Application of Magnetic Resonance Spectroscopic Imaging (MRSI) to Produce Molecular Images of Mouse Brain Tumors

H-MRS is one of the most promising non-invasive techniques used to support radiology and MRI in the diagnosis of numerous diseases. Single voxel (SV) MRS helps in measuring the biochemical content of living tissues and offers metabolic data complementary to anatomical changes observed in MRI and radiological exams.

Moreover, magnetic resonance spectroscopic imaging (MRSI) offers metabolic data which can be processed and displayed as density maps of several metabolites from preclinical models or humans. For instance, it enables the differentiation of spatial metabolic and tumor grade heterogeneity within brain tumors, and the delimitation of regions having metabolic anomalies.

MRSI helps in studying large regions that surround the lesion and this data is quite useful for delineating the tumor’s border for neurosurgery purposes and for ascertaining the infiltration of cells prior to choosing a suitable radiotherapy protocol.

The detection of infiltration is also correlated to the concentration of infiltrating cells and to the relative amount of the infiltration biomarker utilized, for instance identification of mobile lipids in a normal brain parenchyma region would instantly indicate possible infiltration.

With continuous enhancement and refinement in IVRS methods, including better methods of data acquisition and processing and higher magnetic field strength magnets, spatial as well as temporal resolution have improved significantly. High magnetic fields offer better spectral resolution and higher sensitivity.

The main aim of this article is to describe the application of MRS to create molecular images of preclinical brain tumors, such as oligodendroglioma (ODG, grade II) and glioblastoma (GBM, grade V) utilizing their spectroscopic profile, and the efficacy of Perturbation Enhanced MRS (PE-MRSI) for tumor characterization.

Methodology

Ten female C57BL/6 mice, weighing approximately 20 to 22g, were accommodated at the Universitat Autonoma de Barcelona animal facility (Cerdanyola del Valles, Spain). The S100p-v-erbB/lnk4a-Arf(+/-) GEM (Genetically Engineered Mice) model was obtained from the MMHCC, NCI-Frederick, USA repository and accommodated at the same facility. Two GEM (ODG II) were then added to this analysis.

Next, C57BL/6 animals were vaccinated with GL261 glioma cells CO5 in 4pl of RPMI culture media by means of stereotactic intracranial injection into the striatum of the right hemisphere.

MRSI experiments were then carried out at 7 Tesla in a Bruker Biospec 70/30 USR fitted with a B-GA 12 gradient coil inserted into the traditional GA 20 S gradient coil set up. This spectrometer hardware offers gradient strengths of 400 mT/m with rise times of 80 ms and slew rates of 5,500 T/m/s. It has high shim currents delivering shim strengths up to 240 Hz/cm2/A.

High duty cycles are utilized to reduce B0 field drift as a result of temperature fluctuations that could be induced by high shim currents. A coil with 7.2 cm inner diameter was employed for excitation along with a dedicated mouse brain quadrature H surface coil, with a shape modified to the mouse head for improved B1 homogeneity and sensitivity in the brain area, for signal reception.  

IsoFlurane 0.5- 5% in O2 was used to anesthetize the animals, keeping their respiratory frequency between 60 and 80 breaths per minute during MR scans. A recirculating water-system was integrated in the animal bed and this was utilized to control the body temperature, calculated by a rectal probe and kept between 37°C and 38°C. Temperature and breathing were continuously monitored.  

For MRSI studies, six animals were utilized. Two reference high-resolution T2W images having different TE and 12 ms and 136 ms TE basal MRSI were obtained. Acquisition parameters for T2w sequences were as follows:

• Echo train length, 8
• Field of view (FOV), 17.6 x 17.6 mm
• Matrix (MTX), 256 x 256 (75 x 75 pm/pixel)
• Number of slices (MS)
• Number of averages (NA), 3
• Slice thickness (ST), 1 mm
• TR, 3000 ms: TE, 36 and 132 ms
• Total acquisition time (TAT), 3 min and 36 sec.

For tumor diagnosis, ODG afflicted-mice brains were obtained, fixed in 4% formaldehyde, and examined by histopathology (H&E staining).

For Ferturbation-Enhanced-MRSI (PE-MRSI) studies, four animals were utilized. Their body temperature was stabilized between 28.5°C and 29.5°C, 45 mm after inducing anesthesia. This is because moderate brain hypothermia (~30°C) is an important requirement in the GL261 tumor model to increase the pattern perturbation induced by hyperglycemia. MRSI sequence parameters were the same as described above.

At first, two reference MRSI scans were obtained, one with short echo time (12 ms TE), followed by another one at long echo time (36 ms TE). About 90 min after the induction of anesthesia, an i.p. bolus injection of 10 mL/g D-glucose 25% (w/v) in saline was used to induce acute hyperglycemia. Following this, five successive short TE MRS scans were obtained.

CSI Dashboard Tool Software

Bruker's CSI Dashboard Tool software (within Paravision v 5.0) was used to process MRS grid spectra, with an exponential line widening of 4 Hz. In the central spectrum of the grid, first-order phase correction was carried out and then applied to the rest.

Bruker’s 2dseq files were post-processed with the 3diCSI software to acquire text files containing all the data obtained with MRS sequences, including phase correction and apodization. These text files were processed with the software called the Dynamic MRSI Processing Module (DMPM), which runs some scripts in Matlab to normalize each spectrum to unit length (UL2) and changing the text files to color-coded maps that represent the relative intensities of the user-selected metabolites.  

Results and Discussion

All tumors could be clearly viewed and were distinct in the high-resolution T2w images utilized as references for MRSI acquisition, as illustrated by two representative examples in figure 2.

The shimming carried out with the FASTMAP process prior to MRSI acquisition delivered good quality spectra with a linewidth less than 20-22 Hz for tissue water in all cases. The MRSI spectra acquired displayed a good signal to noise ratio, considering a threshold of 7.8.

In the above figure, the mean spectra from an ODG (grade I) and GL261 G8M (grade IV) were obtained from the tumor voxels chosen from their particular MRSI grids (VOI) and compared. A visual analysis of the typical tumor spectral pattern shows for instance that the ratio choline/creatine is obviously higher in the GBM, when compared to the ODG-II. ML+Lactate at 1.3 ppm and mobile Lipids (ML) resonances at 0.9 ppm also reach high levels in both tumors.

The variations identified at typical spectra level in both cases can be assessed for heterogeneity in the metabolic profile of both tumors, producing color-coded maps for the metabolites of interest, as shown in figure 3. These maps denote the normalized intensity of one metabolite chosen from the spectral pattern, for instance Lactate or ML+Lactate.

In all the voxels contained in the VOI, ML+ Lactate signals were high as anticipated in the tumor with higher heterogeneity in the ODG. In the case of Lactate, the highest signal intensity was also identified in the tumor in both the cases. This kind of variations may be of interest for additional study of tumor spectral pattern discrimination.

Figure 4 displays the different accumulation of glucose in three parts of the MRSI VOI at four time-points post injection. The peritumoral/normal brain parenchyma remains unaffected; however, glucose+taurine signal intensity clearly varies with respect to time.

Conclusion

In figure 4, the variations shown at a spectral level can be transformed to differences in color coded maps acquired from these MRSI sequences. In Figure 5, the color coded maps correlate to glucose signal increase in intensity at four different time-points after injecting the four animals with glucose. This validates that glucose accumulation is obviously higher within the tumors.

Acknowledgement

Application of magnetic resonance spectroscopic imaging (MRSI) to produce molecular images of mouse brain tumors - Produced from articles authored by Delgado-Goni T, Simoes RV, Lope-Piedrafita S, Arus C

Sources

1. Kwock L, Smith JK, Castillo M, Ewend MG, Collichio F, Morris DE, Bouldin TW, Cush S. Clinical role of proton magnetic resonance spectroscopy in oncology: brain, breast, and prostate cancer. Lancet Oncol 2006;7(10):859-868.
2. Fellows GA, Wright AJ, Sibtain NA, Rich P, Opstad KS, McIntyre DJ, Bell BA, Griffiths JR, Howe FA. Combined use of neuroradiology and 1H-MR spectroscopy may provide an intervention limiting diagnosis of glioblastoma multiforme. J Magn Reson Imaging; 32(5):1038-1044.
3. Majos C, Alonso J, Aguilera C, Serrallonga M, Perez-Martin J, Acebes JJ, Arus C, Gili J. Proton magnetic resonance spectroscopy ((1)H MRS) of human brain tumours: assessment of differences between tumour types and its applicability in brain tumour categorization. Eur Radiol 2003;13(3):582-591.
4. Sibtain NA, Howe FA, Saunders DE. The clinical value of proton magnetic resonance spectroscopy in adult brain tumours. Clin Radiol 2007;62(2):109-119.
5. Le HC, Lupu M, Kotedia K, Rosen N, Solit D, Koutcher JA. Proton MRS detects metabolic changes in hormone sensitive and resistanthuman prostate cancer models CWR22 and CWR22r. Magn Reson Med 2009;62(5):1112-1119.
6. Simonetti AW, Melssen WJ, van der Graaf M, Postma GJ, Heerschap A, Buydens LM. A chemometric approach for brain tumor classification using magnetic resonance imaging and spectroscopy. Anal Chem 2003;75(20):5352-5361.
7. Chuang CF, Chan AA, Larson D, Verhey LJ, McDermott M, Nelson SJ, Pirzkall A. Potential value of MR spectroscopic imaging for the radiosurgical management of patients with recurrent high-grade gliomas. Technol Cancer Res Treat 2007;6(5):375-382.
8. Martinez-Bisbal MC, Celda B. Proton magnetic resonance spectroscopy imaging in the study of human brain cancer. Q J Nucl Med Mol Imaging 2009;53(6):618-630.
9. Grutter R. Automatic, localized in Vivo adjustment of all first and second-order shim coils. Magnetic Resonance in Medicine 1993;29(6):804-811.
10. Kanayama S, Kuhara S, Satoh K. In vivo rapid magnetic field measurement and shimming using single scan differential phase mapping. Magn Reson Med 1996;36(4):637-642.
11. Simoes RV, Garcia-Martin ML, Cerdan S, Arus C. Perturbation ofmouse glioma MRS pattern by induced acute hyperglycemia. NMR Biomed 2008;21(3):251-264.
12. Simoes RV, Delgado-Goni T, Lope-Piedrafita S, Arus C. 1H-MRSI pattern perturbation in a mouse glioma: the effects of acute hyperglycemia and moderate hypothermia. NMR Biomed 2010;23(1):23-33.
13. https://www.uab.cat/
14. Ortega-Martorell S, Olier I, Julia-Sape M, Arus C. SpectraClassifier 1.0: a user friendly, automated MRS-based classifier-development system. BMC Bioinformatics 2010;11:106.
15. Simoes RV, Ortega-Martorell S, Delgado-Goni T, Le Fur J, Pumarola M, Candiota AP, Martin J, Stoyanova R, Cozzone PJ, Julia- Sape M, Arus C. Improving the classification of brain tumors in mice with perturbation enhanced (PE)-MRSI. Integr Biol (Camb). 2012; 4(2):183-191.
16. Ziegler A, von Kienlin M, Decorps M, Remy C. High glycolytic activity in rat glioma demonstrated in vivo by correlation peak 1H magnetic resonance imaging. Cancer Res 2001;61(14):5595-5600.

About Bruker BioSpin - NMR, EPR and Imaging

Bruker BioSpin offers the world's most comprehensive range of NMR and EPR spectroscopy and preclinical research tools. Bruker BioSpin develops, manufactures and supplies technology to research establishments, commercial enterprises and multi-national corporations across countless industries and fields of expertise.

Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.