and has already been implemented on 3T and 7T clinical MRI scanners. Concerns about long additional scan time have been mitigated by acquiring DMI and MRI simultaneously. Because of these features and the ease of use of the method, glucose DMI can become the robust metabolic imaging method for brain tumors that so far has been missing. The primary significance of this proposal is fulfilling the need for key studies required before DMI can be translated from a promising research-only method to where it can be used in clinical imaging trials to fully establish its diagnostic and prognostic utility. In Aim 1 we will validate the specificity of DMI for detecting active tumor by comparison with histopathology of excised tumor tissue from patients with GBM. In Aim 2 we will perform test-retest studies in patients with GBM to establish the repeatability of glucose DMI. Aim 3 is focused on the use of DMI to detect true disease progression in a clinical setting. These aims will potentially have a high impact on clinical research and could shift disease management of GBM by providing a robust and specific method for imaging tumor progression that can be integrated -without time penalty- in standard MRI protocols.