Brain Tumor Pathol (2004).21:39-46 DOI 10.1007/s10014-004-0151-4

Hitoshi Izumiyama Ataru Fukuda

9 Takumi

9 Norihiko

Abe

9 The Japan Society of Brain Tumor Pathology2004

9 Daisuke

Tanioka

Kunii

Clinicopathological examination of glioma by proton magnetic resonance spectroscopy background

Received: February 10, 2004 / Accepted: February 25, 2004 Automation of proton magnetic resonance spectroscopy (MRS) in recent years has made it possible for MRS measurement to be performed in a shorter time than before, and the number of reports of its usefulness for the assessment of glioma malignancy has been increasing in the past several years. We studied the efficacy of proton MRS when used for glioma and conducted clinicopathological examination of glioma. The subjects were 15 patients who had received a pathological diagnosis of glioma at our hospital (6 cases of glioblastoma, 1 case of anaplastic astrocytoma, 4 cases of low-grade astrocytoma, and 4 cases of radiation necrosis); Siemens Magnetom Vision 1.5T was used for the study. Regions of interest (ROIs) were defined as the areas where abnormal signals were found on magnetic resonance imaging (MRI). Areas of primary peaks, such as choline (Cho), N-acetylaspartate (NAA), and lactate (Lac), were measured, and the ratios to normal brain tissue were examined. This study revealed a tendency of increased malignancy of glioma with a decrease in NAA. Some cases also displayed a decrease in Cho with an increase in malignancy. Assessment of malignancy must not be based on a single ROI alone, but several ROIs should be assessed comprehensively. Measurement was difficult when the tumor volume was small. Because diagnosis of very early glioma by MRS seemed difficult, other adjunctive diagnoses may be necessary. Proton MRS is very useful for diagnosis of glioblastoma. Abstract

Key

Words

Proton MRS

9 MRI

9 Glioma

-

Radiation

necrosis

Introduction Automation of proton magnetic resonance spectroscopy (MRS) in recent years has made it possible for MRS measurement to be performed in a shorter time than before, and the number of reports of its usefulness for the assessment of glioma malignancy has increased in the past several years. We studied the efficacy of proton MRS when used for glioma and conducted clinicopathological examination of glioma.

Subjects and materials The subjects were 15 patients who had received a pathological diagnosis of glioma at our hospital (6 cases of glioblastoma, 1 case of anaplastic astrocytoma, 4 cases of low-grade astrocytoma, and 4 cases of radiation necrosis); Siemens Magnetom Vision 1.5T (Siemens, Munich, Germany) was used for the study. Regions of interest (ROIs) were defined as the areas where abnormal signals were found on magnetic resonance imaging (MRI). Areas of primary peaks, such as choline (Cho), N-acetylaspartate (NAA), and lactate (Lac), were measured, and the ratios to normal brain tissue were examined. In addition, the relationships between the results of proton MRS (1H-MRS) performed before surgery (in some cases, after surgery also), pathological examination, and MIB-1 were examined.

Results

H. Izumiyama (~:~) . T. Abe 9D. Tanioka 9A. Fukuda 9N. Kunii Department of Neurosurgery, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8666, Japan Tel. +81-3-3784-8671;Fax +81-3-3784-8671 e-mail: [email protected]

Case 1: glioblastoma in a 57-year-old man; MIB-1 32.9%. A marked decrease in Cho, a marked decrease in NAA, and a marked increase in Lac were noted within the ring. At the ring margin, a marked increase in Cho (• a decrease in N A A (• and a mild increase in Lac were noted, displaying the pattern of either a low-grade glioma or a high-grade glioma. In T2 high-signal areas, on the other

40 Fig. 1. Case 1: Glioblastoma in a 57-year-old man; MIB-1 32.9%. A A m a r k e d decrease in choline (Cho), a marked decrease in N-acetylaspartate ( N A A ) , and a m a r k e d increase in lactate (Lac) were noted within the ring. B A t the ring margin, a m a r k e d increase in Cho ( x l . 5 ) , a decrease in N A A (• and a mild increase in Lac were noted, displaying the pattern of either a low-grade glioma or a high-grade glioma. C In Ta high-signal areas, off the other hand, a decrease in N A A (• and a decrease in C h o were noted. The p a t t e r n s e e m e d n o r m a l on the surface, but there was a decrease in both Cho and N A A as compared with normal areas

A

B

C

41 Fig. 2. C a s e 2: Glioblastoma in a 61-year-old man; MIB-1 33.0%. A A m a r k e d decrease in Cho, a m a r k e d decrease in N A A (• and a m a r k e d increase in Lac were noted at margin 1 of the ring. B A t margin 2 of the ring, a m a r k e d increase in Cho (• a m a r k e d decrease in N A A (• and a marked increase in Lac were noted, displaying the pattern of a high-grade glioma. C In T 2 high-signal areas, on the other hand, a slight increase was noted in Cho (• but there was no decrease in N A A , and changes in C h o and N A A s e e m e d to present a normal pattern at a glance

A

la

C

42 Fig. 3. Case 3: Anaplastic astrocytoma in a 60-yearold woman; MIB-1 10.0%. A In area 1 of the mass, a marked increase in Cho (• a marked decrease in NAA (• and an increase in Lac were noted, displaying the pattern of a high-grade glioma. B In Area 2 of the mass, a decrease in Cho (• a marked decrease in NAA (• and an increase in Lac were noted

A

B

hand, a decrease in N A A (• and a decrease in Cho were noted. T h e p a t t e r n s e e m e d n o r m a l on the surface, but there was a decrease in b o t h Cho and N A A as c o m p a r e d with n o r m a l areas. (Fig. 1 A - C ) Case 2: glioblastoma in a 61-year-old man; MIB-1 33.0%. A m a r k e d decrease in Cho, a m a r k e d decrease in N A A (• and a m a r k e d increase in Lac were noted at margin 1 of the ring. A t margin 2 of the ring, a m a r k e d increase in Cho (• a m a r k e d decrease in N A A (• and a m a r k e d increase in Lac were noted, displaying the p a t t e r n of a high-grade glioma. In T2 high-signal areas, on the o t h e r hand, a slight increase was n o t e d in Cho (• but there was no decrease in N A A , and changes in Cho and N A A s e e m e d to p r e s e n t a n o r m a l p a t t e r n at a glance (Fig. 2 A - C ) . Case 3: anaplastic a s t r o c y t o m a in a 60-year-old women; MIB-1 10.0%. In area 1 of the mass, a m a r k e d increase in C h o (• a m a r k e d d e c r e a s e in N A A (• and an

increase in Lac were noted, displaying the p a t t e r n of a highgrade glioma. In area 2 of the mass, a decrease in Cho (• a m a r k e d decrease in N A A (• and an increase in Lac were n o t e d (Fig. 3A,B). Case 8: grade II astrocytoma in a 28-year-old man. In a r e a 1 of the mass, a mild increase in C h o (• and a d e c r e a s e in N A A (• were noted, displaying the p a t t e r n of a low-grade glioma. In area 2 of the mass, a m a r k e d increase in Cho (• a m a r k e d decrease in N A A (• and an increase in Lac were noted, displaying the p a t t e r n of a high-grade glioma. MIB-1 at this location was 6.3%, slightly high for a low-grade astrocytoma. In a r e a 3 of the mass, a m a r k e d decrease in Cho (• a m a r k e d decrease in N A A (• and an increase in Lac w e r e n o t e d (Fig. 4 A - C ) . Case 7: grade I! oligoastrocytoma in a 34-year-old woman; MIB-1 3.6%. N o r m a l p a t t e r n s were n o t e d in the

43 Fig. 4. Case 8: G r a d e ii astrocytoma in a 28-year-old man. A In area 1 of the mass, a mild increase in Cho (• and a decrease in N A A (X0.5) were noted, displaying the p a t t e r n of a low-grade glioma. B In area 2 of the mass, a marked increase in Cho (• a m a r k e d decrease in N A A (• and an increase in Lac were noted, displaying the pattern of a high-grade glioma. MIB-1 at this location was 6.3%, slightly high for a low-grade astrocytoma. C In area 3 of the mass, a m a r k e d decrease in Cho (• a m a r k e d decrease in N A A (• and an increase in Lac were noted

A

B

C

44 Fig. 5. Case 7: Grade II oligo-astrocytoma in a 34-year-old woman; MIB-1 3.6%. Normal patterns were noted in the mass area. As for the waveform in the area where radiation necrosis had occurred, a marked decrease in Cho, a marked decrease in N A A , and an increase in Lac were noted.

Fig. 6. Case 4: Gliomatosis cerebri in a 19-year-old woman; MIB-1 40.0%. A In the mass area, a marked increase in Cho and a marked decrease in N A A were noted, but Lac was not found. The results were suggestive of a high-grade glioma. B As for the waveform in the area where radiation necrosis had occurred, a marked decrease in Cho, a marked decrease in N A A , and an increase in Lac were noted.

A

B

45

Fig. 7. A recent case of low-grade astrocytoma in a 54-year-old woman after surgery. MIB-1 was 3.6%. As for the waveform in the area where radiation necrosis had occurred, a marked decrease in Cho, a marked decrease in NAA, and an increase in Lac were noted, as they were in the two cases mentioned earlier

Low grade f

Fig. 8. Proton magnetic resonance spectroscopy (MRS) patterns of typical gliomas

Low grade f

f

As for the waveform in the area where radiation necrosis had occurred, a marked decrease in Cho, a m a r k e d decrease in N A A , and an increase in Lac were noted (Fig. 6A,B). T h e last case is a recent case of low-grade astrocytoma in a 54-year-old w o m a n after surgery. MIB-1 was 3.6%. As for the waveform in the area where radiation necrosis had occurred, a m a r k e d decrease in Cho, a marked decrease in N A A , and an increase in Lac were noted, as they were in the two cases m e n t i o n e d earlier (Fig. 7). 1 3 The study found a t e n d e n c y toward increased malignancy of glioma with a decrease in N A A . Some cases also displayed a decrease in Cho with an increase in malignancy. Assessment of malignancy, must n o t be based on a single R O I alone, but several R O I s should be assessed comprehensively. M e a s u r e m e n t was difficult when the t u m o r volume was small. Because diagnosis of very early glioma by MRS seemed difficult, other adjunctive diagnoses may be necessary.

Discussion

Fig. 9. Proton MRS patterns of gliomas based on our study

mass area. As for the waveform in the area where radiation necrosis had occurred, a m a r k e d decrease in Cho, a m a r k e d decrease in N A A , and an increase in Lac were n o t e d (Fig. 5). Case 4: gliomatosis cerebri in a 19-year-old woman; MIB-1 40.0%. I n the mass area, a m a r k e d increase in Cho and a m a r k e d decrease in N A A were noted, b u t Lac was not found. The results were suggestive of a high-grade glioma.

In general, high-grade gliomas displayed an increase in Cho (Fig. 8), but there were areas in glioblastomas where a decrease in Cho was noted (Fig. 9). 4,5 In this area, a decrease in N A A and an increase in Lac were also n o t e d at the same time, which seemed to reflect the presence of necrosis. In glioblastoma, MRS waveforms can vary greatly, depending on the location of the ROI. I n addition, cases with radiation necrosis similarly displayed a decrease in Cho, a decrease in N A A , and a partial increase in Lac. In particular, among the cases of grade II astrocytoma, one case displayed a highgrade glioma-like pattern in part of the mass in preoperative MRS. In this case, a relatively high value of 6.3% was noted for MIB-1.

46

The results suggest the possibility that cases with a relatively high level of MIB-1 might be distinguished from other grade II gliomas. It is certain that a more detailed study can be performed if diffusion M R I , which we have e n d o r s e d over the past several years, can also be used in addition to MRS. 64 In glioblastoma, high signal intensity was shown in the solid c o m p o n e n t , a n d similar findings ,~ere f o u n d even in residual tumors after surgery and in recurrent cases. In particular, it is i m p o r t a n t to suspect the presence of residual tumors if the location that is e n h a n c e d within a m o n t h following surgery displays as high signal intensity in diffusion-weighted imaging ( D W I ) as it did before surgery. I n addition, D W I can be used as a reference for determination of the scope of after-treatment irradiation. It is also important to continue the use of DWI, because there is a greater chance that r e c u r r e n t lesions will be discovered early with each D W I , e v e n before contrast-enhanced imaging is performed. F u r t h e r m o r e , D W I can be useful for differentiation of radiation necrosis. W e also use D W I before surgery to predict vivid locations in glioblastoma for d e t e r m i n a t i o n of the range of excision. Histopathologically, glioblastoma contains areas of high cellularity as well as low cellularity, and displays a diverse range of findings, such as necrosis, vascular endothelial proliferation, and cellular pleomorphism. Therefore, one must note that the a p p a r e n t diffusion coefficient ( A D C ) can vary greatly, depending on the location of the measurement. Although D W I is extremely useful for diagnosis of ringenhanced masses, it c a n n o t be used as the only m e a n s of j u d g m e n t when it comes to glioma, and this fact necessitates the use of MRS.

Conclusion Proton MRS is very useful for the diagnosis of glioblastoma. This study revealed a t e n d e n c y of increased malignancy of glioma with a decrease in N A A , but some cases also dis-

played a decrease in Cho with an increase in malignancy. High-grade gliomas displayed an increase in Cho, but there were areas in glioblastoma where a decrease in Cho was noted. A s s e s s m e n t of malignancy must not be based on a single R O I alone, but several R O I s should be assessed comprehensively. M e a s u r e m e n t was difficult when the tumor volume was small. Because diagnosis of very early g l i o m a by M R S seemed difficult, other adjunctive diagnoses may be necessary, s'9

References 1. Isobe T, Matsumura A, Anno I, et al (2003) Changes in 1H-MRS in glioma patients before and after irradiation: the significance of quantitative analysis of choline-containing compounds (in Japanese). No Shinkei Geka 31:167-172 2. Shinoda J, Yano H, Ando H, et al (2002) Radiological response and histological changes in malignant astrocytic tumors after stereotactic radiosurgery. Brain Tumor Pathol 19:83-92 3. Nakaiso M, Uno M, Harada M, et al (2002) Brain abscess and glioblastoma identified by combined proton magnetic resonance spectroscopy and diffusion-weighted magnetic resonance imaging two case reports. Neurol Med Chir (Tokyo) 42:346-348 4. Law M, Yang S, Wang H, et al (2003) Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. AJNR Am J Neuroradiol 24:1989-1998 5. Gajewicz W, Papierz W, SzymczakW, et al (2003) The use of proton MRS in the differential diagnosis of brain tumors and tumor-like processes. Med Sci Monit 9:MT97-105 6. Kuznetsov YE, Caramanos Z, Antel SB, et al (2003) Proton magnetic resonance spectroscopic imaging can predict length of survival in patients with supratentorial gliomas. Neurosurgery 53:565-574; discussion, 574-576 7. Nafe R, Herminghaus S, Raab P, et al (2003) Preoperative proton-MR spectroscopy of gliomas - correlation with quantitative nuclear morphology in surgical specimen. J Neurooncol 63:233245 8. Nelson SJ, McKnight TR, Henry RG (2002) Characterization of untreated gliomas by magnetic resonance spectroscopic imaging. Neuroimaging Clin N Am 12:599-613 9. Ishimaru H, Morikawa M, Iwanaga S, et al (2001) Differentiation between high-grade glioma and metastaticbrain tumor using singlevoxel proton MR spectroscopy. Eur Radiol 11:1784-1791