Seizure 21 (2012) 153–159

Contents lists available at SciVerse ScienceDirect

Seizure
journal homepage: www.elsevier.com/locate/yseiz

Review

The pathogenesis of tumor-related epilepsy and its implications for clinical

treatment
Gan You a, Zhiyi Sha b, Tao Jiang a,*
a
Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing 100050, China
b
Department of Neurology, Medical School, University of Minnesota, Minneapolis, MN 55455, USA

A R T I C L E I N F O A B S T R A C T

Article history: Approximately 30–50% of patients with brain tumors present with seizures as the initial symptom.
Received 18 October 2011 Seizures play a very important role in the quality of life, particularly in patients with slow-growing
Received in revised form 28 December 2011 primary brain tumors. Tumor-related seizures are often refractory to antiepileptic treatment. Despite the
Accepted 29 December 2011
importance of this subject to the fields of neurology, neurosurgery and neurooncology, the pathogenesis
of tumor-related epilepsy remains poorly understood. This review summarizes possible mechanisms
Keywords: underlying the pathogenesis of tumor-related epilepsy, including both tumoral and peri-tumoral
Brain tumor
aspects. Tumor cells themselves may create intrinsic epileptogenicity, and inadequate homeostasis in
Epilepsy
Pathogenesis
the peri-tumoral tissues may lead to seizure susceptibility. Other local changes in electrolytes, perfusion,
Seizure metabolism, and enzymes could also contribute. It is generally accepted that changes in amino acid
Treatment neurotransmission are the most important mechanism underlying tumor-related seizures, and changes
in extracellular ions also play an important role. Hypoxia, acidosis, and metabolic, immunological, and
inflammatory changes may also be involved in the occurrence of seizures. Knowledge of these
mechanisms may provide guidance in the search for new strategies for the surgical and medical
treatment of tumor-related epilepsy.
ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.

1. Introduction
Epileptic seizures due to brain tumors have been described
since the 19th century. John Hughling Jackson first reported the
direct relationship between seizures and brain tumor.1 Approxi-
mately 30–50% of patients with brain tumors present with seizures
as the initial symptom.2,3 Seizures play a very important role in the
quality of life, particularly in patients with slow-growing primary
brain tumors, for whom seizure incidence reaches 80–90%.2,4
These seizures often manifest as focal seizures with secondary
generalization and are often refractory to antiepileptic treatment.
A number of studies have established an association between
epileptic seizures at disease onset and a more favorable progno-
sis.4,5 Seizure onset at a later time, however, is frequently
associated with neurosurgical procedures or progression of the
lesion.6,7
Despite the importance of this subject to the fields of neurology,
neurosurgery, and neurooncology, the pathogenesis of tumor-
related epilepsy remains poorly understood. At present, there
seem to be two schools of thought regarding the pathogenesis of
tumor-related epilepsy. One theory is based on the tumor origin:
* Corresponding author. Tel.: +86 010 67036083; fax: +86 010 67036038.
E-mail address: taojiang1964@yahoo.com.cn (T. Jiang).
the tumor itself may excrete molecules that could make the tumor
tissue epileptogenic, or it could change the peri-tumoral microen-
vironment and turn this into an epileptogenic zone. The other
theory is that the tumor mechanically compresses the surrounding
normal tissue, which eventually becomes epileptogenic after
suffering from ischemia and hypoxia. Both of these processes could
potentially cause secondary changes, such as changes in neuro-
transmitters and their receptors, metabolic changes, and inflam-
matory responses, eventually leading to epileptic seizures.
However, neither of these theories can fully explain the patterns
of seizure incidence. This review attempts to summarize the
possible mechanisms of tumor-related epilepsy, based on the
hypotheses mentioned above. This knowledge may provide
guidance in the search for new strategies for the surgical and
medical treatment of tumor-related epilepsy.
2. Mechanisms
2.1. Tumor factors
2.1.1. Histology
Seizures are common in patients with low-grade tumors such
as dysembryoblastic neuroepithelial tumors (DNETs), ganglioglio-
mas (GGs), and oligodendrogliomas (OGs). The incidence of
epilepsy in high-grade brain tumors such as glioblastoma multi-

1059-1311/$ – see front matter ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.seizure.2011.12.016
154 G. You et al. / Seizure 21 (2012) 153–159
formes (GBMs) and metastatic tumors is lower. The pathogenesis
of seizure development is likely to be different for brain tumors
with a different histology.8,9 Developmental tumors consist of
well-differentiated cells, which are able to release neurotrans-
mitters and other modulators that are involved in epileptogen-
esis.10 These tumors may also be associated with structural
epileptogenic abnormalities of the cortex. The highly infiltrative
growth of fast-growing high-grade brain tumors may damage the
subcortical network essential for electrical transmission,11
whereas slow-growing tumors have been suggested to induce
partial deafferentation of cortical regions, causing denervation
hypersensitivity10,12 and producing an epileptogenic milieu. In
addition, gliosis and chronic inflammatory changes in the peri-
tumoral regions may lead to epileptic seizures. Brain tumors with
the same grade but a different histology may have different
seizure incidences.6 Furthermore, not all patients with similar
tumor localization and histology have seizures.9 This strongly
suggests that genetic factors may play a role in tumor develop-
ment and tumor-related epilepsy.
2.1.2. Tumor location
The location of the tumor is also an important determinant of
tumor-associated epilepsy. Chang et al.6 reported that subcortical
location with tumor growth in deep midline structures was less
likely to result in seizures. Some authors13 also suggested that
location and proximity to the cortical gray matter were also
important factors in the development of epilepsy in patients with
glioma. In general, tumors in the frontal and temporal lobes, as
well as in the limbic system, are more likely to cause seizures than
tumors in other locations.6,14 The location of tumors is closely
related to their histology. The majority of glioneuronal tumors
occur in the temporal lobe. Duffau and Capelle15 have provided
some evidence that diffuse low-grade astrocytomas and oligoden-
drogliomas may also have preferred locations. In contrast to
malignant gliomas, low-grade gliomas (LGGs) tend to grow in
secondary functional areas that are close to, but rarely within, the
primary eloquent parts of the brain. Our unpublished data have
shown that oligodendroglial tumors were more likely to be located
in frontal lobe, whereas astrocytomas were more commonly found
in temporal locations. This might be explained by the fact that
tumor histology is to some extent associated with the lobe where
tumors occur during brain development. Loss of heterozygosity at
1p/19q is closely related to tumor histology in gliomas. In a recent
study,16 we hypothesized that some genes on 19q might be
susceptibility genes of glioma-related seizures. It may be
postulated that tumor-related seizures have unique character-
istics, which may share some common genetic pathways with
tumorigenesis.
2.1.3. Blood–brain barrier disruption
The cellular components of the blood–brain barrier (BBB)
include endothelial cells, astrocytes, pericytes, neurons, and
junctional complexes formed by transmembrane junctional
proteins including occludin, claudins, and junctional adhesion
molecules.17 Human and animal studies have suggested that
perturbations in neurovascular integrity and breakdown of the BBB
lead to neuronal hypersynchronization and epileptiform activity.
Relevant molecular changes in brain tumors that affect BBB
structure and function include decreased expression of transmem-
brane junctional proteins18 and heightened release of vascular
endothelial growth factor (VEGF).19 Diffusion of VEGF into the peri-
tumoral brain may aggravate the edema surrounding the lesion.
Stewart et al.20 reported structural defects in endothelial tight cell
junctions surrounding human gliomas. A more recent study21
suggested that this may be mediated by transforming growth
factor b (TGF-b) receptor stimulation, causing activity-dependent
accumulation of extracellular potassium, facilitation of N-methyl-
D-aspartate (NMDA) receptor-mediated neuronal hyperexcitabili-
ty, and eventually epileptiform activity. Another study demon-
strated22 that blockade of TGF-b receptors in vivo reduced the
likelihood of epileptogenesis.
Taken together, these results suggest that pathological disrup-
tion of the BBB in brain tumor patients may contribute to seizure
activity. It is difficult to define the extent to which BBB
disturbances may lead to seizure induction. However, it is
generally considered that tumors such as LGGs, which cause
BBB disturbances but do not destroy the subcortical network, are
likely to cause seizures.
2.1.4. Gap junctions
Communication between adjacent glial cells can take place
through transmembrane proteins known as connexins (CX).
Astrocytic gap junctions contain CX43, whereas oligodendrocytes
express CX32, and neurons express CX26 and CX32.23 Recently,
alterations in glial gap-junctional coupling have been found to be
associated with tumoral epilepsy. Aronica et al.23 demonstrated
increased CX43 membrane immunoreactivity in low-grade glio-
mas when compared with both high-grade lesions and control
brain tissue. Furthermore, the perilesional epileptogenic cortex
surrounding low-grade lesions contained reactive astrocytes that
expressed higher levels of CX43 than control cortex.23 Interesting-
ly, high-grade gliomas exhibited aberrant intracytoplasmic locali-
zation of CX43,23 which might explain the lower risk of tumor-
related seizures among patients with this type of cancer.
2.1.5. Molecular genetic changes
Genetic factors inside the tumor may also play a role in the
epileptogenic process. For example, expression of tumor-suppres-
sor gene LGI1, which could contribute to glioma progression by
increasing cell growth and migration when down-regulated,24 is
low or absent in cell lines from high-grade gliomas.25 Based on the
possible association between seizure recurrence and glioma
progression,6,7 Brodtkorb et al.26 suggested that LGI1 may be
correlated with epileptic susceptibility in patients with brain
tumors. However, some other studies have not supported a tumor-
suppressor function of LGI1,27 and further investigations are
needed to confirm such a role.
Tumors have genomic and chromosomal instability, including
DNA strand breaks and rearrangements. These alterations may be
associated with changes in gene expression, with negative effects
on the stability of DNA repair mechanisms, and this may lead to
mutations. Under these conditions, the tumor cells might become
epileptogenic. Bordey and Sontheimer28 reported that astrocytic
tumor cells might generate action potentials, which might
themselves be the source of epileptic activity. Ye and Sontheimer29
found that glioma cells released glutamate and caused excitotoxic
cell death of the peri-tumoral neurons. These findings suggest that
the tumor itself could be an independent origin of epileptogenesis.
Further studies of the possible candidate susceptibility genes for
tumor-related seizures are encouraged (Figs. 1 and 2).
2.2. Peri-tumoral factors
The microenvironment in brain tumors is substantially differ-
ent from that in normal brain tissue. Contemporary imaging
techniques provide testimony to the remarkable differences
between the peri-tumoral brain and normal tissue. Recent
advances in magnetic resonance spectroscopy (MRS) have
demonstrated decreased levels of N-acetylaspartate, a marker of
neuronal viability and function, in lesional epileptogenic cortex.30
Focal abnormalities of low-frequency magnetic activity on
magnetoencephalography (MEG) in peri-tumoral regions have
 G. You et al. / Seizure 21 (2012) 153–159 155

Fig. 1. MRI (T2-weighted images) of two patients with astrocytoma (WHO grade II), suggesting different underlying biological characteristics of tumors with similar location
and histology. (a) Image from a patient that presented with secondary generalized seizures as the initial symptom. (b) Image from a patient with long-standing headache
before admission.

Fig. 2. Immunohistochemical staining for the astrocytic marker GFAP (glial fibrillary acidic protein) in the marginal areas of samples from two different patients with diffuse
astrocytomas (WHO grade II). (a) High GFAP expression in the tumor of one patient with refractory seizures. (b) Low GFAP expression in another astrocytoma of a patient who
became seizure-free on antiepileptic drugs after an initial seizure (original magnification, 400).

been found in 13 of 20 patients with tumor-related seizures.31
Douw et al. also indicated that pathologically increased theta band
connectivity was related to a higher number of epileptic seizures in
brain tumor patients, suggesting that elevated theta band
connectivity is a hallmark of tumor-related epilepsy.32 Whether
these associations reflect local metabolic imbalances, functional
deafferentation, or profound structural alterations remains un-
clear.
2.2.1. Morphologic changes
Certain morphologic changes in the peri-tumoral brain tissue,
such as persistent neurons in the white matter, inefficient neuronal
migration, and changes in synaptic vesicles, are also believed to
contribute to seizure generation.8,33 It is possible, and indeed
likely, that peri-tumoral cells have an altered or anomalous
phenotype, which is commonly seen in glioneuronal or dysplastic
brain tumors. This could be related to the origin of the tumor or
could constitute a predisposing factor for seizures in people with
brain tumors. Comparison of the ultrastructure of the peri-tumoral
cortex in patients with and without epilepsy has demonstrated
statistically significant changes in the form, size, distribution, and
number of synaptic vesicles.34 Another type of cellular change
suggested to be associated with tumor growth is a decrease in
inhibitory synapses and increase in excitatory synapses in peri-
tumoral pyramidal neurons.35 Dysfunctional astrocytes in peri-
tumoral regions may also contribute to epilepsy through different
mechanisms.36,37
2.2.2. Hypoxia, acidosis and metabolic changes
Hypoxia and acidosis are produced as a result of intratumoral
decreased perfusion and elevated metabolism, respectively.
However, the relationships between perfusion, hypoxia, metabo-
lism, and acidosis are complex. The growth of brain tumors
depends on the establishment of an adequate blood supply.38 On
one hand, tumors with insufficient blood supply often cause
interstitial hypoxia, which subsequently contributes to acidosis.
This intratumoral hypoxia and acidosis may extend to the
surrounding tissue. On the other hand, tumors with large size
usually cause peri-tumoral hypoxia because of direct compression.
Both these factors could cause glial cell swelling and damage.39
This is of particular interest because astroglial cells control the
acidity of the environmental fluid. Under these conditions, the
astrocytic cell membrane becomes prone to inward sodium
currents, leading to risk of epilepsy.40
Furthermore, hypoxia causes acidosis as a consequence of both
heightened metabolic requirements of the proliferating tissue and
156 G. You et al. / Seizure 21 (2012) 153–159
impaired oxidative energy metabolism. The normal response to
hypoxia would be glucose catabolism with lactate production,
leading to acidosis in metabolically active tissue. With the advent
of magnetic resonance imaging (MRI), MRS, and positron emission
tomography, many studies have evaluated the metabolic activity
in both epileptic and neoplastic brain. Increased lactate levels41
and decreased glucose metabolism were observed in the peri-
tumoral white matter. An elevated level of lactate compared to
healthy tissue has also been reported in tumor tissue.42 Thus, the
epileptogenic changes must have a component that is transferable
to the surrounding neurons. However, it is unclear how the
metabolic abnormalities described above could be responsible for
this, unless they have an effect on neuronal morphology or
function.
2.2.3. Ionic changes
Ionic changes in the peri-tumoral zone may influence neuronal
activity. Hossmann et al.43 described elevated sodium and calcium
levels in the extracellular peri-tumoral space, which may
contribute to neuronal hyperexcitability. Kraft et al.44 also
demonstrated that, under acidic conditions, astrocytoma cell
membranes are susceptible to inward sodium currents following
sodium channel activation. The relevant family of ion channels has
been termed the brain sodium channel complex.45 Likewise,
increased densities and altered gating of calcium channels have
been reported in epileptic tissue. Potassium channel mutations
(e.g., KCNQ) have been found in certain epileptic syndromes, and
extracellular potassium concentrations are thought to play a role in
ictogenicity due to their impact on membrane potential.46,47
Magnesium (Mg2+) helps to stabilize neuronal excitability by
blocking calcium influx through NMDA receptor channels. Avoli
et al.48 demonstrated that decreased extracellular concentrations
of Mg2+ can lead to spontaneous epileptiform discharges and
spreading depression in human cortex in vitro. In addition,
increased levels of Fe3+ in intra- or peri-tumoral areas due to
small bleedings from pathological blood vessels may also
contribute to the development of tumor-associated seizures. This
is more likely to occur in high-grade gliomas, since experimental
Fe3+-induced peroxidative injury to neuronal plasma membranes
is related to the development of paroxysmal epileptiform
activity.8,49
Recently, an interesting hypothesis was proposed by Sonthei-
mer,50 who suggested that glioma invasion into the peri-tumoral
zone is in part mediated by chloride (Cl ) channel overexpres-
sion, allowing cells to traverse the extracellular space through
rapid changes in cell shape. Similar mechanisms may be involved
in glioma metastasis, to cause seeding and infiltration of the
narrow extracellular spaces in the brain.51 It remains unclear
whether seizures represent an associated feature of ion channel
overexpression. Further studies are needed to confirm this
hypothesis.
2.2.4. Changes in amino acids and neurotransmitter receptors
Previous studies29,52 have demonstrated that perturbations in
the balance between excitatory and inhibitory compounds may
lead to glioma-related seizures. Dysfunctional astrocytes may send
signals to neurons by releasing a number of gliotransmitters that
can have both excitatory and inhibitory actions.
2.2.4.1. Glutamate neurotransmission. Recent work has demon-
strated a close link between seizure activity and high extracellular
glutamate in tumor-related epilepsy. Patients with glioma and
refractory epilepsy have increased concentrations of glutamate, as
shown by changes in perilesional immunoreactivity of glutamate
decarboxylase.40,52 Glutamate is an excitatory neurotransmitter
that acts on postsynaptic membranes by interacting with
ionotropic and metabotropic glutamate receptors.53 Glutamate
receptors can be present in greater or lesser numbers in tumors,
depending on the degree of differentiation and transcriptional
control in the tumor tissue.
Glutamate activation of ionotropic receptors leads to a rapid
excitatory signal based on cation influx that can cause release of
calcium from intracellular stores. Ionotropic glutamate receptors
include NMDA receptors (NR1, NR2A–C subunits), a-amino-3-
hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors
(GluR1–4 subunits), and kainate receptors. Aronica et al.54
examined 41 cases of gangliogliomas and 16 cases of DNETs, all
with intractable epilepsy. Both tumor types had higher NR2A and
NR2B expression in the tumors than in the nonepileptic cortex. In a
similar study, Lee et al.55 evaluated NMDA receptor expression in
13 patients with DNET and refractory epilepsy. Stronger NR1,
NR2A, and NR2B expression with increased GluR2 and GluR3
immunopositivity was found in the peri-tumoral cortex. Aronica
et al.54 also demonstrated elevated expression of kainate receptors
in the reactive astrocytes of the perilesional zone.
Interaction of glutamate with metabotropic glutamate recep-
tors (mGluR) initiates intracellular signaling through GTP-
binding proteins, which can serve neuromodulatory functions.
Abnormalities of receptor activation could result from differ-
ences in the relative expression in peri-tumoral tissue compared
to normal brains. It has been suggested that mGluR2 expression
was decreased in the neuronal component of GGs and DNETs,
while mGluR5 was expressed at higher levels on reactive
astrocytes surrounding those tumors.54 Rutecki et al.56 showed
that metabotropic receptors might modulate AMPA receptor
activation to generate oscillatory calcium waves. This effect
could be inhibited by AMPA receptor blocker, but not by NMDA
or g-aminobutyric acid A (GABAA) receptor antagonists. In
addition, activation of mGluR5 can lead to phosphoinositol-
based calcium signaling.57 Further investigations of how
metabotropic receptor expression is altered in seizure-related
tumors will be necessary to validate the pathophysiological roles
of these changes.
2.2.4.2. GABA neurotransmission. The inhibitory neurotransmitter
GABA can inhibit neuronal firing. GABA receptor down-regulation
may contribute to hyperactivity of the surrounding microenviron-
ment. GABA receptors are classified as GABAA, GABAB and GABAC
receptors. In peri-tumoral tissue, changes in the function of these
receptors have been well documented, and mainly involve reduced
GABAergic neurotransmission.55,58 Labrakakis et al.59 investigated
the responsiveness of glial tumor tissue slices to GABA, and
observed a depolarizing response to GABAA receptor activation in
71% of oligodendroglioma cells and 62% of astrocytoma cells. In
contrast, no activation by GABA was found in GBM specimens. This
indicates that the GABAA receptor is expressed at higher levels in
LGGs compared to GBM, and that responses are often excitatory in
tumor tissue.
Interestingly, GABA levels appeared to be higher in human
tumor-inflicted tissues than in control tissue.46,47 Although
GABAergic activity does not directly correlate with seizure
susceptibility, it is likely that the changes in transmitter levels
are causal for tumor-associated epilepsies. To date, the role of
GABA and its receptors in tumor-induced epilepsy is still largely
unclear.
2.2.4.3. Other amino acid transmitters. Other amino acids may also
play a role in tumoral epileptogenesis. Gliomas have been shown to
produce lower levels of kynurenic acid,60 which may lead to
disinhibition of NMDA receptors.61 Decreased levels of noradren-
aline and serotonin, which can have intrinsic inhibitory and anti-
epileptogenic effects, have also been demonstrated in gliomas.58
 G. You et al. / Seizure 21 (2012) 153–159
2.2.5. Immunological and inflammatory changes
Several studies have reported an association between strong
immunological or inflammatory changes with decreased risks of
glioma.62,63 Immunological factors are also likely to play a role in
tumor-associated epilepsy, and proinflammatory cytokines and
their receptors have been suggested to be involved in the
pathogenesis of epilepsy. Cytokines, such as interleukin 4 (IL-4)
and IL-6,62 have modulating effects on neurotoxic neurotransmit-
ters that are released during excitation or inflammation in the
central nervous system. IL-1b and tumor necrosis factor a
interfere with astrocyte gap junction communication. In cell
culture, release of these proinflammatory cytokines from activated
microglia leads to a closure of CX43-mediated gap junctions in
astrocytes.64 Since status epilepticus induces massive microglial
activation in mice in vivo,65 inhibition of gap junction communi-
cation in astrocytes appears to represent a very early alteration in
the process of epileptogenesis. In glioma, immune-mediated
neurochemical changes by microglia in the peri-tumoral brain
regions are well documented.66,67 Immune-mediated neuronal
damage of the peri-tumoral brain area, coupled with changes in the
balance between stimulatory and inhibitory cytokines, may
contribute to the development of tumor-related epilepsy.66,68
3. Treatment implications for tumor-related epilepsy
Seizures caused by brain tumors often manifest as focal seizures
with or without secondary generalization, and approximately one
third of patients are refractory to antiepileptic medication
treatment. There are a number of possible reasons for the
medication resistance.69 First, antiepileptic drugs (AEDs) could
be affected by the biochemical milieu of the peri-tumoral space.
Second, significant drug–drug interactions between AEDs and
chemotherapeutics may affect antiepileptic effectiveness, and
exaggerate side effects by influencing the hepatic cytochrome
P450 system. Third, treatment resistance may also arise from
overexpression of multidrug resistance-related proteins (MRPs) in
tumors, which restrict the penetration of lipophilic substances into
the brain. MRPs were found at heightened levels in brain tumor
capillary endothelial cells and astrocytes.69
3.1. How much resection is required?
Based on the presumption that neurons surrounding the tumor
constitute the epileptogenic zone, removing the tumor alone may
not guarantee a good outcome in seizure control. However, it can
also be argued that if the irritating lesion is removed, the
microenvironment may return to normal and the surrounding
neurons may cease to discharge abnormally. For most surgical
series involving pediatric patients, lesionectomy alone yielded
very good results.70,71 However, studies on adult patients
demonstrated that gross total resection or even extended
lesionectomy could greatly improve seizure prognosis.6,72 One
possible reason for the better results of lesionectomy in children
may be that their seizure history is shorter with less opportunity of
permanent secondary changes such as hippocampal sclerosis. In
general, surgeons are encouraged to minimize the residual tumor
volume when possible.
3.2. Should traditional AEDs be phased out?
No randomized clinical trials have evaluated the efficacy of
traditional AEDs, such as valproic acid (VPA), carbamazepine (CBZ),
phenytoin (PHT) and phenobarbital (PB), in patients with brain
tumors. It is difficult to draw firm conclusions from available
studies. Hence, the decision of which AEDs to administer to
patients with brain tumors is based mainly on individual
157
preference rather than clinical evidence. However, a meta-analysis
by Glantz et al.73 of 12 informative studies investigating the use of
prophylactic anticonvulsants (PHT, PB, or VPA) in patients with
primary and metastatic brain tumors demonstrated a lack of
efficacy in preventing the first seizure or in decreasing the
frequency of initial seizures. Temkin also reported that there was
no evidence that long-term treatment with PHT and CBZ could
protect against late seizures.74 The question of whether or not the
choice of a traditional AED should be solely based on their side
effect profile remains to be answered.
VPA is thought to inhibit epileptic discharges by stabilizing
neuronal membranes and enhancing GABA transmission. It can
induce apoptosis, growth arrest, and cell differentiation of tumor
cells through inhibition of histone deacetylase.75 A recent study
reported that VPA induced autophagy in glioma cells and this
action was independent of apoptosis.76 A study by Weller et al. also
indicated potential anti-tumor activity of VPA in patients with
GBM who required an AED during temozolomide-based chemor-
adiotherapy.77 The fact that both tumoral and peri-tumoral factors
contribute to the pathogenesis of tumor-related epilepsy suggests
that VPA should be considered as a first line therapy in treating
tumor-related epilepsy.
3.3. What advantages will new AEDs bring?
Newer generation AEDs, such as levetiracetam (LEV), lamo-
trigine (LTG), topiramate (TPX), gabapentin (GBP), and pregabalin
(PGB), have been recommended for patients with brain tumors.
These drugs have different antiepileptic mechanisms, including
GABA receptor agonism, calcium channel modulation, and NMDA
receptor antagonism. Vecht and van Breemen12 have reviewed
treatment recommendations for patients with brain tumors, and
suggested that first-line anticonvulsants should include LEV and
LTG, because they lack significant drug–drug interactions with
chemotherapy agents. The antiepileptic mechanisms of LTG
include Na channel inhibition, inhibition of Ca2+ currents through
neuron-specific, high-voltage-activated (N-type) Ca2+ channels,
enhancement of GABA receptor activity, and inhibition of
pathological release of glutamate and NMDA receptor-mediated
excitatory synaptic transmission. Studies evaluating the effects of
LEV have found it beneficial for both monotherapy and adjunct
therapy in tumor-related epilepsy.78,79 The synaptic vesicle protein
2A (SV2A) is considered to be the binding site for LEV.80,81 Recently,
Lee et al.82 demonstrated that enhancing the activity of ROMK1
channels, which stabilize the resting membrane potential, may be
an important mechanism of LEV in preventing seizure initiation.
LEV has been shown to be very effective and well tolerated in
patients with brain tumor-associated seizures.
In a recent study, the gap junction inhibitor carbenoxolone has
been evaluated in organotypic hippocampal slice cultures as a
potential antiepileptic agent. It was found that it inhibited both
spontaneous and evoked seizure-like events.83 The clinical utility
of this strategy remains uninvestigated, but it is possible that it
may be valuable in the context of low-grade gliomas accompanied
by elevated CX43 expression.
4. Conclusions
Epileptic seizures are common in patients with brain tumors.
The underlying pathogenesis of tumor-related epilepsy is not
effectively addressed by conventional AEDs. To date, the epilepto-
genic mechanisms of tumor-related epilepsy remain largely
unclear. Tumor cells may create internal epileptogenicity through
their intrinsic characteristics, and inadequate homeostasis in the
peri-tumoral tissue may cause alterations in the balance between
excitation and inhibition. Other local changes in electrolytes,
158 G. You et al. / Seizure 21 (2012) 153–159
Fig. 3. Schematic of observed mechanisms underlying the pathogenesis and clinical strategies of tumor-related epilepsy, showing the possible intra- and peri-tumoral factors
causing tumor-related seizures and how current treatments take effects. (Factors are ranked in the top left corner based on their presumed importance in epileptogenesis.)
perfusion, metabolism, and enzymes could also contribute. It is
generally accepted that changes in neurotransmission and
extracellular ion concentrations are the most important mecha-
nisms underlying tumor-related seizures. Hypoxia, acidosis, and
metabolic, immunological, and inflammatory changes also play
potential roles in the induction of seizures.
This review has summarized the possible pathogenesis of
tumor-associated epilepsy and has also created a framework on
which novel therapeutics may be applied (Fig. 3). Further
investigations of both intra- and peri-tumoral pathophysiology
may give guidance in the choice of medical therapy. Increased
understanding of the dynamic processes at the tumor–brain
interface may lead to novel concepts and treatment strategies for
tumor-associated epilepsy in the future.
References
1. Jackson JH. Localized convulsions from tumour of the brain. Brain 1882;5:364–
74.
2. van Breemen MS, Wilms EB, Vecht CJ. Epilepsy in patients with brain tumours:
epidemiology, mechanisms, and management. Lancet Neurol 2007;6:421–30.
3. Lote K, Stenwig AE, Skullerud K, Hirschberg H. Prevalence and prognostic
significance of epilepsy in patients with gliomas. Eur J Cancer 1998;34:98–102.
4. Danfors T, Ribom R, Berntsson SG, Smits A. Epileptic seizures and survival
during early disease in grade 2 gliomas. Eur J Neurol 2009;16:823–31.
5. Piepmeier J, Christopher S, Spencer D, Byrne T, Kim J, Knisel JP, et al. Variations
in the natural history and survival of patients with supratentorial low-grade
astrocytomas. Neurosurgery 1996;38:872–8.
6. Chang EF, Potts MB, Keles GE, Lamborn KR, Chang SM, Barbaro NM, et al. Seizure
characteristics and control following resection in 332 patients with low-grade
gliomas. J Neurosurg 2008;108:227–35.
7. Chaichana KL, Parker SL, Olivi A, Quinones-Hinojosa A. Long-term seizure
outcomes in adult patients undergoing primary resection of malignant brain
astrocytomas. J Neurosurg 2009;111:282–92.
8. Beaumont A, Whittle IR. The pathogenesis of tumour associated epilepsy. Acta
Neurochir (Wien) 2000;142:1–15.
9. Rosati A, Tomassini A, Pollo B, Ambrosi C, Schwarz A, Padovani A, et al. Epilepsy
in cerebral glioma: timing of appearance and histological correlations. J Neu-
rooncol 2009;93:395–400.
10. Wolf HK, Roos D, Blumcke I, Pietsch T, Wiestler OD. Perilesional neurochemical
changes in focal epilepsies. Acta Neuropathol 1996;91:376–84.
11. Norden AD, Blumenfeld H. The role of subcortical structures in human epilepsy.
Epilepsy Behav 2002;3:219–31.
12. Vecht CJ, van Breemen M. Optimizing therapy of seizures in patients with brain
tumors. Neurology 2006;67:S10–3.
13. Schaller B, Ruegg SJ. Brain tumor and seizures: pathophysiology and its impli-
cations for treatment revisited. Epilepsia 2003;44:1223–32.
14. Engel Jr J. Seizures and epilepsy. Philadelphia: FA Davis; 1989. p. 221–39.
15. Duffau H, Capelle L. Preferential brain locations of low-grade gliomas. Cancer
2004;100:2622–6.
16. Huang L, You G, Jiang T, Li G, Li S, Wang Z. Correlation between tumor-related
seizures and molecular genetic profile in 103 Chinese patients with low-grade
gliomas: a preliminary study. J Neurol Sci 2011;302:63–7.
17. Persidsky Y, Ramirez SH, Haorah J, Kanmogne GD. Blood–brain barrier: struc-
tural components and function under physiologic and pathologic conditions. J
Neuroimmune Pharmacol 2006;1:223–36.
18. Liebner S, Fischmann A, Rascher G, Duffner F, Grote EH, Kalbacher H, et al.
Claudin-1 and claudin-5 expression and tight junction morphology are altered
in blood vessels of human glioblastoma multiforme. Acta Neuropathol
2000;100:323–31.
19. Chi OZ, Hunter C, Liu X, Tan T, Weiss HR. Effects of VEGF on the blood–
brain barrier disruption caused by hyperosmolarity. Pharmacology 2008;82:
187–92.
20. Stewart PA, Hayakawa K, Farrell CL, Del Maestro RF. Quantitative study of
microvessel ultrastructure in human peri-tumoral brain tissue. Evidence for a
blood–brain barrier defect. J Neurosurg 1987;67:697–705.
21. Ivens S, Kaufer D, Flores LP, Bechmann I, Zumsteg D, Tomkins O, et al. TGF-beta
receptor mediated albumin uptake into astrocytes is involved in neocortical
epileptogenesis. Brain 2007;130:535–47.
22. Cacheaux LP, Ivens S, David Y, Lakhter AJ, Bar-Klein G, Shapira M, et al.
Transcriptome profiling reveals TGF-beta signaling involvement in epilepto-
genesis. J Neurosci 2009;29:8927–35.
23. Aronica E, Gorter JA, Jansen GH, Leenstra S, Yankaya B, Troost D, et al. Expres-
sion of connexin 43 and connexin 32 gap-junction proteins in epilepsy-associ-
ated brain tumors and in the perilesional epileptic cortex. Acta Neuropathol
2001;101:449–59.
24. Kunapuli P, Chitta KS, Cowell JK. Suppression of the cell proliferation and
invasion phenotypes in glioma cells by the LGI1 gene. Oncogene 2003;22:
3985–91.
25. Besleaga R, Montesinos-Rongen M, Perez-Tur J, Siebert R, Deckert M. Expression
of the LGI1 gene product in astrocytic gliomas: downregulation with malignant
progression. Virchows Arch 2003;443:561–4.
26. Brodtkorb E, Nakken KO, Steinlein OK. No evidence for a seriously increased
malignancy risk in LGI1-caused epilepsy. Epilepsy Res 2003;56:205–8.
 G. You et al. / Seizure 21 (2012) 153–159
27. Piepoli Z, Jakupoglu C, Gu W, Lualdi E, Suarez-Merino B, Poliani PL, et al.
Expression studies in gliomas and glial cells do not support a tumor suppressor
role for LGI1. Neuro-oncol 2006;8:96–108.
28. Bordey A, Sontheimer H. Properties of human glial cells associated with
epileptic seizure foci. Epilepsy Res 1998;32:286–303.
29. Ye ZC, Sontheimer H. Glioma cells release excitotoxic concentrations of gluta-
mate. Cancer Res 1999;59:4383–91.
30. Chernov MF, Kubo O, Hayashi M, Izawa M, Maruyama T, Usukura M, et al.
Proton MRS of the peri-tumoral brain. J Neurol Sci 2005;228:137–42.
31. Baayen JC, de Jongh A, Stam CJ, de Munck JC, Jonkman JJ, Trenite DG, et al.
Localization of slow wave activity in patients with tumor-associated epilepsy.
Brain Topogr 2003;16:85–93.
32. Douw L, van Dellen E, de Groot M, Heimans JJ, Klein M, Stam CJ, et al. Epilepsy is
related to theta band brain connectivity and network topology in brain tumor
patients. BMC Neurosci 2010;11:103.
33. Villemure JG, de Tribolet N. Epilepsy in patients with central nervous system
tumors. Curr Opin Neurol 1996;9:424–8.
34. Chubinidze AI, Gobechiia ZV, Abramishvili VV, Samodurova GV, Chubinidze
MA. Morphological characteristics of cortical synapses in patients with epilep-
sy. Zh Nevropatol Psikhiatr Im S S Korsakova 1985;89:23–6.
35. McNamara JO. Emerging insights into the genesis of epilepsy. Nature 1999;399:
A15–22.
36. De Keyser J, Mostert JP, Koch MW. Dysfunctional astrocytes as key players in the
pathogenesis of central nervous system disorders. J Neurol Sci 2008;267:3–16.
37. Seiferta G, Carmignotob G, Steinhäusera C. Astrocyte dysfunction in epilepsy.
Brain Res Rev 2010;63:212–21.
38. Fidler IJ, Yano S, Zhang RD, Fujimaki T, Bucana CD. The seed and soil hypothesis:
vascularization and brain metastasis. Lancet Oncol 2002;3:53–7.
39. Kempski O, Staub F, Jansen M, Schodel F, Baethmann A. Glial swelling during
extracellular acidosis in vitro. Stroke 1988;19:385–92.
40. Schaller B. Influences of brain tumor-associated pH changes and hypoxia on
epileptogenesis. Acta Neurol Scand 2005;111:75–83.
41. Herholz K, HeindelW. Luyten PR, denHollander JA, Pietrzyk U, Voges J, et al. In
vivo imaging of glucose consumption and lactate concentration in human
gliomas. Ann Neurol 1992;31:319–27.
42. Martinez-Outschoorn UE, Prisco M, Ertel A, Tsirigos A, Lin Z, Pavlides S, et al.
Ketones and lactate increase cancer cell stemness, driving recurrence, metas-
tasis and poor clinical outcome in breast cancer. Cell Cycle 2011;10:1271–86.
43. Hossmann KA, Seo K, Szymas J, Wechsler W. Quantitative analysis of experi-
mental peri-tumoral edema in cats. Adv Neurol 1990;52:449–58.
44. Kraft R, Basrai D, Benndorf K, Patt S. Serum deprivation and NGF induce and
modulate voltage-gated Na(+) currents in human astrocytoma cell lines. Glia
2001;34:59–67.
45. Berdiev BK, Xia J, McLean LA, Markert JM, Gillespie GY, Mapstone TB, et al. Acid-
sensing ion channels in malignant gliomas. J Biol Chem 2003;278:15023–34.
46. Avoli M, Louvel J, Pumain R, Köhling R. Cellular and molecular mechanisms of
epilepsy in the human brain. Prog Neurobiol 2005;77:166–200.
47. Badawy RA, Harvey AS, Macdonell RA. Cortical hyperexcitability and epilepto-
genesis: understanding the mechanisms of epilepsy—part 1. J Clin Neurosci
2009;16:355–65.
48. Avoli M, Drapeau C, Louvel J, Pumain R, Olivier A, Villemure JG. Epileptiform
activity induced by low extracellular magnesium in the human cortex main-
tained in vitro. Ann Neurol 1991;30:589–96.
49. Singh R, Pathak DN. Lipid peroxidation and glutathione peroxidase, glutathione
reductase, superoxide dismutase, catalase and glucose-6-phosphate dehydro-
genase activities in FeCl3 induced epileptogenic foci in the rat brain. Epilepsia
1990;31:15–26.
50. Sontheimer H. Ion channels and amino acid transporters support the growth
and invasion of primary brain tumors. Mol Neurobiol 2004;29:61–71.
51. Sontheimer H. An unexpected role for ion channels in brain tumor metastasis.
Exp Biol Med (Maywood) 2008;233:779–91.
52. Bateman DE, Hardy JA, McDermott JR, Parker DS, Edwardson JA. Amino acid
transmitter levels in gliomas and their relationship to the incidence of epilepsy.
Neurol Res 1988;10:112–4.
53. Xiao MY, Gustafsson B, Niu YP. Metabotropic glutamate receptors in the
trafficking of ionotropic glutamate and GABAA receptors at central synapses.
Curr Neuropharmacol 2006;4:77–86.
54. Aronica E, Yankaya B, Jansen GH, Leenstra S, van Veelen CW, Gorter JA, et al.
Ionotropic and metabotropic glutamate receptor protein expression in glio-
neuronal tumours from patients with intractable epilepsy. Neuropathol Appl
Neurobiol 2001;27:223–37.
55. Lee MC, Kang JY, Seol MB, Kim HS, Woo JY, Lee JS, et al. Clinical features and
epileptogenesis of dysembryoplastic neuroepithelial tumor. Childs Nerv Syst
2006;22:1611–8.
56. Rutecki PA, Sayin U, Yang Y, Hadar E. Determinants of ictal epileptiform
patterns in the hippocampal slice. Epilepsia 2002;43(Suppl. 5):179–83.
159
57. Coutinho V, Knopfel T. Metabotropic glutamate receptors: electrical and chem-
ical signaling properties. Neuroscientist 2002;8:551–61.
58. Haglund MM, Berger MS, Kunkel DD, Franck JE, Ghatan S, Ojemann GA. Changes
in gamma-aminobutyric acid and somatostatin in epileptic cortex associated
with low-grade gliomas. J Neurosurg 1992;77:209–16.
59. Labrakakis C, Patt S, Hartmann J, Kettenmann H. Functional GABA(A) receptors
on human glioma cells. Eur J Neurosci 1998;10:231–8.
60. Vezzani A, Gramsbergen JB, Speciale C, Schwarcz R. Production of quinolinic
acid and kynurenic acid by human gliomas. Adv Exp Med Biol 1991;294:
691–5.
61. Hwa GG, Avoli M. The involvement of excitatory amino acids in neocortical
epileptogenesis: NMDA & non NMDA receptors. Exp Brain Res 1991;86:248–56.
62. Brenner AV, Butler MA, Wang SS, Ruder AM, Rothman N, Schulte PA, et al.
Single-nucleotide polymorphisms in selected cytokine genes and risk of adult
glioma. Carcinogenesis 2007;28:2543–7.
63. Wiemels JL, Wiencke JK, Kelsey KT, Moghadassi M, Rice T, Urayama KY, et al.
Allergy-related polymorphisms influence glioma status and serum IgE levels.
Cancer Epidemiol Biomarkers Prev 2007;16:1229–35.
64. Meme W, Calvo CF, Froger N, Ezan P, Amigou E, Koulakoff A, et al. Proinflam-
matory cytokines released from microglia inhibit gap junctions in astrocytes:
potentiation by beta-amyloid. FASEB J 2006;20:494–6.
65. Avignone E, Ulmann L, Levavasseur F, Rassendren F, Audinat E. Status epilepti-
cus induces a particular microglial activation state characterized by enhanced
purinergic signaling. J Neurosci 2008;28:9133–44.
66. Chao CC, Molitor TW, Hu S. Neuroprotective role of IL-4 against activated
microglia. J Immunol 1993;151:1473–81.
67. Block ML, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the
molecular mechanisms. Nat Rev Neurosci 2007;8:57–69.
68. Hulkkonen J, Koskikallio E, Rainesalo S, Keranen T, Hurme M, Peltola J. The
balance of inhibitory and excitatory cytokines is differently regulated in vivo
and in vitro among therapy resistant epilepsy patients. Epilepsy Res
2004;59:199–205.
69. Löscher W. How to explain multidrug resistance in epilepsy? Epilepsy Curr
2005;5:107–12.
70. Kim SK, Wang KC, Hwang YS, Kim KJ, Cho BK. Intractable epilepsy associated
with brain tumors in children: surgical modality and outcome. Childs Nerv Syst
2001;17:445–52.
71. Giulioni M, Galassi E, Zucchelli M, Volpi L. Seizure outcome of lesionectomy in
glioneuronal tumors associated with epilepsy in children. J Neurosurg
2005;102:288–93.
72. Hildebrand J, Lecaille C, Perennes J, Delattre JY. Epileptic seizures during follow-
up of patients treated for primary brain tumors. Neurology 2005;65:212–5.
73. Glantz MJ, Cole BF, Forsyth PA, Recht LD, Wen PY, Chamberlain MC, et al.
Practice parameter: anticonvulsant prophylaxis in patients with newly diag-
nosed brain tumors. Report of the Quality Standards Subcommittee of the
American Academy of Neurology. Neurology 2000;54:1886–93.
74. Temkin NR. Antiepileptogenesis and seizure prevention trials with antiepileptic
drugs: meta-analysis of controlled trials. Epilepsia 2001;42:515–24.
75. Li XN, Shu Q, Su JM, Perlaky L, Blaney SM, Lau CC. Valproic acid induces growth
arrest, apoptosis, and senescence in medulloblastomas by increasing histone
hyperacetylation and regulating expression of p21Cip1, CDK4, and CMYC. Mol
Cancer Ther 2005;4:1912–22.
76. Fu J, Shao CJ, Chen FR, Ng HK, Chen ZP. Autophagy induced by valproic acid is
associated with oxidative stress in glioma cell lines. Neuro-oncol 2010;12:
328–40.
77. Weller M, Gorlia T, Cairncross JG, van den Bent MJ, Mason W, Belanger K, et al.
Prolonged survival with valproic acid use in the EORCT/NCIC temozolomide
trial for glioblastoma. Neurology 2011;77:1156–64.
78. Dinapoli L, Maschio M, Jandolo B, Fabi A, Pace A, Sperati F, et al. Quality of life
and seizure control in patients with brain tumor-related epilepsy treated with
levetiracetam monotherapy: preliminary data of an open-label study. Neurol Sci
2009;30:353–9.
79. Lee JW, Bromfield EB, Kesari S. Antiemetic properties of the antiepileptic drug
levetiracetam. N Engl J Med 2008;359:1853.
80. Gillard M, Chatelain P, Fuks B. Binding characteristics of levetiracetam to
synaptic vesicle protein 2A (SV2A) in human brain and in CHO cells expressing
the human recombinant protein. Eur J Pharmacol 2006;536:102–8.
81. de Groot M, Aronica E, Heimans JJ, Reijneveld JC. Synaptic vesicle protein 2A
predicts response to levetiracetam in patients with glioma. Neurology
2011;77:532–9.
82. Lee CH, Lee CY, Tsai TS, Liou HH. PKA-mediated phosphorylation is a novel
mechanism for levetiracetam, an antiepileptic drug, activating ROMK1 chan-
nels. Biochem Pharmacol 2008;76:225–35.
83. Samoilova M, Wentlandt K, Adamchik Y, Velumian AA, Carlen PL. Connexin 43
mimetic peptides inhibit spontaneous epileptiform activity in organotypic
hippocampal slice cultures. Exp Neurol 2008;210:762–75.