Development and Validation of Advanced MRI Methods for Clinical Applications

The objectives of this study are to (1) Develop and optimize acquisition methods for MRI-based biomarkers that are indicative of brain tumor pathophysiology. These methods include, but are not limited to, DWI, CEST, and Dynamic Susceptibility Contrast (DSC) and Dynamic Contrast Enhanced (DCE) MRI; and (2) Validate DSC-MRI accuracy by comparison to an intravascular reference standard. Developmental studies: During the development of new imaging-based biomarkers, it is critical to optimize acquisition parameters, systematically characterize performance and contrast in the pathology of interest, validate with histopathology and establish test-retest repeatability. Thus, the first goal of this study is to develop advanced Diffusion Weighted Imaging (DWI), CEST and DSC/DCE-MRI methods for application to brain tumors. Objective 1 is a single-center study of up to 60 subjects being done to explore and optimize imaging signatures indicative of altered functional tumor states. These advanced imaging methods will allow the investigators to probe neuropathological tumor correlates, including cellular characteristics, molecular and metabolic changes, and vascular characteristics. Compared to existing conventional anatomic imaging, the researchers hypothesize that these experimental methods will be better able to characterize brain tumors and will have the potential to serve as new biomarkers of diagnostic and therapeutic importance. The researchers approach brings together biophysical and physiological information obtained from MRI and correlates this with clinical diagnoses and outcomes. Validation of DSC-MRI: Despite DSC-MRI's potential impact on clinical care, its broad scale integration has been slow, in large part from a lack of consensus about methodology and how to prevent potential CBV inaccuracies. Although DSC-MRI relies on the assumption that gadolinium-based contrast agents remain within the vascular lumen, this condition is often violated in vivo. If not corrected for, contrast agent leakage effects lead to CBV inaccuracy, misdiagnosis, and potentially mistreatment. While there exist numerous leakage correction strategies that have been shown to clearly improve DSC-MRI's clinical utility (e.g., predicting therapeutic response), a key limitation has prevented the standardization and wide-spread adoption of DSC-MRI methodology: To date, no study has validated the accuracy of leakage corrected CBV measures in patients. So while leakage corrected CBV values may be used, for example, to differentiate tumor recurrence from post-treatment effects, it is unknown whether this clinical benefit is a consequence of the complex combination of pulse sequence parameters, kinetics, dosing scheme, relaxivities, and leakage correction strategy or if the computed CBV actually reflects the underlying vascular density. This distinction is critical because it has implications for DSC-MRI standardization, establishing CBV thresholds for clinic use, multi-site comparisons and clinical trials. The investigators believe this limitation represents the most critical and clinically relevant challenge in the field of brain tumor DSC-MRI that urgently needs to be addressed. Objective 2 is a single-center study of up to 160 subjects being done to validate the accuracy of DSC-MRI measures of CBV. To validate the DSC-MRI measures derived from small molecular weight gadolinium-based (Gd) contrast agents, the investigators will compare rCBV maps to those derived from the intravascular contrast agent, Ferumoxytol. Ferumoxytol has been evaluated in humans as a potential DSC-MRI contrast agent but not for the purposes of validating leakage correction techniques. Since ferumoxytol-based DSC-MRI is not influenced by leakage effects, it enables the assessment of the most reliable perfusion metrics that can be expected from DSC-MRI and, as such, is the most reliable reference standard by which to evaluate CBV accuracy. Note that while ferumoxytol is undergoing clinical trials as a potential DSC-MRI contrast agent, it is unlikely to replace Gd-based contrast agents because it cannot provide the signal enhancement expected on conventional post-contrast T1-weighted images and is therefore unsuitable for use with standard response criteria.