Pia C. Sundgren, MD, PhD

MR spectroscopy (MRS) techniques suitable for in vivo use developed parallel to the introduction of MRI into routine clinical practice. The possibility of combining the advantages of MRI in delineating anatomical and pathological information with the ability of MRS to provide insight into the biochemical changes underlying different pathological conditions has received increasing attention from clinicians. Today, in vivo 1 H MRS has become an important applied tool in the evaluation of central nervous system (CNS) diseases and in the workup of brain lesions in particular. The use of MRS to evaluate, classify, and grade brain tumors, differentiate cystic lesions in the brain, monitor treatment, and differentiate recurrent neoplasm from radiation injury will be discussed in this short overview of MRS applications in oncology.

Suresh K. Mukherji, MD

MRS TECHNIQUES

Single voxel spectroscopy and 2D-chemical shift imaging . MRS can be performed using different techniques depending on clinical question and localization of the lesion. The spatial resolution of in vivo 1 H MRS is limited. Spectra can be acquired using single voxel spectroscopy (SVS) or multiple voxels (multivoxel spectroscopy) also referred to as chemical shift imaging (CSI). Both techniques are commonly used in evaluation of brain lesions. There are several limitations  of MRS regardless of the technique used. MRS cannot be performed in or adjacent to bone, air, large vessels, and hemorrhagic lesions. Susceptibility artifacts from metal and shunts may obscure the spectra.

The decision on which type of sequence – SVS versus 2D-CSI – and which parameters – ie, echo time (TE) and repetition time (TR) – to use depends on the location of the lesion and the clinical question. To best evaluate a brain lesion with spectroscopy, it is important to be able to evaluate different portions of the lesion: the cystic/necrotic center and the contrast-enhancing rim, or solid contrast-enhancing mass and the surrounding normal-appearing brain tissue. Therefore, multivoxel technique – 2D-CSI – can be advantageous, since the volume of interest can be large and multiple voxels provide a greater coverage of the lesion and surrounding brain tissue.

Recent studies have demonstrated the feasibility of performing 2D-CSI in the posterior fossa despite it being a smaller compartment with adjacent air and bone structures. 1,2

When SVS technique is used, multiple examinations are required to evaluate the lesions and compare the metabolic ratios to normal brain tissue. This may often be in the contralateral hemisphere to ensure the brain is not affected by the lesion as could occur if tissue directly adjacent to the brain was sampled. This direct spread may not be visible on routine anatomic imaging. A second examination has to be performed in order to accomplish this. One advantage of 2D-CSI is that the contralateral brain can be evaluated at a higher spectral resolution in a single study.

Metabolites in the brain . Metabolites in the central nervous system (CNS) that are commonly evaluated in brain lesions include N-acetylaspartate (NAA), choline (Cho), creatine (Cr), and the presence of lactate and lipids.

NAA serves as a measure of normal healthy neuronal tissue. Cho has been implicated as a marker of cellularity and cell turnover and, therefore, may be used to infer a neoplastic process. Cr is a relatively stable metabolite and is used by most investigators as an internal control for quantifying other metabolites (metabolite ratios often include creatine as the denominator). 3 Lactate and lipids are normally not present in the brain. These metabolites are shown to increase in anaerobic metabolism and are present in some tumors such as high-grade astrocytomas and meningiomas. 4

MRS APPLICATIONS

Brain neoplasms . MR spectroscopy has proven useful in the evaluation of many CNS neoplasms. 5 The clinical impact of spectroscopy on the assessment of contrast-enhancing lesions that have indeterminate imaging characteristics, distinguishing neoplasm from non-neoplastic CNS tissue, has been reported. 6,7 Spectral changes often shown in brain tumors are an increase in  Cho, a decrease or absence of NAA, and the presence of lactate and lipids. It is not only the individual peaks and spectral changes that are important, but their relation toward each other as well. Such relationships can be evaluated by calculating ratios. Cho/Cr, NAA/Cr, and Cho/NAA ratios have commonly been used by many investigators to distinguish neoplasm from non-neoplastic tissue in the central nervous system. 8 It has been shown that the results of MR spectroscopy of radiation necrosis9 and intracranial neoplasms 10 correlate with histopathology from biopsy specimens, when measuring choline, creatine, and NAA metabolites.

a) 2D-CSI VOI was placed over the contrast- enhancing lesion and surrounding brain tissue.

Figure 1 a-b. Axial T1-weighted images after contrast administration demonstrate areas of irregular contrast enhancement on 36-month follow-up MRI at the site of prior resection in this 7-year-old boy previously resected for an ependymoma in the left frontal lobe. The patient underwent additional resection, and recurrent ependymoma was confirmed by histopathology.

b) The MRS demonstrated pathological spectra with increased Cho/NAA and Cho/Cr ratios, respectively, and a decrease in the NAA/Cr ratio consistent with tumor recurrence.

Guide surgical brain biopsy . MR spectroscopy will not be able to replace brain biopsy to get histological diagnosis, but preoperative MR spectroscopy of primary brain neoplasms is increasingly performed to guide tumor biopsy and surgical planning. 11,12 MRS has the potential to identify areas with higher Cho/Cr ratios from tissue with normal ratios. These areas may be indistinguishable on routine anatomic images. The technique is particularly useful in patients who have already been treated. In addition, the possibility of distinguishing abnormal normal-appearing brain tissue from completely normal brain tissue and also, of course, neoplastic tissue from non-neoplastic tissue will help in surgical planning.

Grade brain tumors . MRS has also been used in attempts to grade tumors prior to surgery. It is known that conventional MRI seems to have limited accuracy in defining tumor boundaries or differentiating mild and moderate tumor infiltration from normal brain tissue. It is also known that gadolinium contrast enhancement does not always correlate with the highest cellularity, and infiltrative tumor may extend far beyond the contrast-enhancing area. In this context, alternatives for tumor grading and evaluation of tumor extension, such as MRS, become interesting tools in the workup. Preoperative proton-MRS has been shown to correlate with histology and morphology of surgical resection specimens in gliomas, with increased Cho content correlating directly with increased tumor grade. 11,13-15 In many cases, however, the published findings are contradictory, and it has been pointed out that the correlation between the spectral patterns of individual brain tumors and the histological diagnosis is often imperfect. 16-20 Most of these reports have been using SVS technique. Because the SVS technique with volume of interest (VOI) often between 5 and 30 cm 3 is the most commonly used in brain tumor examinations, some of these variations might be due to partial-volume effects. Because of the large size of the voxel, spectral evaluation of small lesions will include uninvolved brain adjacent to the lesion. This results in partial volume changes that may affect the integrity of the resultant spectra. Due to the size of the lesion, it may not be possible to place the VOI completely over mass, especially if the lesion is close to bone. In such a situation, the VOI has to be placed so that it covers parts of the lesion and parts of the adjacent brain tissue, which will introduce partial-volume changes into the spectra. In the evaluation of a heterogeneous lesion, with both solid contrast-enhancing areas and necrotic/cystic centers, it is advantageous to obtain spectra from the solid contrast-enhancing portion that is the more active malign part of the lesion. The large voxel size used in SVS precludes limiting the spectral sampling to just the enhancing solid areas of the mass. These partial-volume effects associated with SVS again emphasize the need for proper MRS technique to obtain the most accurate spectral changes. SVS certainly is valuable, especially to evaluate the content of a cystic brain lesion in order to separate cystic primary and secondary neoplasm from brain abscess. The opinion among those who routinely perform spectroscopy is that a combination of both techniques is required in the workup of a suspected brain neoplasm.

The main advantages of 2D-CSI is the possibility to manually, or automatically, place multiple 1 x 1 x 1 cm voxels over smaller regions of interest and obtain individual spectra that have a lower chance of being affected by partial averaging changes.

Cystic brain lesions and abscess . Cystic masses may in some cases present a diagnostic dilemma. The ability to differentiate a primary cystic neoplasm from a cystic metastasis or even an abscess can be difficult based on conventional MRI alone. Other clinically interesting diseases that may present as cystic masses are lymphoma, toxoplasmosis, and HIV lesions. All these lesions may present as well-circumscribed cystic lesions with peripheral contrast ring enhancement. The possibility of differentiating between anaerobic and aerobic abscesses may be useful in facilitating prompt and appropriate treatment of patients with brain abscesses. In this clinical setting, MRS has been shown to be beneficial. The spectral changes seen in an untreated brain abscess are significantly different from those found in both primary and secondary brain lesions. In an acute untreated bacterial abscess, the specific changes in spectra include absence of NAA and resonances representing metabolic end products arising from microorganisms such as lactate, succinate, and acetate, the latter presenting as a signal intensity peak at 1.98 ppm. 21,22 Apart from lactate, which is a nonspecific marker in many diseases, these metabolites have not been reported in brain tissue other than in conjunction with bacterial infections, 21-23 hydatid cyst, 24 or cysticercosis. 25 It is important to be aware that these spectral changes are present only in untreated bacterial abscesses and will dramatically change under antibiotic treatment. 26 It has been shown that it is possible to differentiate anaerobic from aerobic or sterile brain abscesses based on the metabolite patterns observed at in vivo MR spectroscopy. 27

Figure 2 a-b. A new contrast-enhancing brainstem lesion was present in a 60-year-old male who had previously undergone radiation therapy for suspected brainstem glioma. SVS was performed with VOI placed on axial FLAIR images (a). There was significant tenfold elevated signal intensity of the choline metabolite and marked decrease of NAA consistent with recurrent tumor (b) that was confirmed by progression on subsequent imaging. (Images reprinted after permission Neuroradiology, Weybright et al. In press, DOI:10.1007/s00234-004-1195-1, 2004.)

Monitoring treatment . MRS can also be used to monitor treatment of brain lesions. This can be done in, for example, brain abscesses, in which a normalization of the spectral changes will occur during successful treatment. Another potential area in which monitoring of treatment would be of interest is in treatment of neoplasm. It might be possible with MRS to monitor the effect of radiation therapy and chemotherapy of brain neoplasms. It could be assumed that a decrease in the abnormal Cho/Cr and Cho/NAA ratios and decline in the Cho peak would occur during successful treatment. On the contrary, if the treatment is failing, the Cho/Cr and Cho/NAA peaks would remain highly abnormal or even further increase not only in the contrast- enhancing lesion but also in surrounding brain tissue as an indication for tumor progression and therapy failure.

Radiation injury versus recurrent brain tumor . Contrast-enhancing lesions that arise on routine brain MR imaging at the site of a previously identified and treated primary intracranial neoplasm present a significant diagnostic dilemma. These lesions are in regions that have been subjected to radiation, with or without chemotherapy, and in some instances surgical resection. Many do not have specific imaging characteristics that will enable the neuroradiologist to discriminate tumor recurrence from the inflammatory or necrotic changes that can result from treatment with radiation. 28 Both recurrent tumors and radiation injury typically demonstrate enhancement with gadolinium. Specific spectroscopic changes that occur with radiation-induced necrosis have been reported and include slight depression of NAA and variable changes in choline. 29-32 A recent study reported a 93% success rate in differentiating recurrent tumor from radiation injury with significantly increased Cho/NAA and Cho/Cr ratios (P < 0.001) in areas of recurrent tumor compared to areas of radiation injury and compared to normal adjacent brain tissue. 33 These results also are supported by previous work utilizing SVS, where the authors claim that a Cho/Cr ratio over 1.79 or lipid-lactate/Cho ratio less than 0.75 has a sevenfold increased odds of indicating tumor compared to pure necrosis. 9

Unfortunately, the results are not always definitive, and we have to assume that in many of the new contrast-enhancing lesions, both tumor cells and radiation injury are present. This might be more of a problem when using SVS compared to 2D-CSI. SVS of an enhancing lesion that contains a small focus of recurrent tumor in a bed of much larger radiation necrosis would likely be “averaged out” such that the choline and NAA metabolite profiles suggested only inflammatory changes and the recurrent tumor would be missed. In addition, an enhancing lesion containing heterogeneous areas of normal CNS tissue and recurrent tumor could be averaged into a spectral profile suggestive also of only inflammatory changes, with the recurrent tumor being missed. Averaging normal or radiation-injured brain tissue with tumor tissue will tend to lower Cho/Cr and Cho/NAA ratios over pure tumor and lower its conspicuity, which is a matter of major clinical importance. 2D-CSI allows coverage of both contrast-enhancing tissue and surrounding tissue, and of normal-appearing white matter in the contralateral hemisphere. This enables a sampling of multiple discrete regions, which may be necessary to discern the subtle differences between tumor recurrence and radiation injury and the identification of areas of both tumor and inflammatory changes in the same enhancing lesion.

SUMMARY

MRS plays an increasingly important role in the workup of brain lesions and may be used to distinguish primary brain neoplasm from metastases and other cystic lesions, such as abscess in patients with a rim-enhancing cystic brain lesion. MRS may have an important role to play in guiding the surgeon to the area of highest abnormal spectral changes in the brain lesion prior to stereotactic surgery for histological diagnosis and tumor staging.

MRS cannot replace brain biopsy for histological diagnosis, but might be able to better delineate and define tumor boundaries and separate an infiltrative growing glioma from normal brain tissue, thereby helping in presurgical planning of brain neoplasm. Lately, MRS has been shown to be a helpful tool in evaluation of a new contrast-enhancing lesion at the site of a previously identified and treated primary intracranial neoplasm, differentiating recurrent tumor from radiation injury.

Pia C. Sundgren, MD, PhD, is associate professor, Department of Radiology, division of neuroradiology.

Suresh K. Mukherji, MD, is professor and chief of neuroradiology, and professor of radiology and otolaryngology head & neck surgery, University of Michigan Health System, Ann Arbor, Mich.

References:

  1. Weybright P, Maly P, Gomez-Hassan D, Blaesing C, Sundgren PC. MR spectroscopy in the evaluation of recurrent contrast enhancing lesions in the posterior fossa after tumor treatment. Neuroradiology. In press.
  2. Galanaud D, Le Fur Y, Nicoli F, et al. Regional metabolite levels of the normal posterior fossa studied by proton chemical shift imaging. MAGMA. 2001;13:127-133.
  3. Burtscher IM, Holtas S. Proton MR spectroscopy in clinical routine. J Magn Reson Imaging. 2001;13:560-567. 4. Castillo M, Kwock L, Mukherji SK. Clinical applications of proton MR spectroscopy. AJNR Am J Neuroradiol. 1996;17:1-15.
  4. Castillo M, Kwock L, Mukherji SK. Clinical applications of proton MR spectroscopy. AJNR Am J Neuroradiol. 1996;17:1-15.
  5. Mukherji SK, ed. Clinical Applications of MR Spectroscopy. New York: Wiley-Liss; 1998.
  6. Norfray JF, Tomita T, Byrd SE, Ross BD, Berger PA, Miller RS. Clinical impact of MR spectroscopy when MR imaging is indeterminate in pediatric brain tumors. AJR Am J Roentgenol. 1999;173:119-125.
  7. Lin A, Bluml S, Mamelak AN. Efficacy of proton magnetic resonance spectroscopy in clinical decision making for patients with suspected malignant brain tumors. J Neurooncol. 1999;45:69-81.
  8. Fulham MJ, Bizzi A, Dieta MJ, et al. Mapping of brain tumor metabolites with proton MR spectroscopic imaging: clinical relevance. Radiology. 1992;185:675-686.
  9. Rock JP, Hearshen D, Scarpace L, et al. Correlations between magnetic resonance spectroscopy and image-guided histopathology, with special attention to radiation necrosis. Neurosurgery. 2002;51:912-919.
  10. Croteau D, Scarpace L, Hearschen D, et al. Correlation between magnetic resonance spectroscopy imaging and image-guided biopsies: semi quantitative and qualitative histopathological analyses of patients with untreated glioma. Neurosurgery. 2001;49:823-829.
  11. Preul MC, Caramanos Z, Leblanc R, Villemure JG, Arnold DL. Using pattern analysis of in vivo proton MRSI data to improve the diagnosis and surgical management of patients with brain tumors. NMR Biomed. 1998;11:192-200.
  12. Hall WA, Martin A, Liu H, Truwit CL. Improving diagnostic yield in brain biopsy: coupling spectroscopic targeting with real-time needle placement. J Magn Reson Imaging. 2001;13:12-15.
  13. Dowling C, Bollen AW, Noworolski SM, et al. Preoperative proton MR spectroscopic imaging of brain tumors: correlation with histopathologic analysis of resection specimens. AJNR Am J Neuroradiol. 2001;22:604-612.
  14. Nafe R, Herminghaus S, Raab P, et al. Preoperative proton-MR spectroscopy of gliomas—correlation with quantitative nuclear morphology in surgical specimen. J Neurooncol. 2003;63:233-245.
  15. Moller-Hartmann W, Herminghaus S, Krings T, et al. Clinical application of proton magnetic resonance spectroscopy in the diagnosis of intracranial mass lesions. Neuroradiology. 2002;44:371-381.
  16. Ott D, Henning J, Ernst T. Human brain tumor: assessment with in vivo proton MR spectroscopy. Radiology. 1993;186:745-752.
  17. Demaerel P, Johannik K, Van Hecke P, et al. 1H NMR spectroscopy in fifty cases of newly diagnosed intracranial tumors. J Comput Assist Tomogr. 1991;15:67-76.
  18. Demaerel P. In vivo localized single-voxel proton magnetic resonance spectroscopy of intracranial tumors. Int J Neuroradiol. 1997;3:94-110.
  19. Sutton LN, Wehrli SL, Gennarelli L, et al. High resolution 1H magnetic resonance spectroscopy of pediatric posterior fossa tumors in vivo. J Neurosurg. 1994;81:443-448.
  20. Sijens PE, Knopp MV, Brunnetti A, et al. 1H MR spectroscopy in patients with metastatic brain tumors; a multicenter study. Magn Reson Med. 1995;33:818-826.
  21. Martinez-Perez I, Moreno A, Alonso J, et al. Diagnosis of brain abscess by magnetic resonance spectroscopy. J Neurosurg. 1997;86:708-713.
  22. Remy C, Grand S, Lai ES, et al. 1H MRS of human brain abscesses in vivo and in vitro. Magn Reson Med. 1995;34:508-514.
  23. Potani H, Gupta RK, Jain VK, Roy R, Pandey R. Cystic intracranial mass lesions: possible role of in vivo MR spectroscopy in its differential diagnosis. Magn Reson Imaging. 1995;13:1019-1029.
  24. Kohli A, Gupta RK, Potani H, Roy R. In vivo magnetic resonance spectroscopy in a case of intracranial hydatid cyst. Neurology. 1995;45:562-564.
  25. Chang KH, Song IC, Kim SH, et al. In vivo single-voxel proton MR spectroscopy in intracranial cystic masses. AJNR Am J Neuroradiol. 1998;19:401-405.
  26. Burtscher IM, Holtas S. In vivo 1H-MR spectroscopy in untreated and treated brain abscesses. AJNR Am J Neuroradiol. 1999;20:1049-1053.
  27. Garg M, Gupta RK, Husain M, et al. Brain abscesses: etiologic categorization with in vivo proton MR spectroscopy. Radiology. 2004;230:519-527
  28. Bonavita S, Di Salle F, Tedeschi G. Proton MRS in neurological disorders. Eur J Radiol. 1999;30:125-131.
  29. Chong VF, Rumpel H, Fan YF, Mukherji SK. Temporal lobe changes following radiation therapy: imaging and proton MR spectroscopic findings. Eur J Radiol. 2001;11:317-324.
  30. Chong VF, Rumpel H, Aw YS, Ho GL, Fan YF, Chua EJ. Temporal lobe necrosis following radiation therapy for nasopharyngeal carcinoma: 1H MR spectroscopic findings. Int J Radiat Oncol Biol Phys. 1999;45:699-705.
  31. Schlemmer JP, Bachert P, Henze M, et al. Differentiation of radiation necrosis from tumor progression using proton magnetic resonance spectroscopy. Neuroradiology. 2002;44:216-222.
  32. Schlemmer HP, Bachert P, Herfarth K, Zuna I, Debus J, van Kaick G. Proton MR spectroscopic evaluation of suspicious brain lesions after stereotactic radiotherapy. ANJR Am J Neuroradiol. 2001;22:1316-1324.
  33. Weybright P, Maly P, Gomez-Hassan D, Nan B, Sundgren PC. MR spectroscopy in the differentiation of recurrent neoplasm versus tumor necrosis in recurrent intracranial contrast enhancing lesions. ASNR 2004 Proceedings. O164:121.