Original article
Measurement of glycine in a brain and brain tumors by means of 1H MRS

Introduction

Proton magnetic resonance spectroscopy (1H MRS) is mainly used in diagnostic of brain and prostate diseases [6,13,26,35], but in recent years there have been attempts of using this method in diagnostic of breast [2] and other organs diseases.
The application of 1H MRS has been used for brain tumor differentiation from other diseases, determination of malignancy or histopathologic types and metabolite profile of neoplastic tissue [3,9,16,18,19,30,31,34,39].
Glycine is an amino acid synthesized from serine in the synaptic vesicles which acts as a neuromodulator or antioxidant. It is distributed throughout different parts of central nervous system [11,23,39]. Glycine has two methylene-protons (–CH2) gropus that co- resonate at 3.55 ppm as a single peak [7,8,11,14,30,37]. It should be noted that at 3.52-3.57 ppm in short echo time Gly overlaps with myo-inositol (mI) [12,30,37]. On the other hand in long echo time (TE 100-170 ms) mI has short T2 relaxation time so its contribution to magnetic resonance spectrum is reduced considerably. Therefore in long echo time peak at 3.55-3.57 ppm is assigned to Gly rather than to mI [11,14,41].
Gly concentration in the normal brain range from 0.4 to 1.0 mmol/kgww [8,30]. The concentration of Gly is elevated in patients with hyperglycinemia and tumors such as glioblastoma multiforme, medulloblastoma, ependymoma and central neurocytoma [8,14,37,41]. It has been also found in other pathologies of a central nervous system like stroke, adrenoleukodystrophia or hamartoma [34]. On the contrary it has been recomended that Gly presence is a typical feature of CNC [5,16,14,15,40,41], though its absence do not exclude CNC [5,15,18,19].
CNC consist of uniform round cells with neuronal differentiation comparable with oligodendrocytes [24,27,39,41]. Calcifications, necrosis and tiny cysts may occur in both of the tumors [14,24,26,32,39]. Because of great morphological similarity to oligodendroglioma or ependymoma it makes difficult to differentiate these tumors and extraventicular neurocytoma in imaging and histopathologic examinations [24,27,32,39,40]. Since the presence of glycine has been described as pathognomonic for CNC [5,14, 15,19,41]. 1H MRS is considered as helpful, non-invasive method providing additional information which is useful to make CNC recognition, especially in extraventricular location.
The aim of this study was to evaluate the peak at 3.55 ppm in a long echo time recognized as glycine in central neurocytomas and WHO grade II gliomas by means of 1H MRS.

Material and methods

Material

21 consecutive patients with primary brain tumor were included in the analysis. All patients included in the analysis had MR examination at Radiodiagnostics Department at Comprehensive Cancer Centre Maria Sklodowska-Curie Memorial Institute Branch Gliwice between January 2006 and June 2008. WHO grade II glioma group comprised of 19/21 patients (6 females and 13 males, average age 38 ± 9 years). The characterisation of the group is presented in Table 1. Central neurocytoma group comprised of 2/21 patients (female and male, average age 44). The tumor was localized intraventricular in both cases. All of the patients underwent surgery after MR examination and resected tumor was histopathologically diagnosed.
61 healthy volunteers (39 females and 22 males, average age 31 ± 11 years) were consecutively examined at Radiodiagnostics Department at Comprehensive Cancer Centre Maria Sklodowska-Curie Memorial Institute Branch Gliwice or Department of Radiology at Oncology Centre in Bydgoszcz.
All of the MR examinations were performed with the consent of the Local Ethics Committee.

Methods

MR examination of the brain was performed with the 1.5 T scanner. A commercial head coil was used for imaging and spectroscopy. One patient with CNC had a brain examination with use of 3.0 T scanner.
Conventional MR imaging consisted of T1-weighted, T2-weighted images, FLAIR (Fluid-Attenuated Inver­sion Recovery), PD (Proton Density) and T1-weighted images before and after CE (contrast enhancement).
Proton MR spectroscopy(1H MRS)
1.5 T : 3D CSI PRESS: long TE (TR/TE 1500/135 ms), NSA 1, acquisition time 7.47 min and mean voxel volume 1.3 mL.
3.0 T : 3D CSI PRESS: long TE (TR/TE 2000/144 ms) NSA 1, acquisition time 7.13 min and mean voxel volume 2.7 mL. The T1 relaxation time increase at higher fields, leading to increased signal saturation for a given TR. To maintain the same level of T1 relaxation, we used higher TR for 3.0 T scanner.
In tumors without contrast enhancement, voxels were placed in a central, solid part of area of high signal intesity on T2-weighted images, which corresponds to mass effect. In tumors with contrast enhancement, voxels were placed in a central, solid part of contrast enhancing area.
787 voxels were included into the analysis in the group of WHO grade II glioma and 74 voxels in the group of central neurocytoma.
We took into account such metabolites as: choline (Cho), creatine (Cr), N-acetylaspartate (NAA) and a sin­gle peak at 3.55 ppm in long TE defined as glicyne (Gly). Ratios of Gly/Cr, Gly/Cho and Gly/NAA were calculated for WHO grade II glioma and central neurocytoma group separately. Additionally such metabolites ratios as Cho/Cr, NAA/Cr and Cho/NAA were calculated. Obtained data were analyzed with the LCModel version 6.1-4F. The LCmodel algorithm analyzes the in vivo spectrum as a linear combination of individual in vitro metabolite spectra that constitute a basis set. Each metabolite’s signal intesity value is assigned to 8 ml voxel size, so it is independent from raw voxel size obtained during in vivo measurements. The estimated standard deviation (%SD) below 20% has been used as a rough criterion for estimates of acceptable reliability. Therefore the presence of the peak at 3.55 ppm was accepted when its SD was less than 20%.
All of the 61 healthy volunteers were examined SVS PRESS, long TE (TR/TE 1500/135 ms, NSA 192, and the average voxel volume 5.0 mL. In 41 out of 61 volunteers an additional 1H MRS was performed with 2D CSI, long TE (TR/TE 1500/135 ms, NSA 4 and the average voxel volume 1.7 mL). Only diagnostic spectra were analysed.
Analysis contained:
– 156 voxels obtained from SVS localized in hipocampus 45/156, in cerebellum 74/156, in frontal and tem­poral lobe 37/156;
– 2456 voxels obtained from 2D CSI localized in thalamus 1048/2456 and in frontal and parietal lobes 1408/2456.

Statistical analysis

Statistical calculations and analyses were performed with Statistical PL software version 7.1 by StatSoft, Inc. Estimation of metabolites ratios between analysed groups was performed with Kolmogorov-Smirnov test for independent experiments. Statistically significant p-levels were assumed as < 0.05.

Results

In 12/19 (63%) WHO grade II gliomas Gly was detected whereas in 7/19 (37%) was not. In these 12 WHO grade II gliomas 646 voxels were analysed, but Gly was observed only in a part of the voxels – 169, which comprise approximately 26% of the tumor volume. Figures 1 and 2 shows the spectrum obtained from 3D CSI in astrocytoma fibrillare with and without distinguished Gly respectively. In none of oligodenrogliomas Gly was observed (Table 2).
In 2 patients with central neurocytoma 74 voxels from tumors were analysed. Gly was distinguished in both tumors in 32 out of 74 voxels which comprise approximately 43% of the tumor volume. Figure 3 shows the spectrum obtained from 3D CSI in central neurocytoma with distinguished Gly.
Mean value of Gly/Cr, Gly/Cho and Gly/NAA ratios were calculated from all voxels in which Gly was distinguished separately for each group (Tables 3A and 3B, Figs. 4-6).
The ratio of Gly/Cr in central neurocytomas was statistically significantly higher than in WHO grade II gliomas (meanCNC 0.62 ± 0.18 vs. meanWHO II 0.37 ± 0.10; p < 0.001) but the ratio of Gly/Cho was statistically significantly lower (meanCNC 0.18 ± 0.04 vs. meanWHO II 0.24 ± 0.07; p < 0.001). There was no difference between analysed groups in terms of Gly/NAA ratio (meanCNC 0.36 ± 0.09 vs. meanWHO II 0.36 ± 0.14; p = NS). For better understanding of the results additional analysis was conducted and Cho/Cr, NAA/Cr and Cho/NAA ratios were calculated in both analysed groups. It was proved that mean Cho/Cr, NAA/Cr and Cho/NAA ratios were statistically significantly higher in central neurocytomas than in WHO grade II gliomas. Results were presented in Tables 4A and 4B. In the group of 61 volunteers 2612 spectra were analysed. Glycine was found in 7 out of 61 (11.5%) people but only in 8 out of 2612 analysed voxels which comprise 0.3%.

Discussion

Glycine is an amino acid synthesized from serine in the synaptic vesicles which acts as a neuromodulator or antioxidant. It is distributed throughout different parts of central nervous system [11,13,39]. Its concentration in the normal brain range from 0.4 to 1.0 mM [8,30]. Conventional 1H MRS might not be enough to distinguish Gly peak in normal brain. Some authors who distinguished Gly by means of 1H MRS used scanners with magnetic field higher than 3.0 T [7], echo-time-averaged [30] or triple-refocusing filtration [4].
In 1H MRS peak at 3.52-3.57 ppm is assigned to mI and/or Gly [4,8,9,11,14,23]. In short echo time mI resonance as multiplet and overlaps with Gly’s methylene-protons (–CH2) groups that co-resonate at 3.55 ppm as a single peak [7,8,11,14,30,37]. It is impossible to separate these metabolites by means of 1.5 T and 3.0 T 1H MRS in short echo time. Gambarot et al. proved in 7.0 T 1H MRS with TE 30 ms single peak of Gly reso­nating between mI multiplet at 3.55 ppm [7]. mI has short and shorter than Gly T2 relaxation time so its contribution to magnetic resonance spectrum is reduced considerably, in longer echo time (TE 100-170 ms). Therefore in long echo time peak at 3.55-3.57 ppm is assigned to Gly rather than to mI [4,7,8,11,14,30,37].
Gly is considered as characteristic feature for CNC, but it might be detected in a low and high grade gliomas [5,11,14-17,19,20,36,41]. Higher level of Gly was detected in high grade gliomas than in low ones, which was claimed by Hattingten et al. [11], by Lehnhardt’a et al. as well as Tugnoli et al. during in vitro studies [20,36]. In our material Gly in long TE was observed in 12 out of 19 (63%) WHO grade II gliomas. Gly was present in astrocytomas – 9/12 (75%) as well as in oligoastrocytomas – 3/12 (25%). We didn’t find Gly in none of 2 oligodendrogliomas. It might be cau­sed by small number of these specimen in our mate­rial. Other authors also didn’t find Gly in oligodendrogliomas [36,17].
Of note is the fact that the peak of Gly was detected only in a part of the tumor volume (approximately in 26% of the ananlysed voxels in WHO grade II gliomas). To our konowledge a report about heterogeneity of WHO II gliomas considering glycine presen­ce/absence has not been yet published.
Central neurocytoma is a rare tumor of central nervous system mostly occurring in young adults [5,21,24]. Approximately 70% of CNC is diagnosed in patients between the ages of 20 and 40 years [28]. It makes up 0.25-0.5% of intraaxial brain tumors [39,41] and corresponds to WHO grade II [24]. It was described for the first time in 1982 by Hassoun [10]. CNC is typically located in the lateral ventricles and/or the third ventricle, in the foramen of Monro and its characteristic feature is the attachment to the septum pellucidum [5,22,24,27,28,39,42]. Extraventricular location is rather rare [24,40]. Possibility of diffe­rentiation would be particularly precious for rare – extraventricular localisation of infiltration [1,29,32,33]. As previously mentioned, the Gly peak at 3.55 ppm might be detected in different primary brain tumors, but it is more characteristic feature for CNC [5,14-16,19,41]. On the other hand lack of distinct glycine in 1H MRS doesn’t exclude central neurocytoma type [15,18]. Using SVS 1H MRS Kocaoglu et al. found Gly in only 1 out of 7(14.3%) CNC and Kanamori et al. in 1 out of 3 (33%) CNC [15,18]. The reason for the absence of Gly peak at 3.55 ppm in 1H MRS examination at patients with CNC, can be the fact that glycine can exist only in a part of tumor volume. Our mate­rial consist two patients with CNC in intraventricular localisation. Gly peak appeared in both tumors but only in part of the tumor volume (approximately in 43% of the ananlysed voxels). Because of this SVS might not be enough to prove presence or absence of Gly. Krishnamoorthy et al. distinguished peak at 3.55 ppm in long echo time (TE 135 ms) in all three analysed central neurocytomas. For these tumors MR spectroscopy was performed with a multivoxel method but there isn’t any information whether Gly was observed in all voxels [19].
The analysis of the Gly/Cr, Gly/Cho, Gly/NAA ratio in central neurocytomas as well as in WHO grade II gliomas was also proposed by other authors but comparison between those two groups was not conducted [16,20,39]. Kim et al. reported higher Gly/Cr, Gly/Cho and Gly/NAA ratios, Ueda et al. higher Gly/Cr ratio in CNC than our results. It might be caused by different 1H MRS methods: Kim et al., Ueda et al. used single voxel spectroscopy whereas we used multivoxel one. Lehnhardt et al. presented lower Gly/Cr ratio in low grade gliomas than we obtained. But Lehnhardt et al. carried in vitro study and we in vivo one. We obtained higher mean Gly/Cr ratio in the central neurocytoma group than in the WHO grade II glioma one. Furthermore Gly/Cho ratio was lower in the central neurocytomas than in the WHO grade II gliomas. It might be explained by higher Gly and Cho peak in the central neurocytomas in comparsion with WHO grade II gliomas. In the additionally performed analysis we obtained higher Cho/Cr ratio in central neurocytomas group than WHO grade II one. We also observed equal Gly/NAA ratio in both compared groups. These results might be due to higher NAA peak in central neurocytomas group than WHO grade one as it was proved by higher NAA/Cr and Cho/NAA ratios in the central neurocytoma group.
In our analysis we estimated the presence of Gly at 3.55 ppm in 1H MRS in long echo time only in only in 8 out of 2616 voxels which comprise 0.3% and it seems rather coincidental result, which should be avoided.

Conclusions

Glycine is found in WHO II grade gliomas as well as in central neurocytomas, but only in a part of a tumor volume. It is necessary to perform 1H MRS of the whole tumor volume to confirm/exclude the presence of glycine. Glycine in a normal brain can not be identified by means of conventional 1H MRS performed by means of 1.5 T or 3.0 T scanners.

References

 1. Andrychowski J, Taraszewska A, Czernicki Z, Jurkiewicz J, Netczuk T, Dąbrowski P. Ten years observation and treatment of multifocal pilocytic astrocytoma. Folia Neuropathol 2009; 47: 362-370.  
2. Bolan PJ, Meisamy S, Baker EH, Lin J, Emory T, Nelson M, Everson LI, Yee D, Garwood M. In vivo quantification of choline compounds in the breast with 1H MR spectroscopy. Magn Reson Med 2003; 50: 1134-1143.  
3. Castillo M, Smith KJ, Kwock L. Correlation of Myo-inositol levels and grading of cerebral astrocytomas. AJNR Am J Neuroradiol 2000; 21: 1645-1649.  
4. Cho C, Bhardwaj PP, Seres P, Kalra S, Tibbo PG, Coupland NJ. Measurement of glycine in human brain by triple refocusing 1H-MRS in vivo at 3T. Magn Reson Med 2009; 62: 1305-1310.  
5. Chuang MT, Lin WC, Tsai HY, Liu GC, Hu SW, Chiang IC. 3-T proton magnetic resonance spectroscopy of central neurocytoma: 3 case reports and review of the literature. J Comput Assist Tomogr 2005; 29: 683-688.  
6. Czernicki T, Szeszkowski W, Marchel A, Gołebiowski M. Spectral changes in postoperative MRS in high-grade gliomas and their effect on patient prognosis. Folia Neuropathol 2009; 47: 43-49.  
7. Gambarota G, Mekle R, Xin L, Hergt M, van der Zwaag W, Kru­ger G, Gruetter R. In vivo measurement of glycine with short echo – time 1H MRS in human brain at 7T. MAGMA 2009; 22: 1-4.  
8. Govindaraju V, Young K, Maudsley AA. Proton NMR chemical shift and coupling constants for brain metabolites. NMR Biomed 2000; 13: 129-153.  
9. Gutowski NJ, Gómez-Ansón B, Torpey N, Revesz T, Miller DH, Rudge P. Oligodendroglial gliomatosis cerebri: 1H-MRS suggests elevated glycine/inositol levels. Neuroradiology 1999; 41: 650-653.
10. Hassoun J, Gambarelli D, Grisoli F, Pellet W, Salmon G, Pelis­sier JF, Toga M. Central neurocytoma: an electron-microscopic study of two cases. Acta Neuropathol 1982; 56: 151-156.
11. Hattingen E, Lanfermann H, Quick J, Franz K, Zanella FE, Pilatus U. 1H MR spectroscopic imaging with short and long echo time to discriminate glycine in glial tumours. MAGMA 2009; 22: 33-41.
12. Hattingen E, Raab P, Franz K, Zanella FE, Lanfermann H, Pilatus U. Myo-Inositol: a marker of reactive astrogliosis in glial tumors? NMR Biomed 2008; 21: 233-241.
13. Jacobs MA, Ouwerkerk R, Petrowski K, Macura KJ. Diffusion-weighted imaging with apparent diffusion coefficient mapping and spectroscopy in prostate cancer. Top Magn Reson Imaging 2008; 19: 261-272.
14. Jayasunder R, Shsh T, Vaishya S, Singh VP, Sarkar C. In vivo and in vitro spectroscopic profile of central neurocytomas. J Magn Reson Imaging 2003; 17: 256-260.
15. Kanamori M, Kumabe T, Shimizu H, Yoshimoto T. (201) TI-SPECT, (1)H-MRS, and MIB-1 labeling index of central neurocytomas: three case reports. Acta Neurochir (Wien) 2002; 144: 157-163; discussion 163.
16. Kim DG, Choe WJ, Chang KH, Song IC, Han MH, Jung HW, Cho BK. In Vivo Proton Magnetic Resonance Spectroscopy of Central Neurocytomas. Clinical Studies. Neurosurgery 2000; 46: 329-333; discussion 333-334.
17. Kinishita Y, Yokota A. Absolute concentrations of metabolites in human brain tumors using in vitro proton magnetic resonance spectroscopy. NMR Biomed 1997; 10: 2-12.
18. Kocaoglu M, Ors F, Bulakbasi N, Onguru O, Ulutin C, Secer HI. Central neurocytoma: proton MR spectroscopy and diffusion weighted MR imaging findings. Magn Reson Imaging 2009; 27: 434-440.
19. Krishnamoorthy T, Radhakrishnan VV, Thomas B, Jeyadevan ER, Menon G, Nair S. Alanine peak in central neurocytomas on proton MR spectroscopy. Neuroradiology 2007; 49: 551-554.
20. Lehnhardt FG, Bock C, Röhn G, Ernestus RI, Hoehn M. Meta­bolic differences between primary and recurrent human brain tumors: a 1HNMR spectroscopic investigation. NMR Biomed 2005; 18: 371-382.
21. Liberski P, Mossakowski MJ. Nowotwory neuroepithelialne, nowotwory neuronalne i mieszane neuronalno-glejowe. In: Mossakowski MJ, Liberski PP. Guzy układu nerwowego. Zakład Narodowy im. Ossolińskich, Wydawnictwo Polskiej Akademii Nauk, 1997; pp. 138-141.
22. Liebert W, Szymaś J, Majewski T, Paprzycki W. Central neurocytoma of the right parietal and occipital lobe. Case report. Neurol Neurochir Pol 1998; 32: 191-199.
23. London~o A, Castillo M, Armao D, Kwock L, Suzuki K. Unusual MR Spectroscopic imaging pattern of an astrocytoma: lack of elevated choline and high myo-inositol and glycine levels. AJNR Am J Neuroradiol 2003; 24: 942-945.
24. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, Scheithauer BW, Kleihues P. The 2007 WHO Classification of Tumors of the Central Nervous System. Acta Neuropathol 2007; 114: 97-109.
25. Majós C, Julia`-Sapé M, Alonso J, Serrallonga M, Aguilera C, Acebes JJ, Arús C, Gili J. Brain tumor classification by proton MR spectroscopy: comparison of diagnostic accuracy at short and long TE. AJNR Am J Neuroradiol 2004; 25: 1696-1704.
26. Matulewicz L, Sokół M, Wydmański J, Hawrylewicz L. Could lipid CH2/CH3 analysis by in vivo 1H MRS help in differentiation of tumor recurrence and post-radiation effects? Folia Neuropathol 2006; 44: 116-124.
27. Mena H, Morrison AL, Jones RV, Gyure KA. Central neurocytomas express photoreceptor differentiation. Cancer 2001; 91: 136-143.
28. Möller-Hartmann W, Krings T, Brunn A, Korinth M, Thron A. Proton magnetic resonance spectroscopy of neurocytoma outside the ventricular region – case report and review of the literature. Neuroradiology 2002; 44: 230-234.
29. Nowak S, Zukiel R, Barciszewska AM, Barciszewski J. The diagnosis and therapy of brain tumours. Folia Neuropathol 2005; 43: 193-196.
30. Prescot AP, de B Frederick B, Wang L, Brown J, Jensen JE, Kaufman MJ, Renshaw PF. In vivo detection of brain glycine with echo-time-averaged 1H magnetic resonance spectroscopy at 4.0 T. Magn Reson Med 2006; 55: 681-686.
31. Rijpkema M, Schuuring J, van der Meulen Y, van der Graaf M, Bernsen H, Boerman R, van der Kogel A, Heerschap A. Characterization of oligodendriogliomas using short echo time 1H MR spectroscopy imaging. NMR Biomed 2003; 16: 12-18.
32. Schild SE, Scheithauer BW, Haddock MG, Schiff D, Burger PC, Wong WW, Lyons MK. Central neurocytomas. Cancer 1997; 79: 790-795.
33. Schmidt MH, Gottfried ON, von Koch CS, Chang SM, McDermott MW. Central Neurocytoma: a review. J Neurooncol 2004; 66: 377-384.
34. Sener RN. The glycine peak in brain diseases. Comput Med Imaging Graph 2003; 27: 297-305.
35. Swindle P, McCredie S, Russell P, Himmelreich U, Khadra M, Lean C, Mountford C. Pathologic characterization of human prostate tissue with proton MR spectroscopy. Radiology 2003; 228: 144-151.
36. Tugnoli V, Tosi MR, Barbarella G, Bertoluzza A, Ricci R, Trevisan C. In vivo 1H MRS and in vitro multinuclear MR study of human brain tumors. Anticancer Res 1996; 16: 2891-2899.
37. Walecki. J. Postępy Neuroradiologii. Wyd. 1. Oświata UN-O, War­szawa 2007, pp. 44-300.
38. Walecki J, Ziemiński A. Rezonans magnetyczny i tomografia komputerowa w praktyce klinicznej. Springer PWN, Warszawa 1998.
39. Ueda F, Suzuki M, Matsui O, Uchiyama N. Automated MR spectroscopy of intra- and extraventricular neurocytomas. Magn Reson Med Sci 2007; 6: 75-81.
40. Yang GF, Wu SY, Zhang LJ, Lu GM, Tian W, Shah K. Imaging findings of extraventricular neurocytoma: report of 3 cases and review of the literature. AJNR Am J Neuroradiol 2009; 30:581-585.
41. Yeh IB, Xu M, Ng WH, Ye J, Yang D, Lim CC. Central neurocytoma: typical magnetic resonance spectroscopy findings and atypical ventricular dissemination. Magn Reson Imaging 2008; 26: 59-64.
42. Zhang B, Luo B, Zhang Z, Sun G, Wen J. Central neurocytoma: a clinicopathological and neuroradiological study. Neuroradiology 2004; 46: 888-895.