Volume 8, Issue 1 pp. 221-234
CONCEPTS AND HYPOTHESES
Open Access

Systemic inflammation as a biomarker of seizure propensity and a target for treatment to reduce seizure propensity

Coral Stredny

Coral Stredny

Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA

Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA

Search for more papers by this author
Alexander Rotenberg

Alexander Rotenberg

Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA

Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA

Search for more papers by this author
Alan Leviton

Corresponding Author

Alan Leviton

Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA

Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA

Correspondence

Alan Leviton, Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital, Boston Massachusetts, USA.

Email: [email protected]

Search for more papers by this author
Tobias Loddenkemper

Tobias Loddenkemper

Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital, Boston, Massachusetts, USA

Department of Neurology, Harvard Medical School, Boston, Massachusetts, USA

Search for more papers by this author
First published: 15 December 2022
Citations: 1

Abstract

People with diabetes can wear a device that measures blood glucose and delivers just the amount of insulin needed to return the glucose level to within bounds. Currently, people with epilepsy do not have access to an equivalent wearable device that measures a systemic indicator of an impending seizure and delivers a rapidly acting medication or other intervention (e.g., an electrical stimulus) to terminate or prevent a seizure. Given that seizure susceptibility is reliably increased in systemic inflammatory states, we propose a novel closed-loop device where release of a fast-acting therapy is governed by sensors that quantify the magnitude of systemic inflammation. Here, we review the evidence that patients with epilepsy have raised levels of systemic indicators of inflammation than controls, and that some anti-inflammatory drugs have reduced seizure occurrence in animals and humans. We then consider the options of what might be incorporated into a responsive anti-seizure system.

1 INTRODUCTION

Some seizures can be fairly reliably forecasted1 and even prevented.2 This usually requires the placement of intracranial electrodes and a responsive neurostimulation system. While non-invasive forecasting of seizures may be feasible in some patients with a combination of neurophysiological and clinical data,1, 3, 4 a systemic, chemical marker of seizure propensity is not available. We suggest that a complementary and potentially less invasive approach modeled on the closed-loop systems for release of insulin in response to continuously monitored blood glucose levels might achieve the goals of forecasting and intervention.5

In light of evidence that some seizures are associated with inflammation identified peripherally,6-15 we suggest that a closed-loop system that identifies rising concentrations of an inflammation indicator in a body fluid and delivers an anti-inflammatory agent (either highly specific or broad-spectrum) or more anti-seizure medication16-18 might be more acceptable than closed-loop electroencephalographic systems requiring a neurosurgical procedure, and may complement other means of non-invasive seizure forecasting, such as clinical information, seizure diaries, and non-invasive neurophysiological data. We provide more detail about this option.

2 PREDICTION

We envision two time scales of probability assessment. The shorter time frame is likely seconds or minutes, and the presumed best early indicator of state shift is the EEG.

The longer time frame is likely hours or days, and the frame of reference is “forecasting.”19-21 This orientation takes advantage of the increasing availability of online seizure diaries allowing “cycles” (circadian, multi-day) to be identified,3, 22-25 and machine-learning to be applied to identifying correlates of increased seizure propensity,4, 26, 27 similar to what has been done for predicting the likelihood of sustained hypoglycemia.28 Identification of the trajectory that best describes a child's pre-ictal course would allow an intervention minutes to hours before seizure risk rises rapidly.

3 PROOF OF CONCEPT

Proof of concept that seizures can be forecasted for up to 3 days comes from a study of EEG and clinical data of 157 adults with drug-resistant focal epilepsy followed at 35 US medical centers.19 Proof of principle/concept that seizures can be aborted within a very short time through a closed-loop system after receipt of a signal of heightened seizure propensity comes from documentation of the success of neurostimulation.2, 29, 30

4 INFLAMMATION

A plethora of recent reviews has linked inflammation and epilepsy.6-15, 31, 32 Although elevated body temperature is probably the best-known indicator of inflammation that provides predictive information about increased risk of a seizure, we do not know of any study that has assessed how well serum/plasma levels of inflammation-related proteins convey information about impending seizures.

Inflammation-associated hereditary characteristics33-36 can contribute to epileptogenesis. Later inflammation-related exposures probably account for some of the seizures associated with auto-immune encephalitides,37 febrile infection-related epilepsy syndrome (FIRES), and new-onset refractory status epilepticus (NORSE),15 as well as the late seizures that follow head trauma.38, 39 In addition, some/many children and adults with seizures of presumed idiopathic origin are more likely than others to have genetic propensities to inflammation and other immune-related processes.40, 41

4.1 Both “chicken and egg”?

Some of the inflammation associated with seizures might be compensatory, reflecting the recruitment of the “clean-up squad” (e.g., microglia, astrocytes) to remove debris and initiate repair.42 On the other hand, some reactive glia can contribute to epileptogenesis.10 Although sterile inflammation of the brain need not require white cell infiltration,43-46 activated peripheral mononuclear immune cells can contribute to seizure generation.47

In addition, pre-clinical studies document that the inflammation can come before, during, and after the seizure.48 Studies in humans are needed to determine to what extent inflammation precedes seizures, and what biomarkers of inflammation are an appropriate signal for intervention. While inflammation may not be the sole predictor of seizures in all patients, evaluation of the contribution of inflammation, including the gain of additional predictive information for seizures, is crucial.

4.2 NLRP3 inflammasome

Inflammasomes are multiprotein complexes that promote the availability of pro-inflammatory cytokines, such as interleukin-1β and interleukin-18.49, 50 Among various inflammasome complexes, the NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome can contribute to secondary brain damage51 and is activated in rodent models of epilepsy,52-57 and in humans with a seizure disorder.36, 57-61 A recently-published review provides additional details,62 and another suggests inflammasomes might be good targets for therapeutic intervention.63

4.3 Non-inflammatory stimuli that induce inflammation-related epileptogenesis

Support for the hypothesis that inflammation contributes to ictogenesis comes from rodent models of pilocarpine-induced, kainic acid-induced, and pentylenetetrazole-induced seizure propensities that include a robust inflammatory response.64-70 Anti-inflammatory approaches involving inhibition of prostaglandin EP2 receptors,65, 66 inhibition of mTOR signaling,64 and inhibition of neurotensin receptor 2 reduce the inflammation-associated phenomena (including seizure occurrence) attributed to pilocarpine, and thus offer support for the hope that these anti-inflammatory approaches might reduce seizure occurrence in humans.

5 MEASUREMENT OF INFLAMMATION—WHAT TO MEASURE?

The first decision point is whether the biomarker should be a general indicator of inflammation or one more specific to an individual patient.

If a general indicator will do, perhaps the least invasive way to identify a systemic inflammatory process contributing to seizure onset is to measure body temperature continuously with a wearable device. We are not aware of any assessments of the value measuring body temperature continuously. Most likely this reflects the previous unavailability of non-invasive wearable devices capable of measuring a surrogate for body temperature. Wristbands are now available that can continuously measure and record skin temperature.1, 4 Their usefulness for identifying heightened risk of a seizure, however, remains to be determined.

Tables 1 and 2, which list inflammation-related biomarkers that have been measured in human peripheral blood near the time of a seizure, include broad-spectrum indicators of inflammation along with specific cytokines, chemokines, adhesion molecules, growth factors, and other biomarkers of inflammation.

TABLE 1. Products of inflammation documented repeatedly in the peripheral blood of patients with recurrent seizures in reports published since a 2016 review7
CRP Various seizure disorders Case-control Children, adults [71-76]
IL-6 Various seizure disorders Case-control Children, adults [58, 72, 77-81]
HMGB-1 Various seizure disorders Case-control Children [82-85]
IL-1β Various seizure disorders Case-control Children, adults [58, 77, 82, 85]
TNF-α Various seizure disorders Case-control Children, adults [82, 86, 87]
interferon-γ Various seizure disorders Case-control Children, adults [80, 87, 88]
miRNAs Temporal lobe epilepsy Case-control Children, adults [84, 89, 90]
IL-17A Various seizure disorders Case-control Adults [80, 87]
MMP-9 TLE + Generalized Seizures Case-control Children, adults [72, 73]
prolactin TLE and febrile seizures Case-control Children, adults [73, 91]
Caspase 3 Generalized Seizures Case-control Adults [77, 86]
  • Abbreviations: CRP, C-Reactive Protein; HMGB-1, High Mobility Group Box 1; IL, Interleukin; miRNA, micro-ribonucleic acids; MMP, Matrix-MetaloProtein; TLE, Temporal Lobe Epilepsy; TLE-hs, Temporal Lobe Epilepsywith hippocampus sclerosis; TNF-α, Tumor necrosis factor-alpha.
TABLE 2. Products of inflammation measured in peripheral blood reported since publication of a review7 in 2016. Elevated concentrations of these analytes have been associated with seizures in single reports only
IL-R1 Refractory seizures Case-WCE control Children [82]
IL-5 drug-resistant TLE Case-control Adults [79]
IL-8 Generalized Seizures Case-control Adults [92]
IL-10 (lower levels) TLE-hs Case-other sz Adults [93]
IL-18 + IL-18BP Generalized Seizures Case-control Adults [94]
IL-22 Drug-resistant epilepsy case-WCE control Adults [87]
IL-33 Epilepsy Case-control Adults [95]
sTNFr2 Epilepsy Case-control Adults [96]
TRAIL Generalized Seizures Case-control Teens, adults [13]
TLR4 Refractory seizures Case-WCE control Children [82]
NLRP3 Febrile Seizures Case-feb control Children [97]
ICAM-1 Generalized Seizures Case-control Teens, adults [13]
MCP-2 Generalized Seizures Case-control Teens, adults [13]
Eotaxin (CCL11) Refractory seizures Case-control Children [98]
NFκB mRNA Focal seizures Case-control Adults [58]
tRNA Focal epilepsy Case-control Adults [99]
BDNF Epilepsy Case-control Adults [96]
GMCSF Drug-resistant epilepsy Case-WCE control Adults [87]
NGF Epilepsy Case-control Adults [96]
NT3 Epilepsy Case-control Adults [96]
Caspase 1 Generalized Seizures Case-control Adults [77]
AACT New onset epilepsy Case-control Children [85]
homocysteine Symptomatic epilepsy Case-control Adults [72]
kallikrein Temporal lobe epilepsy Case-control Adults [74]
bradykinin Temporal lobe epilepsy Case-control Adults [74]
α-synuclein Generalized seizures Case-control Children [100]
CD4 + CD38+ drug-resistant TLE Case-control Adults [79]
Adiponectin Febrile seizures Case-control Children [81]
Copeptin Febrile seizures Case-control Children [81]
GDF-15 Status with fever Case-control Children [101]
Zfas1 Temporal lobe epilepsy Case-control Adults [102]
P2X7R Temporal lobe epilepsy Case-control Adults [76]
  • Note: blood CD14+ mononuclear cells. AACT α1-antichymotrypsin. Zfas1 long non-coding RNA Zfas1. P2X7R P2X7 receptor.
  • Abbreviations: BDNF, Brain-Derived Neurotrophic Factor; GDF-15, Growth/differentiation factor-15 (macrophage inhibitory cytokine-1); GMCSF, Granulocyte-Macrophage Colony-Stimulating Factor; ICAM-1, intercellular adhesion molecule 1; IL, Interleukin; MCP-2, monocyte chemoattractant protein-2; mRNA, micro-ribonucleic acids; NFκB, nuclear factor kappa-light-chain-enhancer of activated B cells mRNA measured in peripheral; NGF, Nerve Growth Factor; NLRP3, nucleotide-binding domain, leucine-rich repeat family, pyrin-domain containing 3 inflammasome; NT3, Neurotrophic Factor-3; sTNFr2 soluble TNF-α, Tumor necrosis factor-receptor; TLE, Temporal Lobe Epilepsy; TLE-hs, Temporal Lobe Epilepsy with hippocampus sclerosis; TLR4, Toll-Like Receptor-4; TRAIL, Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand; tRNA, transfer RNA; WCE, well-controlled epilepsy.
  • a measured in peripheral blood mononuclear cells and not in serum or plasma.
  • b and an acquired demyelinating disorder.
  • c T and B regulatory cells.

5.1 High-sensitivity C-reactive protein

Serum levels of high-sensitivity C-reactive protein, (hs-CRP), apparently the most frequently studied inflammation biomarker in people with epilepsy,71 not only convey information about acute inflammation but can also provide information about chronic processes.103-108

The five identical monomers (mCRP) of the circulating pentameric (pCRP) are capable of activating the complement cascade, and thereby able to amplify inflammation.109 This recognition has led to the exploration of ways to minimize the inflammatory capabilities of mCRP.110

A meta-analysis of 16 case-control studies (comprised 1918 individuals) found that the CRP blood levels “were significantly increased in epileptic patients compared to healthy controls, indicating a significant association between inflammation and epilepsy.”71 Subsequent individual studies have confirmed this.72-75, 111

5.2 Procalcitonin

Procalcitonin (PCT), another frequently studied inflammation biomarker of seizures,112, 113 is now considered superior to CRP for evidence of infection and inflammation.114-116 Nevertheless, its specificity is also less than desired.114, 117 Decreases in elevated PCT levels were once thought to mirror improvement so well that they could be a source of information for decisions about whether or not to discontinue antibiotic treatment.118 This approach, however, is now viewed with some skepticism.119

5.3 Other biomarkers of inflammation

Among the more interesting candidates are microRNA (miRNA)120-123 and circular RNA (circRNA),124, 125 which can play a prominent role in regulating the immune system.121, 126, 127 In vivo measurement capabilities are not yet available.

Transfer RNAs (tRNAs) constitute another group of candidates.128, 129 tRNAs function as adaptor molecules that help ribosomes decode messenger RNA (mRNA), transfer amino acids, and synthesize proteins,130 and are therefore capable of influencing inflammatory processes. Pre-seizure plasma samples obtained from patients with focal epilepsy had higher levels of three tRNA fragments than did plasma from healthy controls.99 A point-of-care electrochemical system that measures three specific tRNA fragments is likely to be available to patients and their families very soon.131

6 MEASUREMENT OF INFLAMMATION—HOW TO MEASURE IN VIVO?

To identify changes in seizure propensity as soon as they happen requires continuous surveillance. The surveillance we consider most appropriate for the most rapid identification of a change in seizure propensity is an EEG.132, 133 Integrating EEG and inflammation signals has the potential to enhance progress in working out the best signals for initiating prophylaxis.

Forecasting for less urgent intervention might not need to rely on electrophysiologic indicators. Surveillance for this task might do as well, or potentially better, relying on frequent assessments of a circulating indicator of inflammation. The closest analogy of the ideal we envision is the continuous measurement of blood glucose by a sensor attached to an insulin pump worn by some people who have relatively unstable diabetes mellitus.134 By adjusting drug release in response to a biosignal, such closed-loop delivery strategies are designed to approximate homeostasis.16-18

Although frequent intermittent static assessments (“snapshots”) might be adequate, they are less desirable than continuous assessments (“video”), which can identify fluctuations over time from a given baseline, and also identify periodic patterns.135 The best candidate for this task remains to be determined.

Proof of concept that systemic inflammation is associated with seizure propensity comes from literature reviews6-15 and Tables 1-3, which were prepared to provide more recently published support for inflammation in peripheral blood accompanying or soon following seizures.

TABLE 3. Potential anti-inflammation therapies intended to reduce seizure occurrence reported since a 2016 review7
Focus Drug Disorder Recipients Reference
Human
IL-1 Anakinra NORSE, FIRES Children [136-144]
Rasmussen's Children [145]
Rilonacept CAPS Children [34]
Canakinumab CAPS Children [35, 36]
IL-6 Tocilizumab FIRES Children [146-154]
IL-6, CRP Statins Post-stroke szs Adults [155-159]
IL-6, IL-2, CRP VPA + LEV Epilepsy Children [160]
TNF-α Adalimumab Rasmussen's Children [161]
HMGB1 Melatonin Epilepsy Adults (+rodents) [162]
α-4 integrin Natalizumab Drug-resistant Adults [163]
NMDAR antibody Rituximab NMDARE Children + adults [164, 165]
COX-2 Aspirin Sturge-Weber Children [166-171]
Focal seizures Adults [172]
Pre-clinical
microRNA ASOs Drug-resistant Rats + mice [173]
HMGB1 mAb Epilepsy Rats + mice [174]
PGE2 PGE2 RA Status epilepticus Rats + mice [171]
IL-1β and IL-6 Lacosamide LPS + pilocarpine Mice [175]
IL-1β and IL-6 serotonin PTZ-induced Rats [176]
PEA Multiple models Mice and rats [177]
Minocycline Multiple models Mice and rats [178]
  • Abbreviations: ASOs, Anti-Sense Oligonucleotides (antagomirs); CAPS, Cryopyrin-Associated Periodic Syndrome; CD20, Cluster of differentiation 20; FIRES, Febrile Infection-Related Epilepsy Syndrome; LEV, levetiracetam; mAb, monoclonal antibody; NMDAR, N-methyl-d-aspartate receptor; NMDARE, N-methyl-d-aspartate receptor (NMDAR) antibody encephalitis; NORSE, New Onset Refractory Status Epilepticus; PEA, palmitoylethanolamide; PGE2 RA, Prostaglandin EP2 Receptor Antagonist; PTZpentylenetetrazole; VPA, valproate.

Proof of concept that “prodromal biomarkers” … [might] be used to predict seizure risk comes from a report that concentrations of a marker of systemic inflammation (e.g., tRNA fragments) are higher before a seizure than afterward.99

Proof of concept that seizures can be predicted based on rising concentrations of an inflammation biomarker is not yet available.

7 TIMING

Almost all the reports documenting circulating indicators of inflammation in people with epilepsy provide measurements after a recent or remote seizure. Indeed, ongoing systemic inflammation (elevated CXCL8 [IL-8] concentrations) has been documented in adults a week after the last overt seizure,75 and even months later.92

Levels of tRNA fragments were higher in pre-seizure plasma than in plasma collected soon after the seizure.99 This is the only evidence that suggests inflammation predates seizure onset in humans. But even this would not be enough. Documentation is needed that levels of inflammation indicators increase prior to seizure onset.

Missing are reports of increasing concentrations of an inflammation indicator preceding a seizure.121 Without this type of information everything presented here is rather indirect support for the putative benefits of a closed-loop system. Consequently, studies documenting increasing concentrations of a biomarker of inflammation is essential.

8 HOW BEST TO INTERVENE

8.1 Closed-loop systems

8.1.1 The sensor

The first closed-loop system for seizures we know of delivered an anti-seizure medication in response to EEG discharges.179 The closed-loop system we envision will work best if measurements of inflammation indicators are made continuously in vivo.

Although cytokines and other proteins are now measured ex vivo as part of a panel,13, 180, 181 potentially wearable nanosensors offer the promise of continuous measurement in vivo.182, 183 The availability of fluorescence sandwich immunoassays for continuous measurements/detection of cytokines and other proteins in the blood184 adds to the hope of being able to continuously measure what conveys helpful information about seizure propensity.

Just as efforts are underway to create a bionic pancreas,185 in which beta cells sense glucose levels, the hope is that implanted inflammation-sensitive cells that sense inflammation might be able to function as the ideal sensors and responsive interventions.

8.1.2 Set points

The level of inflammation that best triggers prophylaxis probably needs to be determined for each patient individually. Some propose variable setpoints, at least initially.186

The information collected by a sensor provides current data, but does not provide any information about the occurrence of a subsequent event. The closed-loop system we consider most appropriate also needs to have the capability of taking the sensor-collected data and anticipating a delayed future event.

Most attempts to predict seizure occurrence utilize information extracted from electroencephalograms.187 We are not aware of any report of efforts to predict seizure occurrence based on the concentration of any blood component. Consequently, the most appropriate model for doing so would seem to come from the literature about predicting sustained hypoglycemia based on blood glucose levels collected continuously by an in situ sensor.188, 189 In essence, machine-learning algorithms would determine the magnitude or slope of concentration changes of the inflammation biomarker that best predicts/anticipates the next seizure.190

9 CANDIDATES FOR DELIVERY

9.1 Broad spectrum

Anti-inflammatory treatments are applied in clinical practice, and albeit the working mechanism may not be solely tied to anti-inflammatory properties, reduction of inflammation may play a role. This includes steroid application in epilepsy syndromes such as infantile spasms, Landau-Kleffner Syndrome, Continuous Spikes and Slow Waves during Sleep, or Lennox Gastaut syndrome,191 potential anti-inflammatory properties of vigabatrin192 and metformin64 through their effects on the mTOR pathway, anti-inflammatory properties of the ketogenic diet,193 or use of (broad-spectrum) immunomodulatory drugs (as outlined in Table 3).194

9.2 Specific anti-inflammatory drugs

Recent reviews and Table 3 document inflammation pathways activated in drug-resistant epilepsy,11 and provide proof of concept that anti-inflammatory agents and strategies can reduce seizure occurrence.48

The more specific, the more personalized the approach,195, 196 the closer the realization of “precision medicine.”197 In the spirit of “one size does not fit all,” we offer as candidates the anti-inflammatory agents first those listed in Table 2, but consider other options that have yet to be tested. This personalized approach may also entail application of anti-seizure or anti-inflammatory interventions only when needed. Some might appear to target specific steps in the inflammatory process. However, inflammatory cascades are complex incorporating interconnected sets of networks.50, 198 Consequently, interfering in the function of just one component can sometimes appreciably reduce the extent and severity of inflammation.199

More than a decade ago, antibodies that neutralize a single cytokine were known to have broad anti-inflammatory effects.199, 200 More recently, small-molecule kinase inhibitors were shown to have broad anti-inflammatory effects.201 With the multiple effects of inflammasomes,202, 203 anti-inflammasome drugs have the potential to affect several targets,63 each with some potential to initiate an inflammatory cascade.49, 204 Thus, single-target drugs, or those expected to have limited effects sometimes/often achieve their broad potential by limiting inflammation in multiple ways.205

Because some/many of the inflammation-related biomarkers of seizure risk are not specific, and because some/many children with epilepsy might not have appreciable change in any of them as seizure risk rises, our vision might have limited application.

9.3 The future

A recently proposed roadmap for biomarker research has five phases.206 Tables 1 and 2 and related text document some of the progress made in four of the phases in identifying and quantifying circulating biomarkers of seizure diathesis. These four phases include preclinical exploratory studies (phase 1), initial clinical assessments (phase 2), retrospective studies of data in repositories (phase 3), and prospective studies to determine diagnostic accuracy (phase 4). The main component of phase five is assessing the reduction in mortality, morbidity, and disability associated with biomarker testing and the therapeutic intervention. Table 3 offers guidance about potential therapeutic interventions. However, we are not aware of any study that has assessed to what extent an anti-inflammation intervention in response to an elevated blood concentration of a biomarker of inflammation before a seizure occurs reduces the probability of a seizure.

10 CONCLUSION

We have gathered support for our vision that people with epilepsy will be able to benefit from a wearable closed-loop system similar to those that benefit people with “difficult-to-control” epilepsy. This has included documentation that epilepsy is associated with inflammation, that both specific and broad-spectrum anti-inflammatory therapies have reduced seizure occurrence, that a wearable closed-loop system might be capable of identifying increasing concentrations of inflammation indicators in the blood or other sources, and that machine learning programs have the potential to identify the slope of these increasing levels of inflammation proteins most predictable of an impending seizure.

Yet, despite all this support, what is missing is documentation that circulating concentrations of an inflammation biomarker increase shortly before an impending seizure. Until indwelling inflammation biomarker sensors are available, this documentation will most likely come from animals whose blood is sampled sequentially.

AUTHOR CONTRIBUTIONS

Alan Leviton prepared the first draft; Coral Stredny, Alexander Rotenberg, and Tobias Loddenkemper then provided multiple edits iteratively. All authors read and approved the final draft.

FUNDING INFORMATION

Preparation of this report was supported by the Epilepsy Research Foundation. The sponsors had no involvement in the writing of this report or in the decision to submit it for publication.

CONFLICT OF INTEREST

The authors declare that they wrote this essay in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Coral Stredny is on the advisory board of the NORSE Institute and receives active research support from the Pediatric Epilepsy Research Foundation. Alexander Rotenberg is co-founder of Prevep Inc. and Neuromotion Labs, is on the advisory boards of Gamify and Neurorex, and has active research support or consulting agreements with Biomarin, CRE Medical, Encoded, Loulou Foundation, National Football League, Neuroelectrics, Roche and Praxis. Tobias Loddenkemper discloses pending and approved patents and intellectual property related to epilepsy management, seizure detection and prediction, device loans from Empatica, and receives active research support from Upsher-Smith.

ETHICS STATEMENT

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

    The full text of this article hosted at iucr.org is unavailable due to technical difficulties.