Application Value of 3.0-T Multivoxel 1H-MR Spectroscopy in the Peritumoral Tissue of Brain Astrocytic Tumors
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.280633
Source of Support: None, Conflict of Interest: None
Keywords: Astrocytoma, brain neoplasms, magnetic resonance spectroscopyKey Messages: In this study, we sought to analyze the 3.0T multivoxel 1HMRS ratio values in the peritumoral tissues of the high-grade and low-grade astrocytoma. The information presented in this article may be useful to improve the diagnostic accuracy and define the infiltration area in surrounding tissues of brain astrocytoma.
Gliomas are the most common primary intracranial tumor. According to the 2016 World Health Organization (WHO) Classification of Tumors of the Central Nervous System, gliomas consist of all brain tumors that are of glial cell origin, such as astrocytoma, oligodendroglioma, oligoastrocytoma, and ependymoma, accounting for almost 80% of primary malignant brain tumors. Among them, astrocytoma is the most common architectural pattern.
It is remarkably important to accurately investigate whether astrocytomas infiltrate the surrounding tissue and the range of their infiltration for developing rational treatment regimens. The 3.0-T magnetic resonance spectroscopy (1 H-MRS) can measure the metabolism of brain tissues in vivo and provide information such as cellular metabolism and cell membrane disintegration. Moreover, it can potentially provide new strategies for studying the biological behavior of brain tumors. This study retrospectively analyzed 3.0-T multivoxel 1 H-MRS ratio values in the surrounding tissue of different-level brain astrocytomas to further explore its clinical application in classifying astrocytomas and indicating the scope of invasion.
The 3.0-T MRI data were collected from 82 patients with astrocytoma of different severity from January 2010 to January 2017. All cases were confirmed using histopathological and immunohistochemical methods. According to the 2016 WHO central nervous system tumor classification, 32 cases were categorized as low-grade (WHO I–II) astrocytoma while 50 cases as high-grade (WHO III–IV grade) astrocytoma. All participants (including 37 males and 45 females, aged 18–79 years) underwent routine MRI scanning, diffusion-weighted imaging, and enhanced scanning. The study was approved by the Institutional Review Board, and all patients provided informed consent.
MRI was performed using a 3.0-T scanner (Achieva, Philips X-series, Netherlands) along with double gradient field and 16-channel head coil-assisted imaging. Multivoxel 1 H-MRS used 2D-point-resolved spectroscopy (2D-PRESS) method: repetition time (TR), 2000 ms; echo time (TE), 144 ms; field of view, 230 × 190 mm 2; layer thickness, 10 mm; number of excitation (NEX), one time. The size of each voxel was 10 × 10 × 10 mm 3. Automatic pre-scanning was used to complete shimming and water suppression.
Image post-processing and analyses
Two senior physicians analyzed all images independently. Region of interest (ROI) was measured on the workstation according to the characteristics of the lesion. ROI size should cover the parenchymal region, surrounding tissue area, and contralateral normal brain tissue, while avoiding bone, ventricle, blood vessel, necrosis, cystic change, bleeding, gas, and calcification. Finally, the measurement of 1 H-MRS ROI was sent to the Picture Archiving and Communication System (PACS). Choline (Cho)/creatine (Cr), N-acetyl aspartate (NAA)/Cr, and Cho/NAA ratios in the tumor parenchyma, tumor surrounding tissues, and contralateral normal brain tissues were calculated. Lac (Lac), glutamic acid (Glu) and glutamate (Glx), inositol (mI) and lipid (Lip), and other metabolites, along with their corresponding ratios, were also investigated and recorded.
The statistical analyses were performed using SPSS 19.0 statistical software (SPSS, IL, USA) with single-factor analysis of variance. The ratios of Cho/Cr, NAA/Cr, Cho/NAA, and Glx/Cr in the tumor parenchyma, surrounding tissue, and normal regions were calculated. The results were demonstrated as x̄± s, and the boundary value was P < 0.05.
3.0-T multivoxel 1H-MRS ratio values in the surrounding tissue of low-grade brain astrocytomas
No significant increase was observed in the Cho peak in the surrounding tissue area of the low-grade brain astrocytomas. Moreover, the decreases in Cr and NAA peaks were not remarkable. The ratios of Cho/Cr, NAA/Cr, and Cho/NAA in the surrounding tissue areas of low-grade astrocytomas were statistically significant compared with the ratios in the tumor parenchyma (P < 0.01). The ratios of Cho/Cr, NAA/Cr, and Cho/NAA in the surrounding tissue area of low-grade astrocytomas were similar to those in the contralateral normal tissue. The metabolite ratios in different regions of low-grade astrocytomas from 32 cases are shown in [Table 1] and [Figure 1].
3.0-T multivoxel 1H-MRS ratio values in the surrounding tissue of high-grade brain astrocytomas
The peak of Cho in the surrounding tissue area of glioblastomas increased while those of Cr and NAA decreased. The expression of different ROIs was inconsistent, suggesting that the tumor infiltrated into the surrounding tissue asymmetrically. At 2.8 ppm, the Glx peak increased, reached, or nearly approached half of the Cr peak. The changes in Cho, Cr, and NAA in the surrounding tissue area of anaplastic astrocytomas were uncertain. Significant differences were found in the ratios of Cho/Cr, NAA/Cr, and Cho/NAA between the surrounding tissue area and tumor parenchyma in high-grade astrocytomas (P = 0.00). [Table 2] and [Figure 2] show the metabolite ratios in different regions of high-grade astrocytomas from 50 cases.
Comparison of metabolite ratios in different regions of high- and low-grade brain astrocytomas
A significant difference was observed in Cho/Cr ratios between the high-grade astrocytoma and the low-grade astrocytoma (t = 4.274, P < 0.01). The difference in the Cho/NAA ratio was statistically significant (t = 2.413, P < 0.05), but the NAA/Cr ratio was not statistically significant (t = 1.757, P > 0.05). The metabolite ratios in the parenchyma and surrounding tissues of high-trade and low-grade astrocytomas are demonstrated in [Table 3].
Advantages of 3.0-T multivoxel 1H-MRS application in brain tumors
The 3.0-T multivoxel 1 H-MRS is sensitive for detecting low levels of metabolites in vivo such as Glx and mI. The benefits of the 2D-PRESS method are as follows: (1) an obvious improvement in signal-to-noise ratio; (2) shortened signal acquisition time; (3) higher spectrum resolution; (4) more stable baseline; (5) larger coverage of a collection; (6) less susceptibility to the influence of fat; and (7) more reliable quantification of various metabolites. The 3.0-T multivoxel 1 H-MRS can also be used to observe the metabolic index in certain areas such as tumor parenchyma, tumor surrounding tissue, and normal area., The application of 1 H-MRS in the diagnosis of brain tumors is to differentiate between benign and malignant brain tumors, intracranial tumors and non-neoplastic lesions, brain tumor necrosis and brain abscess, primary brain tumors and metastases, and clear tumor type and degree of malignancy. The sensitivity, specificity, and accuracy in distinguishing benign and malignant tumors were 100%, 86%, and 96%, respectively. The 1 H-MRS ratio, which is the most widely used currently, emerges as the most valuable standardized ratio. The 3.0-T, compared with 1.5-T, multivoxel 1 H-MRS can better distinguish between Cho and Cr with a difference of 0.2 ppm. It can also detect 2.05 ppm of Glu, 2.35 ppm of Glu, and 3.8 ppm of Glx in short TE, which is more favorable for the diagnosis of tumor levels and the judgment regarding the extent of infiltration to surrounding tissue. In this group of patients, the 3.0-T multi-voxel 1 H-MRS has shown high success rate, accurate positioning, voxel volume reduction, and high safety and repeatability. Therefore, it can be considered as the main inspection method used in the study and diagnosis of brain tumors.
Practicality of 1H-MRS quantitative ratio in peripheral tissues of cerebral astrocytomas
The brain astrocytoma entities often undergo bleeding, necrosis, cystic degeneration, and other pathological changes, remarkably affecting the accuracy of 1 H-MRS in the tumor diagnosis. The metabolic changes in tumor surrounding tissues reflect the main biological behavior of the tumors. Therefore, the 1 H-MRS examination of the tumor surrounding tissue is an effective complementary diagnostic method. Studies have shown that 100% tumor invasion in pathological specimens can be seen when the Cho peak is two times higher than that in the normal brain tissue and the NAA peak is less than half of that in the normal brain tissue. 1H-MRS is more accurate than conventional MRI in determining the tumor demarcation and quantitatively analyzing tumor infiltration into the surrounding tissues. Astrocytoma is a neuroepithelial tumor common in the central nervous system. Infiltration into the surrounding tissues and no capsule between normal tissues are the main biological characteristics. Pathological immunohistochemical analysis shows glial fibrous acid protein positivity, high cell proliferation index, and high expression of vimentin. High expression of vimentin indicates tumor invasion. NAA decreases but Co and ratios of Cho/Cr and Cho/NAA increase with the increase in the grade of brain astrocytoma. The results showed that Cho/Cr and Cho/NAA were significantly different between high- and low-grade astrocytomas (P < 0.001), consistent with the values reported in the literature. Further, the Glx peak and Glx/Cr were elevated in the tumor parenchyma and surrounding tissue of the high-grade astrocytomas compared with the low-grade ones, which might be associated with malignant tumor cell proliferation and needs to be further investigated. High-grade astrocytes asymmetrically invade to the surrounding tissue. A certain overlap exists between the variability of MRS in the astrocytomas and the surrounding astrocytomas.
Clinical importance of 1H-MRS in determining the infiltration range of brain astrocytoma surrounding tissues
It is important to accurately define the infiltration range of brain astrocytoma surrounding tissues for surgical resection and radiotherapy vision. It also helps in detecting tumor recurrence and radiation brain injury, assessing the tumor progression, and monitoring the treatment efficacy. The extent of conventional MRI enhancement of brain tumor is not able to accurately reflect the tumor range, which is mainly affected by many factors such as blood–brain barrier, change in tissue reactivity, and radioactive brain injury. At present, MRI application of 1 H-MRS and perfusion-weighted imaging can clearly reflect the scope of brain tumor invasion. The results of the present study suggested the clinical value of 1 H-MRS in defining the infiltration area in surrounding tissues of brain astrocytoma.
In addition to conventional MR imaging techniques, advanced techniques, including diffusion tensor imaging (DTI), perfusion weighted imaging (PWI), and 1 H-MRS, can provide more information than that regarding anatomy.1 H-MRS can provide metabolic information, including that regarding tissue activity or brain damage correlating with pathological changes. Stadlbauer et al. demonstrated that 1 H-MRS can improve delineation of tumor borders compared to routine imaging strategies. These techniques have been commonly used in the clinical field. Combining images from 1 H-MRS with those from other advanced techniques is reported to improve diagnostic accuracy for gliomas.,,,,
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]