Abstract
Phosphorus magnetic resonance spectroscopic imaging (31P MRSI) is of particular interest for investigations of patients with brain tumors as it enables to non-invasively assess altered energy and phospholipid metabolism in vivo. However, the limited sensitivity of 31P MRSI hampers its broader application at clinical field strengths. This study aimed to identify the additional value of 31P MRSI in patients with glioma at ultra-high B0 = 7T, where the increase in signal-to-noise ratio may foster its applicability for clinical research. High-quality, 3D 31P MRSI datasets with an effective voxel size of 5.7 ml were acquired from the brains of seven patients with newly diagnosed glioma. An optimized quantification model was implemented to reliably extract an extended metabolic profile, including low-concentrated metabolites such as extracellular inorganic phosphate, nicotinamide adenine dinucleotide [NAD(H)], and uridine diphosphoglucose (UDPG), which may act as novel tumor markers; a background signal was extracted as well, which affected measures of phosphomonoesters beneficially. Application of this model to the MRSI datasets yielded high-resolution maps of 12 different 31P metabolites, showing clear metabolic differences between white matter (WM) and gray matter, and between healthy and tumor tissues. Moreover, differences between tumor compartments in patients with high-grade glioma (HGG), i.e., gadolinium contrast-enhancing/necrotic regions (C+N) and peritumoral edema, could also be suggested from these maps. In the group of patients with HGG, the most significant changes in metabolite intensities were observed in C+N compared to WM, i.e., for phosphocholine +340%, UDPG +54%, glycerophosphoethanolamine −45%, and adenosine-5′-triphosphate −29%. Furthermore, a prominent signal from mobile phospholipids appeared in C+N. In the group of patients with low-grade glioma, only the NAD(H) intensity changed significantly by −28% in the tumor compared to WM. Besides the potential of 31P MRSI at 7T to provide novel insights into the biochemistry of gliomas in vivo, the attainable spatial resolutions improve the interpretability of 31P metabolite intensities obtained from malignant tissues, particularly when only subtle differences compared to healthy tissues are expected. In conclusion, this pilot study demonstrates that 31P MRSI at 7T has potential value for the clinical research of glioma.
Highlights
Gliomas are the most common primary brain tumors in adults, with a generally poor prognosis due to persisting challenges in treatment [1]
The aim of this study is to identify the additional value of 31P magnetic resonance spectroscopic imaging (MRSI) at ultra-high fields (UHF) for application in clinical research of brain tumors in terms of (I) the detection of novel tumor markers and (II) the improved interpretability of results via increased spatial resolution
Besides the typical signals observed in brain parenchyma arising from higherconcentrated metabolites, like PCr, Adenosine-5′ Triphosphate (ATP), GPC, GPE, intracellular inorganic phosphate (Pi), and PE (Figures 1D,E), resonances from lower-concentrated metabolites were detectable, i.e., extracellular inorganic phosphate (ePi) in the frequency range between 5.0 and 5.4 ppm (Figures 1G,H), and a resonance dominated by NAD+ and NADH around δ ≈ −8.3 ppm (Figures 1J,K)
Summary
Gliomas are the most common primary brain tumors in adults, with a generally poor prognosis due to persisting challenges in treatment [1]. 31P MRS(I) has been applied to patients with glioma for some time in several studies at clinical field strengths B0 ≤ 3 T and has recently been shown to potentially predict therapy response via phospholipid metabolites [5] and the site of tumor progression via intracellular pH [6]. One of the challenges is the large effective voxel sizes applied at clinical field strength, typically on the order of about 50 ml at B0 = 3 T [8] These voxel sizes compromise the clear separation of healthy and tumor tissue spectra [5, 6, 9], even more so for the separation of metabolically different tumor compartments, which in turn complicates the interpretation of 31P MRSI results obtained in patients. The increase in signal-to-noise ratio (SNR) at ultra-high fields (UHF) B0 ≥ 7T may foster the clinical applicability of 31P MRSI by allowing the spatial resolution to be increased within
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
