Magnetic Resonance Spectroscopy of Cystathionine and 2‐Hydroxyglutarate in Brain Tumors

TL;DR

This study uses magnetic resonance spectroscopy to show that cystathionine levels are higher in certain brain tumors with specific genetic changes.

Contribution

The study introduces a new PRESS sequence to improve detection of cystathionine in brain tumors and confirms its potential as a biomarker.

Findings

Cystathionine levels were significantly higher in IDH-mutant 1p/19q-codeleted gliomas compared to non-codeleted ones.

2HG levels were significantly higher in IDH-mutant gliomas than in IDH-wildtype gliomas.

Cystathionine measures showed high sensitivity and specificity for predicting 1p/19q codeletion status.

Abstract

Codeletion of chromosome 1p and 19q arms in a subset of gliomas has been shown to associate with the accumulation of cystathionine in the tumor. Here, we report the analyses of cystathionine and 2‐hydroxyglutarate (2HG) in 38 biopsy‐proven glioma patients, as measured with TE 97‐ms point‐resolved spectroscopy (PRESS) at 3 T. Following the confirmation of in‐house calculated cystathionine basis signals in a phantom solution, LCModel spectral fitting was performed to decompose metabolite signals, and the millimolar concentrations of metabolites were estimated with reference to water. Cystathionine was estimated to be significantly higher in IDH mutated 1p/19q codeleted gliomas than in noncodeleted gliomas (1.4 ± 1.1 vs. 0.3 ± 0.4 mM; n = 15 and 14; p = 0.002). 2HG estimation was significantly higher in IDH‐mutant gliomas compared to IDH‐wildtype gliomas (3.7 ± 3.3 vs. 0.1 ± 0.1 mM; n = 29 and 9; p = 0.002). Using cutoff values from receiver operating characteristic curve analysis, the sensitivity and specificity of the cystathionine measures with respect to 1p/19q status were estimated to be 0.8 and 0.79 (cutoff at 0.6 mM), and those of 2HG measures with respect to IDH status were both unity (cutoff at 0.5 mM). A TE 113‐ms PRESS sequence was designed to improve detection of the cystathionine C4‐proton resonance at 2.72 ppm. Data from two glioma patients showed that cystathionine can be detected with minimal interference from the overlapping aspartate signal. 2D chemical‐shift imaging of cystathionine using the TE 113‐ms PRESS is demonstrated. Our MRS data confirm elevation of cystathionine in 1p/19q codeleted gliomas, as demonstrated in prior studies, and suggest that relatively high estimates of cystathionine may provide a biomarker of 1p/19q codeleted gliomas.

Cystathionine and 2‐hydroxyglutarate were evaluated in 38 glioma patients with TE 97ms PRESS at 3T. Cystathionine level was significantly higher in IDH‐mutant 1p/19q‐codeleted gliomas than in IDH‐mutant non‐codeleted gliomas (p = 0.002). The sensitivity and specificity of the cystathionine measures with respect to 1p/19q status were 0.8 and 0.79, while those of 2‐hydroxyglutarate measures with respect to IDH status were both unity. Our data suggested that relatively high estimates of cystathionine may be predictive of 1p/19q codeleted gliomas.

Figures