1
Introduction
Epilepsy is a common symptom in patients with primary central nervous system (CNS)
tumors. Seizures are reported in 15–95% of patients with brain tumors, depending on
the type of tumor [1]. Seizures are the presenting symptom in 15–50% of patients with
gliomas, and up to 75% will have at least one seizure at some point in the disease
course [2].
The recognition and classification of epileptic seizures and providing effective treatment
are essential to preserve the quality of life of patients with glial tumors.
In a previous study, we found that 63% of patients with glial tumors experienced seizures
at some point in the disease course. Sixty two percent of seizures were focal onset,
while 38% were generalized onset seizures [3]. However, considering the tumor etiology
of seizures, we were unable to verify the results. Therefore, the goal of this study
was to present the results of a larger cohort of patients with primary glial tumors
and epilepsy based on a more rigorous examination approach.
The primary objectives of this study were to classify and quantify epileptic seizures
as a symptom of onset or later in the course of the disease and the correlation between
ILAE 1981 and 2017 classifications [5,6]. The secondary objectives were to quantify:
1) the cases in which epileptic seizures were reclassified; 2) patients who received
antiepileptic drugs (AEDs) as monotherapy; 3) patients whose antiepileptic treatment
on admission was modified; and (4) patients whose seizures were controlled with the
prescribed AEDs.
2
Patients and methods
This is a prospective, descriptive, observational, clinical research study of a cohort
of 82 patients with glial tumors obtained from the follow-up of 168 patients diagnosed
with primary CNS tumor from January 1, 2010 through March 31, 2013 at our institution.
The diagnosis of primary tumor was according to WHO classification of CNS tumors [4].
Characteristics considered include: histopathological diagnosis, epileptic seizure
type according to the International League Against Epilepsy (ILAE) classification
[5,6], antiepileptic treatment and response to it, as well as the modification of
antiepileptic drugs (AEDs), if it has been done. For didactic and chronological purposes,
we have maintained the description of the seizures according to both the 1981 and
2017 ILAE classifications. In all patients, seizure characterization and antiepileptic
treatment were compared with those conducted at other medical centers. This study
was approved by the Institutional Review Board.
3
Results
Eighty-two patients were included in the study: 49 men and 33 women. The majority
of cases (29%) occurred between the third and fourth decade (Fig. 1).
Fig. 1
Cases distribution by age.
Fig. 1
Histological types (WHO 2007) [4] are shown in Table 1.
Table 1
Histopathological diagnostic (WHO 2007) [4].
Table 1
Astrocytics tumors
Glioblastoma
28
Diffuse astrocytoma (GII)
22
Anaplastic astrocytoma (GIII)
8
Brainstem glioma (not biopsied)
2
Oligodendroglials tumors
Oligodendroglioma (GII)
14
Anaplastic oligodendroglioma (GIII)
7
Other neuroepithelial tumors
Astroblastoma
1
Eighty-eight percent of cases had lobular location with 59% of cases limited to a
single lobe: 19 frontal, 17 temporal and 9 parietal. The involvement of two or more
lobes, with frontotemporal predominance location, was observed in 29% of cases (Fig.
2). A gliomatosis cerebri growth pattern was observed in 4% of cases. The remaining
cases were distributed as the following: 5 basal ganglia, 2 brainstem, 2 spinal cord
and 1 cerebellum. Infratentorial cases (brainstem, cerebellum, and spinal cord) were
excluded.
Fig. 2
Cases distribution by sites.
Fig. 2
Seventy four percent (57/77) of cases had epilepsy. The types of epileptic seizures
diagnosed according to the ILAE 1981 and ILAE 2017 classifications were simple partial
(focal aware) 39%, partial complex seizure (focal impaired awareness) 13%, partial
secondarily generalized tonic-clonic (focal to bilateral tonic-clonic) 11%, and generalized
convulsive (generalized onset motor tonic-clonic) 37% (Fig. 3). In 15/57 patients
(26%), the type of seizure was reclassified, and generalized onset motor seizures
decreased from 37% to 17%. As a result, focal to bilateral tonic-clonic seizures increased
from 11% to 31%. Thus, focal onset seizures accounts for 83% of the types of seizures.
Fig. 3
Seizure types according to the reclassification.
Fig. 3
As expected, some patients presented more than one type of epileptic seizure at the
time of the first consultation. Table 2 shows the figures corresponding to each one
of them.
Table 2
Seizure types according to ILAE 2017 classification [6].
Table 2
Seizure types
n/% (75/100)
Focal onset
Aware
Motor onset
12/16
Nonmotor onset
17/23
Sensory
9/12
Autonomic
4/5
Cognitive
4/5
Impaired awareness
10/13
Focal to bilateral tonic-clonic
23/31
Generalized onset
13/17
Seizures were the presenting symptom in 79% of patients, and in the rest during the
course of the disease.
According to the histological type, epilepsy was observed in 49% and 51% of patients
with low-grade and high-grade glial tumors, respectively (Fig. 4).
Fig. 4
Seizures according to 2007 WHO grading gliomas.
Fig. 4
Seventy five out of 77 patients (97%) were on treatment with AEDs. Of these, 20 (27%)
discontinued AEDs treatment because they never experienced seizures. All neurosurgical
patients were already on treatment or received prophylactic treatment with phenytoin.
Fifty five of 57 patients with seizures (96%) were on AEDs at the time of the first
consultation at our institution, and 6/55 (11%) were on monotherapy with levetiracetam.
The two patients who were not on treatment started treatment with levetiracetam. Among
patients treated with AEDs other than levetiracetam, 31/49 (63%) were on phenytoin
monotherapy. AEDs were switched to levetiracetam in 35/49 (71%) cases. Finally, 43/57
(75%) received levetiracetam at a dose range of 1000 to 2500 mg per day. During follow-up
visits, 95% of patients with levetiracetam were seizure-free. Poor seizure control
in 2 patients required the addition of a second and a third drug; one with the addition
of lacosamide presented full seizure control, and the other with phenobarbital and
lamotrigine had occasional breakthrough seizure (drug-resistant epilepsy). Ultimately,
57/77 (74%) patients had seizures and received AEDs.
4
Discussion
In this cohort of patients with gliomas we observed a higher prevalence of male cases
and patients between 35 and 40 years old, similar to what was observed in a study
performed previously by our neurology department in 2010 [3]. Astrocytic tumors were
the most common type, especially glioblastoma.
More than half of gliomas were located within the frontal, temporal, and parietal
lobes. These findings were similar to previous studies on primary tumors of the glial
series [[7], [8], [9]], and also similar to our previous study [3].
Young age and long duration of illness are associated with an increased risk of secondary
epileptogenesis. The exact nature of epileptogenesis in patients with brain tumors
is unclear. It is probably multifactorial and caused by the different tumor types
and changes in the properties of the tumor-cell membranes that generate action potentials,
which affect neuronal excitability (e.g., raised concentrations of amino acids, neuroreceptors
disturbance, increased gap junctions between tumor cells, and low pH microenvironment)
[[10], [11], [12]].
Tumor type, histological grade and site are determining factors in the incidence and
frequency of epileptic seizures. Thus, slow growing low-grade gliomas with longer
survival times are more frequently associated with epilepsy than high-grade ones [10,13].
In our study, seizures were distributed almost equally among patients with high and
low-grade gliomas. This finding differs from previously published figures in the literature,
where seizures were observed in 65–95% of patients with low-grade gliomas and 15–34%
with high-grade gliomas [1,8,9,13,14]. Nevertheless, other studies reported up to
62% seizure frequency in glioblastomas [15]. In our cohort, 28 patients had glioblastoma
and 16/28 (57%) of them had seizures.
Intra-axial tumors with involvement of the cerebral cortex are more epileptogenic
than extra-axial and deep tumors. Tumors within frontal, temporal, and parietal lobes
are more linked to seizures than those with occipital location [10,13]. We believe
that focal onset seizures of this location may be well underdiagnosed.
In this cohort of patients seizures were prevalent (57/77, 74%), as well as in our
previous study [3]. Fifty-eight percent (45/77) had seizure as the presenting symptom
of the disease, which is more than what was reported by Glantz et al. (14–51%); however,
primary brain tumor and metastases were included in the meta-analysis [16]. Moreover,
the study mentions that the onset of seizures during the course of the disease (10–45%)
depends on factors such as tumor histology, location, age, and treatment [16]. In
this study, 16% of patients developed seizures at some point in the disease course.
These differences in both percentages are probably due to better information ascertainment
from patients and/or witnesses as data collection was comprehensive in our study.
Another previous study that included only patients with supratentorial glioblastomas
and seizures, reported seizure as the presenting symptom of the disease in 68% of
cases, and 32% developed seizures at some point in the disease course [15]. This is
similar to what we observed in our cohort: 62.5% and 37.5% for each group of patients,
respectively. It seems that in patients with high-grade gliomas, the presence of epilepsy
is more frequent than previously expected.
Focal onset seizures were the most common type in our study: focal aware, focal impaired
awareness, and focal to bilateral tonic-clonic. Prior to the reclassification of epileptic
seizures, the percentage of generalized onset motor tonic-clonic was 37%, similar
to that obtained in a previous study [3]. However, after the implementation of a rigorous
examination to patients, relatives, and/or witnesses, this number went down by 20%.
Another explanation for the overestimation of generalized onset motor tonic-clonic
seizures is that some patients may have experienced a rapid seizure evolution from
focal to bilateral tonic-clonic; and therefore, the focal phase was unnoticed. It
should be noted that from a practical point of view we had no difficulties incorporating
the ILAE 2017 classification for the purpose of this study [6]. We reinforce the concept
of using the actual classification.
In patients, de novo or with the diagnosis of an encephalic tumor, the priority for
the patients themselves and much more for the family is to begin the specific treatment
as soon as possible. The diagnosis of convulsion and AED indicated as prophylaxis
or treatment, goes into the background. Several of the authors have reexamined patients
and family members to better classify their epileptic seizures, especially among those
to whom AEDs prophylaxis were indicated and their potential withdrawal would have
resulted in great concerns by the patient and their families.
Seizure control was excellent (seizure-free) in almost all patients treated with levetiracetam.
We started AED monotherapy with levetiracetam, a drug we use since the early 2000s
that has been shown to be effective for focal and generalized onset seizures [3,[17],
[18], [19], [20], [21]], Currently, there is no evidence of levetiracetam interaction
with any known drug, especially drugs administered to neuro-oncologic patients, as
it does not have deleterious effects on cognition [[22], [23], [24], [25], [26]].
Reasons why patients with seizures did not start treatment on levetiracetam (25%)
were: 1) accessibility to medication; 2) poor health and death before treatment initiation;
and 3) transfer of care to other medical centers.
We acknowledge that phenytoin was the most common used drug for the treatment of neuro-oncologic
seizures outside our institution. Phenytoin was the most commonly discontinued AED
in favour of levetiracetam, as observed in our cohort and several previous studies
[9,20,27]. Also, we found that all patients without a history of epilepsy that went
on to have surgery were on long-term phenytoin treatment. In patients with newly diagnosed
brain tumor, AEDs are not recommended as epilepsy prophylaxis. Furthermore, AEDs tapering
and discontinuing 7-10 days after neurosurgery applies for patients who did not have
seizures [16,22,[28], [29], [30]].
This study is not without limitations. All cases were from a single medical center,
the Department of Neurology and Neuro-Oncology University Center, Oncologic Institute
“Ángel H. Roffo “, University of Buenos Aires, which might limit the generalizability
of our findings. Additionally, while our studies had a relatively larger sample size
compared to previous studies, it only analyzed data from 82 patients.
5
Conclusions
A rigorous examination of the patient and family/witnesses undoubtedly allows for
a better diagnosis and characterization of epileptic seizures in addition to the possibility
of withdrawing treatment in seizure-free patients.
One fifth of epileptic patients referred with generalized onset motor tonic-clonic
seizure were reclassified as focal to bilateral tonic-clonic. Focal onset seizures
are the most common type in patients with glial tumors.
Levetiracetam used as monotherapy is suitable for patients with epilepsy secondary
to glial tumors.
The transition of the ILAE classification from 1981 to 2017 was favorable, coherent,
and with phenomenological improvement.
Source of funding
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
Conflict of interest
None of the authors has any conflict of interest to disclose.
Ethical standards
This study was approved by the Institutional Review Board, Oncologic Institute “Ángel
H. Roffo” and conducted in accordance with the Declaration of Helsinki and current
Argentine legislation.