|Epilepsy Research Benchmarks|
| The 2007 Epilepsy Research Benchmarks are now available
|Judith Hoyer Lecture on Epilepsy|
|Anticonvulsant Screening Program (ASP)
NIH RePORTER is an electronic tool that allows users to search a repository of NIH-funded research projects and access publications and patents resulting from NIH funding.
|Brandy Fureman, Ph.D.
Program Director, Channels Synapses & Circuits Cluster
Epilepsy Benchmark IA1
Benchmark Area I: Understanding basic mechanisms of epileptogenesis
Section A: Discover the range of anatomical, physiological, and molecular substrates associated with the epilepsies; define unambiguous markers of epileptogenicity.
Specific Benchmark 1: Successfully develop a non-invasive, dynamic imaging system or physiological monitoring system that reliably identifies an epileptogenic region in at least one form of human epilepsy.
2005 Report submitted by Benchmark Steward:
Jerome Engel, M.D., Ph.D. (UCLA School of Medicine)
Background of the benchmark goal:
Epilepsy is a chronic disorder characterized by recurrent epileptic seizures, resulting from a persistent localized or diffuse disturbance of brain function. Although interictal (between seizures) EEG recordings can reveal epileptiform abnormalities, such as spikes, which help to diagnose the existence of an epileptic disorder, and to localize the epileptogenic region, these interictal electrographic events are relatively nonspecific: they do not occur in all people with epilepsy; they do occur in some people without epilepsy; their location does not accurately coincide with the region generating spontaneous seizures; no features of these events have been demonstrated to correlate with the severity of the epileptic condition; and they can not predict when an epileptic seizure will occur. There is a great need for noninvasive, interictal electrophysiological and/or neuroimaging surrogate markers that reliably indicate the existence of an epileptic disorder, the location and degree of epileptogenicity, and predict when and whether a next seizure will occur.
A noninvasive, interictal surrogate marker that reliably indicates the location and degree of epileptogenicity would: 1) greatly enhance our ability to screen potential epileptic compounds in experimental models of epilepsy (Benchmark IC3); 2) permit the effectiveness of antiepileptic drugs to be determined in individual patients without the need to wait for another seizure to occur (Benchmark IIA); 3) provide a substrate for evaluating new approaches to the treatment of epilepsy (IIIA); 4) provide an accurate and cost-effective means of localizing the epileptogenic region for resective surgical therapy (IIIF), and 5) if also indicative of epileptogenesis, allow determination of the risk for developing epilepsy, so that preventive treatments can be introduced (IIA). A surrogate marker that reliably predicts when seizures will occur will not only enhance patients’ ability to protect themselves from injury during these events and perhaps institute self-abortive measures, but will also provide opportunities to devise automated interventional feed-back systems to prevent spontaneous seizure occurrence (Benchmark IIIE).
Current status of field:
Overview of Interictal Surrogate Markers of Epileptogenicity and Epileptogenesis. Until recently, very little research was devoted to identifying reliable surrogate markers of epileptogenicity. Initial research focused on the interictal EEG epileptiform spike. Spike mapping during intraoperative electrocorticography was a common technique for determining the size of brain resections to treat intractable focal epilepsy, but the high degree of false localization based on this method was well-recognized. Theodore Rasmussen, a prominent neurosurgeon at the Montreal Neurologic Institute, noted that spike mapping should provide more reliable information because there are red spikes and green spikes; the problem is that electrophysiologists are not able to tell them apart. Attempts to distinguish red from green spikes based on aspects of spike morphology, such as rise time, sharpness, or aftercoming slow waves, by patterns of recurrence such a frequency, or by response to suppression by drugs such as methohexitol, were generally unsuccessful.
Recent studies suggest that EEG spikes with high frequency oscillations (250-600 Hz), termed Fast Ripples, may be unique to structures capable of generating spontaneous seizures in patients with mesial temporal lobe epilepsy, and also in animal models of this disorder. There has been considerable interest in this phenomenon in the past fourteen months, elucidating the specificity and potential neuronal substrates of epileptiform high frequency oscillations (1-14). Some slower oscillations in the “ripple” frequency (100-200 Hz) may also be abnormal epileptiform events with a similar significance with respect to epileptogenicity and these epileptiform high frequency oscillations occur early after an epileptogenic insult, long before seizures begin. High frequency oscillations, therefore, appear to be putative surrogate markers not only for epileptogenicity, but also for epileptogenesis (1).
Although these putative red spikes can so far only be detected by direct brain recordings, which require depth electrode implantation, they indicate that the pathophysiological mechanisms responsible for interictal EEG spikes originating in tissue capable of generating spontaneous seizures is different from those of propagated spikes, or spikes originating in secondary epileptogenic areas that are not capable of generating spontaneous seizures. Consequently, research has now begun to determine whether higher resolution noninvasive EEG and/or magnetoencephalography (MEG) can identify these high frequency epileptogenic oscillations. Other techniques such as functional magnetic resonance imaging (fMRI) have a high likelihood of detecting the metabolic consequences of these red spikes. A number of laboratories have successfully used EEG spike-triggered fMRI averaging to anatomically localize the source of interictal EEG spikes, and studies in the past 14 months continue to provide hope that the fMRI image produced by spikes originating in primary epileptogenic regions will be different from those that are propagated, or originate in secondary epileptogenic regions (15-31).
Positron emission tomography (PET) may provide another novel approach to identification of seizure-generating brain tissue. In patients with tuberous sclerosis who have multiple brain tubers and epileptic seizures, alpha methyl tryptophan (AMT) was found to accumulate preferentially in the tubers that generated spontaneous seizures (32). AMT uptake appears to detect the presence of increased activity in the kynurenine pathway, which may be important for the epileptogenicity of this condition. Two studies on AMT-PET were published in the past 14 months (33-34). In addition, 5-HT1A receptor ligands have been identified as showing changes with epilepsy. [18F]FCWAY binding was lower in the epileptic hippocampus of patients with mesial temporal lobe epilepsy (35), and [18F]MPPF also showed decreased binding in epileptic hippocampus (36). If these binding changes reflect dynamic disturbances, rather than merely cell loss, it is possible that labeling with a shorter half-life radionuclide would permit continuous studies over time to demonstrate changes in epileptogenicity.
Structural imaging does not indicate epileptogenicity and does not, per se, provide information that could be considered surrogate markers of epileptogenicity or epileptogenesis; however, MRI of rats following lithium-pilocarpine-induced status epilepticus suggests that changes identified in piriform and entorhinal cortices may predict which animals will develop epilepsy (37). I am unaware of any further work in this area since the last benchmark report.
A potentially exciting new development for functional MRI is the use of magnetized nanoparticles (MNP), extremely small magnetized particles that are visible on MRI, and that can be attached to a wide variety of bioactive molecules (38-42). Consequently, MNP could be used to measure localized alterations in neurotransmitter activities that reflect brain excitation and inhibition, cerebral metabolism, immune responses, and drug distribution with extremely high spatial resolution. The potential for this new technology to be used to identify surrogate markers of epileptogenesis and epileptogenicity is enormous.
Transcranial magnetic stimulation (TMS) is currently being investigated as a possible diagnostic, as well as therapeutic, tool for epilepsy at a number of laboratories. TMS, particularly paired-pulse stimulation, can be used to measure cortical excitability, a potential surrogate marker of epileptogenicity (43-52). It is conceivable that such TMS measures could eventually be employed to assess, for instance, the effect of an antiepileptic drug on reducing epileptogenicity without the need to wait for another seizure to occur.
Many investigators are now using gene microarray technology to examine alterations in genetic profile of animal and human brain regions that may correlate with an epileptic disturbance. Gene expression profiles reflect dynamic processes that could cause, or result from, changes in epileptogenicity and serve as surrogate markers not only of epilepsy, but of the severity of an epileptic condition. Genetic expression fingerprints of brain disorders can also be measured in peripheral blood (53-55), suggesting the possibility that genomic surrogate markers for epileptogenesis and epileptogenicity may eventually be monitored from a finger stick. If gene expression profile surrogate markers are identified in the brain, but not in peripheral blood, there may be ways to image them noninvasively with PET or MNP-MRI.
Optical intrinsic signal (OIS) imaging is an invasive technique for measuring changes in brain associated with neuronal activity. Several investigators are applying this technique to animal models of epilepsy and in patients during surgery, and it may yield results in future that could become surrogate markers of epileptogenicity. Only modest progress has been made in the past year (56-58).
Surrogate markers of epileptogenicity specifically for the purpose of screening potential antiepileptic compounds that are more relevant to current medically refractory epilepsy syndromes, and sufficiently cost-effective to permit rapid throughput of tens of thousands of compounds, would revolutionize drug discovery. Preliminary efforts in this direction might include the development of assays in small mammals such as mice, using Fast Ripples, but more innovative approaches are also being investigated with zebrafish and fruit flies. Ideally, drug screening ought to be accomplished using integrated circuits constructed from cultured neurons.
Overview of Surrogate Markers for Seizure Prediction. The use of depth-recorded epileptiform EEG events to predict the onset of epileptic seizures and automatically activate seizure-aborting interventions has a long history, and was popularized by the Michael Crichton novel and film “The Terminal Man.” Nevertheless, such efforts remained in the realm of science fiction until recent publications from several laboratories demonstrating that non-linear dynamic analysis of ongoing depth-recorded EEG activity can reliably identify changes in neuronal synchronization that precede epileptic seizure occurrence by several minutes. One of these laboratories has successfully applied this technique to scalp EEG recordings, demonstrating that noninvasive seizure prediction and automatic seizure abortion is feasible. Work in this area over the past 14 months provides additional evidence for dynamical surrogate markers defining the preictal state (59-69).
Interictal Surrogate Markers of Epileptogenicity and Epileptogenesis: NINDS research grants currently support work on Fast Ripples, PET with AMT, and OIS, while various sources, including private foundations and venture capital, are supporting MEG and fMRI-EEG spike imaging studies.
Surrogate Markers for Seizure Prediction: In addition to research grants from government and private foundations, at least two medical instrument companies (NeuroPace and Medtronics) are interested in a feed-back approach to seizure control and are currently funding investigations, not only to refine techniques for seizure prediction, but to create effective seizure-aborting interventions.
Top priorities for next 5-10 years:
The recent evidence that identification of reliable, noninvasive, surrogate markers is feasible should stimulate an increased interest in both basic and clinical research on these important topics. The NINDS could further this research direction by encouraging investigators to consider how their ongoing research may contribute to these goals and objectives. The time is now right for an international workshop on surrogate markers.
The most fruitful areas of investigations over the next five to ten years would include work to: 1) further establish and characterize the role of Fast Ripples as an interictal surrogate marker of epileptogenicity; 2) determine if EEG, MEG and/or fMRI can identify these events noninvasively; 3) develop novel PET tracers to measure compounds in the brain that are associated with epileptogenicity; 4) develop MNP technology in order to identify surrogate markers that can be imaged with MRI; 5) search for gene expression fingerprints measurable in peripheral blood that reflect epileptogenicity; and 6) pursue noninvasive computerized EEG analysis to further define pattern changes that precede seizure onset.
Roadblocks to progress:
The major roadblock to progress is the high risk/high payoff nature of these projects, which makes it extremely difficult to get NIH funding to pursue the most creative concepts.
Last updated January 12, 2010