Functional and Anatomic Imaging of Tumor Vasculature: High-Resolution MR Spectroscopic Imaging Combined with a Superparamagnetic Contrast Agent

Abstract

The advantages of incorporating spectroscopic information into magnetic resonance imaging (MRI) data sets were demonstrated by Dixon (1) and later by others (2,3). This early work focused on imaging water and fat signals with low spectral resolution combined with high spatial resolution. More recently, improvements in MR instrumentation have encouraged detection of water and fat with both high spatial and spectral resolution (HiSS) MR to improve anatomic detail and contrast (4–8) in tissues with broad and complex water proton resonances. This is particularly important in tumors but may also be applicable to imaging regions in the brain that have been damaged by stroke and other conditions associated with hemorrhage and necrosis. HiSS can improve imaging of the effects of endogenous and injected contrast agents. For example, in this laboratory, HiSS has been used to image changes in deoxyhemoglobin levels in tumors caused by tumor-oxygenating agents and hemodynamic challenges (6,7,9). HiSS is also useful for imaging effects of synthetic contrast agents. The data presented here illustrate the use of HiSS to detect superparamagnetic contrast agents. Superparamagnetic MR contrast agents provide a unique opportunity to detect tumors, define tumor anatomy, delineate tumor boundaries, and evaluate vascular structure and function of tumors. The complexity of the effects of these agents provides new mechanisms for contrast in MR images but also poses a challenge for MRI. Iron oxide particles with small diameters have very large Rl and R2* (longitudinal and transverse relaxivities), and both effects may be spectrally inhomogeneous (ie, different components of the water resonance in each voxel are affected differently). The results presented here suggest that to measure these effects independently, accurately, and with a high signal-to-noise ratio, it is advantageous to acquire spectroscopic images with high spectral and spatial resolution.