SEARCH

SEARCH BY CITATION

Keywords:

  • Brain tumor;
  • Tumor-related epilepsy;
  • Surgical treatment;
  • Medical treatment

Summary

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References

Epilepsy surgery represents the main treatment option for epileptogenic brain tumors. Scalp video–electroencephalography (EEG) and magnetic resonance imaging (MRI) may suffice for defining lesional area and seizure-onset zone in discrete, surgically resectable lesions. The choice of timing for surgery requires a multidisciplinary evaluation, especially in children, when a “wait and see” approach is chosen. Discordant electroclinical and neuroimaging data and an ill-defined epileptogenic lesion require invasive investigations. A multimodal integrated approach may maximize the extent of resection while preserving cerebral function in the eloquent cortex. Radical removal of the tumor is the most important predictor of seizure freedom. Additional predictors include histopathology, age at surgery, duration of epilepsy, and seizure type. Patients with brain tumors are highly vulnerable in relation to the frequent drug-resistance of seizures, the potential interactions between antiepileptic drugs (AEDs) and chemotherapeutic agents (CMTs), and the risk of AED-related cognitive adverse events (24% higher than in the rest of the epilepsy population), in addition to brain damage resulting from tumor itself, surgery, and radiotherapy. No robust, randomized, controlled evidence supports the choice of AEDs for the treatment of seizures in patients with brain tumors. Newer AEDs have limited or no enzyme-inducing profile, prevalent renal excretion, lower plasma protein binding and, consequently, fewer interactions with CMTs. Enzyme-inducing AEDs can lower serum levels of concomitantly administered CMTs. Class I evidence suggests that in patients with brain tumors who do not have a history of seizures, prophylactic use of AEDs is neutral or ineffective.

Seizures complicate the clinical course of >80% of patients with low-grade gliomas (Rudà et al., 2012). In children, tumoral etiology accounts for the 17–38% of syndromes amenable to epilepsy surgery and represents a positive predictive factor for postoperative seizure freedom (Spencer & Huh, 2008). Glioneuronal tumors, such as gangliogliomas (GGs) and dysembryoplastic neuroepithelial tumors (DNTs), are typically associated with pharmacoresistant epilepsy in up to 90% of patients (Rudà et al., 2012). They occur predominantly in children and young adults and in the temporal lobe. Other rare grade I gliomas, such as supratentorial pilocytic astrocytomas, pleomorphic xanthoastrocytomas, and angiocentric gliomas—which prevail in children and young adults—are also frequently accompanied by drug-resistant seizures (Rudà et al., 2012). Diffuse low grade gliomas (LGGs) (World Health Organization [WHO] grade II astrocytomas, oligodendrogliomas, and oligoastrocytomas) are associated with seizures in 60%–88% of patients (Rudà et al., 2012).

Perspectives on Surgical Treatment

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References

Epilepsy surgery usually represents the first-choice treatment option for epileptogenic brain tumors. When the lesion is well defined and surgically resectable, presurgical investigations including history, scalp video–electroencephalography (EEG), and magnetic resonance imaging (MRI) may help with better defining the relationship between lesional area and seizure onset zone. There is no consensus as to whether additional noninvasive investigations, such as single-photon emission computed tomography (SPECT) and positron emission tomography (PET), are of use in defining extent of resection. Moreover, access to these techniques is limited to a few epilepsy surgery centers with dedicated expertise (Harvey et al., 2008). In pediatric surgical candidates, neuropsychological assessment is crucial for monitoring possible deleterious consequences of long-lasting drug-resistant seizures (Van Schooneveld & Braun, 2013). In one study, children with tumoral epilepsy exhibited higher presurgical intelligence quotient (IQ) scores compared to other etiologies, with limited postoperative changes (D'Argenzio et al., 2011). The choice of optimal timing for surgery requires a multidisciplinary evaluation by a specialized pediatric team, especially in patients in whom a “wait and see” approach is chosen (see Fig. 1 for an illustrative case report).

image

Figure 1. Case 1. Fifteen-year-old girl with uneventful personal and family history and normal development. Two generalized tonic–clonic seizures occurred at age 10, followed by daily drug-resistant complex partial seizures, during which she was unable to understand “what people are saying.” She exhibited eye and head deviation to the left, staring, unresponsiveness, and postictal aphasia. MRI revealed a left temporal lesion that was suspected to be a low-grade glial tumor. Neuropsychology revealed normal cognition and fluent aphasia. At age 11, lesionectomy was performed. Histopathology revealed WHO II astrocytoma. Three years after surgery the patient is seizure-free on carbamazepine. (A, B) Preoperative MRI (axial FLAIR and coronal T2-weighted images) show a left temporal subcortical cystic lesion (white arrows). (C) Functional MRI discloses activated areas in the left opercular and posterior temporal regions during a word comprehension task. (D) Preoperative interictal scalp EEG showed subcontinuous sharp and slow waves over the left temporal leads (see red arrow). (E, F) Postoperative MRI (axial FLAIR and coronal T2-weighted images) demonstrate complete removal of the lesion (white arrows). (G) Histology: hematoxylin-eosin (H&E) staining reveal moderately increased cellularity of astrocytes. (H, I, J) Verbal fluency, naming, and verbal memory evaluated preoperatively and at 6, 12, and 24 months after surgery. Language and verbal memory deficits appear significantly improved at 2-year follow-up.

Download figure to PowerPoint

In complex cases in which electroclinical data are discordant with neuroimaging or the borders of the epileptogenic lesion are ill-defined (Spencer & Huh, 2008), invasive investigations are often needed to define extent of resection. Intracranial recordings can also be helpful to better define the relationship between the lesion and the epileptogenic zone in focal cortical dysplasia (FCD) type IIIb (i.e., abnormal cortical lamination adjacent to a glial or glioneuronal tumor) (Blümcke et al., 2011). The incomplete removal of the cortical dysplasia adjacent to GGs or DNTs represents the major cause of surgical failures in these tumors (Rudà et al., 2012; Barba et al., 2011; see Figs. 2 and 3 for an illustrative case). The role of acute electrocorticography (ECoG) in defining the area of resection remains to be established, as no specific electrographic pattern has been described in tumoral tissues (Palmini et al., 1995; Barba et al., 2011). Epileptogenic tumors can also be associated with hippocampal sclerosis (HS) (dual pathology) or to a second independent lesion other than HS (double pathology) (Blümcke et al., 2011). In such cases a comprehensive presurgical evaluation is needed to identify the “true” epileptogenic lesion.

image

Figure 2. Case 2. Eight-year-old girl with uneventful personal and family history. At age 6 months, weekly episodes of brief, motionless staring appeared. At age 17 months, spasms also appeared. EEG revealed bilateral asynchronous spikes and spike and wave discharges. MRI revealed a right temporal cystic lesion suspected to be a tumor. A right anterior temporal lobectomy was planned, but occurrence of a subdural hematoma prevented the neurosurgeons from completing the procedure. Histology revealed FCD type IB. The patient remained seizure-free for 8 months, and then both spasms and complex partial seizures reappeared. Neuropsychology demonstrated mild delay and impaired language skills with attention deficit. To better define the extent of the resection, invasive recordings were performed. At age 3, she underwent an extended right temporal lobectomy. Histopathology revealed a ganglioglioma. The patient is seizure-free 5 years after surgery and is off medication. (A) Preoperative MRI: axial FLAIR image showing grey–white matter junction blurring, abnormal gyration, and a cystic abnormality in the right temporal lobe (white arrow). (B) Axial T1-weighted postsurgical MRI (first surgery) showing the remnant lesional area in the medial and posterior temporal lobe (white arrrow). (C) Histology (first surgery): abnormal radial lamination with microcolumnar arranged neurons with small diameter (black arrows). Nissl staining. (D) Sagittal T1-weighted postsurgical MRI (second surgery) showing complete right temporal lobectomy. (E) Histology (second surgery): biphasic pattern with neoplastic glial cells and dysplastic neurons (black arrows). H&E staining. (F) CD34 immunoreactive tumor cells.

Download figure to PowerPoint

image

Figure 3. Case 2. (A) Ictal scalp EEG showing spasms (red arrows). (B) Postimplantation x-ray disclosing intracranial electrodes. (C) Intracranial EEG recordings. T and L contacts: very low voltage background activity with rhythmic bursts of spikes. T0 leads: subcontinuous slow activity intermingled with spikes. The red arrow indicates a spasm. The contacts are labeled as in the three-dimensional (3D) image. (D) 3D image by digital reconstruction of MRI sequences for neuronavigation: the exact location of each contact of the two grids and of the depth electrode are indicated by the green full circles. The area of resection is in violet and includes the tumor (black line), the epileptogenic zone (red line), and part of the irritative zone (green line).

Download figure to PowerPoint

When epileptogenic tumors are localized in proximity to eloquent areas, a multimodal integrated diagnostic approach may help maximize the extent of resection while preserving, or even improving, cerebral function (Duffau et al., 2003). Intraoperative mapping, during sedation or while awake in collaborative patients, is still the gold standard for surgery of gliomas in eloquent areas, as it can help identifying functional cortical areas as well as subcortical tracts (Duffau et al., 2003). However, functional neuroimaging and fiber tractography are increasingly used to identify eloquent areas before surgery and can be integrated into a multimodal neuronavigational system during resection, thus reducing the need for intraoperative monitoring in selected cases (Duffau et al., 2003). It is likely that ultrahigh field MRI, through improved yield of functional study and higher signal-to-noise ratio, will increase the likelihood of complete removal and accuracy of presurgical assessment in the eloquent cortex.

Prognosis

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References

The prognostic factors for seizure freedom after surgery of epileptogenic brain tumors include histopathology, age at surgery, duration of epilepsy, complex versus simple partial seizures, and completeness of resection (Aronica et al., 2001; Englot et al., 2011; Southwell et al., 2012).

The histologic types usually associated with long-lasting epilepsy are low grade tumors such as glioneuronal tumors (i.e., gangliogliomas and DNTs), WHO grade I and II astrocytomas, and pilocytic astrocytomas. The average percentage of seizure freedom after surgery of DNTs and gangliogliomas is between 87.7% (range 58–90%) (Minkin et al., 2008) and 79.2% (range 45–100%) (Southwell et al., 2012). The overall high incidence of seizure freedom after surgery of low grade glioneuronal tumors is confirmed by almost all reports (Giulioni et al., 2005; Englot et al., 2011, 2012; Southwell et al., 2012).

An extended review including 773 patients with epileptogenic glioneuronal tumors (Englot et al., 2011) indicated radical removal of the tumor as the most important predictor of seizure freedom, followed by good preoperative control of seizures with antiepileptic drugs (AEDs), complex versus simple partial seizures, absence of secondarily generalized seizures, and a history of epilepsy of <1 year duration. Southwell et al. (2012) analyzed prognostic factors for seizure freedom in their series of 66 gangliogliomas and pointed out that factors such as preoperative history and type of seizures, tumor location, and intraoperative ECoG had no predictive value, whereas gross total removal (GTR) was the strongest predictor with 94% of patients becoming seizure-free after GTR as compared to 54% after subtotal removal (STR). A crucial issue in the surgical management of epileptogenic tumors in the temporal lobe is whether or not to couple amygdalo-hippocampectomy with GTR. According to some authors, amygdalohippocampectomy would be recommended only in temporomesial tumors (Giulioni et al., 2009), whereas in temporolateral lesions it should be considered in relation to clinical and electrographic findings. Clusmann et al. (2002) suggest that amygdalohippocampectomy should not be systematically performed in temporomesial tumors. Englot et al. (2012) reviewed published data on 1,181 patients with low-grade temporal lobe tumors from 41 studies, and they found that GTR with amygdalohippocampectomy resulted in higher rates of seizure freedom as compared to GTR alone (87% vs. 79% seizure-free). These findings probably reflect the frequent association between epileptogenic tumors and mesial temporal sclerosis (Blümcke et al., 2011).

The series of epileptogenic brain tumors operated in our department includes 128 surgeries in 104 patients (60 male, 44 female; mean age 13 years). Surgical procedures were GTR (81), STR (33), temporal lobectomy plus amygdalohippocampectomy (13), and frontal lobectomy (1). Histology was mostly represented by ganglioglioma (21; 20.2%), DNT (19; 18.3%) and low grade gliomas such as pilocytic astrocytoma (5; 4.8%), and WHO grade I (13; 12.5%) and II astrocytoma (7; 6.7%). FCD IIIB was observed in 10 patients (9.6%). Average follow-up was 36.3 months (range 6 months–17 years). At last follow-up, 85 patients (81.7%) were seizure-free (Engel class I); 25 (25.9%) of 85 were off AEDs. Ten, six, and two children were Engel class II, III, and IV, respectively. The analysis of the rate of seizure freedom in relation to extent of resection, tumor topography, and histology reveals that of 85 Engel class I patients, 77 (90.5%) underwent GTR and 11 (12.9%) underwent temporal lobectomy plus amygdalohippocampectomy. The best results were observed in patients with tumor location in the temporal lobe, either lateral (28) or mesial (22), with 50 (58.8%) of 85 of them achieving seizure freedom (Engel class I).

Perspectives on Medical Treatment

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References

Epilepsy is an important contributor to morbidity in patients with brain tumors, as seizures and AEDs significantly worsen quality of life (Klein et al., 2003; Klein, 2012). Managing seizures in these patients requires diligent comprehensive care to limit unnecessary side effects and improve quality of life.

When treating tumoral epilepsy with AEDs, aspects that are ordinarily taken into account deserve even more attention, considering the high vulnerability of such patients in relation to the frequent drug-resistance of seizures, potential interactions between AEDs and chemotherapeutic agents (CMT), and the risk of cognitive adverse events from use of AEDs, in addition to earlier brain damage resulting from the tumor itself, surgery, and radiotherapy.

Tumor-related seizures, especially those related to low-grade tumors, are difficult to control with AEDs (Rudà et al., 2012). Several hypotheses have been proposed to explain resistance to AEDs in brain tumors: (1) “intrinsic severity” due to underlying mechanisms of seizure generation that are not covered by AED action; (2) altered expression or activity of target molecules of AEDs; and (3) altered expression or activity of drug transporters at the blood–brain barrier, limiting penetration to the tissue of seizure origin (de Groot et al., 2012).

There is no robust, randomized, controlled evidence to support choice of AEDs for the treatment of seizures in patients with brain tumors, and different experts may make different choices (Kerrigan & Grant, 2011). Although some authors support the use of non–enzyme-inducing AEDs, reliable comparative evidence justifying this is limited. Several studies have investigated the efficacy of the newer AEDs in tumor-related epilepsy, but most such studies are retrospective or include small numbers of patients (Kerrigan & Grant, 2011). Moreover, the few studies comparing the efficacy of newer versus older AEDs (Kerrigan & Grant, 2011) report comparable results but less adverse events with newer AEDs (Kerrigan & Grant, 2011). Newer AEDs (oxcarbazepine, topiramate, lamotrigine, levetiracetam, zonisamide, and lacosamide) have poor or no enzyme-inducing properties, predominant renal excretion, lower plasma protein binding and, consequently, fewer interactions with other medications (e.g., CMT).

Enzyme-inducing AEDs, especially those affecting the cytochrome P450 system, are generally not recommended because they can lead to insufficient serum levels of concomitantly administered CMTs. Usually, the addition of phenytoin, carbamazepine, phenobarbital, and other inducers, reduces levels or efficacy of several CMTs and corticosteroids (Glantz et al., 2000). Valproic acid, being an inhibitor, can have the opposite effect and increase levels of CMT agents. The use of valproic acid, which exhibits inherent antitumor activity through inhibition of histone deacetylase, has been recommended in patients with glioblastoma multiforme (Weller et al., 2011).

Treatment-related adverse events appear to be more frequently reported in patients with tumor-related epilepsy (24%) than in the rest of the epilepsy population (0.5–12%) (Glantz et al., 2000). AEDs of earlier generations exhibit more adverse events (Klein et al., 2003). Skin rash due to carbamazepine, phenobarbital, and phenytoin are more frequent in patients with brain tumors (Glantz et al., 2000). Patients receiving radiotherapy and concomitant oxcarbazepine have a higher risk of developing serious skin rashes such as Stevens-Johnson syndrome and toxic epidermal necrolysis (Maschio et al., 2010). Adverse central nervous system events can strongly affect patients’ quality of life, with a sixfold increase in deficits affecting attention, psychomotor speed, or executive functions as compared to the severity of adverse events secondary to radiation of the brain (Klein et al., 2003; Klein, 2012). This makes it difficult to assess the role of seizures, the clinical response to CMT, and can even mimic tumor progression. Considering the risks of AED-related adverse events and life-threatening complications due to drug interactions between AEDs and CMTs, prophylactic use of AEDs for patients who have never experienced a seizure should be discouraged.

Several prospective, randomized trials have investigated AED prophylaxis of late tumor-related seizures. No study has proven benefit and a consensus statement from the American Academy of Neurology Clinical Practice Guidelines, based on studies performed with conventional AEDs, recommends against routine use of AEDs as prophylaxis in patients with brain tumors (Glantz et al., 2000). Class I evidence suggests that prophylactic use of AEDs in patients with brain tumors who do not have a history of seizures is neutral or ineffective (Tremont-Lukats et al., 2008).

Disclosures

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References

The authors have no conflicts of interest to disclose. The authors confirm that they have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Perspectives on Surgical Treatment
  4. Prognosis
  5. Perspectives on Medical Treatment
  6. Disclosures
  7. References
  • Aronica E, Leenstra S, van Veelen CW, van Rijen PC, Hulsebos TJ, Tersmette AC, Yankaya B, Troost D. (2001) Glioneuronal and medically intractable epilepsy: a clinical study with long-term follow-up of seizure outcome after surgery. Epilepsy Res 43:179191.
  • Barba C, Coras R, Giordano F, Buccoliero AM, Genitori L, Blümcke I, Guerrini R. (2011) Intrinsic epileptogenicity of gangliogliomas may be independent from co-occurring focal cortical dysplasia. Epilepsy Res 97:20813.
  • Blümcke I, Thom M, Aronica E, Armstrong DD, Vinters HV, Palmini A, Jacques TS, Avanzini G, Barkovich AJ, Battaglia G, Becker A, Cepeda C, Cendes F, Colombo N, Crino P, Cross JH, Delalande O, Dubeau F, Duncan J, Guerrini R, Kahane P, Mathern G, Najm I, Ozkara C, Raybaud C, Represa A, Roper SN, Salamon N, Schulze-Bonhage A, Tassi L, Vezzani A, Spreafico R. (2011) The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52:158174.
  • Clusmann H, Schramm J, Kral T, Helmstaedter C, Ostertun B, Fimmers R, Haun D, Elger CE. (2002) Prognostic factors and outcome after different types of resection for temporal lobe epilepsy. J Neurosurg 97:11311141.
  • D'Argenzio L, Colonnelli MC, Harrison S, Jacques TS, Harkness W, Vargha-Khadem F, Scott RC, Cross JH. (2011) Cognitive outcome after extratemporal epilepsy surgery in childhood. Epilepsia 52:196672.
  • de Groot M, Reijneveld JC, Aronica E, Heimans JJ. (2012) Epilepsy in patients with a brain tumor: focal epilepsy requires focused treatment. Brain 135:10021016.
  • Duffau H, Capelle L, Denvil D, Sichez N, Gatignol P, Taillandier L, Lopes M, Mitchell MC, Roche S, Muller JC, Bitar A, Sichez JP, van Effenterre R. (2003) Usefulness of intraoperative electrical subcortical mapping during surgery for low-grade gliomas located within eloquent brain regions: functional results in a consecutive series of 103 patients. J Neurosurg 98:764778.
  • Englot DJ, Berger MS, Barbaro NS, Chang EF. (2011) Predictors of seizure freedom after resection of supratentorial low-grade gliomas. A review. J Neurosurg 115:240244.
  • Englot DJ, Han SJ, Berger MS, Barbaro NM, Chang EF. (2012) Extent of surgical resection predicts seizure freedom in low-grade temporal lobe brain tumors. Neurosurgery 70:921928.
  • Giulioni M, Galassi E, Zucchelli M, Volpi L. (2005) Seizure outcome of lesionectomy in glioneuronal tumors associated with epilepsy in children. J Neurosurg Pediatr 102:288293.
  • Giulioni M, Rubboli G, Marucci G, Martinoni M, Volpi L, Michelucci R, Marliani AF, Bisulli F, Tinuper P, Castana L, Sartori I, Calbucci F. (2009) Seizure outcome of epilepsy in focal epilepsies associated with temporomesial glioneuronal tumors: lesionectomy compared with tailored resection. J Neurosurg 111:12751282.
  • Glantz MJ, Cole BF, Forsyth PA, Recht LD, Wen PY, Chamberlain MC, Grossman SA, Cairncross JG. (2000) Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 54:18861893.
  • Harvey AS, Cross JH, Shinnar S, Mathern BW, ILAE Pediatric Epilepsy Surgery Survey Taskforce. (2008) Defining the spectrum of international practice in pediatric epilepsy surgery patients. Epilepsia 49:146155.
  • Kerrigan S, Grant R. (2011) Antiepileptic drugs for treating seizures in adults with brain tumors. Cochrane Database Syst Rev 10(8):CD008586.
  • Klein M. (2012) Neurocognitive functioning in adult WHO grade II gliomas: impact of old and new treatment modalities. Neuro Oncol 14(Suppl. 4):1724.
  • Klein M, Engelberts NH, van der Ploeg HM, Kasteleijn-Nolst Trenite′ DG, Aaronson NK, Taphoorn MJ, Baaijen H, Vandertop WP, Muller M, Postma TJ, Heimans JJ. (2003) Epilepsy in low-grade gliomas: the impact on cognitive function and quality of life. Ann Neurol 54:514520.
  • Maschio M, Dinapoli L, Vidiri A, Muti P. (2010) Rash in four patients with brain tumor-related epilepsy in monotherapy with oxcarbazepine, during radiotherapy. J Neurol 257:19391940.
  • Minkin K, Klein O, Mancini J, Lena G. (2008) Surgical strategies and seizure control in pediatric patients with dysembrioblastic neuroepithelial tumors: a single-institution experience. J Neurosurg Pediatr 1:206210.
  • Palmini A, Gambardella A, Andermann F, Dubeau F, da Costa JC, Olivier A, Tampieri D, Gloor P, Quesney F, Andermann E, Paglioli E, Paglioli-Neto E, Coutinho Andermann L, Leblanc R, Kim H (1995) Intrinsic epileptogenicity of human dysplastic cortex as suggested by corticography and surgical results. Ann Neurol 37:476487.
  • Rudà R, Bello L, Duffau H, Soffietti R. (2012) Seizures in low-grade gliomas: natural history, pathogenesis, and outcome after treatments. Neuro Oncol 14(Suppl. 4):iv5564.
  • Southwell DG, Garcia PA, Berger MS, Barbaro NM, Chang EF. (2012) Long-term seizure control outcomes after resection of gangliogliomas. Neurosurgery 70:14061414.
  • Spencer S, Huh L. (2008) Outcomes of epilepsy surgery in adults and children. Lancet Neurol 7:525537.
  • Tremont-Lukats IW, Ratilal BO, Armstrong T, Gilbert MR. (2008) Antiepileptic drugs for preventing seizures in people with brain tumors. Cochrane Database Syst Rev (2):CD004424.
  • Van Schooneveld MM, Braun KP. (2013) Cognitive outcome after epilepsy surgery in children. Brain Dev 35:721729.
  • Weller M, Gorlia T, Cairncross JG, van den Bent MJ, Mason W, Belanger K, Brandes AA, Bogdahn U, Macdonald DR, Forsyth P, Rossetti AO, Lacombe D, Mirimanoff RO, Vecht CJ, Stupp R. (2011) Prolonged survival with valproic acid use in the EORTC/NCIC temozolomide trial for glioblastoma. Neurology 77:11561164.