Abstract
Elevated levels of choline are generally emphasized as marker of increased cellularity and cell membrane turnover in gliomas. In this study, we investigated the incidence rate of lack of choline/creatine and choline/water elevation in a population of grade I–III gliomas. A cohort of 41 patients with histopathologically confirmed gliomas underwent multi-voxel proton magnetic resonance spectroscopy on a 3 T magnetic resonance system prior to treatment. Peak areas for choline and myoinositol were measured from all voxels that exhibited hyperintensity on fluid-attenuated inversion recovery images and were normalized to creatine and unsuppressed water from each voxel. The average metabolite/creatine and metabolite/water ratios from these voxels were then computed. Similarly, average metabolite ratios were computed from normal brain parenchyma. Gliomas were considered for lack of choline elevation when choline/creatine and choline/water ratios from neoplastic regions were less than those from normal brain parenchyma regions. Six of 41 (14.6%) grade I–III gliomas showed lack of elevation for choline/creatine and choline/water ratios compared to normal brain parenchyma. Four of these six gliomas also demonstrated elevated levels of myoinositol/creatine ratio. All other gliomas (n = 35) had elevated choline levels from neoplastic regions relative to normal parenchyma. The sensitivity of choline/creatine or choline/water in determining a grade I–III glioma was 85.4%. These findings suggest that a lack of choline/creatine or choline/water elevation may be seen in some gliomas and low choline levels should not prevent us from considering the possibility of a grade I–III glioma.
Keywords: Glioma, proton magnetic resonance spectroscopy, choline, myoinositol
Introduction
Proton magnetic resonance spectroscopy (1H MRS) allows noninvasive assessment of metabolic alterations within a tissue of interest.1,2 A number of studies3–9 have reported the utility of 1H MRS for studying brain tumor metabolism. These studies have reported promising results in evaluation of glioma grade, identification of glioma genotypes, differentiation of neoplasm types, differentiation of recurrent tumors from radiation injury, and assessment of the effect of whole brain radiation therapy on normal brain parenchyma in patients with metastases. In general, an elevation in the levels of choline (Cho) resonance centered at 3.2 ppm on 1H MRS has been emphasized as a putative marker for brain neoplasms, regardless of histologic subtype.10,11 An increase in Cho content is generally considered to be due to increased cellularity and increased turnover of cell membranous structures within the neoplasms.10,11 Some 1H MRS studies have shown a positive correlation between Cho levels and the histological grade of brain neoplasms.1,2,10,11 In addition, differential levels of Cho have been shown to be useful in discriminating between neoplastic and non-neoplastic lesions in the brain.12 Taken together, these studies suggest that Cho is valuable adjunct biomarker to conventional magnetic resonance (MR) imaging for characterization of brain neoplasms.
It has also been reported that Cho levels are generally higher in neoplastic regions compared to contralateral normal parenchyma, regardless of the histological grade of gliomas.13 However, few case studies14,15 have reported lower Cho levels (choline/creatine (Cho/Cr)) in glial neoplasms, suggesting caution should be observed while interpreting the normal-appearing spectra from intracranial brain masses. However, the frequency of normal-appearing Cho levels from brain masses is generally unknown.
Therefore, the purpose of this study was to determine the frequency of lack of Cho elevation in a heterogeneous population of histologically proven grade I–III gliomas.
Materials and methods
Patients
This retrospective study was approved by the Institutional Review Board and was compliant with the Health Insurance Portability and Accountability Act. At our institution, a total of 89 patients with brain neoplasms who underwent 1H MRS between January 2009–October 2011 were identified through database searching. The primary inclusion and exclusion criteria for the selection of patients in this study are presented as a flow chart in Figure 1. Based upon the inclusion and exclusion criteria, a cohort of 41 patients (19 males and 22 females, mean age ± standard deviation = 41.6 ± 12.14 years, minimum age = 21 years and maximum age = 67 years), with grade I–III gliomas was included for the final data analysis. The histological diagnosis of grade I–III gliomas from these patients according to the World Health Organization (WHO) grading system is presented in Table 1. All patients had undergone MR evaluation prior to surgery, radiation, or chemotherapy.
Figure 1.
Flow chart of patients who met inclusion/exclusion criteria for the study population.
The flow chart shows the primary inclusion and exclusion criteria along with the total number of patients who were finally selected in the data analysis. 1H MRS: proton magnetic resonance spectroscopy.
Table 1.
Histologic diagnosis of grade I–III gliomas.
| Histopathology of neoplasms | Number of patients |
|---|---|
| Oligodendroglioma grade-III | 2 |
| Oligodendroglioma grade-II | 7 |
| Anaplastic Astrocytoma, grade-III | 7 |
| Fibrillary astrocytoma, grade-II | 10 |
| Pilocytic astrocytoma | 2 |
| Ganglioglioma | 3 |
| DNET | 3 |
| Extraventricular neurocytoma | 2 |
| Ependymoma | 1 |
| Subependymoma | 1 |
| Subependymal giant cell astrocytoma | 1 |
| Hemangioblastoma | 1 |
| Gliomatosis cerebri | 1 |
| Total | 41 |
DNET: dysembryoplastic neuroepithelial tumor.
Data acquisition
Conventional MR imaging
All patients underwent MR imaging on a 3 T Tim-Trio whole body MR scanner (Siemens, Erlangen, Germany) equipped with a 12-channel phased array head coil. The anatomical imaging protocol included a three-plane scout localizer, three-dimensional (3D)-T1-weighted magnetization-prepared rapid acquisition of gradient echo (MPRAGE) imaging (repetition time (TR)/echo time (TE)/inversion time (TI) = 1620/3.9/950 ms); matrix size = 192 × 256; section thickness = 1 mm; number of sections per slab = 192; flip angle = 15°; number of excitations (NEX) = 1; bandwidth (BW) = 150 Hz/pixel), an axial T2-fluid-attenuated inversion recovery (T2-FLAIR) image (TR/TE/TI = 9420/141/2500 ms); section thickness = 3 mm; number of sections = 60; flip angle = 170°; NEX = 1; BW = 287 Hz/pixel. The post-contrast T1-weighted images were acquired with the same parameters as the pre-contrast acquisition after administration of a standard dose (0.1 mmol/kg) of gadobenate dimeglumine contrast agent (MultiHance; Bracco Diagnostics, Monroe Township, New Jersey, USA) using a power injector (Medrad, Idianola, Pennsylvania, USA).
1H MRS imaging
Single-slice two-dimensional (2-D) multivoxel 1H MRS imaging was performed using a spin echo (point-resolved spectroscopy) sequence with water suppression by means of a chemical shift selective saturation (CHESS) sequence. Sequence parameters were: TR/TE = 1700/30 ms, NEX = 3, slice thickness = 15–20 mm, BW = 1200 Hz, matrix size = 16 × 16, vector size = 1024, BW = 1200 Hz. The size of the voxel varied from 10 × 10 × 15 mm3 (volume: 1.5 cm3) to 10 × 12.5 × 20 mm3 (volume: 2.5 cm3) depending on the dimension of the neoplasms. The volume of interest (VOI) was selected based on FLAIR images to include the largest cross section of neoplasm and contra-lateral normal-appearing brain parenchyma (NABP) in the same plane. Areas of scalp, skull base, and sinuses were avoided in the selection of the VOI. Additionally, eight outer volume saturation slabs (30 mm thick) were placed outside the VOI to suppress lipid signals from the bone and scalp. The data set was acquired using elliptical k-space sampling with weighted-phase encoding to reduce the acquisition time. To minimize the effect of increased nominal voxel size by elliptical k-space sampling, a Hamming filter (50%) was applied. Manual shimming was performed to achieve an optimal full width half maximum (FWHM) of the water signal. In general, a linewidth of < 20 Hz was achieved on the magnitude signal of the water resonance.
Data analysis
Contrast-enhanced images are generally used to select the voxels of interest for 1H MRS imaging studies. However, in this study, most of the gliomas (n = 29/41) did not demonstrate contrast enhancement on post-contrast T1-weighted images. On the other hand, hyperintense abnormality on FLAIR images was observed in all of the patients. Therefore, 1H MRS imaging data were analyzed from voxels that exhibited FLAIR hyperintensity and exhibited good spectral resolution (magnitude value of FWHM of unsuppressed water signal<20 Hz) using a Syngo software on Leonardo workstation (Siemens, Erlangen, Germany). Voxels encompassing the NABP regions were analyzed for comparison. For each spectrum, the acquired 1H MRS signal (free induction decay) was zero-filled (2048 data points), smoothed (Hanning filter, width 200 ms), and Fourier-transformed. This was followed by phase (zero and first order polynomial) and baseline correction for optimal linear frequency dependence. The Syngo software analyzes the 1H MRS data in frequency domain by fitting the individual metabolite peaks. The Syngo software has predefined model fits for most resonances observed from the human brain. The fitting of individual metabolite peaks can be optimized by adjusting the chemical shift, amplitude, and line-width interactively. The quality of spectral fitting can be estimated by the difference spectrum (fitted spectrum subtracted from original spectrum).
The region between 0.2–4.0 ppm of the spectrum was processed. The following metabolite signals were evaluated: N-acetyl aspartate (NAA), 2.02 ppm; creatine (Cr), 3.02 ppm; Cho, 3.22 ppm; myoinositol (mI), 3.56 ppm. Peak areas for Cho and mI were measured from all voxels that exhibited hyperintensity on T2-FLAIR images and were normalized to ipsilateral Cr, NAA, and unsuppressed water from each voxel. The average metabolite/Cr, metabolite/NAA, and metabolite/water ratios from these voxels were computed. Similarly, average metabolite ratios were computed from 2–8 voxels that encompassed contralateral NABP regions if a glioma was located in one cerebral hemisphere or encompassed NABP regions included within the VOI if a glioma was located on the midline. Gliomas were considered for lack of Cho elevation when Cho/Cr and Cho/water ratios from neoplastic regions were less than those from NABP regions.
Statistical analysis
Paired sample t-tests were performed to investigate the differences in Cho/Cr and Cho/water ratios between neoplastic and NABP regions from all grade I–III gliomas (n = 41). Additionally, receiver operating characteristic (ROC) curve analyses were performed to determine the optimal threshold values of Cho/Cr and Cho/water in separating a neoplastic region from NABP region. The sensitivity, specificity, and accuracy (area under the curve (AUC)) for each parameter were computed. We also computed the sensitivity of Cho/Cr and Cho/water ratios in detecting a grade I–III glioma. All statistical analyses were performed using PASW Statistics, Version 18.0 (IBM, Armonk, New York, USA).
Results
Six of 41 (14.7%) gliomas including one grade-II oligodendroglioma, two grade-III astrocytomas, one dysembryoplastic neuroepithelial tumor (DNET), one ganglioglioma, and one case of gliomatosis cerebri demonstrated spectra with lack of elevation of Cho/Cr and Cho/water from different voxels encompassing hyperintense abnormality on FLAIR images compared to NABP regions. Representative T2-FLAIR images and spectra from one case of grade-II oligodendroglioma and one case of grade-III astrocytoma demonstrating lack of elevated Cho levels from gliomas are shown in Figures 2 and 3 respectively. However, all of these six gliomas showed elevation of Cho/NAA with respect to NABP. Moreover, four of these six gliomas also showed prominent mI resonances. Mean values of Cho/Cr, Cho/Water, Cho/NAA, and mI/Cr from different voxels are presented in Table 2. All other gliomas (n = 35) had elevated Cho levels from neoplastic regions relative to NABP regions (Table 3). The sensitivity of Cho/Cr or Cho/water in detecting a grade I–III glioma was 85.4%.
Figure 2.
A 39-year-old female patient with grade-II oligodendroglioma.
Axial T2-fluid-attenuated inversion recovery (T2-FLAIR) image demonstrates a hyperintense mass in the left splenium and posterior body of corpus callosum. A multivoxel proton magnetic resonance spectroscopy (1H MRS) grid is overlaid on hyperintense mass and normal normal-appearing brain parenchyma (NABP) regions. Representative voxels from neoplasm (red color) and NABP (yellow color) and corresponding spectra exhibiting various metabolites are shown. Please note the presence of lower choline (Cho) resonance from the neoplastic region compared to that of the NABP region. Cr: creatine; mI: myoinositol; NAA: N-acetyl aspartate.
Figure 3.
A 44-year-old male patient with grade-III astrocytoma.
Axial T2-fluid-attenuated inversion recovery (T2-FLAIR) image demonstrates a hyperintense mass in the inferior cerebellar hemisphere. A multivoxel proton magnetic resonance spectroscopy (1H MRS) is overlaid on hyperintense mass and normal normal-appearing brain parenchyma (NABP) regions. Representative voxels from neoplasm (red color) and NABP (yellow color) and corresponding spectra exhibiting various metabolites are shown. Please note the presence of lower choline (Cho) resonance from the neoplastic region compared to that of the NABP region. Cr: creatine; mI: myoinositol; NAA: N-acetyl aspartate.
Table 2.
Demographic features and proton magnetic resonance spectroscopy (1H MRS) findings of grade I–III gliomas with lack of elevation of choline (Cho)/creatine (Cr) and Cho/water ratios with respect to the normal-appearing brain parenchyma (NABP). The average Cho/Cr and Cho/water ratios from neoplastic regions were less than that of NABP regions.
| Case | Age (years) | Gender | Histopathological type | Cho/Cr | Cho/water (×10–3) | Cho/NAA | mI/Cr |
|---|---|---|---|---|---|---|---|
| 1 | 39 | Male | Oligodendroglioma grade II | Tumor: 1.02 ± 0.18 NABP: 1.04 ± 0.11 | Tumor: 0.33 ± 0.06 NABP: 0.58 ± 0.02 | Tumor: 0.79 ± 0.16 NABP: 0.74 ± 0.10 | Tumor: 0.50 ± 0.17 NABP: 0.24 ± 0.07 |
| 2 | 51 | Female | Anaplastic astrocytoma, grade III | Tumor: 0.84 ± 0.16 NABP: 0.91 ± 0.03 | Tumor: 0.32 ± 0.08 NABP: 0.51 ± 0.02 | Tumor: 0.69 ± 0.19 NABP: 0.44 ± 0.08 | Tumor: 0.54 ± 0.16 NABP: 0.38 ± 0.09 |
| 3 | 42 | Male | Gliomatosis cerebri | Tumor: 0.78 ± 0.09 NABP: 0.80 ± 0.07 | Tumor: 0.48 ± 0.10 NABP: 0.62 ± 0.09 | Tumor: 1.29 ± 0.28 NABP: 0.77 ± 0.11 | Tumor: 0.50 ± 0.11 NABP: 0.44 ± 0.05 |
| 4 | 27 | Male | Ganglioglioma | Tumor: 0.66 ± 0.09 NABP: 0.77 ± 0.08 | Tumor: 0.37 ± 0.11 NABP: 0.59 ± 0.05 | Tumor: 0.78 ± 0.08 NABP: 0.36 ± 0.04 | Tumor: 0.85 ± 0.21 NABP: 0.33 ± 0.06 |
| 5 | 44 | Female | Anaplastic astrocytoma, grade III | Tumor: 0.69 ± 0.19 NABP: 0.75 ± 0.06 | Tumor: 0.42 ± 0.08 NABP: 0.64 ± 0.05 | Tumor: 0.69 ± 0.18 NABP: 0.39 ± 0.05 | Tumor: 0.36 ± 0.17 NABP: 0.50 ± 0.11 |
| 6 | 32 | Female | DNET | Tumor: 0.85 ± 0.12 NABP: 0.90 ± 0.05 | Tumor: 0.45 ± 0.14 NABP: 0.52 ± 0.06 | Tumor: 0.62 ± 0.15 NABP: 0.42 ± 0.09 | Tumor: 0.44 ± 0.09 NABP: 0.49 ± 0.06 |
DNET: dysembryoplastic neuroepithelial tumor; mI: myoinositol; NAA: N-acetyl aspartate.
Table 3.
Proton magnetic resonance spectroscopy (1H MRS) findings of grade I–III gliomas (n = 35) exhibiting higher choline (Cho)/creatine (Cr) and Cho/water ratios with respect to the normal-appearing brain parenchyma (NABP).
| Case | Histopathological type | Cho/Cr | Cho/water (×10–3) | Cho/NAA | mI/Cr |
|---|---|---|---|---|---|
| 1 | Oligodendroglioma grade III | Tumor: 0.95 | Tumor: 0.66 | Tumor: 0.66 | Tumor: 0.76 |
| NABP: 0.74 | NABP: 0.50 | NABP: 0.45 | NABP: 0.55 | ||
| 2 | Pilocytic astrocytoma | Tumor: 1.02 | Tumor: 0.67 | Tumor: 0.73 | Tumor: 0.88 |
| NABP: 0.71 | NABP: 0.62 | NABP: 0.45 | NABP: 0.49 | ||
| 3 | Oligodendroglioma grade II | Tumor: 0.98 | Tumor: 0.44 | Tumor: 0.59 | Tumor: 0.65 |
| NABP: 0.81 | NABP: 0.41 | NABP: 0.52 | NABP: 0.62 | ||
| 4 | Fibrillary astrocytoma, grade II | Tumor: 1.03 | Tumor: 0.61 | Tumor: 0.66 | Tumor: 0.52 |
| NABP: 0.96 | NABP: 0.44 | NABP: 0.64 | NABP: 0.29 | ||
| 5 | Anaplastic astrocytoma, grade III | Tumor: 1.95 | Tumor: 0.58 | Tumor: 1.91 | Tumor: 0.61 |
| NABP: 0.77 | NABP: 0.52 | NABP: 1.54 | NABP: 0.27 | ||
| 6 | DNET | Tumor: 0.91 | Tumor: 0.64 | Tumor: 0.74 | Tumor: 0.45 |
| NABP: 0.81 | NABP: 0.41 | NABP: 0.68 | NABP: 0.25 | ||
| 7 | Fibrillary astrocytoma, grade II | Tumor: 1.85 | Tumor: 0.53 | Tumor: 1.23 | Tumor: 0.57 |
| NABP: 0.82 | NABP: 0.49 | NABP: 1.04 | NABP: 0.33 | ||
| 8 | Extraventricular neurocytoma | Tumor: 2.18 | Tumor: 0.50 | Tumor: 1.28 | Tumor: 0.86 |
| NABP: 1.04 | NABP: 0.47 | NABP: 0.62 | NABP: 0.62 | ||
| 9 | Ganglioglioma | Tumor: 1.94 | Tumor: 0.69 | Tumor: 1.63 | Tumor: 0.37 |
| NABP: 0.79 | NABP: 0.65 | NABP: 1.41 | NABP: 0.27 | ||
| 10 | Pilocytic astrocytoma | Tumor: 1.10 | Tumor: 0.47 | Tumor: 0.72 | Tumor: 0.67 |
| NABP: 0.91 | NABP: 0.44 | NABP: 0.68 | NABP: 0.43 | ||
| 11 | Anaplastic astrocytoma, grade III | Tumor: 1.46 | Tumor: 0.46 | Tumor: 0.90 | Tumor: 0.49 |
| NABP: 0.75 | NABP: 0.45 | NABP: 0.58 | NABP: 0.42 | ||
| 12 | Oligodendroglioma grade II | Tumor: 1.40 | Tumor: 0.54 | Tumor: 0.97 | Tumor: 0.60 |
| NABP: 1.11 | NABP: 0.48 | NABP: 0.61 | NABP: 0.56 | ||
| 13 | Anaplastic astrocytoma, grade III | Tumor: 1.06 | Tumor: 0.72 | Tumor: 0.65 | Tumor: 0.39 |
| NABP: 0.92 | NABP: 0.62 | NABP: 0.51 | NABP: 0.49 | ||
| 14 | Extraventricular neurocytoma | Tumor: 1.32 | Tumor: 0.62 | Tumor: 0.94 | Tumor: 0.68 |
| NABP: 0.96 | NABP: 0.47 | NABP: 0.51 | NABP: 0.59 | ||
| 15 | Fibrillary astrocytoma, grade II | Tumor: 1.96 | Tumor: 0.58 | Tumor: 1.04 | Tumor: 0.33 |
| NABP: 1.01 | NABP: 0.48 | NABP: 0.89 | NABP: 0.30 | ||
| 16 | Oligodendroglioma grade III | Tumor: 3.18 | Tumor: 0.52 | Tumor: 1.08 | Tumor: 0.98 |
| NABP: 1.05 | NABP: 0.47 | NABP: 0.84 | NABP: 0.72 | ||
| 17 | Subependymal giant cell astrocytoma | Tumor: 1.40 | Tumor: 0.47 | Tumor: 0.65 | Tumor: 0.74 |
| NABP: 0.76 | NABP: 0.45 | NABP: 0.57 | NABP: 0.73 | ||
| 18 | Fibrillary astrocytoma, grade II | Tumor: 1.01 | Tumor: 0.59 | Tumor: 0.65 | Tumor: 0.67 |
| NABP: 0.84 | NABP: 0.57 | NABP: 0.54 | NABP: 0.79 | ||
| 19 | Oligodendroglioma grade II | Tumor: 1.32 | Tumor: 0.59 | Tumor: 1.46 | Tumor: 0.58 |
| NABP: 0.86 | NABP: 0.52 | NABP: 0.95 | NABP: 0.45 | ||
| 20 | DNET | Tumor: 1.17 | Tumor: 0.55 | Tumor: 1.08 | Tumor: 0.48 |
| NABP: 0.94 | NABP: 0.53 | NABP: 0.82 | NABP: 0.47 | ||
| 21 | Ganglioglioma | Tumor: 1.15 | Tumor: 0.62 | Tumor: 0.80 | Tumor: 0.59 |
| NABP: 1.08 | NABP: 0.59 | NABP: 0.63 | NABP: 0.63 | ||
| 22 | Anaplastic astrocytoma, grade III | Tumor: 1.51 | Tumor: 0.65 | Tumor: 1.25 | Tumor: 0.60 |
| NABP: 0.85 | NABP: 0.57 | NABP: 0.94 | NABP: 0.27 | ||
| 23 | Fibrillary astrocytoma, grade II | Tumor: 1.20 | Tumor: 0.72 | Tumor: 1.23 | Tumor: 0.96 |
| NABP: 0.91 | NABP: 0.68 | NABP: 0.82 | NABP: 0.94 | ||
| 24 | Fibrillary astrocytoma, grade II | Tumor: 1.05 | Tumor: 0.56 | Tumor: 0.80 | Tumor: 0.39 |
| NABP: 0.89 | NABP: 0.54 | NABP: 0.50 | NABP: 0.45 | ||
| 25 | Oligodendroglioma grade II | Tumor: 2.00 | Tumor: 0.62 | Tumor: 1.18 | Tumor: 0.58 |
| NABP: 1.06 | NABP: 0.57 | NABP: 0.95 | NABP: 0.51 | ||
| 26 | Anaplastic astrocytoma, grade III | Tumor: 2.13 | Tumor: 0.69 | Tumor: 0.99 | Tumor: 0.72 |
| NABP: 0.86 | NABP: 0.65 | NABP: 0.81 | NABP: 0.80 | ||
| 27 | Fibrillary astrocytoma, grade II | Tumor: 1.77 | Tumor: 0.59 | Tumor: 1.05 | Tumor: 0.64 |
| NABP: 0.88 | NABP: 0.41 | NABP: 0.89 | NABP: 0.66 | ||
| 28 | Subependymoma | Tumor: 0.98 | Tumor: 0.74 | Tumor: 0.54 | Tumor: 0.67 |
| NABP: 0.77 | NABP: 0.61 | NABP: 0.45 | NABP: 0.56 | ||
| 29 | Oligodendroglioma grade II | Tumor: 0.96 | Tumor: 0.50 | Tumor: 1.16 | Tumor: 0.55 |
| NABP: 0.88 | NABP: 0.48 | NABP: 0.99 | NABP: 0.27 | ||
| 30 | Fibrillary astrocytoma, grade II | Tumor: 1.18 | Tumor: 0.58 | Tumor: 0.82 | Tumor: 0.33 |
| NABP: 0.97 | NABP: 0.54 | NABP: 0.52 | NABP: 0.50 | ||
| 31 | Ependymoma | Tumor: 0.92 | Tumor: 0.50 | Tumor: 1.22 | Tumor: 0.51 |
| NABP: 0.75 | NABP: 0.47 | NABP: 0.96 | NABP: 0.38 | ||
| 32 | Fibrillary astrocytoma, grade II | Tumor: 1.29 | Tumor: 0.71 | Tumor: 0.76 | Tumor: 0.67 |
| NABP: 0.66 | NABP: 0.68 | NABP: 0.46 | NABP: 0.59 | ||
| 33 | Oligodendroglioma grade II | Tumor: 1.38 | Tumor: 0.48 | Tumor: 0.55 | Tumor: 0.83 |
| NABP: 1.01 | NABP: 0.47 | NABP: 0.48 | NABP: 0.44 | ||
| 34 | Hemangioblastoma | Tumor: 1.03 | Tumor: 0.75 | Tumor: 0.72 | Tumor: 0.40 |
| NABP: 0.95 | NABP: 0.73 | NABP: 0.45 | NABP: 0.40 | ||
| 35 | Fibrillary astrocytoma, grade II | Tumor: 1.19 | Tumor: 0.74 | Tumor: 0.79 | Tumor: 0.36 |
| NABP: 0.73 | NABP: 0.64 | NABP: 0.63 | NABP: 0.33 |
DNET: dysembryoplastic neuroepithelial tumor; mI: myoinositol; NAA: N-acetyl aspartate.
When 1H MRS data from 41 patients were pooled together, a significantly higher Cho/Cr (1.31 ± 0.50 vs 0.87 ± 0.11, p < 0.001) was observed from the neoplastic region compared to the NABP region. Additionally, the neoplastic region had higher Cho/water (0.59 ± 0.11 × 10–3 vs 0.51 ± 0.08 × 10–3) than the NABP region. However, only a trend towards significance (p = 0.075) was observed for the Cho/water ratio in separating these two regions. ROC analyses revealed that Cho/Cr was a better predictive parameter to separate a neoplastic region from the NABP region with an accuracy 85%; a sensitivity of 81% and a specificity of 80% at a threshold value of 0.97. On the other hand, Cho/water separated the two regions with a moderate accuracy of 60%; a moderate sensitivity of 66% and a moderate specificity of 54% at a threshold value of 0.58 × 10–3.
Discussion
In this study, we retrospectively evaluated 1H MRS in a heterogeneous group of grade I–III gliomas to evaluate the frequency of absent elevation of Cho levels. We observed that a substantial proportion of gliomas demonstrated absence of Cho/Cr and Cho/water elevation. Although a few case reports14,15 have been published on the lack of Cho elevation in gliomas, we are not aware of any previous systematic study specifically dedicated to investigating the frequency of absence of Cho elevation in grade I–III gliomas. We believe that this finding needs to be taken into consideration in the preoperative differential evaluation of brain masses.
The precise reason for a lack of substantial Cho elevation is not known, however, it may be speculated that the presence of relatively fewer numbers of neoplastic cells in the active division phase of the cell cycle in these gliomas as has been suggested previously14,16 may be responsible for lower Cho levels. The normal cell cycle consists of an orderly sequence of phases (growth 1 (G1), synthesis (S), G2, mitotic (M), and quiescent (G0)) and several external and internal signals regulate the checkpoints that may stimulate or inhibit the cell growth.17 However, in neoplastic cells, these checkpoints are not rigorous and hence are not followed strictly, allowing production of daughter cells with genetic alterations and subsequent entering of these cells to non-proliferative phase (G0) of cell cycle.17
Reviewing the literature dealing with 1H MRS of gliomas reveals that Cho/Cr and Cho/NAA are the two ratios that have been mainly used for the purpose of grading gliomas. Many of the published articles measured both ratios and demonstrated that both of these ratios may be helpful in glioma grading;1,10,18–20 however some studies only measured Cho/Cr ratio as the main ratio.8 In the present study, all of the cases demonstrated elevated levels of Cho/NAA from glioma regions compared to NABP regions in spite of lower Cho/Cr and Cho/water. Several studies have reported a reduction of NAA concentration in gliomas compared to normal brain regions. It has also been reported that in the normal brain, T2 relaxation time of NAA is longest and followed by Cho and Cr in descending order. This is in contrast to the reports that in gliomas, T2 relaxation time of Cho tends to be longer than that of Cr and NAA in gliomas. We believe an elevated Cho/NAA from different regions of glioma as observed in the present case series may be a combined effect of lower concentration and shorter T2 values of NAA from gliomas compared to normal parenchyma regions.
An increased Cho/Cr ratio, which is generally considered to be due to increased cellularity and increased membrane turnover in brain neoplasms, has been extensively used to investigate multiple aspects of brain tumors. One area of investigation has been to differentiate the histological grade of brain tumors based on the degree of Cho/Cr elevation, and this concept has been demonstrated by multiple studies.1,2,10,11,21,22 Based on the same notion, Cho/Cr has been commonly used in algorithms along with Cho/NAA to determine the transition zone between neoplasm (elevated Cho/Cr) and edema (normal Cho/Cr).23,24 Another use of the Cho/Cr ratio has been to differentiate between similar appearing neoplastic and non-neoplastic lesions on conventional MR imaging. Hourani et al.13 found that the best discriminator to differentiate between non-neoplastic lesions and brain neoplasms in children was the Cho/Cr ratio. In a study by Vuori et al.,12 1H MRS was used for differentiation of low-grade gliomas and focal cortical developmental malformations. The investigators of this study found a more pronounced increase of Cho and decrease of NAA in low-grade gliomas than in focal cortical developmental malformations.12
A prominent resonance at 3.5–3.6 ppm signifies the presence of mI or glycine or combinations of these two metabolites. The metabolite mI is mainly located within astrocytes and is presumed to act as an osmolyte, and its concentration is altered in many brain disorders.25,26 In a study, Castillo et al.26 reported a trend toward elevated mI levels in low-grade astrocytomas compared to high-grade astrocytomas. In addition, mI has been shown to be useful biomarker for differentiating neoplastic from non-neoplastic entities on short TE 1H MRS.27
In the present study, four out of our six gliomas with a lack of Cho elevation showed prominent mI resonance. In addition, the Cho/NAA ratio was lower compared to the NABP in all six patients. These findings emphasize the importance of analyzing metabolite ratios other than Cho/Cr before considering the possibility of eliminating a case of grade I–III glioma. Another important issue is awareness of the tumor types that may show a lack of Cho elevation. Some authors have reported only a moderate increase of Cho in patients with gliomatosis cerebri28,29 whereas others have described normal values.15,17,29 A positive correlation between the Cho/Cr ratio and survival has also been described in gliomatosis cerebri.30,31 The Cho/Cr ratio has also been reported to be lower in gangliogliomas than in gliomas.21,22,32 We had one case of gliomatosis cerebri that showed a lack of Cho elevation. Although, we had two DNET and one ganglioglioma patients who did not show a Cho/Cr elevation, we had other patients with these gliomas who demonstrated elevated Cho/Cr levels. In addition, a lack of Cho/Cr elevation was observed in two of our patients with oligodendrogliomas, a neoplasm that is not generally associated with lack of Cho contents.
A potential limitation of the current study is that only a short TE of 30 ms was used to acquire 1H MRS data. In the past, different TE values have been used for the acquisition of 1H MRS to study brain tumors.33,34 Using a short TE, it is possible to detect metabolites with short T2 relaxation times such as mI, glutamate and glutamine, and lipids. In our study, we used a short TE (30 ms), and care was taken to avoid contamination of lipid/macromolecular signals from the scalp and bone by placing outer volume saturation slabs around the VOI. The signals from short T2 components are generally attenuated or completely disappeared at higher or intermediate TE. However, these short T2 metabolites play an important role in the diagnosis and characterization of brain tumors.3–9 A study evaluating the diagnostic value of 3 T 1H MRS in grading cerebral gliomas using short and long echo times demonstrated that shorter TE 1H MRS has higher sensitivity in predicting glioma grade compared to longer TE.18 In another study, Kim et al.25 compared the application of different TEs in the grading of cerebral gliomas. At both short and intermediate TEs, there were significant differences in Cho/Cr ratio between low and high-grade gliomas, but Cho/Cr and Cho/NAA ratios were lower at short TE compared with those at intermediate TE. They proposed that the T2 relaxation time of Cho tends to be longer than that of Cr and NAA in cerebral gliomas, and they also advocated that if only a single 1H MRS sequence is to be acquired, then short TE should be preferred because of the slightly higher diagnostic accuracy associated with this sequence.
Another shortcoming of the present study was the relatively small sample size which was due to multiple reasons. While some patients had only single voxel 1H MRS from the gliomas and, as such, there was no metabolite information available from NABP regions for comparison, some other patients had poor quality multivoxel 1H MRS data. As such, these patients were excluded from the final data analysis. Moreover, we did not have availability of final histopathological diagnosis in a few cases. In spite of these shortcomings, our finding may be useful as it may caution clinical neuroradiologists in rejecting the differential diagnosis of a grade I–III gliomas even if Cho/Cr or Cho/water ratios are not elevated within the neoplastic regions. However, prospective studies in a large population of patients with grade I–III gliomas are required for further characterization of 1H MRS findings in these patients.
Conclusion
This study documents the absence of Cho/Cr and Cho/water elevation in a proportion (14%) of patients with grade I–III gliomas. Attention to other 1H MRS measures such as Cho/NAA, mI/Cr, and conventional diffusion and perfusion MR imaging features is important in the evaluation of brain masses, in order to avoid the pitfall of excluding grade I–III glioma from the differential diagnosis of brain lesions when there is no elevation of Cho/Cr and Cho/water ratios.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
- 1.Brandão LA, Castillo M. Adult brain tumors: Clinical applications of magnetic resonance spectroscopy. Magn Reson Imaging Clin N Am 2016; 24: 781–809. [DOI] [PubMed] [Google Scholar]
- 2.Bertholdo D, Watcharakorn A, Castillo M. Brain proton magnetic resonance spectroscopy: Introduction and overview. Neuroimaging Clin N Am 2013; 23: 359–380. [DOI] [PubMed] [Google Scholar]
- 3.Chawla S, Krejza J, Vossough A, et al. Differentiation between oligodendroglioma genotypes using dynamic susceptibility contrast perfusion-weighted imaging and proton MR spectroscopy. Am J Neuroradiol 2013; 34: 1542–1549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chawla S, Wang S, Kim S, et al. Radiation injury to the normal brain measured by 3D-echo-planar spectroscopic imaging and diffusion tensor imaging: Initial experience. J Neuroimaging 2015; 25: 97–104. [DOI] [PubMed] [Google Scholar]
- 5.Chawla S, Wang S, Wolf RL, et al. Arterial spin-labeling and MR spectroscopy in the differentiation of gliomas. Am J Neuroradiol 2007; 28: 1683–1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chawla S, Zhang Y, Wang S, et al. Proton magnetic resonance spectroscopy in differentiating glioblastomas from primary cerebral lymphomas and brain metastases. J Comput Assist Tomogr 2010; 34 836–841. [DOI] [PubMed] [Google Scholar]
- 7.Crain ID, Elias PS, Chapple K, et al. Improving the utility of 1H-MRS for the differentiation of glioma recurrence from radiation necrosis. J Neurooncol 2017; 133: 97–105. [DOI] [PubMed] [Google Scholar]
- 8.Morales H, Alfaro D, Martinot Carlos, et al. MR spectroscopy of intracranial tuberculomas: A singlet peak at 3.8 ppm as potential marker to differentiate them from malignant tumors. Neuroradiol J 2015; 28: 294–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Naveed MA, Goyal P, Malhotra A, et al. Grading of oligodendroglial tumors of the brain with apparent diffusion coefficient, magnetic resonance spectroscopy, and dynamic susceptibility contrast imaging. Neuroradiol J 2018; 31: 379–385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fountas KN, Kapsalaki EZ, Vogel RL, et al. Noninvasive histologic grading of solid astrocytomas using proton magnetic resonance spectroscopy. Stereotact Funct Neurosurg 2004; 82: 90–97. [DOI] [PubMed] [Google Scholar]
- 11.Zhang J, Zhuang DX, Yao CJ, et al. Metabolic approach for tumor delineation in glioma surgery: 3D MR spectroscopy image-guided resection. J Neurosurg 2016; 124: 1585–1593. [DOI] [PubMed] [Google Scholar]
- 12.Vuori K, Kankaanranta L, Hakkinen AM, et al. Low-grade gliomas and focal cortical developmental malformations: Differentiation with proton MR spectroscopy. Radiology 2004; 230: 703–708. [DOI] [PubMed] [Google Scholar]
- 13.Hourani R, Horska A, Albayram S, et al. Proton magnetic resonance spectroscopic imaging to differentiate between nonneoplastic lesions and brain tumors in children. J Magn Reson Imaging 2006; 23: 99–107. [DOI] [PubMed] [Google Scholar]
- 14.Londono A, Castillo M, Armao D, et al. Unusual MR spectroscopic imaging pattern of an astrocytoma: Lack of elevated choline and high myo-inositol and glycine levels. Am J Neuroradiol 2003; 24: 942–945. [PMC free article] [PubMed] [Google Scholar]
- 15.Saraf-Lavi E, Bowen BC, Pattany PM, et al. Proton MR spectroscopy of gliomatosis cerebri: Case report of elevated myoinositol with normal choline levels. Am J Neuroradiol 2003; 24: 946–951. [PMC free article] [PubMed] [Google Scholar]
- 16.Porto L, Kieslich M, Franz K, et al. MR spectroscopy differentiation between high and low grade astrocytomas: A comparison between paediatric and adult tumours. Eur J Paediatr Neurol 2011; 15: 214–221. [DOI] [PubMed] [Google Scholar]
- 17.Bendszus M, Warmuth-Metz M, Klein R, et al. MR spectroscopy in gliomatosis cerebri. Am J Neuroradiol 2000; 21: 375–2. [PMC free article] [PubMed] [Google Scholar]
- 18.Kousi E, Tsougos I, Tsolaki E, et al. Spectroscopic evaluation of glioma grading at 3T: The combined role of short and long TE. Sci World J 2012; 2012: 1–11. [DOI] [PMC free article] [PubMed]
- 19.Liu ZL, Zhou Q, Zeng QS, et al. Noninvasive evaluation of cerebral glioma grade by using diffusion-weighted imaging-guided single-voxel proton magnetic resonance spectroscopy. J Int Med Res 2012; 40: 76–84. [DOI] [PubMed] [Google Scholar]
- 20.Zeng QS, Liu HP, Zhang K, et al. Noninvasive evaluation of cerebral glioma grade by using multivoxel 3D proton MR spectroscopy. Magn Reson Imaging 2011; 29: 25–31. [DOI] [PubMed] [Google Scholar]
- 21.Law M, Yang S, Wang H, et al. Glioma grading: Sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. Am J Neuroradiol 2003; 24: 1989–2. [PMC free article] [PubMed] [Google Scholar]
- 22.Senft C, Hattingen E, Pilatus U, et al. Diagnostic value of proton magnetic resonance spectroscopy in the noninvasive grading of solid gliomas: Comparison of maximum and mean choline values. Neurosurgery 2009; 65: 908–913. [DOI] [PubMed] [Google Scholar]
- 23.Al-Okaili RN, Krejza J, Woo JH, et al. Intraaxial brain masses: MR imaging-based diagnostic strategy - initial experience. Radiology 2007; 243: 539–550. [DOI] [PubMed] [Google Scholar]
- 24.Law M, Cha S, Knopp EA, et al. High-grade gliomas and solitary metastases: Differentiation by using perfusion and proton spectroscopic MR Imaging. Radiology 2002; 222: 715–721. [DOI] [PubMed] [Google Scholar]
- 25.Kim JH, Chang KH, Na DG, et al. 3T H-1-MR spectroscopy in grading of cerebral gliomas: Comparison of short and intermediate echo time sequences. Am J Neuroradiol 2006; 27: 1412–1418. [PMC free article] [PubMed] [Google Scholar]
- 26.Castillo M, Smith JK, Kwock L. Correlation of myo-inositol levels and grading of cerebral astrocytomas. Am J Neuroradiol 2000; 21: 1645–1649. [PMC free article] [PubMed] [Google Scholar]
- 27.Majos C, Aguilera C, Alonso J, et al. Proton MR spectroscopy improves discrimination between tumor and pseudotumoral lesion in solid brain masses. Am J Neuroradiol 2009; 30: 544–551. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mohana-Borges AV, Imbesi SG, Dietrich R, et al. Role of proton magnetic resonance spectroscopy in the diagnosis of gliomatosis cerebri – a unique pattern of normal choline but elevated myo-inositol metabolite levels. J Comput Assist Tomogr 2004; 28: 103–105. [DOI] [PubMed] [Google Scholar]
- 29.Pyhtinen J. Proton MR spectroscopy in gliomatosis cerebri. Neuroradiology 2000; 42: 612–615. [DOI] [PubMed] [Google Scholar]
- 30.Galanaud D, Chinot O, Nicoli F, et al. Use of proton magnetic resonance spectroscopy of the brain to differentiate gliomatosis cerebri from low-grade glioma. J Neurosurg 2003; 98: 269–276. [DOI] [PubMed] [Google Scholar]
- 31.Guzman-De-Villoria JA, Sanchez-Gonzalez J, et al. H-1 MR spectroscopy in the assessment of gliomatosis cerebri. Am J Roentgenol 2007; 188: 710–714. [DOI] [PubMed] [Google Scholar]
- 32.Zhang D, Henning TD, Zou LG, et al. Intracranial ganglioglioma: Clinicopathogical and MRI findings in 16 patients. Clin Radiol 2008; 63: 80–91. [DOI] [PubMed] [Google Scholar]
- 33.Liu M, Yue Q, Isobe T, et al. Proton MR spectroscopy of central neurocytoma using short and long echo time: New proofs for the existence of glycine and glutamate. Acad Radiol 2012; 19: 779–784. [DOI] [PubMed] [Google Scholar]
- 34.Yamamoto T, Isobe T, Akutsu H, et al. Influence of echo time in quantitative proton MR spectroscopy using LC Model. Magn Reson Imaging 2015; 33: 644–648. [DOI] [PubMed] [Google Scholar]