Accuracy of 1H magnetic resonance spectroscopy for quantification of 2-hydroxyglutarate using linear combination and J-difference editing at 9.4 T

• Verfasst von: Neuberger, Ulf [VerfasserIn]
• Verfasst von: Vollmuth, Philipp [VerfasserIn]
• Verfasst von: Fischer, Manuel [VerfasserIn]
• Verfasst von: Bendszus, Martin [VerfasserIn]
• Verfasst von: Heiland, Sabine [VerfasserIn]
• Titel: Accuracy of 1H magnetic resonance spectroscopy for quantification of 2-hydroxyglutarate using linear combination and J-difference editing at 9.4 T
• Paralleltitel: Genauigkeit der 1H-Magnetresonanzspektroskopie zur Quantifizierung von 2-Hydroxyglutarat durch MRS mit kurzer Echozeit und J-Differenz-Editierung bei 9.4 T
• Verf.angabe: Ulf Neuberger, Philipp Kickingereder, Xavier Helluy, Manuel Fischer, Martin Bendszus, Sabine Heiland
• E-Jahr: 2017
• Jahr: 12 May 2017
• Umfang: 10 S.
• Fussnoten: Mit einer Zusammenfassung in englischer und deutscher Sprache ; Gesehen am 24.07.2018
• Weitere Titel: Übersetzung des Haupttitels: Genauigkeit der 1H-Magnetresonanzspektroskopie zur Quantifizierung von 2-Hydroxyglutarat durch MRS mit kurzer Echozeit und J-Differenz-Editierung bei 9.4 T
• Titel Quelle: Enthalten in: Zeitschrift für medizinische Physik
• Ort Quelle: Amsterdam [u.a.] : Elsevier, 1990
• Jahr Quelle: 2017
• Band/Heft Quelle: 27(2017), 4, Seite 300-309
• ISSN Quelle: 1876-4436
• DOI: doi:10.1016/j.zemedi.2017.04.002
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: 2-Hydroxyglutarat (2HG)
• Sach-SW: 2-Hydroxyglutarate (2HG)
• Sach-SW: Glioma
• Sach-SW: Gliome
• Sach-SW: J-difference editing
• Sach-SW: J-Differenz-Editierung
• Sach-SW: LCModel
• Sach-SW: Magnetic resonance spectroscopy (MRS)
• Sach-SW: Magnetresonanzspektroskopie (MRS)
• K10plus-PPN: 1577903498

Abstract

Non-invasive detection of 2-hydroxyglutarate (2HG) by magnetic resonance spectroscopy is attractive since it is related to tumor metabolism. Here, we compare the detection accuracy of 2HG in a controlled phantom setting via widely used localized spectroscopy sequences quantified by linear combination of metabolite signals vs. a more complex approach applying a J-difference editing technique at 9.4T. Different phantoms, comprised out of a concentration series of 2HG and overlapping brain metabolites, were measured with an optimized point-resolved-spectroscopy sequence (PRESS) and an in-house developed J-difference editing sequence. The acquired spectra were post-processed with LCModel and a simulated metabolite set (PRESS) or with a quantification formula for J-difference editing. Linear regression analysis demonstrated a high correlation of real 2HG values with those measured with the PRESS method (adjusted R-squared: 0.700, p<0.001) as well as with those measured with the J-difference editing method (adjusted R-squared: 0.908, p<0.001). The regression model with the J-difference editing method however had a significantly higher explanatory value over the regression model with the PRESS method (p<0.0001). Moreover, with J-difference editing 2HG was discernible down to 1mM, whereas with the PRESS method 2HG values were not discernable below 2mM and with higher systematic errors, particularly in phantoms with high concentrations of N-acetyl-asparate (NAA) and glutamate (Glu). In summary, quantification of 2HG with linear combination of metabolite signals shows high systematic errors particularly at low 2HG concentration and high concentration of confounding metabolites such as NAA and Glu. In contrast, J-difference editing offers a more accurate quantification even at low 2HG concentrations, which outweighs the downsides of longer measurement time and more complex postprocessing.