|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Proton Magnetic Resonance Spectroscopic Imaging in Children With Recurrent Primary Brain TumorsFrom the Pediatric Oncology Branch, National Cancer Institute; National Institute of Neurological Disorders and Stroke, and Laboratory of Diagnostic Radiology Research, National Institutes of Health, Bethesda, MD. Address reprint requests to Katherine E. Warren, MD, Pediatric Oncology Branch, NCI, Bldg 10, Rm 13N240, 10 Center Dr, MSC 1928, Bethesda, MD 20892-1928; email warrenk{at}exchange.nih.gov
PURPOSE: Proton magnetic resonance spectroscopic imaging (1H-MRSI) is a noninvasive technique for spatial characterization of biochemical markers in tissues. We measured the relative tumor concentrations of these biochemical markers in children with recurrent brain tumors and evaluated their potential prognostic significance. PATIENTS AND METHODS: 1H-MRSI was performed on 27 children with recurrent primary brain tumors referred to our institution for investigational drug trials. Diagnoses included high-grade glioma (n = 10), brainstem glioma (n = 7), medulloblastoma/peripheral neuroectodermal tumor (n = 6), ependymoma (n = 3), and pineal germinoma (n = 1). 1H-MRSI was performed on 1.5-T magnetic resonance imagers before treatment. The concentrations of choline (Cho) and N-acetyl-aspartate (NAA) in the tumor and normal brain were quantified using a multislice multivoxel method, and the maximum Cho:NAA ratio was determined for each patient’s tumor. RESULTS: The maximum Cho:NAA ratio ranged from 1.1 to 13.2 (median, 4.5); the Cho:NAA ratio in areas of normal-appearing brain tissue was less than 1.0. The maximum Cho:NAA ratio for each histologic subtype varied considerably; approximately equal numbers of patients within each tumor type had maximum Cho:NAA ratios above and below the median. Patients with a maximum Cho:NAA ratio greater than 4.5 had a median survival of 22 weeks, and all 13 patients died by 63 weeks. Patients with a Cho:NAA ratio less than or equal to 4.5 had a projected survival of more than 50% at 63 weeks. The difference was statistically significant (P = .0067, log-rank test). CONCLUSION: The maximum tumor Cho:NAA ratio seems to be predictive of outcome in children with recurrent primary brain tumors and should be evaluated as a prognostic indicator in newly diagnosed childhood brain tumors.
BRAIN TUMORS ARE the most common solid tumors of childhood, but they represent a diverse group of tumors. The overall 5-year survival rate for children diagnosed with a primary brain tumor between 1989 and 1994 was 64%, but survival rates range from greater than 90% in children with low-grade cystic cerebellar astrocytomas to less than 10% in children with high-grade, diffuse brainstem gliomas.1 In addition to tumor histology and grade, other prognostic factors that are predictive of outcome for childhood brain tumors include tumor location within the brain, extent of disease (eg, the presence of leptomeningeal metastases), and extent of tumor resection.2-6 Identifying tumor characteristics at diagnosis that are predictive for outcome is a critical component of treatment planning for a patient and can provide a basis for stratifying patients on clinical research trials. Ideally, these prognostic indicators should be measured noninvasively and before definitive treatment. Tumor histology, grade, and extent of tumor resection are determined after surgical intervention, but the location and extent of tumor can be delineated by neuroimaging techniques. Magnetic resonance (MR) imaging has become the standard neuroimaging technique for childhood brain tumors. MR images are based on detection of hydrogen nuclei (protons) primarily in water and lipids because these nuclei are present in highest concentrations in tissue.7 Proton magnetic resonance spectroscopy (1H-MRS), which is a technique that can be performed on a conventional MR imager, can quantify the relative concentration of nonwater, proton-containing metabolites from discreet tissue regions by suppressing the signals from water and lipids.8-10 In the brain, the principal metabolite signals that can be measured by 1H-MRS at long echo times are N-acetyl aspartate (NAA), creatine, and choline (Cho) (Fig 1). NAA is a neuronal marker present in normal functioning neurons; creatines, including creatine and phosphocreatine, are important in energy metabolism and vary little in normal brain tissue; and Cho-containing compounds are constituents of cell membranes. Lactate, which is a marker of anaerobic metabolism, is not usually detected in normal brain but may be present in some brain tumors or areas of ischemic injury.11,12 1H-MRS has been used to study the metabolic profiles of various CNS diseases, including ischemia, metabolic diseases, demyelination, and brain tumors.11,13-19
The MR spectra from primary brain tumors are substantially different than the spectra from normal brain tissue. In tumors, the relative concentration of NAA is decreased, and Cho concentration is increased.18 Lactate may also be detected in some brain tumors, depending on the extent of anaerobic glycolysis. Tumor spectra have been used to noninvasively diagnose the most common types of supratentorial brain tumors in adults using pattern recognition analysis.20 In addition, serial monitoring of 1H-MRS in adults with gliomas seems to be capable of detecting malignant degeneration and disease progression.21 A low-resolution metabolite map that reflects the spatial distribution of each of the metabolites over multiple slices in the brain can be generated by collecting MR spectroscopic data from multiple contiguous discreet regions (voxels) of the brain. This technique is called 1H-MRSI.22 The 1H-MRSI metabolite maps can be superimposed onto conventional MR images to correlate structural changes with metabolite distribution (Fig 2). We performed a pilot study in pediatric patients with recurrent brain tumors to determine whether tumor biochemical profiles measured by 1H-MRSI are predictive of disease outcome.
We studied 27 children with recurrent primary brain tumors who were referred to the Pediatric Oncology Branch of the National Cancer Institute for investigational drug treatment trials (Table 1). All patients were studied under a protocol approved by the National Cancer Institute’s Institutional Review Board, and informed consent was obtained from all patients or their legal guardians. Each patient had 1H-MRSI performed at the time of presentation, before the initiation of treatment at our institution. The median age of the patient population was 14 years (range, 5 to 20 years). The patients had a median of two prior tumor recurrences (range, one to five recurrences), and had received a median of one prior chemotherapy regimen (range, zero to five regimens). All patients had previously received radiation treatment, including 17 patients who received local radiation to the tumor and 10 patients who received craniospinal radiation. Diagnoses included high-grade glioma (n = 10), brainstem glioma (n = 7), medulloblastoma/peripheral neuroectodermal tumor (n = 6), ependymoma (n = 3), and pineal germinoma (n = 1).
1H-MRSI was performed on a 1.5-T whole-body imager (Signa; GE Medical Systems, Milwaukee, WI) equipped with self-shielded gradients and a standard quadrature head coil. Spectroscopy data was acquired using a multislice, multivoxel technique that simultaneously collects spectra from multiple voxels, each with a nominal volume of 0.84 mL, over approximately 20 minutes.22 Phase encoding was used to obtain a 32 x 32 matrix of 1,024 spectra for each of four axial 15-mm thick slices. After 1H-MRSI acquisition was complete, T2-weighted MR images using the same field of view and angulation were obtained to allow the MR images to be coregistered with the 1H-MRSI. Spectroscopy data was analyzed on a Sun Workstation (Sun Microsystems, Mountain View, CA) with a customized software package developed at the NIH using Interactive Data Language (Research Systems, Inc, Boulder, CO) data processing language.21,22 Quantitative analysis of metabolites was performed for each voxel to yield relative concentrations. The relative concentration of each metabolite in a voxel was derived by integrating the area under the signal intensity peak for that metabolite,23 and the concentrations for each metabolite in all voxels were displayed in two-dimensional metabolite maps (Fig 2). The relative concentrations of metabolites were normalized by expressing them as ratios. For our study the Cho:NAA ratio was derived from each voxel, and the maximum Cho:NAA ratio for each patient’s tumor was used in the survival analysis. The 27 patients were split into two groups according to whether their maximum tumor Cho:NAA ratio was greater than the median value for the entire group or less than or equal to this median value. The duration of survival was derived from the date that the spectroscopy was performed to the date of death or the date of last follow-up for surviving patients. The survival of each group was analyzed graphically using the Kaplan-Meier method, and the significance of the difference in survival between the two groups was assessed by the log-rank test for censored survival data.24
In all patients, the maximum Cho:NAA ratio occurred within the tumor, which was identified from the T2-weighted MR image, or in the brain immediately surrounding the tumor. The maximum Cho:NAA ratio for each patient’s tumor according to the histologic diagnosis is shown in Fig 3. The median Cho:NAA ratio for the entire population of 27 patients was 4.5 (range, 1.1 to 13.2). There was a wide range of values within each histologic subtype of brain tumor and considerable overlap in the values across tumor types. The median Cho:NAA ratio in the patients with recurrent high-grade gliomas was 6.0, the median metabolite ratio in patients with brainstem gliomas was 5.1, and the median ratio in patients with medulloblastoma or peripheral neuroectodermal tumor was 5.4.
Approximately equal numbers of each histologic subtype of brain tumor fell above and below the median Cho:NAA ratio for the entire group (Fig 3); therefore, the population was split into two groups based on the median Cho:NAA ratio for the entire group, which resulted in approximately equal representation of each tumor type in the two groups. In addition, the two groups had a similar age profile. The median age for the group with a Cho:NAA ratio equal to 4.5 was 15 years (range, 5 to 20 years). For the group with a Cho:NAA ratio greater than 4.5, the median age was 12 years (range, 5 to 18 years). Six of 14 patients with a low Cho:NAA ratio had previously received craniospinal radiation compared with four of 13 patients with a high Cho:NAA ratio. The number of prior recurrences and the number of prior chemotherapy regimens were also similar in the two groups (Fig 4).
The postspectroscopy treatment was also comparable in the two groups. Twenty-six of the 27 patients were treated on one or more phase I investigational treatment protocols after their spectroscopy was performed. Twenty-two patients initially received a carboplatin-based investigational treatment regimen, and, of these 22 patients, 13 had a maximum Cho:NAA less than or equal to 4.5 and the remaining nine patients had a ratio greater than the median. Four of 22 patients treated on this regimen had objective responses, including one complete and two partial responses in the patients with Cho:NAA greater than 4.5 and one partial response in a patient with a Cho:NAA less than or equal to 4.5. Four of the remaining five patients were treated on other phase I trials, and one patient received no further treatment. The Kaplan-Meier estimated cumulative survival of children with recurrent primary brain tumors that were split into two groups based on the maximum tumor Cho:NAA ratio is shown in Fig 5. Patients with a Cho:NAA ratio greater than 4.5 had a median post1H-MRSI survival time of 22 weeks, and all 13 patients had died by 63 weeks. In the patients with a Cho:NAA ratio less than or equal to 4.5, the projected survival rate at 63 weeks exceeds 50%. Eight of the 14 patients are alive 22 to 136 weeks after 1H-MRSI. Using the log-rank test, the survival of patients with a Cho:NAA ratio below the median was significantly greater than the survival in the patients with a high Cho:NAA ratio (P = .0067).
In a heterogeneous population of children with recurrent primary brain tumors, our study demonstrated that 1H-MRSI, specifically the maximum tumor Cho:NAA ratio, seems to be predictive of outcome. A high Cho:NAA ratio within the tumor was associated with a shorter duration of survival. Although the small size of this study precluded a multivariate analysis assessing the independence of the Cho:NAA ratio as a prognostic factor, the two groups, which were split based on the median Cho:NAA ratio for the entire population, were balanced for histologic diagnosis, age, number of recurrences, prior therapy (radiation and chemotherapy), and the type of treatment administered after 1H-MRSI. The multislice, multivoxel technique used in this study measures the relative concentrations of Cho, creatine, and NAA in four axial slices of the brain. Calculating ratios of these metabolites normalizes the relative concentrations and allows for comparisons across patients and between groups. Absolute quantification of metabolite concentration requires the use of external standards and specific acquisition and postprocessing procedures, which would substantially prolong the scan time.25 An alternative to using ratios of metabolites within the same voxel is comparing individual metabolite levels within the tumor with metabolite levels in the contralateral normal brain or with metabolite levels in a selected area of the brain, such as the centrum semiovale. In our patient population, the effects of prior therapy, such as craniospinal radiation, on metabolite levels in the brain are not well studied, and this method would not be applicable to midline lesions, such as brainstem gliomas. Cho and NAA are known to fluctuate with age in normal brain tissue. Cho is the highest peak in normal newborns and gradually decreases until 2 to 3 years of life. NAA increases steadily until adolescence and is the major normal metabolite peak after 4 months of age.26 To avoid these potentially confounding developmental changes, we excluded children less than 3 years of age from our study. The Cho:NAA ratio accentuates the differences between brain tumors and normal brain tissue because brain tumors typically have higher Cho concentrations and lower NAA concentrations than normal brain.11,15,16 Cho:NAA ratios in normal brain tissue are less than 1.0.27,28 In our study, the maximum tumor Cho:NAA ratio was greater than 1.0 in all patients. The maximum Cho:NAA ratio was selected for this analysis because a prior study in pediatric patients with low-grade astrocytomas demonstrated considerable heterogeneity in Cho:NAA ratios within a single brain tumor.29 The concept of worst voxel analysis has been used in a previous study, in which the highest tumor Cho level correlated with malignant degeneration of cerebral gliomas.21 The relationship between high Cho:NAA ratio and short duration of survival may have a biologic basis. The Cho signal is derived from choline, phosphocholine, and glycerophosphocholine, which are the constituents of phospholipid metabolism and components of cell membranes. Increased choline has been associated with an increased number of cells, a greater rate of membrane synthesis, and increased cell turnover.21,30 NAA is found only in neurons and axons.31 Neoplasms are thought to replace or destroy the NAA-containing cells, thereby accounting for the decreased NAA.18 The majority of prior MR spectroscopy studies have used single voxel techniques in which metabolite concentration is measured in a single preselected volume of interest that incorporates the tumor volume.32-35 These single voxel techniques are more widely available, use short acquisition times, and provide absolute quantification of metabolite concentrations. However, single voxel techniques require preselection of a defined region of interest (ROI) from which spectroscopic data is obtained. The volume of the ROI is usually large (8 mL), and areas of surrounding normal brain tissue or CSF may be included in the ROI. The voxel size of the multislice, multivoxel technique is 0.84 mL, which lowers the chance for partial volume effects. The range of Cho:NAA ratios for different histologic subtypes of recurrent childhood brain tumors from our study overlapped considerably. Therefore, this ratio does not seem to be useful for providing a noninvasive method for determining tumor histology in childhood brain tumors. The potential value of 1H-MRSI as a clinical tool has been enhanced by advances in technology that enable patients to be scanned in a reasonable time period and allow for more efficient data analysis. Conventional MR imaging hardware that is found in most medical centers can be used to perform this 1H-MRSI technique with only additional software requirements. Our study suggests that the biochemical profile obtained from 1H-MRSI can complement the anatomic images from MR imaging scans and may be predictive of the clinical behavior of childhood brain tumors. Based on the results of this pilot study, 1H-MRSI should be prospectively evaluated as a prognostic tool in newly diagnosed childhood brain tumors.
1. Surveillance, Epidemiology, and End-Results (SEER): SEER Cancer Statistics Review, 1973-1996: Childhood Cancer by the ICCC, 1999. Bethesda, MD, NIH publication, 1999 2. Campbell J, Pollack IF, Martinez AJ, et al: High grade astrocytomas in children: Radiologically complete resection is associated with an excellent long-term prognosis. Neurosurg 38:258-264, 1996[Medline] 3. Duffner PK, Krischer JP, Sanford RA, et al: Prognostic factors in infants and very young children with intracranial ependymomas. Pediatr Neurosurg 28:215-222, 1998[Medline] 4. Giordana MT, Schiffer P, Schiffer D: Prognostic factors in medulloblastoma. Childs Nerv Syst 14:256-262, 1998[Medline] 5. Packer RJ: Brain tumors in children. Curr Opin Pediatr 8:549-557, 1996[Medline] 6. Zeltzer PM, Boyett JM, Finlay JL, et al: Metastasis stage, adjuvant treatment, and residual tumor are prognostic factors for medulloblastoma in children: Conclusions from the Children’s Cancer Group 921 randomized phase III trial. Oncol 17:832-845, 1999 7. Moonen CT, van Zijl PCM, Frank JA, et al: Functional magnetic resonance imaging in medicine and physiology. Science 50:53-61, 1990
8.
Hanstock CC, Rothman DL, Prichard JW, et al: Spatially localized 1H NMR spectra of metabolites in the human brain. Proc Natl Acad Sci USA 85:1821-1825, 1988 9. Frahm J, Bruhn H, Gyngell ML, et al: Localized high-resolution proton NMR spectroscopy using stimulated echoes: Initial applications to human brain in vivo. Reson Med 9:79-93, 1989 10. Tzika AA, Vajapeyam S, Barnes PD: Multivoxel proton MR spectroscopy and hemodynamic MR imaging of childhood brain tumors: Preliminary observations. Am J Neuroradiol 18:203-218, 1997[Abstract]
11.
Bruhn H, Frahm J, Gyngell ML, et al: Noninvasive differentiation of tumors with use of localized 1H MR spectroscopy in vivo: Initial experience in patients with cerebral tumors. Radiology 172:541-548, 1989 12. Brunetti A, Alfano B, Soricelli A, et al: Functional characterization of brain tumors: An overview of the potential clinical value. Nuclear Med Biol 23:699-715, 1996 13. van der Knaap MS, van der Grond J, Luyten PR, et al: 1H and 31P magnetic resonance spectroscopy of the brain in degenerative cerebral disorders. Ann Neurol 31:202-211, 1992[Medline] 14. Kruse B, Barker PB, van Zijl PCM, et al: Multislice proton magnetic resonance spectroscopic imaging in X-linked adrenoleukodystrophy. Ann Neurol 36:595-608, 1994[Medline]
15.
Alger JR, Frank JA, Bizzi A, et al: Metabolism of human gliomas: Assessment with 1HMR spectroscopy and 18F fluorodeoxyglucose PET. Radiology 177:633-641, 1990
16.
Fulham MJ, Bizzi A, Dietz MJ, et al: Mapping of brain tumor metabolites with proton MR spectroscopic imaging: Clinical relevance. Radiology 185:675-686, 1992 17. Taylor JS, Ogg RJ, Langston JW: Proton MR spectroscopy of pediatric brain tumors. Neuroimaging Clin N Am 8:753-779, 1998[Medline] 18. Ross B, Michaelis T: Clinical applications of magnetic resonance spectroscopy. Magn Reson Q 10:191-247, 1994[Medline] 19. Tate AR, Griffiths JR, Martinez-Perez I, et al: Towards a method for automated classification of 1H MRS spectra from brain tumours. NMR Biomed 11:177-191, 1998[Medline] 20. Preul MC, Caramanos Z, Collins DL, et al: Accurate, noninvasive diagnosis of human brain tumors by using proton magnetic resonance spectroscopy. Nat Med 2:323-325, 1996[Medline] 21. Tedeschi G, Lundbom N, Raman R, et al: Increased choline signal coinciding with malignant degeneration of cerebral gliomas: A serial proton magnetic resonance spectroscopy imaging study. J Neurosurg 87:516-524, 1997[Medline]
22.
Duyn JH, Gillen J, Sobering G, et al: Multisection proton MR spectroscopic imaging of the brain. Radiology 188:277-282, 1993 23. Novotny E, Ashwal S, Shevell M: Proton magnetic resonance spectroscopy: An emerging technology in pediatric neurology research. Pediatr Res 44:1-10, 1998[Medline] 24. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163-170, 1963 25. Christiansen P, Henriksen O, Stubgaard M, et al: In vivo quantification of brain metabolites by 1H-MRS using water as an internal standard. Magn Reson Imaging 11:107-118, 1993[Medline] 26. Byrd SE, Tomita T, Palka PS, et al: Magnetic resonance spectroscopy (MRS) in the evaluation of pediatric brain tumors, part I: Introduction to MRS. J Natl Med Assoc 88:649-654, 1996[Medline] 27. Soher BJ, van Zijl PC, Duyn JH, et al: Quantitative proton MR spectroscopic imaging of the human brain. Magn Reson Med 35:356-363, 1996[Medline] 28. Tedeschi G, Bertolino A, Righini A, et al: Brain regional distribution pattern of metabolite signal intensities in young adults by proton magnetic resonance spectroscopic imaging. Neurology 45:1384-1391, 1995[Abstract] 29. Lazareff JA, Olmstead C, Bockhorst KH, et al: Proton magnetic resonance spectroscopic imaging of pediatric low-grade astrocytomas. Childs Nerv Syst 12:130-135, 1996[Medline] 30. Miller BL: A review of chemical issues in 1H NMR spectroscopy: N-acetyl-aspartate, creatine and choline. NMR Biomed 4:47-52, 1991[Medline] 31. Birken DL, Oldendorf WH: N-acetyl-L-aspartic acid: A literature review of a compound prominent in 1H-NMR spectroscopic studies of the brain. Neurosci Biobehav Rev 13:23-31, 1989[Medline] 32. Byrd SE, Tomita T, Palka PS, et al: Magnetic resonance spectroscopy (MRS) in the evaluation of pediatric brain tumors, part II: Clinical analysis. J Natl Med Assoc 88:717-723, 1996[Medline] 33. Sutton LN, Wang Z, Gusnard D, et al: Proton magnetic resonance spectroscopy of pediatric brain tumors. Neurosurg 31:195-202, 1992[Medline] 34. Negendank WG, Sauter R, Brown TR, et al: Proton magnetic resonance spectroscopy in patients with glial tumors: A multicenter study. J Neurosurg 84:449-458, 1996[Medline] 35. Taylor JS, Langston JW, Reddick WE, et al: Clinical value of proton magnetic resonance spectroscopy for differentiating recurrent or residual brain tumor from delayed cerebral necrosis. Oncol Biol Phys 36:1251-1261, 1996 Submitted July 30, 1999; accepted October 7, 1999.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
) and the high Cho:NAA ratio group (
).
