FULL-LENGTH ORIGINAL RESEARCH
IDH1 mutation is associated with seizures and protoplasmic subtype in patients with low-grade gliomas *†Simon V. Liubinas, †Giovanna M. D’Abaco, ‡Bradford M. Moffat, §Michael Gonzales, §Frank Feleppa, ¶Cameron J. Nowell, **Alexandra Gorelik, *†Katharine J. Drummond, ††‡‡Terence J. O’Brien, *†Andrew H. Kaye, and *†Andrew P. Morokoff Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
SUMMARY
Dr. Simon Liubinas is a neurosurgical trainee in Melbourne, Australia.
Objective: The isocitrate dehydrogenase 1 (IDH1) R132H mutation is the most common mutation in World Health Organization (WHO) grade II gliomas, reported to be expressed in 70–80%, but only 5–10% of high grade gliomas. Low grade tumors, especially the protoplasmic subtype, have the highest incidence of tumor associated epilepsy (TAE). The IDH1 mutation leads to the accumulation of 2-hydroxyglutarate (2HG), a metabolite that bears a close structural similarity to glutamate, an excitatory neurotransmitter that has been implicated in the pathogenesis of TAE. We hypothesized that expression of mutated IDH1 may play a role in the pathogenesis of TAE in low grade gliomas. Methods: Thirty consecutive patients with WHO grade II gliomas were analyzed for the presence of the IDH1-R132H mutation using immunohistochemistry. The expression of IDH1 mutation was semiquantified using open-source biologic-imaging analysis software. Results: The percentage of cells positive for the IDH1-R132H mutation was found to be higher in patients with TAE compared to those without TAE (median and interquartile range (IQR) 25.3% [8.6–53.5] vs. 5.2% [0.6–13.4], p = 0.03). In addition, we found a significantly higher median IDH1 mutation expression level in the protoplasmic subtype of low grade glioma (52.2% [IQR 19.9–58.6] vs. 13.8% [IQR 3.9–29.4], p = 0.04). Significance: Increased expression of the IDH1-R132H mutation is associated with seizures in low grade gliomas and also with the protoplasmic subtype. This supports the hypothesis that this mutation may play a role in the pathogenesis of both TAE and low grade gliomas. KEY WORDS: Glioma, IDH1, Seizure, Epilepsy, Protoplasmic astrocytoma, Glutamate.
Accepted April 17, 2014; Early View publication June 5, 2014. *Department of Neurosurgery, The Royal Melbourne Hospital, Parkville, Victoria, Australia; †The Department of Surgery (RMH), The University of Melbourne, Parkville, Victoria, Australia; ‡The Department of Radiology (RMH/WH), The University of Melbourne, Parkville, Victoria, Australia; §Department of Pathology, The Royal Melbourne Hospital, Parkville, Victoria, Australia; ¶Ludwig Institute for Cancer Research, Melbourne, Parkville, Victoria, Australia; **The Melbourne Epicentre, The Royal Melbourne Hospital, Parkville, Victoria, Australia; ††The Department of Medicine (RMH), The University of Melbourne, Parkville, Victoria, Australia; and ‡‡Department of Neurology, The Royal Melbourne Hospital, Parkville, Victoria, Australia Address correspondence to Andrew Morokoff, Department of Surgery, Royal Melbourne Hospital, Parkville,VIC 3050, Australia. E-mail: morokoff @unimelb.edu.au Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
Epilepsy is a common and disabling symptom of many types of brain tumors, but its pathogenesis is poorly understood.1 Factors such as tumor location, patient age, and histopathological subtype have been linked to the risk of seizures. In particular, low grade (World Health Organization [WHO] II) gliomas are known to have the highest risk of tumor-associated epilepsy (TAE), with rates of 80– 100% reported, whereas high grade (WHO IV) gliomas have a lower risk of 30–40%. Point mutation in the arginine 132 position of the isocitrate dehydrogenase 1 (IDH1) gene (leading to histidine substitution IDH1-R132H) has been reported in 70–80% of WHO grade II gliomas and secondary glioblastomas.2–6 Conversely, the IDH1
1438
1439 IDH1 and Seizures in Low Grade Gliomas mutation is reported in only 3–21% of high grade primary WHO grade IV gliomas.2,6–9 The presence of IDH1 mutation appears to be an independent positive prognostic marker for survival.4,6,10 Although the cause of TAE is multifactorial, there is increasing evidence to implicate glutamate, the major excitatory neurotransmitter in the mammalian central nervous system, in its pathogenesis.11–14 The IDH1-R132H mutation, as well as the rarer homologous IDH2 mutation, both lead to marked reduction of normal enzymatic activity and the accumulation of 2-hydroxyglutarate (2HG), a metabolite that bears a close structural similarity to glutamate.6,15 Recently, an association between IDH1 mutation and seizures in low grade gliomas was reported.16 Herein we hypothesize that the level of expression of IDH1-R132H is linked to seizure risk as well as histopathologic subtype, and report this association in a series of 30 patients with grade II gliomas.
Materials and Methods Patient cohort, sample collection, and clinical data Patients with newly diagnosed supratentorial low-grade (WHO grade II) gliomas admitted to The Royal Melbourne Hospital (RMH) or Melbourne Private Hospital (MPH) from April 1, 2010 to August 31, 2012 were included in the study. Patients were entered prospectively and data were analyzed retrospectively. Patients were excluded if they had a hemorrhagic tumor or had a history of other malignancy elsewhere. Clinical variables collected were the following: patient demographics, presenting signs and symptoms, seizure history and characteristics, epilepsy risk factors, antiepileptic drug (AED) use, and date of surgery. Pathologic variables collected were the following: histopathologic classification and WHO grade, side, and lobar location. Glioma classification and grading was determined by an experienced neuropathologist (M.G.) in the Department of Pathology, The Royal Melbourne Hospital using the WHO Classification of Tumors of the Central Nervous System. TAE was defined as seizures that could be attributable directly to the presence of a supratentorial glioma, and we restricted patients to those with preoperative seizures in order to avoid any potential bias from surgery and postoperative treatment. The study protocol was approved by the Melbourne Health Human Research and Ethics Committee (HREC number 2006.199). Written, informed consent for research participation was obtained from patients directly, or their next-of-kin if the patient was unable to consent. Immunohistochemistry Gliomas were analyzed for the presence of the IDH1R132H mutation in the RMH Department of Pathology, by researchers who were blinded to the clinical presentation. Surgical pathology specimens were fixed in formalin for a minimum of 6 h and up to 48 h. The fixed tissue was processed to paraffin through a series of ethanol solutions
ranging from 70% 100% ethanol followed by xylene and molten paraffin wax using a Tissue-Tek VIP (Sakura-Finetek, Torrance, CA , U.S.A.) tissue processor. Paraffin sections were cut at 3 lm thickness. Slides were routinely stained with hematoxylin and eosin (H&E) using an ST5020 automated stainer (Leica Biosystems, Wetzlar, Hesse, Germany). Immunohistochemistry staining was performed using a Leica Bond Max automated immunohistochemistry stainer (Leica). Slides were air dried at 37°C overnight and then baked at 60°C for 1 h prior to staining. Antigen retrieval using Epitope Retrieval Solution 2 (Leica) for 20 min at 100°C was performed prior to staining with anti-IDH1-R132H antibody (Dianova catalog number DIAH09, Hamburg, Germany) at a 1:600 dilution, and the DS9800 detection system (Leica). The slides were scanned using a ScanScope AT slide scanner (Aperio, Vista, CA) with a 209 objective (Planar Corrected, Apochromatic, Numerical Aperture 0.6) (Fig. 1). The scanned images were visualized using ImageScope software (Aperio), and for each sample, three regions of interest (ROIs) were selected so that the majority of cells in that ROI of H&E section were assessed as being tumor cells, not vessels, microglial cells, or nonneoplastic brain tissue. The resulting images were separated into two images (one representing the hematoxylin/ nuclei stain, the other the DAB-IDH1-R132H stain, Fig. 2) using the Colour Deconvolution module that is part of the Fiji distribution of the ImageJ open-source biologic-imaging analysis software (www.fiji.sc).17 The individual channel images were then inverted, and the number of positive and negative cells was counted using the Multi Wavelength Cell Scoring application module from MetaMorph v7.7.9.0 (Molecular Devices, Sunnyvale, CA, U.S.A.) and the percentage of IDH1 positive tumor cells was calculated. IDH1 mutation was also scored based on immunohistochemistry (IHC) staining using the same antibody as part of routine histopathologic workup by another experienced neuropathologist who was also blinded to the study outcomes. Statistical analysis To compare glioma groups with and without TAE, MannWhitney tests were performed on the semiquantified amounts of IDH-R132H. Sensitivity analysis was used to determine the optimum cutoff level of IDH-R132H in prediction of TAE. Interrater agreement between standard histopathologic versus semiquantitative methods of scoring IDH1 was assessed using the kappa statistic. Analyses were performed using the program Stata12 (StataCorp, College Station, TX, U.S.A.).
Results The expression of the IDH1-R132H was semiquantified in tumor samples of 30 consecutive patients with histologically diagnosed WHO grade II gliomas. The mean age was 35.4 years (range 17–70 years). There Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
1440 S. V. Liubinas et al.
Figure 1. (Left panel) Surgical pathology specimen of a World Health Organization (WHO) grade II astrocytoma prepared as formalin fixed, paraffin embedded section, and stained using mouse monoclonal mouse primary anti-IDH1-R132H antibody (Medical & Biological Laboratories, Nagoya, Chubu, Japan) and 3,3’ diamobenzidine aividin-biotin complex (DAB-ABC) secondary antibody detection kit (Dako, Glostrup, Denmark). Brown staining shows the location of the mutant enzyme in the cytoplasm of tumor cells. The scanned slides were separated into two images using the Colour Deconvolution module that is part of the Fiji distribution of the ImageJ open-source biologicimaging analysis software. One represents the nuclei/nuclei stain (center panel) the other the DAB-IDH1-R132H stain (right panel). Epilepsia ILAE
Figure 2. Immunohistochemical staining for the IDH1-R132H mutation in 30 patients with WHO grade II gliomas. The percentage of positive cells was found to be higher in patients with TAE compared to those without (median and IQR 25.3% {8.6–53.5] vs. 5.2% [0.6– 13.4], p = 0.03). Epilepsia ILAE
were 23 patients with preoperative TAE (77%) and 7 patients without preoperative TAE (23%). There were 22 astrocytomas (73%), 6 oligoastrocytomas (OAs) (23%), one mixed OA and protoplasmic astrocytoma, and one oligodendroglioma. Table 1 summarizes the relevant characteristics of the 30 patients. The percentage expression of IDH1-R132H ranged from zero to 83%. The percentage of IDH1-R132H positive cells was found to be significantly higher in patients with TAE compared to those without TAE (median and IQR 25.3% [8.6–53.5] vs. 5.2% [0.6–13.4], p = 0.03) (Fig. 2). An IDH1-R132H expression level of 14% was determined as the optimum cutoff point (based on Youden’s index) with 69.5% sensitivity and 85.7% specificity in detecting TAE. When the semi-quantitative expression level cutoff point for defining the presence of IDH1 mutation was set at 14%, there was 93.3% agreement with the Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
IDH1 mutation determination by a blinded neuropathologist based on IHC staining with the same antibody (kappa = 0.864). Based on either scoring method, this group of low grade gliomas as a whole, had an IDH1 mutation rate of 17/30 (57%), which is lower but still consistent with reported rates in the literature. The majority of tumors with IDH1 mutation were located in the frontal lobe; however, there was no significant relationship between lobar location and IDH1 mutation (p = 0.58). However, IDH1-R132H mutation–positive tumors were significantly associated with a higher risk of TAE (94.1% vs. 53.9%, p = 0.03). There were eight tumors (27%) with a protoplasmic component. There was no difference in median patient age between the protoplasmic group versus the other types of tumors (p = 0.45). This histologic subtype had a significantly higher median level of mutant IDH1 expression compared to other histologic subtypes (52.2% [IQR 19.9–58.6] vs. 13.8% [IQR 3.9–29.4], p = 0.04). An IDH1R132H level of 32.6% represented the optimum cut off point (75% sensitivity and 86.4% specificity) for protoplasmic component detection. Of the eight tumors that contained an oligodendroglial component, these did not have a significantly different level of IDH1 mutation expression (p = 0.79); however, most of the oligoastrocytomas were observed to be in the midrange of IDH1-R132H expression level (Table 1).
Discussion Seizures are very common in the clinical course of patients with gliomas, and are often not improved by tumor removal or controlled with the use of antiepileptic drugs.18–20 It has been hypothesized, therefore, that molecular alterations within and around the glioma permanently affect the peritumoral brain, from where tumor-associated seizures are thought to arise.13 IDH1 mutations were
1441 IDH1 and Seizures in Low Grade Gliomas Table 1. List of low grade glioma patients ranked by IDH1% S. no.
Sex
Age
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
F F F M F F M F F F M F F M M M M M M F M M F M M F M M F F
28 35 41 67 70 24 39 32 22 32 39 48 25 20 25 27 32 41 28 51 25 37 38 43 41 30 17 27 43 37
TAE + + + + + + + + + + + + + + + + + + + + + + +
Protoplasmic component
Lobe
Histopathology
Temporal Temporal Temporal Temporal Frontal Frontal Frontal Deep Frontal Frontal Frontal Frontal Frontal Occipital Parietal Temporal Frontal Frontal Frontal Parietal Temporal Parietal Frontal Temporal Frontal Frontal Frontal Occipital Frontal Temporal
OA Protoplasmic Diffuse astrocytoma Fibrillary Diffuse astrocytoma Diffuse astrocytoma Fibrillary Diffuse astrocytoma Protoplasmic Diffuse astrocytoma Diffuse astrocytoma OA Fibrillary Diffuse astrocytoma Diffuse astrocytoma OA Diffuse astrocytoma OA OA Oligodendroglioma Diffuse astrocytoma OA + protoplasmic Diffuse astrocytoma Protoplasmic Protoplasmic Diffuse astrocytoma Protoplasmic Protoplasmic OA Protoplasmic
Oligo component
IDH1mut
Y Y
Y
Y
Y Y Y Y Y
Y
Y Y Y Y Y Y
+ + + + + + + + + + + + + + + + +
IDH1% 0.07 0.20 0.20 0.41 0.64 0.95 3.88 5.20 7.30 7.44 8.60 11.30 13.40 14.10 14.60 19.85 22.50 25.30 29.40 29.80 30.06 32.58 49.20 50.83 53.50 53.82 54.73 62.56 68.80 83.40
OA, oligoastrocytoma; IDH1mut, standard scoring by neuropathologist; IDH1%, semiquantitative expression level.
initially sequenced in 22 glioblastomas in 2008 and found to be heterozygous in all cases.21 IDH1/2 mutations were recently reported to be an independent predictor for seizures, as a presenting feature in 79 patients with low-grade gliomas.16 Another study found that the presence of IDH1 mutations did not correlate with seizure incidence, although this study was possibly confounded by the inclusion of some higher grade tumors.10 In both of these studies, IDH1 expression was reported qualitatively as either positive or negative, based on IHC staining. In our current study, we focused on only low grade (WHO II) gliomas and attempted to semi-quantitatively assess the expression of the mutant IDH1-R132H enzyme using automated computer imaging software, and demonstrated that there is significantly increased expression of IDH1-R132H in patients with TAE compared to those with no TAE (p = 0.03). Because this study was based on IHC using a specific IDH1 mutation antibody, we focused only on the most common IDH1 mutation. Future studies using polymerase chain reaction (PCR) to include the mutually exclusive IDH2 mutation in other tumors may show a stronger relationship with TAE as in the previous literature.16 Semiquantitative expression of IDH1
mutation correlated well with standard scoring techniques based on immunohistochemistry performed by a neuropathologist. Whether measurement of the expression percentage is either biologically or clinically meaningful, however, is an open question. The variation in detectable expression levels between tumor cells would be dependent on relative gene and protein expression variation between the normal and mutated alleles; however, the effects of the IDH1 mutation overall would be expected to depend on the cumulative quantitative effects of enzyme function on levels of metabolites. The precise role of the IDH1 mutation in gliomagenesis is not known but is most likely related to the resulting changes in its enzymatic function. Rather than converting isocitrate to a-KG, mutant IDH1 consumes a-KG and produces 2-hydroxyglutarate (2HG) in an NADPH-dependent reaction.15,22 2HG production is found in all IDH1 and IDH2 mutation types.23–26 The role of glutamate in promoting glioma progression, as well as TAE, has been supported by a number of studies.27–30 We recently reported that glutamate levels were elevated and glutamate transporter (EAAT2, system Xc) levels were altered in a large series of Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
1442 S. V. Liubinas et al. human high grade glioma patients with TAE.14 However, metabolic profiling of changes in human oligodendroglioma cells harboring IDH1/2 mutations found lower levels of glutamate, most likely due to depletion during its conversion to a-ketoglutarate and then 2HG.31 It has also been reported that low grade gliomas, especially those with loss of heterozygosity of 1p, have lower activity of system Xc and therefore would be expected to have lower extracellular glutamate.26 These findings suggest that the high incidence of seizures in low grade tumors, unlike in high grade tumors, is not directly related to glutamate levels. One possible explanation is that 2HG itself could be mimicking glutamate: It bears a close structural similarity to glutamate; is able to activate NMDA receptors, leading to increased intracellular Ca2+ and neurodegeneration; and excess 2HG has been indirectly implicated in epileptogenesis. For instance, D-2-hydroxyglutaric aciduria is a rare inherited metabolic disease characterized by an accumulation of 2HG, in which patients present as neonates or in early infancy with developmental delay, hypotonia, and epilepsy.32,33 We also report for the first time, a high level of IDH1 mutation expression in protoplasmic astrocytoma. This rare subtype has regular, small round cells, prominent microcyst formation, ‘bland’ histology, and distinctive magnetic resonance imaging (MRI) characteristics and is known to be associated with a more indolent clinical course and a very high incidence of seizures.34 Protoplasmic astrocytes mature in the cerebral cortex during late embryonic development and early infancy, are known to modulate cortical neuronal function by synchronizing neuronal firing and have recently been shown to express glutamate transporters EAAT1/ 2.35,36 Protoplasmic astrocytes and oligodendrocytes are both thought to have a common origin from NG2+ cells,37 and it has been suggested that IDH1 mutation is an early stem cell event during the development of oligodendroglial and protoplasmic tumors.6 The fact that these cells are involved in glutamate metabolism and the modulation of surrounding neuronal activity makes them likely candidates for involvement in the extreme seizure susceptibility of these tumors. Further studies are needed to better elucidate the complex molecular and metabolic mechanisms involved in seizures associated with gliomas, in particular the role of IDH1 mutation in specific populations of astrocytic progenitors.
Funding National Health and Medical Research Council (Australia) grant number 628301. National Health and Medical Research Council (Australia) postgraduate research scholarship 2012. Royal Australasian College of Surgeons Foundation for Surgical Research Scholarship, 2011 and 2012. Melville Hughes Scholarship, The University of Melbourne, 2010. Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Beaumont A, Whittle IR. The pathogenesis of tumour associated epilepsy. Acta Neurochir (Wien) 2000;142:1–15. 2. Gupta R, Webb-Myers R, Flanagan S, et al. Isocitrate dehydrogenase mutations in diffuse gliomas: clinical and aetiological implications. J Clin Pathol 2011;64:835–844. 3. Balss J, Meyer J, Mueller W, et al. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol 2008;116:597–602. 4. Hartmann C, Hentschel B, Wick W, et al. Patients with IDH1 wild type anaplastic astrocytomas exhibit worse prognosis than IDH1-mutated glioblastomas, and IDH1 mutation status accounts for the unfavorable prognostic effect of higher age: implications for classification of gliomas. Acta Neuropathol 2010;120:707–718. 5. Ichimura K, Pearson D, Kocialkowski S, et al. IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro-Oncology 2009;11:341–347. 6. Yan H, Parsons D, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med 2009;360:765–773. 7. Sanson M, Marie Y, Paris S, et al. Isocitrate dehydrogenase 1 codon 132 mutation is an important prognostic biomarker in gliomas. J Clin Oncol 2009;27:4150–4154. 8. Watanabe T, Nobusawa S, Kleihues P, et al. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol 2009;174:1149–1153. 9. Weller M, Felsberg J, Hartmann C, et al. Molecular predictors of progression-free and overall survival in patients with newly diagnosed glioblastoma: a prospective translational study of the German Glioma Network. J Clin Oncol 2009;27:5743–5750. 10. Dubbink HJ, Taal W, van Marion R, et al. IDH1 mutations in lowgrade astrocytomas predict survival but not response to temozolomide. Neurology 2009;73:1792–1795. 11. Sontheimer H. A role for glutamate in growth and invasion of primary brain tumors. J Neurochem 2008;105:287–295. 12. Shamji MF, Fric-Shamji EC, Benoit BG. Brain tumors and epilepsy: pathophysiology of peritumoral changes. Neurosurg Rev 2009;32: 275–284. 13. Buckingham S, Campbell S, Haas B, et al. Glutamate release by primary brain tumors induces epileptic activity. Nat Med 2011;17:1269–1274. 14. Yuen TI, Morokoff AP, Bjorksten A, et al. Glutamate is associated with a higher risk of seizures in patients with gliomas. Neurology 2012;79:883–889. 15. Dang L, White DW, Gross S, et al. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature 2009;462:739–744. 16. Stockhammer F, Misch M, Helms H-J, et al. IDH1/2 mutations in WHO grade II astrocytomas associated with localization and seizure as the initial symptom. Seizure 2011;21:194–197. 17. Schindelin J, Arganda-Carreras I, Frise E, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 2012;9:676–682. 18. Pace A, Bove L, Innocenti P, et al. Epilepsy and gliomas: incidence and treatment in 119 patients. J Exp Clin Cancer Res 1998;17:479– 482. 19. Behin A, Hoang-Xuan K, Carpentier AF, et al. Primary brain tumours in adults. Lancet 2003;361:323–331. 20. Bartolomei JC, Christopher S, Vives K, et al. Low-grade gliomas of chronic epilepsy: a distinct clinical and pathological entity. J Neurooncol 1997;34:79–84. 21. Parsons D, Jone S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science 2008;321:1807–1812. 22. Prensner J, Chinnaiyan A. Metabolism unhinged: IDH mutations in cancer. Nat Med 2011;17:291–293. 23. Reitman ZJ, Olby NJ, Mariani CL, et al. IDH1 and IDH2 hotspot mutations are not found in canine glioma. Int J Cancer 2010;127:245– 246.
1443 IDH1 and Seizures in Low Grade Gliomas 24. Frezza C, Tennant D, Gottlieb E. IDH1 mutations in gliomas: when an enzyme loses its grip. Cancer Cell 2010;17:7–9. 25. Metellus P, Figarella-Branger D. Magnetic resonance metabloic imaging of glioma. Sci Transl Med 2012;4:1–4. 26. Stockhammer F, von Deimling A, van Landeghem FK. Decreased expression of the active subunit of the cystine/glutamate antiporter xCT is associated with loss of heterozygosity of 1p in oligodendroglial tumours WHO grade II. Histopathology 2009;54:241–247. 27. Roslin M, Henriksson R, Bergstrom P, et al. Baseline levels of glucose metabolites, glutamate and glycerol in malignant glioma assessed by stereotactic microdialysis. J Neurooncol 2003;61:151–160. 28. de Groot JF, Liu TJ, Fuller G, et al. The excitatory amino acid transporter-2 induces apoptosis and decreases glioma growth in vitro and in vivo. Cancer Res 2005;65:1934–1940. 29. Sontheimer H. Ion channels and amino acid transporters support the growth and invasion of primary brain tumors. Mol Neurobiol 2004;29:61–71. 30. Ye ZC, Sontheimer H. Glioma cells release excitotoxic concentrations of glutamate. Cancer Res 1999;59:4383–4391.
31. Reitman ZJ, Jin G, Karoly ED, et al. Profiling the effects of isocitrate dehydrogenase 1 and 2 mutations on the cellular metabolome. Proc Natl Acad Sci USA 2011;108:3270–3275. 32. Kolker S, Pawlak V, Ahlemeyer B, et al. NMDA receptor activation and respiratory chain complex V inhibition contribute to neurodegeneration in d-2-hydroxyglutaric aciduria. Eur J Neurosci 2002;16:21–28. 33. van der Knaap MS, Jakobs C, Hoffmann GF, et al. D-2hydroxyglutaric aciduria: further clinical delineation. J Inherit Metab Dis 1999;22:404–413. 34. Tay KL, Tsui A, Phal PM, et al. MR imaging characteristics of protoplasmic astrocytomas. Neuroradiology 2011;53:405–411. 35. DeSilva TM, Borenstein NS, Volpe JJ, et al. Expression of EAAT2 in neurons and protoplasmic astrocytes during human cortical development. J Comp Neurol 2012;520:3912–3932. 36. Fellin T, Haydon PG. Do astrocytes contribute to excitation underlying seizures? Trends Mol Med 2005;11:530–533. 37. Trotter J, Karram K, Nishiyama A. NG2 cells: properties, progeny and origin. Brain Res Rev 2010;63:72–82.
Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
IDH1 mutation is associated with seizures and protoplasmic subtype in patients with low-grade gliomas *†Simon V. Liubinas, †Giovanna M. D’Abaco, ‡Bradford M. Moffat, §Michael Gonzales, §Frank Feleppa, ¶Cameron J. Nowell, **Alexandra Gorelik, *†Katharine J. Drummond, ††‡‡Terence J. O’Brien, *†Andrew H. Kaye, and *†Andrew P. Morokoff Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
SUMMARY
Dr. Simon Liubinas is a neurosurgical trainee in Melbourne, Australia.
Objective: The isocitrate dehydrogenase 1 (IDH1) R132H mutation is the most common mutation in World Health Organization (WHO) grade II gliomas, reported to be expressed in 70–80%, but only 5–10% of high grade gliomas. Low grade tumors, especially the protoplasmic subtype, have the highest incidence of tumor associated epilepsy (TAE). The IDH1 mutation leads to the accumulation of 2-hydroxyglutarate (2HG), a metabolite that bears a close structural similarity to glutamate, an excitatory neurotransmitter that has been implicated in the pathogenesis of TAE. We hypothesized that expression of mutated IDH1 may play a role in the pathogenesis of TAE in low grade gliomas. Methods: Thirty consecutive patients with WHO grade II gliomas were analyzed for the presence of the IDH1-R132H mutation using immunohistochemistry. The expression of IDH1 mutation was semiquantified using open-source biologic-imaging analysis software. Results: The percentage of cells positive for the IDH1-R132H mutation was found to be higher in patients with TAE compared to those without TAE (median and interquartile range (IQR) 25.3% [8.6–53.5] vs. 5.2% [0.6–13.4], p = 0.03). In addition, we found a significantly higher median IDH1 mutation expression level in the protoplasmic subtype of low grade glioma (52.2% [IQR 19.9–58.6] vs. 13.8% [IQR 3.9–29.4], p = 0.04). Significance: Increased expression of the IDH1-R132H mutation is associated with seizures in low grade gliomas and also with the protoplasmic subtype. This supports the hypothesis that this mutation may play a role in the pathogenesis of both TAE and low grade gliomas. KEY WORDS: Glioma, IDH1, Seizure, Epilepsy, Protoplasmic astrocytoma, Glutamate.
Accepted April 17, 2014; Early View publication June 5, 2014. *Department of Neurosurgery, The Royal Melbourne Hospital, Parkville, Victoria, Australia; †The Department of Surgery (RMH), The University of Melbourne, Parkville, Victoria, Australia; ‡The Department of Radiology (RMH/WH), The University of Melbourne, Parkville, Victoria, Australia; §Department of Pathology, The Royal Melbourne Hospital, Parkville, Victoria, Australia; ¶Ludwig Institute for Cancer Research, Melbourne, Parkville, Victoria, Australia; **The Melbourne Epicentre, The Royal Melbourne Hospital, Parkville, Victoria, Australia; ††The Department of Medicine (RMH), The University of Melbourne, Parkville, Victoria, Australia; and ‡‡Department of Neurology, The Royal Melbourne Hospital, Parkville, Victoria, Australia Address correspondence to Andrew Morokoff, Department of Surgery, Royal Melbourne Hospital, Parkville,VIC 3050, Australia. E-mail: morokoff @unimelb.edu.au Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
Epilepsy is a common and disabling symptom of many types of brain tumors, but its pathogenesis is poorly understood.1 Factors such as tumor location, patient age, and histopathological subtype have been linked to the risk of seizures. In particular, low grade (World Health Organization [WHO] II) gliomas are known to have the highest risk of tumor-associated epilepsy (TAE), with rates of 80– 100% reported, whereas high grade (WHO IV) gliomas have a lower risk of 30–40%. Point mutation in the arginine 132 position of the isocitrate dehydrogenase 1 (IDH1) gene (leading to histidine substitution IDH1-R132H) has been reported in 70–80% of WHO grade II gliomas and secondary glioblastomas.2–6 Conversely, the IDH1
1438
1439 IDH1 and Seizures in Low Grade Gliomas mutation is reported in only 3–21% of high grade primary WHO grade IV gliomas.2,6–9 The presence of IDH1 mutation appears to be an independent positive prognostic marker for survival.4,6,10 Although the cause of TAE is multifactorial, there is increasing evidence to implicate glutamate, the major excitatory neurotransmitter in the mammalian central nervous system, in its pathogenesis.11–14 The IDH1-R132H mutation, as well as the rarer homologous IDH2 mutation, both lead to marked reduction of normal enzymatic activity and the accumulation of 2-hydroxyglutarate (2HG), a metabolite that bears a close structural similarity to glutamate.6,15 Recently, an association between IDH1 mutation and seizures in low grade gliomas was reported.16 Herein we hypothesize that the level of expression of IDH1-R132H is linked to seizure risk as well as histopathologic subtype, and report this association in a series of 30 patients with grade II gliomas.
Materials and Methods Patient cohort, sample collection, and clinical data Patients with newly diagnosed supratentorial low-grade (WHO grade II) gliomas admitted to The Royal Melbourne Hospital (RMH) or Melbourne Private Hospital (MPH) from April 1, 2010 to August 31, 2012 were included in the study. Patients were entered prospectively and data were analyzed retrospectively. Patients were excluded if they had a hemorrhagic tumor or had a history of other malignancy elsewhere. Clinical variables collected were the following: patient demographics, presenting signs and symptoms, seizure history and characteristics, epilepsy risk factors, antiepileptic drug (AED) use, and date of surgery. Pathologic variables collected were the following: histopathologic classification and WHO grade, side, and lobar location. Glioma classification and grading was determined by an experienced neuropathologist (M.G.) in the Department of Pathology, The Royal Melbourne Hospital using the WHO Classification of Tumors of the Central Nervous System. TAE was defined as seizures that could be attributable directly to the presence of a supratentorial glioma, and we restricted patients to those with preoperative seizures in order to avoid any potential bias from surgery and postoperative treatment. The study protocol was approved by the Melbourne Health Human Research and Ethics Committee (HREC number 2006.199). Written, informed consent for research participation was obtained from patients directly, or their next-of-kin if the patient was unable to consent. Immunohistochemistry Gliomas were analyzed for the presence of the IDH1R132H mutation in the RMH Department of Pathology, by researchers who were blinded to the clinical presentation. Surgical pathology specimens were fixed in formalin for a minimum of 6 h and up to 48 h. The fixed tissue was processed to paraffin through a series of ethanol solutions
ranging from 70% 100% ethanol followed by xylene and molten paraffin wax using a Tissue-Tek VIP (Sakura-Finetek, Torrance, CA , U.S.A.) tissue processor. Paraffin sections were cut at 3 lm thickness. Slides were routinely stained with hematoxylin and eosin (H&E) using an ST5020 automated stainer (Leica Biosystems, Wetzlar, Hesse, Germany). Immunohistochemistry staining was performed using a Leica Bond Max automated immunohistochemistry stainer (Leica). Slides were air dried at 37°C overnight and then baked at 60°C for 1 h prior to staining. Antigen retrieval using Epitope Retrieval Solution 2 (Leica) for 20 min at 100°C was performed prior to staining with anti-IDH1-R132H antibody (Dianova catalog number DIAH09, Hamburg, Germany) at a 1:600 dilution, and the DS9800 detection system (Leica). The slides were scanned using a ScanScope AT slide scanner (Aperio, Vista, CA) with a 209 objective (Planar Corrected, Apochromatic, Numerical Aperture 0.6) (Fig. 1). The scanned images were visualized using ImageScope software (Aperio), and for each sample, three regions of interest (ROIs) were selected so that the majority of cells in that ROI of H&E section were assessed as being tumor cells, not vessels, microglial cells, or nonneoplastic brain tissue. The resulting images were separated into two images (one representing the hematoxylin/ nuclei stain, the other the DAB-IDH1-R132H stain, Fig. 2) using the Colour Deconvolution module that is part of the Fiji distribution of the ImageJ open-source biologic-imaging analysis software (www.fiji.sc).17 The individual channel images were then inverted, and the number of positive and negative cells was counted using the Multi Wavelength Cell Scoring application module from MetaMorph v7.7.9.0 (Molecular Devices, Sunnyvale, CA, U.S.A.) and the percentage of IDH1 positive tumor cells was calculated. IDH1 mutation was also scored based on immunohistochemistry (IHC) staining using the same antibody as part of routine histopathologic workup by another experienced neuropathologist who was also blinded to the study outcomes. Statistical analysis To compare glioma groups with and without TAE, MannWhitney tests were performed on the semiquantified amounts of IDH-R132H. Sensitivity analysis was used to determine the optimum cutoff level of IDH-R132H in prediction of TAE. Interrater agreement between standard histopathologic versus semiquantitative methods of scoring IDH1 was assessed using the kappa statistic. Analyses were performed using the program Stata12 (StataCorp, College Station, TX, U.S.A.).
Results The expression of the IDH1-R132H was semiquantified in tumor samples of 30 consecutive patients with histologically diagnosed WHO grade II gliomas. The mean age was 35.4 years (range 17–70 years). There Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
1440 S. V. Liubinas et al.
Figure 1. (Left panel) Surgical pathology specimen of a World Health Organization (WHO) grade II astrocytoma prepared as formalin fixed, paraffin embedded section, and stained using mouse monoclonal mouse primary anti-IDH1-R132H antibody (Medical & Biological Laboratories, Nagoya, Chubu, Japan) and 3,3’ diamobenzidine aividin-biotin complex (DAB-ABC) secondary antibody detection kit (Dako, Glostrup, Denmark). Brown staining shows the location of the mutant enzyme in the cytoplasm of tumor cells. The scanned slides were separated into two images using the Colour Deconvolution module that is part of the Fiji distribution of the ImageJ open-source biologicimaging analysis software. One represents the nuclei/nuclei stain (center panel) the other the DAB-IDH1-R132H stain (right panel). Epilepsia ILAE
Figure 2. Immunohistochemical staining for the IDH1-R132H mutation in 30 patients with WHO grade II gliomas. The percentage of positive cells was found to be higher in patients with TAE compared to those without (median and IQR 25.3% {8.6–53.5] vs. 5.2% [0.6– 13.4], p = 0.03). Epilepsia ILAE
were 23 patients with preoperative TAE (77%) and 7 patients without preoperative TAE (23%). There were 22 astrocytomas (73%), 6 oligoastrocytomas (OAs) (23%), one mixed OA and protoplasmic astrocytoma, and one oligodendroglioma. Table 1 summarizes the relevant characteristics of the 30 patients. The percentage expression of IDH1-R132H ranged from zero to 83%. The percentage of IDH1-R132H positive cells was found to be significantly higher in patients with TAE compared to those without TAE (median and IQR 25.3% [8.6–53.5] vs. 5.2% [0.6–13.4], p = 0.03) (Fig. 2). An IDH1-R132H expression level of 14% was determined as the optimum cutoff point (based on Youden’s index) with 69.5% sensitivity and 85.7% specificity in detecting TAE. When the semi-quantitative expression level cutoff point for defining the presence of IDH1 mutation was set at 14%, there was 93.3% agreement with the Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
IDH1 mutation determination by a blinded neuropathologist based on IHC staining with the same antibody (kappa = 0.864). Based on either scoring method, this group of low grade gliomas as a whole, had an IDH1 mutation rate of 17/30 (57%), which is lower but still consistent with reported rates in the literature. The majority of tumors with IDH1 mutation were located in the frontal lobe; however, there was no significant relationship between lobar location and IDH1 mutation (p = 0.58). However, IDH1-R132H mutation–positive tumors were significantly associated with a higher risk of TAE (94.1% vs. 53.9%, p = 0.03). There were eight tumors (27%) with a protoplasmic component. There was no difference in median patient age between the protoplasmic group versus the other types of tumors (p = 0.45). This histologic subtype had a significantly higher median level of mutant IDH1 expression compared to other histologic subtypes (52.2% [IQR 19.9–58.6] vs. 13.8% [IQR 3.9–29.4], p = 0.04). An IDH1R132H level of 32.6% represented the optimum cut off point (75% sensitivity and 86.4% specificity) for protoplasmic component detection. Of the eight tumors that contained an oligodendroglial component, these did not have a significantly different level of IDH1 mutation expression (p = 0.79); however, most of the oligoastrocytomas were observed to be in the midrange of IDH1-R132H expression level (Table 1).
Discussion Seizures are very common in the clinical course of patients with gliomas, and are often not improved by tumor removal or controlled with the use of antiepileptic drugs.18–20 It has been hypothesized, therefore, that molecular alterations within and around the glioma permanently affect the peritumoral brain, from where tumor-associated seizures are thought to arise.13 IDH1 mutations were
1441 IDH1 and Seizures in Low Grade Gliomas Table 1. List of low grade glioma patients ranked by IDH1% S. no.
Sex
Age
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
F F F M F F M F F F M F F M M M M M M F M M F M M F M M F F
28 35 41 67 70 24 39 32 22 32 39 48 25 20 25 27 32 41 28 51 25 37 38 43 41 30 17 27 43 37
TAE + + + + + + + + + + + + + + + + + + + + + + +
Protoplasmic component
Lobe
Histopathology
Temporal Temporal Temporal Temporal Frontal Frontal Frontal Deep Frontal Frontal Frontal Frontal Frontal Occipital Parietal Temporal Frontal Frontal Frontal Parietal Temporal Parietal Frontal Temporal Frontal Frontal Frontal Occipital Frontal Temporal
OA Protoplasmic Diffuse astrocytoma Fibrillary Diffuse astrocytoma Diffuse astrocytoma Fibrillary Diffuse astrocytoma Protoplasmic Diffuse astrocytoma Diffuse astrocytoma OA Fibrillary Diffuse astrocytoma Diffuse astrocytoma OA Diffuse astrocytoma OA OA Oligodendroglioma Diffuse astrocytoma OA + protoplasmic Diffuse astrocytoma Protoplasmic Protoplasmic Diffuse astrocytoma Protoplasmic Protoplasmic OA Protoplasmic
Oligo component
IDH1mut
Y Y
Y
Y
Y Y Y Y Y
Y
Y Y Y Y Y Y
+ + + + + + + + + + + + + + + + +
IDH1% 0.07 0.20 0.20 0.41 0.64 0.95 3.88 5.20 7.30 7.44 8.60 11.30 13.40 14.10 14.60 19.85 22.50 25.30 29.40 29.80 30.06 32.58 49.20 50.83 53.50 53.82 54.73 62.56 68.80 83.40
OA, oligoastrocytoma; IDH1mut, standard scoring by neuropathologist; IDH1%, semiquantitative expression level.
initially sequenced in 22 glioblastomas in 2008 and found to be heterozygous in all cases.21 IDH1/2 mutations were recently reported to be an independent predictor for seizures, as a presenting feature in 79 patients with low-grade gliomas.16 Another study found that the presence of IDH1 mutations did not correlate with seizure incidence, although this study was possibly confounded by the inclusion of some higher grade tumors.10 In both of these studies, IDH1 expression was reported qualitatively as either positive or negative, based on IHC staining. In our current study, we focused on only low grade (WHO II) gliomas and attempted to semi-quantitatively assess the expression of the mutant IDH1-R132H enzyme using automated computer imaging software, and demonstrated that there is significantly increased expression of IDH1-R132H in patients with TAE compared to those with no TAE (p = 0.03). Because this study was based on IHC using a specific IDH1 mutation antibody, we focused only on the most common IDH1 mutation. Future studies using polymerase chain reaction (PCR) to include the mutually exclusive IDH2 mutation in other tumors may show a stronger relationship with TAE as in the previous literature.16 Semiquantitative expression of IDH1
mutation correlated well with standard scoring techniques based on immunohistochemistry performed by a neuropathologist. Whether measurement of the expression percentage is either biologically or clinically meaningful, however, is an open question. The variation in detectable expression levels between tumor cells would be dependent on relative gene and protein expression variation between the normal and mutated alleles; however, the effects of the IDH1 mutation overall would be expected to depend on the cumulative quantitative effects of enzyme function on levels of metabolites. The precise role of the IDH1 mutation in gliomagenesis is not known but is most likely related to the resulting changes in its enzymatic function. Rather than converting isocitrate to a-KG, mutant IDH1 consumes a-KG and produces 2-hydroxyglutarate (2HG) in an NADPH-dependent reaction.15,22 2HG production is found in all IDH1 and IDH2 mutation types.23–26 The role of glutamate in promoting glioma progression, as well as TAE, has been supported by a number of studies.27–30 We recently reported that glutamate levels were elevated and glutamate transporter (EAAT2, system Xc) levels were altered in a large series of Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
1442 S. V. Liubinas et al. human high grade glioma patients with TAE.14 However, metabolic profiling of changes in human oligodendroglioma cells harboring IDH1/2 mutations found lower levels of glutamate, most likely due to depletion during its conversion to a-ketoglutarate and then 2HG.31 It has also been reported that low grade gliomas, especially those with loss of heterozygosity of 1p, have lower activity of system Xc and therefore would be expected to have lower extracellular glutamate.26 These findings suggest that the high incidence of seizures in low grade tumors, unlike in high grade tumors, is not directly related to glutamate levels. One possible explanation is that 2HG itself could be mimicking glutamate: It bears a close structural similarity to glutamate; is able to activate NMDA receptors, leading to increased intracellular Ca2+ and neurodegeneration; and excess 2HG has been indirectly implicated in epileptogenesis. For instance, D-2-hydroxyglutaric aciduria is a rare inherited metabolic disease characterized by an accumulation of 2HG, in which patients present as neonates or in early infancy with developmental delay, hypotonia, and epilepsy.32,33 We also report for the first time, a high level of IDH1 mutation expression in protoplasmic astrocytoma. This rare subtype has regular, small round cells, prominent microcyst formation, ‘bland’ histology, and distinctive magnetic resonance imaging (MRI) characteristics and is known to be associated with a more indolent clinical course and a very high incidence of seizures.34 Protoplasmic astrocytes mature in the cerebral cortex during late embryonic development and early infancy, are known to modulate cortical neuronal function by synchronizing neuronal firing and have recently been shown to express glutamate transporters EAAT1/ 2.35,36 Protoplasmic astrocytes and oligodendrocytes are both thought to have a common origin from NG2+ cells,37 and it has been suggested that IDH1 mutation is an early stem cell event during the development of oligodendroglial and protoplasmic tumors.6 The fact that these cells are involved in glutamate metabolism and the modulation of surrounding neuronal activity makes them likely candidates for involvement in the extreme seizure susceptibility of these tumors. Further studies are needed to better elucidate the complex molecular and metabolic mechanisms involved in seizures associated with gliomas, in particular the role of IDH1 mutation in specific populations of astrocytic progenitors.
Funding National Health and Medical Research Council (Australia) grant number 628301. National Health and Medical Research Council (Australia) postgraduate research scholarship 2012. Royal Australasian College of Surgeons Foundation for Surgical Research Scholarship, 2011 and 2012. Melville Hughes Scholarship, The University of Melbourne, 2010. Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Beaumont A, Whittle IR. The pathogenesis of tumour associated epilepsy. Acta Neurochir (Wien) 2000;142:1–15. 2. Gupta R, Webb-Myers R, Flanagan S, et al. Isocitrate dehydrogenase mutations in diffuse gliomas: clinical and aetiological implications. J Clin Pathol 2011;64:835–844. 3. Balss J, Meyer J, Mueller W, et al. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol 2008;116:597–602. 4. Hartmann C, Hentschel B, Wick W, et al. Patients with IDH1 wild type anaplastic astrocytomas exhibit worse prognosis than IDH1-mutated glioblastomas, and IDH1 mutation status accounts for the unfavorable prognostic effect of higher age: implications for classification of gliomas. Acta Neuropathol 2010;120:707–718. 5. Ichimura K, Pearson D, Kocialkowski S, et al. IDH1 mutations are present in the majority of common adult gliomas but rare in primary glioblastomas. Neuro-Oncology 2009;11:341–347. 6. Yan H, Parsons D, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med 2009;360:765–773. 7. Sanson M, Marie Y, Paris S, et al. Isocitrate dehydrogenase 1 codon 132 mutation is an important prognostic biomarker in gliomas. J Clin Oncol 2009;27:4150–4154. 8. Watanabe T, Nobusawa S, Kleihues P, et al. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol 2009;174:1149–1153. 9. Weller M, Felsberg J, Hartmann C, et al. Molecular predictors of progression-free and overall survival in patients with newly diagnosed glioblastoma: a prospective translational study of the German Glioma Network. J Clin Oncol 2009;27:5743–5750. 10. Dubbink HJ, Taal W, van Marion R, et al. IDH1 mutations in lowgrade astrocytomas predict survival but not response to temozolomide. Neurology 2009;73:1792–1795. 11. Sontheimer H. A role for glutamate in growth and invasion of primary brain tumors. J Neurochem 2008;105:287–295. 12. Shamji MF, Fric-Shamji EC, Benoit BG. Brain tumors and epilepsy: pathophysiology of peritumoral changes. Neurosurg Rev 2009;32: 275–284. 13. Buckingham S, Campbell S, Haas B, et al. Glutamate release by primary brain tumors induces epileptic activity. Nat Med 2011;17:1269–1274. 14. Yuen TI, Morokoff AP, Bjorksten A, et al. Glutamate is associated with a higher risk of seizures in patients with gliomas. Neurology 2012;79:883–889. 15. Dang L, White DW, Gross S, et al. Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature 2009;462:739–744. 16. Stockhammer F, Misch M, Helms H-J, et al. IDH1/2 mutations in WHO grade II astrocytomas associated with localization and seizure as the initial symptom. Seizure 2011;21:194–197. 17. Schindelin J, Arganda-Carreras I, Frise E, et al. Fiji: an open-source platform for biological-image analysis. Nat Methods 2012;9:676–682. 18. Pace A, Bove L, Innocenti P, et al. Epilepsy and gliomas: incidence and treatment in 119 patients. J Exp Clin Cancer Res 1998;17:479– 482. 19. Behin A, Hoang-Xuan K, Carpentier AF, et al. Primary brain tumours in adults. Lancet 2003;361:323–331. 20. Bartolomei JC, Christopher S, Vives K, et al. Low-grade gliomas of chronic epilepsy: a distinct clinical and pathological entity. J Neurooncol 1997;34:79–84. 21. Parsons D, Jone S, Zhang X, et al. An integrated genomic analysis of human glioblastoma multiforme. Science 2008;321:1807–1812. 22. Prensner J, Chinnaiyan A. Metabolism unhinged: IDH mutations in cancer. Nat Med 2011;17:291–293. 23. Reitman ZJ, Olby NJ, Mariani CL, et al. IDH1 and IDH2 hotspot mutations are not found in canine glioma. Int J Cancer 2010;127:245– 246.
1443 IDH1 and Seizures in Low Grade Gliomas 24. Frezza C, Tennant D, Gottlieb E. IDH1 mutations in gliomas: when an enzyme loses its grip. Cancer Cell 2010;17:7–9. 25. Metellus P, Figarella-Branger D. Magnetic resonance metabloic imaging of glioma. Sci Transl Med 2012;4:1–4. 26. Stockhammer F, von Deimling A, van Landeghem FK. Decreased expression of the active subunit of the cystine/glutamate antiporter xCT is associated with loss of heterozygosity of 1p in oligodendroglial tumours WHO grade II. Histopathology 2009;54:241–247. 27. Roslin M, Henriksson R, Bergstrom P, et al. Baseline levels of glucose metabolites, glutamate and glycerol in malignant glioma assessed by stereotactic microdialysis. J Neurooncol 2003;61:151–160. 28. de Groot JF, Liu TJ, Fuller G, et al. The excitatory amino acid transporter-2 induces apoptosis and decreases glioma growth in vitro and in vivo. Cancer Res 2005;65:1934–1940. 29. Sontheimer H. Ion channels and amino acid transporters support the growth and invasion of primary brain tumors. Mol Neurobiol 2004;29:61–71. 30. Ye ZC, Sontheimer H. Glioma cells release excitotoxic concentrations of glutamate. Cancer Res 1999;59:4383–4391.
31. Reitman ZJ, Jin G, Karoly ED, et al. Profiling the effects of isocitrate dehydrogenase 1 and 2 mutations on the cellular metabolome. Proc Natl Acad Sci USA 2011;108:3270–3275. 32. Kolker S, Pawlak V, Ahlemeyer B, et al. NMDA receptor activation and respiratory chain complex V inhibition contribute to neurodegeneration in d-2-hydroxyglutaric aciduria. Eur J Neurosci 2002;16:21–28. 33. van der Knaap MS, Jakobs C, Hoffmann GF, et al. D-2hydroxyglutaric aciduria: further clinical delineation. J Inherit Metab Dis 1999;22:404–413. 34. Tay KL, Tsui A, Phal PM, et al. MR imaging characteristics of protoplasmic astrocytomas. Neuroradiology 2011;53:405–411. 35. DeSilva TM, Borenstein NS, Volpe JJ, et al. Expression of EAAT2 in neurons and protoplasmic astrocytes during human cortical development. J Comp Neurol 2012;520:3912–3932. 36. Fellin T, Haydon PG. Do astrocytes contribute to excitation underlying seizures? Trends Mol Med 2005;11:530–533. 37. Trotter J, Karram K, Nishiyama A. NG2 cells: properties, progeny and origin. Brain Res Rev 2010;63:72–82.
Epilepsia, 55(9):1438–1443, 2014 doi: 10.1111/epi.12662
