If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Neurostimulation is making its way into the therapeutic armamentarium of the epileptologists, with several invasive neurostimulation modalities available today and several less invasive modalities under investigation. Clinicians will soon face a choice that should not be made randomly. We introduce the concept of a prestimulation evaluation protocol, consisting of a series of rationally chosen investigations that evaluate the presence of biomarkers for response to various neurostimulation therapies. These biomarkers should reflect the susceptibility of the individual’s epileptic network to a given neurostimulation technique. This will require elucidation of the specific mechanism(s) of action of the different neurostimulation modalities. This manuscript provides a hypothetical framework that may be more applicable in the near future when pre-clinical research progresses and can be translated into human applications.
Neurostimulation is making its way into the therapeutic armamentarium of epileptologists treating patients with drug-resistant epilepsy. For vagus nerve stimulation (VNS) [
Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group.
], efficacy and side effects profile have been demonstrated in large multicenter RCTs (see Table 1). During the blinded phase of the randomized trials seizure frequency was reduced with 30% in VNS [
Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group.
Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System Pivotal trial.
Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group.
Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System Pivotal trial.
Several other neurostimulation modalities are currently under investigation in a pre-clinical or clinical setting: DBS in other brain targets (e.g., hippocampus) and non-invasive neurostimulation techniques such as transcutaneous VNS (tVNS), non-invasive VNS (nVNS), trigeminal nerve stimulation (TNS), repetitive transcranial magnetic stimulation (rTMS) and transcutaneous direct current stimulation (tDCS).
Neurostimulation therapies are currently available to patients who are considered unsuitable candidates for epilepsy surgery based on the investigations performed during the presurgical evaluation protocol. For a long time, VNS was the only available neurostimulation treatment option. More recently ANT-DBS and RNS have also become available for unsuitable surgery candidates. To avoid a merely negative selection procedure and in view of an increasing number of neurostimulation options becoming available, we introduce in this manuscript the concept of a prestimulation evaluation protocol. We envisage a protocol consisting of a series of rationally chosen investigations that evaluate the presence of biomarkers for response to various neurostimulation therapies. A patient-tailored approach is warranted also for neurostimulation and may optimize the potential therapeutic success of neurostimulation as well as help physicians to rationally propose a specific therapy at a given moment in an epilepsy patient’s life.
2. Current modus operandi
In the absence of head-to-head comparative trials with VNS, ANT-DBS and RNS, the results of the RCTs suggest that in choosing a therapy for unsuitable surgery candidates none of the neurostimulation options are superior with regards to efficacy. Differences in invasiveness and adverse events may therefor direct the choice between these options [
Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group.
]. VNS is associated with stimulation-related side effects such as hoarseness, coughing, dyspnea or a pain sensation in the throat; side effects that typically wear off after long-term treatment [
Apart from taking into account absolute and relative contra-indications, in clinical practice the choice between the three neurostimulation options is often based on a combination of considerations and personal clinical experience of the physician or the epilepsy center where the patient is being treated. In Table 2 we have summarized some arguments in favor of or against the available treatment options. In analogy to choosing between one of many available anti-epileptic drugs, comorbidities may play a role in the choice of neurostimulation therapies. Comorbidities are prevalent in refractory epilepsy patients, with some medical and psychiatric conditions occurring up to eight times more often compared to the general population [
Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 clinical guidelines for the management of adults with major depressive disorder: section 4. Neurostimulation treatments.
]. The SANTE trial not only reported higher rates of self-reported depression but also of subjective memory impairment in the active treatment group. For patients with pre-existing cognitive problems other neurostimulation options may be preferred. Of notice is that in the open label follow-up ANT-DBS trial neuropsychological tests scores did not differ significantly and the long-term follow-up report of Salanova et al. even showed an improvement on multiple neuropsychological domains at the group level [
]. The presence of obstructive sleep apnea warrants caution with regards to VNS, since vagal stimulation may increase the apnea-hypopnea index (AHI). Therefor it may be recommended to perform screening for excessive daytime sleepiness (e.g., the Epworth sleepiness scale or ESS) with or without polysomnography to diagnose sleep apnea before implantation, as well as to follow-up AHI after implantation [
FDA approved, reimbursement varies in United States, not reimbursed in Europe.
Generalized epilepsy
+
/
/
Partial epilepsy
≤2 foci
+
+
+
Multifocal
+
+
/
Children
+
/
/
Invasiveness
+
−
−
Occupation
Voice
−
Comorbidities
Depression
+
−
Memory impairment
+
−
Sleep apnea
−
SUDEP risk
−
+ in favor of; − against; / not available.
Abbreviations: VNS, vagus nerve stimulation; ANT-DBS, deep brain stimulation of the anterior thalamic nucleus; RNS, Responsive Neurostimulation System; SUDEP, sudden unexpected death in epilepsy; FDA, Food and Drug Administration.
a FDA approval is pending, reimbursed in Europe, not in United States.
b FDA approved, reimbursement varies in United States, not reimbursed in Europe.
Regulatory issues like reimbursement procedures may also interfere with today’s treatment selection. In children and in patients with generalized epilepsy, VNS is the only approved modality. RNS is approved by the Food and Drug Administration (FDA) in the United States, with variable reimbursement, but not available in Europe, whereas ANT-DBS is only available in Europe; both RNS and ANT-DBS are indicated in patients with refractory partial epilepsy.
3. Future strategies
The currently reported efficacy outcome (30–40% response rate) of the different neurostimulation modalities are still considered to be modest. Different mechanisms of action (MOA) are hypothesized for the different neurostimulation modalities. We can therefor assume that outcome is affected by (1) a variability in underlying pathophysiology or a difference in brain networks involved in different types of epilepsy and (2) the potential of specific types of neurostimulation to affect a given underlying disturbance. Choosing a given neurostimulation modality on the basis of a particular MOA in relation to a particular type of epilepsy, may enhance outcome and decrease the number of non-responders.
Patient selection for a given treatment and the investigations to be performed should focus on a priori response prediction. In the past 2 decades, large patient series have been treated with VNS but retrospective correlation analyses between patient characteristics and outcome have been disappointing. Overall, VNS seems more effective in patients with younger age and shorter duration of epilepsy [
Vagus nerve stimulation in children with drug-resistant epilepsy: age at implantation and shorter duration of epilepsy as predictors of better efficacy?.
] and importantly none of these currently described ‘predictors’ for response allows to make predictions at the individual level. In depth investigations performed in patients who went through the presurgical evaluation process, have focused on the identification of an epileptogenic lesion, preferably an anatomically identifiable lesion corresponding to a neurophysiological ictal onset. It is unlikely that successful response prediction will arise from the analysis of factors that are unrelated to the MOA of the applied neurostimulation technique. In the field of neurostimulation the identification of dynamic biomarkers may be more appropriate. Such biomarkers should reflect one or more key features of the MOA of a given neurostimulation treatment. Accordingly the prestimulation investigations to be performed should evaluate whether the epileptic network in a particular patient is likely to be modulated by a given mode of action. Preferably this assessment should be evidence-based and standardized allowing to define a prestimulation evaluation protocol, in analogy to the presurgical evaluation in drug resistant epilepsy patients. A prestimulation protocol should provide strategical guidance to clinicians in objectively choosing the most optimal neurostimulation therapy at a given moment in a patient’s treatment process. However, the MOA of the various neurostimulation techniques remains to be elucidated. In fact, only for VNS and more than twenty years after its initial use in patients an evidence-based hypothesis on the working mechanism and translation of the findings to patient treatment is currently under investigation [
4. A prestimulation evaluation protocol applied to VNS
Several lines of evidence support the VNS-induced activation of the vagus nerve–locus coeruleus–norepinephrenic (VN–LC–NE) system as the key mechanism in its seizure-reducing effect. The afferent fibers of the VN project to the nucleus of the solitary tract (NTS) and either directly or indirectly to the LC, the main NE nucleus of the brain [
]. In a limbic seizure model we previously demonstrated a strong positive correlation between this VNS-induced increase in NE and the seizure-suppressing effect of VNS [
The level of VNS-induced NE release may therefor be predictive of the seizure-suppressing effect, whereas a lack of NE increase following VNS could be indicative of non-response. Unlike in experimental animals, intracerebral NE monitoring in humans is impossible. However, there are some non-invasive indirect measures of NE in the brain that are currently being investigated as biomarkers for response prediction.
The P3 (or P300) component of the event-related potential (ERP) may represent such a non-invasive NE measure [
] and it is hypothesized that the P3 is modulated differently in VNS responders as opposed to non-responders. Two separate studies have evidenced a significant P3 enhancement following VNS in responders only, one in depression [
], was found to be affected by VNS. In an experimental animal set-up, electrical stimulation of the afferent VN induced bilateral intensity-dependent pupillary dilatation [
] and Desbeaumes Jodoin et al. found a significantly larger resting pupillary diameter while VNS was ON compared to OFF in 21 patients (7 with major depression, 14 with epilepsy) chronically treated with VNS [
], which could be illustrative of either sympathetic activation or parasympathetic inhibition. Unfortunately this study showed no relationship between the percentage of symptom improvement and pupillary measures, but further assessment of this parameter is ongoing.
A second approach to identify dynamic biomarkers for VNS relates to the heart–brain connection, of which the vagus nerve is considered to represent the anatomical correlate [
Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology.
]. In comparison to healthy subjects, the autonomic regulation of HR in epilepsy patients appears dysfunctional, reflected by changes in heart rate (HR), heart rate variability (HRV) and morphological ECG changes [
] and are thought to result from ictal activation of cortical autonomic centers, giving rise to either ictal tachycardia or ictal bradycardia, depending on a shift to either sympathetic or parasympathetic predominance [
]. Research is being conducted based on the hypothesis that VNS, which entails electrical stimulation of a major autonomic structure, may be particularly suited to affect seizures that involve cortical areas that are implicated in the autonomic nervous system. A second hypothesis is that the presence of impaired HRV in epilepsy patients could be reflective of autonomic disbalance that is susceptible to modulation with VNS. Both hypothesis imply that the presence of ictal HR changes in epilepsy patients may be predictive of VNS response.
A third potential biomarker approach comprises the measurement of cortical excitability by means of transcranial magnetic stimulation (TMS). In 2004, Di Lazzaro et al. showed that acute VNS produced a selective and pronounced increase in intracortical inhibition measured with single- and paired-pulse TMS over the motor cortex [
]. Moreover, they found a correlation between optimal clinical outcome and VNS-induced cortical modulation. These findings remain to be replicated in larger populations, but represent opportunities for VNS response prediction.
There is growing evidence for a genetic predisposition for susceptibility to neuromodulation. A common single nucleotide polymorphism (SNP) in the gene for BDNF (brain-derived neurotrophic factor), Val66Met, present in up to 33% in a predominantly Caucasian population [
]. Poor response to rTMS and tDCS in Met carriers has been described, and this finding may apply for neuromodulation in general, and thus also for VNS.
As one of the main goals of response prediction is to prevent idle risky and expensive interventions in likely non-responders, biomarker assessment ideally takes place before VNS is implanted. Both tVNS and nVNS were developed as techniques to non-invasively stimulate the VN [
], with a MOA similar to conventional VNS, making them potentially suitable to combine non-invasive stimulation and biomarker assessment. Modulation of the P3 component, pupillary diameter or TMS measures of cortical excitability may as such be predictive of subsequent VNS response. A similar strategy may be applicable for TNS, since the affected trigeminal nerve fibers also project to the aforementioned brain stem nuclei [
A trial period with the non-invasive neurostimulation techniques, although still under investigation, may be part of the prestimulation evaluation protocol. In theory, a clinical response induced by the non-invasive neurostimulation modality would imply a high positive predictive value for subsequent response to the corresponding invasive neurostimulation modalities, which can be considered in view of long-term compliance. However, a lack of response to non-invasive modalities may not allow to exclude patients from invasive treatment, as the latter may nevertheless be more effective by more efficiently stimulating neural tissue and therefore inducing stronger effects.
5. Conclusion
Due to the rapidly expanding field of neurostimulation, a standardized prestimulation evaluation protocol for drug resistant epilepsy patients to identify a suitable neurostimulation treatment option for individual patients seems to be required. We suggest a tailored-approach in order to improve the outcome of patients treated with neurostimulation, focusing on response prediction making use of dynamic biomarkers.
We introduce the concept of a prestimulation evaluation protocol and advocate that it should be implemented early on in the work-up of drug resistant epilepsy patients, in some patients maybe even in parallel to the presurgical evaluation protocol. We believe neurostimulation is not just a last resort option that deserves a positive choice, rather than being the result of a negative selection process. Due to the limited knowledge of the MOA of neurostimulation and in the absence of validated biomarkers, the full implementation of a prestimulation evaluation protocol is premature, but we do propose a draft version of how such a protocol would look like and shape the therapeutic decision process (Fig. 1). We believe this concept and proposed framework may direct future research. The investigation of the MOA of the different neurostimulation modalities with a translational focus may identify novel strategies to predict individual response and increase individual treatment efficacy. These should be studied in well-designed clinical trials including specific patient populations of adequate size.
Fig. 1Potential design of a prestimulation evaluation protocol.
Sofie Carrette has received a travel grant from Magventure and MedCat to participate in TMS and TMS-EEG workshops. Kristl Vonck has received personal compensation for consulting for LivaNova. Paul Boon has received personal compensation for consulting from LivaNova, Medtronic and UCB. Kristl Vonck and Paul Boon have received research support (including for clinical trials) through their institution from Cerbomed, LivaNova, Medtronic, Neurosigma and UCB.
Acknowledgements
Dr. S. Carrette is supported by an aspirant grant from ‘Fonds voor Wetenschappelijk Onderzoek’ (FWO) Flanders. Prof. Dr. P. Boon is supported by grants from FWO Flanders, grants from BOF and by the Clinical Epilepsy Grant from Ghent University Hospital. Prof. Dr. K. Vonck is supported by a BOF-ZAP grant from Ghent University Hospital.
References
Ben-Menachem E.
Manon-Espaillat R.
Ristanovic R.
Wilder B.J.
Stefan H.
Mirza W.
et al.
Vagus nerve stimulation for treatment of partial seizures: 1. A controlled study of effect on seizures. First International Vagus Nerve Stimulation Study Group.
Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: final results of the RNS System Pivotal trial.
Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 clinical guidelines for the management of adults with major depressive disorder: section 4. Neurostimulation treatments.
Vagus nerve stimulation in children with drug-resistant epilepsy: age at implantation and shorter duration of epilepsy as predictors of better efficacy?.
Heart rate variability. Standards of measurement, physiological interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology.
30208-4&id=gr1.jpg){kind=link}