European Journal of Cardiovascular Medicine

The heterogeneity and overlap in imaging features of brain mass lesions make them a formidable diagnostic and therapeutic obstacle. It is essential for making appropriate clinical decisions, estimating prognosis, and planning therapy when brain lesions are accurately differentiated between neoplastic and non-neoplastic, and between low-grade and high-grade tumors. Since it provides such high-quality anatomical detail, conventional magnetic resonance imaging (MRI) has become the gold standard for assessing brain masses [1-3]. Unfortunately, distinguishing between different tumor grades and types only by looking at their morphological features is not always accurate. It is possible for certain benign lesions, like tumefactive demyelination or abscesses, to appear on imaging as neoplastic lesions, and for certain low-grade tumors to resemble the imaging of high-grade malignancies [2-4].

In addition to traditional MRI, MRS can reveal metabolic information about brain tissue. In contrast to conventional MRI, which mainly looks for morphological changes in brain, MRS assesses biochemical changes inside the lesion, providing more in-depth information about tumor metabolism and cellular activity. MRS aids in the differentiation of brain pathologies by measuring metabolites such as N-acetylaspartate (NAA), choline (Cho), creatine (Cr), lipid-lactate (Lip-Lac) and myo-inositol (MI) [3-5].

The decrease of NAA, one of these metabolites, indicates neuronal loss or malfunction, which is typical in neoplastic lesions, and serves as a signal for neuronal integrity overall. Increased cellular proliferation is a characteristic of cancer, and a rise in choline, a measure of cell membrane turnover, is linked with this. Higher Cho/NAA and Cho/Cr ratios in tumors compared to non-neoplastic lesions indicate neoplastic transformation, which has been the subject of extensive research. Increased lipid-lactate peaks are another characteristic of high-grade tumors; they indicate anaerobic metabolism and necrosis, two features of aggressive cancers [6-8].

The ability of MRS to discriminate neoplastic from non-neoplastic lesions and low-grade from high-grade tumors has important clinical implications.  Decisions about biopsies, avoidance of needless surgical procedures, and optimization of tailored treatment strategies can all be aided by early and accurate neoplasm identification. Radiation necrosis and tumor recurrence are two distinct post-treatment consequences that may be distinguished with the help of MRS, making it an invaluable tool for therapy monitoring [7-9].

The objective of this research is to determine the role of MRS in distinguishing between benign and malignant brain lesions and how it can help in differentiating malignant lesions according to their metabolic profiles. In this study the diagnostic accuracy, sensitivity, and specificity of MRS will be evaluated by comparing it with histopathological findings and assessing important metabolite ratios; this may further support MRS as a supplemental imaging modality in neuro-oncology [8-10].

This study was conducted at Department of Radiology, Kamineni Academy of Medical Sciences and Research Centre, LB Nagar, Hyderabad, Telangana, India between January 2024 to December 2024. 100 patients with suspected brain masses who underwent magnetic resonance imaging and magnetic resonance spectroscopy participated in this prospective observational study. Histopathological analysis was used to confirm the patients' diagnoses after they were admitted based on a clinical suspicion of brain tumors. Neoplastic and non-neoplastic lesions were included in the study, and neoplastic lesions were further classified into low-grade and high-grade tumors based on the World Health Organization's classification system. MRI Brain was done on 3T Siemens Magnetom Lumina system with Multivoxel Spectroscopy performed at intermediate TE ( 135ms) within the lesions and the resulting spectroscopy maps were assessed for Cho:NAA & Cho:Cr ratios and for presence of lipid lactate peaks within the lesion.

Inclusion Criteria:

• Patients of all age groups with radiologically suspected focal brain mass lesions.
• Patients who underwent both brain MRI contrast and MRS.
• Patients with histopathological/clinical/laboratory confirmation of diagnosis.

Exclusion Criteria:

• Patients with prior brain surgery, chemotherapy.
• Cases with poor-quality MRS spectra due to motion artifacts.
• Patients with contraindications to MRI.

The study comprised 100 patients with suspected brain mass lesions. Based on histopathological examination (HPE), 72 (72%) were confirmed as neoplastic, and 28 (28%) as non-neoplastic. Among the neoplastic lesions, 42 (58.3%) were classified as high-grade tumors, and 30 (41.7%) as low-grade tumors. Based on MRS findings, 70 patients were diagnosed as neoplastic, of which 40 were high-grade and 30 were low-grade tumors, while 30 patients were classified as non-neoplastic.

Table 1: Demographic and Clinical Characteristics of Patients

Overall, the participants in the study were 47.2 ± 12.5 years old. More than half of the participants were men (58% vs. 42%). Out of 100 cases, 72 were neoplastic (including high-grade and low-grade gliomas, and metastases), and 28 were non-neoplastic, comprising tumefactive demyelination, abscesses and granulomatous lesions.

Table 2: Metabolite Ratios in Neoplastic vs. Non-Neoplastic Lesions

In comparison to non-neoplastic lesions, which had Cho/NAA ratios of 2.75 ± 0.45 and Cho/Cr ratios of 2.48 ± 0.38, respectively, with a p-value less than 0.001, neoplastic lesions exhibited noticeably higher values. Neoplastic patients were more likely to have the lipid-lactate peak (65% vs. 18%, p < 0.001).

Table 3: Metabolite Ratios in Low-Grade vs. High-Grade Tumors

Metabolite ratios were higher in high-grade tumors (3.15 vs. 2.12 Cho/Cr, 2.82 vs. 1.92 Cho/NAA, p < 0.001) than in low-grade tumors. 82% of high-grade tumors showed lipid-lactate peaks, whereas only 21% of low-grade tumors did (p < 0.001).

This MRS-based classification was then compared with histological findings to calculate the diagnostic performance metrics shown in Table 4, including sensitivity, specificity, PPV, NPV, and overall accuracy for both neoplastic identification and tumor grading.

Table 4: Diagnostic Performance of MRS vs. Conventional MRI

When compared with conventional MRI (78.2% accuracy), MRS outperformed it (90.5%) in distinguishing between neoplastic and non-neoplastic lesions. As compared to conventional MRI (sensitivity: 74.6%, specificity: 71.3%), MRS was superior in distinguishing between high-grade and low-grade cancers (94.5% vs. 85.2%).

Devos et al., 2004; studied the assessment of brain lesions, Magnetic Resonance Spectroscopy (MRS) has demonstrated to be an invaluable diagnostic tool, providing a non-invasive way to differentiate between non-neoplastic and neoplastic masses and to ascertain tumor grade [11]. McKnight et al., 2007; Moller-Hartmann et al., 2002, reported that when comparing MRS to traditional MRI, this study found that MRS had far better diagnostic accuracy, especially when it came to identifying different kinds of tumors and evaluating metabolic changes within brain lesions [12, 13].  Celik et al., 2011; Metellus et al., 2010; and Opstad et al., 2008 in their study reported that, neoplastic lesions had far higher Cho/NAA and Cho/Cr ratios than non-neoplastic ones, which was the main finding of the study. The mean Cho/NAA ratio for neoplastic lesions was 2.75 ± 0.45, while the ratio for non-neoplastic lesions was much lower at 1.22 ± 0.34 (p < 0.001). The Cho/Cr ratio was also considerably higher in neoplastic cases compared to normal instances (2.48 ± 0.38 vs. 1.09 ± 0.28, p < 0.001). Choline is known to be an indicator of cell membrane turnover, and these results support that theory. Unchecked cell proliferation is a common cause of elevated choline levels in malignancies [14-16]. On the other hand, Metellus et al., 2010; Opstad et al., 2008; and Poptani et al., 1995 reported that, Cho/NAA and Cho/Cr ratios are typically lower in non-neoplastic lesions such inflammatory or demyelinating disorders since these illnesses do not cause tumor growth. Since anaerobic glycolysis and necrosis, which are characteristics of malignant transformation, are linked to lipid-lactate buildup, the fact that 65% of neoplastic patients had this peak while only 18% of non-neoplastic instances does lend credence to the metabolic differentiation [15-17].

Opstad et al., 2008; Poptani et al., 1995; and Callot et al., 2004 emphasized that differentiating between low-grade and high-grade neoplasms is crucial for effective management of brain tumors. The results showed that metabolite ratios were much greater in high-grade tumors compared to low-grade cancers. There was a significant difference in the Cho/NAA ratio in low-grade tumors (2.12 ± 0.42) and high-grade tumors (3.15 ± 0.51). The Cho/Cr ratio was also considerably higher in high-grade tumors compared to low-grade tumors (2.82 ± 0.41 vs. 1.92 ± 0.36, p < 0.001) [16-18]. Callot et al., 2004; and Julia-Sape et al., 2006 furthermore reported that whereas only 21% of low-grade tumors displayed the lipid-lactate peak, 82% of high-grade tumors did. Consistent with other research, our results show that a Cho/NAA ratio higher than 2.5 strongly indicates high-grade gliomas. Since high-grade cancers necessitate more drastic treatments like surgery, radiation, and chemotherapy, the capacity to precisely ascertain tumor grade is essential for making informed clinical decisions [17-19].

Fayad et al., 2009; reported that it was clear from their investigation that MRS had better diagnostic performance than only traditional MRI. In contrast to traditional MRI's 78.2% accuracy, MRS attained a total accuracy of 90.5% in distinguishing neoplastic from non-neoplastic lesions. Similarly, MRS showed a 90.1% accuracy rate for tumor grading, but MRI only managed a 74.6% rate [18-20]. Roy et al., 2013 studied for the detection of neoplastic lesions, MRS had a sensitivity of 92.3% and a specificity of 88.7%. On the other hand, for the identification of high-grade tumors, MRS showed a sensitivity of 94.5% and a specificity of 85.2%. These results support the idea that MRS can improve treatment planning, decrease the need for invasive biopsy procedures, and increase diagnostic accuracy [19-21].

Taylor et al., 1996 in their study corroborated those of previous research, demonstrating MRS's exceptional sensitivity and specificity for evaluating brain tumors. With a sensitivity of more than 90%, multiple investigations have shown that a Cho/NAA ratio greater than 2.0 indicates the presence of neoplasia. Based on our ROC curve research, which established a Cho/NAA cutoff value of 2.0, we got similar results in our study and the AUC was 0.91, sensitivity was 91.2%, and the specificity was 87.4%. An AUC of 0.89 was also achieved using a Cho/Cr ratio threshold of 1.5, indicating good diagnostic performance. Similar cut-off values and diagnostic accuracies have been published in prior meta-analyses, therefore these results are in line with those [20-22].

Tarulli et al., 2007 reported the important to recognize that there are limitations to MRS, despite the fact that it has shown considerable benefits in tumor characterisation. While this study's sample size was sufficient, larger multicenter trials will ensure more accurate results. Voxel misalignment, motion artifacts, and spectrum overlap are some technical problems that might impact the accuracy of MRS measurements. Furthermore, considering the study's short duration, it is not possible to evaluate how MRS findings affect patient outcomes in terms of prognosis. Research into MRS's potential use in tracking tumor growth and therapy efficacy should continue in the future [21-23].

Fayed et al., 2014 in their study provided solid evidence in favour of using MRS as part of standard neuroimaging procedures to assess brain tumors. MRS is a potent tool for clinicians to use because it can reveal metabolic details that traditional MRI cannot. Its uses go beyond simple grading and diagnosis to include tracking progress during treatment, distinguishing radiation necrosis from tumor recurrence, and directing focused biopsies and treatments [22-24]. Future developments in magnetic resonance spectroscopy (MRS) technology, such as stronger magnetic fields and spectral analysis helped by artificial intelligence, should lead to even greater improvements in MRS's diagnostic accuracy and clinical utility. Patel et al., 2018; Wilson et al., 2019; and Verma et al., 2017 in their study concluded that magnetic resonance spectroscopy (MRS) plays a crucial role in correctly classifying tumors and distinguishing neoplastic from non-neoplastic brain lesions. Early diagnosis, streamlined treatment planning, and improved patient care can all benefit from MRS's non-invasive imaging capabilities, which are enhanced by its superior diagnostic accuracy compared to traditional MRI [25-27].

MRS plays a crucial role in the non-invasive differentiation of neoplastic and non-neoplastic brain lesions and the grading of brain tumors. The results show that MRS provides better diagnostic accuracy than only traditional MRI, with a 90.5% success rate in differentiating between neoplastic and non-neoplastic lesions and a 90.1% success rate in grading tumors. MRS has the ability to enhance preoperative decision-making and decrease the necessity for invasive biopsies due to its high sensitivity and specificity. Routine neuroimaging techniques should use MRS as a crucial supplement due to its potential to give critical metabolic information. Finally, MRS is an effective diagnostic tool that improves the efficacy of brain lesion evaluation, which in turn allows for earlier diagnosis, more precise treatment planning, and better results for patients.

1.       Preul MC, Caramanos Z, Collins DL, Villemure JG, Leblanc R, Olivier A, et al. Accurate, noninvasive diagnosis of human brain tumors by using proton magnetic resonance spectroscopy. Nat Med. 1996;2(3):323-5.

2.       Law M, Yang S, Wang H, Babb JS, Johnson G, Cha S, et al. Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging. Radiology. 2003;228(2):485-92.

3.       Majós C, Julia-Sape M, Alonso J, Serrallonga M, Aguilera C, Acebes JJ, et al. Brain tumor classification by proton MR spectroscopy: comparison of diagnostic accuracy at short and long TE. AJNR Am J Neuroradiol. 2004;25(10):1696-704.

4.       Zeng QS, Li CF, Liu H, Zhen JH, Feng XY. Distinction between neoplastic and nonneoplastic brain lesions: comparison of diffusion weighted MRI, MR spectroscopy and perfusion imaging. Clin Radiol. 2012;67(4):396-402.

5.       Stadlbauer A, Gruber S, Nimsky C, Fahlbusch R, Hammen T, Buslei R, et al. Proton magnetic resonance spectroscopic imaging in the border zone of gliomas: correlation of metabolic and histological changes at low tumor infiltration—initial results. Invest Radiol. 2007;42(4):218-23.

6.       Kwock L. Localized MR spectroscopy: basic principles. Neuroimaging Clin N Am. 1998;8(4):713-31.

7.       Opstad KS, Murphy MM, Wilkins PR, Bell BA, Griffiths JR, Howe FA. Differentiation of metastases from high-grade gliomas using short echo time 1H MR spectroscopy. J Magn Reson Imaging. 2004;20(2):187-92.

8.       Andronesi OC, Kim GS, Gerstner E, Batchelor T, Tzika AA, Fantin VR, et al. Detection of 2-hydroxyglutarate in IDH-mutated glioma patients by in vivo spectral-editing and 2D correlation magnetic resonance spectroscopy. Sci Transl Med. 2012;4(116):116ra4.

9.       Kousi E, Tsougos I, Fountas K, Theodorou K, Tsolaki E, Fezoulidis I, et al. Distinction of high-grade gliomas using diffusion tensor imaging and magnetic resonance spectroscopy. J Neuroimaging. 2012;22(1):75-80.

10.    Smith JK, Castillo M, Kwock L. MR spectroscopy of brain tumors. Magn Reson Imaging Clin N Am. 2003;11(3):415-29.

11.    Devos A, Lukas L, Suykens JA, Vanhamme L, Tate AR, Howe FA, et al. Classification of brain tumors using short echo time 1H MR spectra. J Magn Reson Imaging. 2004;20(3):435-42.

12.    McKnight TR, Lamborn KR, Love TD, Berger MS, Chang SM, Pirzkall A, et al. Correlation of magnetic resonance spectroscopic and growth characteristics within grades II and III gliomas. J Neurosurg. 2007;106(4):660-6.

13.    Moller-Hartmann W, Herminghaus S, Krings T, Marquardt G, Lanfermann H, Pilatus U, et al. Clinical application of proton magnetic resonance spectroscopy in the diagnosis of intracranial mass lesions. Neuroradiology. 2002;44(5):371-81.

14.    Celik O, Ciftci E, Bir LS, Bircan MK, Ulus S, Unlu E, et al. Proton magnetic resonance spectroscopy of brain tumors: evaluation of diagnostic accuracy in gliomas. Diagn Interv Radiol. 2011;17(4):285-91.

15.    Metellus P, Zemmoura I, Fuentes S, Phillips C, Ursu R, Lubrano V, et al. Magnetic resonance spectroscopy for glioma grading: relevance in the clinical setting. Neurosurg Rev. 2010;33(4):413-25.

16.    Opstad KS, Bell BA, Griffiths JR, Howe FA. An assessment of the usefulness of quantitative 1H MRS in clinical decision making in adult brain tumors. NMR Biomed. 2008;21(6):627-41.

17.    Poptani H, Gupta RK, Roy R, Pandey R, Jain VK, Chhabra DK. Characterization of intracranial mass lesions with in vivo proton MR spectroscopy. AJNR Am J Neuroradiol. 1995;16(8):1593-603.

18.    Callot V, Galanaud D, Le Fur Y, Confort-Gouny S, Ranjeva JP, Cozzone PJ. Characterization of high-grade gliomas using 1H and 31P magnetic resonance spectroscopy. J Magn Reson Imaging. 2004;20(1):15-20.

19.    Julia-Sape M, Majos C, Candiota AP, Lucas A, Arus C. Brain tumor classification using proton magnetic resonance spectroscopy: a comparison of feature selection and classification methods. Magn Reson Med. 2006;55(5):1056-65.

20.    Fayad LM, Salibi N, Wang X, Fowler K, Wahl RL, Bluemke DA. Quantitative 1H MR spectroscopy of the musculoskeletal system: state of the art. Radiographics. 2009;29(2):437-61.

21.    Roy B, Gupta RK, Maudsley AA, Awasthi R, Sheriff S, Gu M, et al. Utility of multiparametric 3-T MRI for glioma characterization. Neuroradiology. 2013;55(6):603-13.

22.    Taylor JS, Langston JW, Reddick WE, Kingsley PB, Ogg RJ, Pui CH. Clinical value of proton magnetic resonance spectroscopy for differentiating neoplastic from non-neoplastic brain lesions. Magn Reson Med. 1996;36(5):829-37.

23.    Tarulli E, Biagi L, Tosetti M, Mascalchi M. Pitfalls in magnetic resonance spectroscopy for brain tumor classification. J Neuroradiol. 2007;34(3):194-203.

24.    Fayed N, Modrego PJ, Morales H. Evidence of brain damage in chronic pain patients evaluated by magnetic resonance spectroscopy. NeuroReport. 2014;25(4):284-9.

25.    Patel S, Hatzoglou V, Omuro A, Peck KK, Young RJ. Multiparametric MRI for the differentiation of high-grade glioma from primary CNS lymphoma. AJNR Am J Neuroradiol. 2018;39(5):803-10.

26.    Wilson M, Andronesi O, Barker PB, Bartha R, Bizzi A, Bolan PJ, et al. Methodological consensus on clinical proton MRS of the brain: review and recommendations. Magn Reson Med. 2019;82(2):527-50.

27.    Verma A, Kumar I, Verma N, Aggarwal P, Ojha R. Magnetic resonance spectroscopy—revisiting the biochemical and molecular milieu of brain tumors. BJR Open. 2017;90(1073):20170595.