Abstract
A Novel, Noninvasive, Predictive Epilepsy Biomarker with Clinical Potential.
Choy M, Dubé CM, Patterson K, Barnes SR, Maras P, Blood AB, Hasso AN, Obenaus A, Baram TZ. J Neurosci 2014;34:8672–8684.
A significant proportion of temporal lobe epilepsy (TLE), a common, intractable brain disorder, arises in children with febrile status epilepticus (FSE). Preventative therapy development is hampered by our inability to identify early the FSE individuals who will develop TLE. In a naturalistic rat model of FSE, we used high-magnetic-field MRI and long-term video EEG to seek clinically relevant noninvasive markers of epileptogenesis and found that reduced amygdala T2 relaxation times in high-magnetic-field MRI hours after FSE predicted experimental TLE. Reduced T2 values likely represented paramagnetic susceptibility effects derived from increased unsaturated venous hemoglobin, suggesting augmented oxygen utilization after FSE termination. Indeed, T2 correlated with energy-demanding intracellular translocation of the injury-sensor high-mobility group box 1 (HMGB1), a trigger of inflammatory cascades implicated in epileptogenesis. Use of deoxyhemoglobin-sensitive MRI sequences enabled visualization of the predictive changes on lower-field, clinically relevant scanners. This novel MRI signature delineates the onset and suggests mechanisms of epileptogenesis that follow experimental FSE.
Commentary
Choy and colleagues address one of the most challenging issues in epilepsy: Who of the patients with febrile seizures will ultimately develop temporal lobe epilepsy (TLE), which in ~30% of cases is refractory to pharmacotherapy? There is a significant link between febrile seizures with complex features occurring in children and development of TLE later in life (1). Children experiencing febrile status epilepticus (FSE), febrile seizures with focal features, or repeated episodes of febrile convulsions have significantly increased risk for developing TLE. The combination of all three factors increases future occurrence (within 8–12 years) of TLE up to almost 50%. Thus, a population of children suffering from febrile seizures with complex features will eventually take two pathways: One cohort of about 50% of these children will go through the process of epileptogenesis resulting in TLE, with the prospect of approximately a third of them being refractory to treatment. The other cohort will not develop epilepsy. Previously, the authors devoted significant efforts to the research on febrile seizures in children as well as on hyperthermia-induced seizures in animal models (2), and they are fully aware of the dire outcomes.
One might suggest that in those children experiencing febrile seizures with complex features, common sense dictates an initiation of antiepileptogenic therapy, that is, a treatment that would prevent epileptogenesis resulting in emergence of (likely) refractory TLE. Unfortunately, this approach has at least two shortcomings: First, antiepileptogenic treatment is still a work in progress; while some drugs may have certain features useful for suppressing epilepsy development, the results are inconclusive (3). Second, if common anti-seizure drugs—such as phenytoin, phenobarbital, and valproate—are used in immature individuals, there is a risk of increased rate of apoptosis in the brain (4). The group of the late Dr. Karen Gale revisited this topic both in the immature brain without seizure experience and after repeated seizures. Early neonatal treatment with phenytoin, phenobarbital, and lamotrigine decreased striatal synaptic connectivity in the naïve brain without seizures (5). However, if seizures had been present before the treatment started, the outcome for at least some drugs improved, resulting in decreased apoptosis rate (6). Thus, blanket administration of anticonvulsant drugs to all children with febrile seizures with complex features raises serious concerns, as the basic research indicates that these drugs may have detrimental effects in one-half of those children, and in those with slowly developing epileptogenesis only uncertain benefits may be accomplished.
Finding a biological marker (“biomarker”), a characteristic that can be objectively (and possibly noninvasively) measured and could become an indicator for future epileptogenesis would be a milestone in the relationship between febrile seizures and TLE. The investigators returned to their well-established model of FSE induced in infant rats by a hot stream of air now combined with early high-intensity magnetic field MRI investigation. In the model of FSE (7), similar to the human condition, about 30 to 40 percent of rats experiencing seizures during the period with increased core temperature will eventually develop TLE. However, in contrast to humans, the progression in a rodent model is more rapid and lasting months, allowing for quick validation of potential biomarkers. Well-thought controls were included: A group of rats not exposed to the hot stream of air (i.e., without an increase in core temperature) and a group of rats exposed to hot air experiencing relevant increase in core temperature but without concurrent seizures that were prevented by administration of pentobarbital.
Thus, the true biomarker in this study should be able to identify the population of rats eventually developing TLE after the initial FSE insult. Translating this all-or-none outcome into quantifiable scale, the biomarker values in those rats with FSE should form two clusters: one not far from the clustered biomarker values and one remote. The former group of rats would be expected not to develop epilepsy, while in the latter, the epileptogenic process should culminate in TLE. In this study, authors used noninvasive high-intensity field (11.7T) MRI to investigate differences among the groups. At two hours after FSE, the whole brain T2 relaxation times were significantly shorter than in either normothermic controls or in rats with hyperthermia with seizure suppression by pentobarbital. Moreover, among those rats with FSE, there was a subgroup with T2 relaxation values more than two SDs lower than others, giving a hope for identification of the epileptogenic subgroup. Unfortunately, long-term follow-up of those animals did not confirm the role of whole brain T2 relaxation times as a predictor of epileptogenesis. Similarly, duration of FSE, length of hyperthermia, average core temperature, threshold temperature, and so on, were not associated with the development of epilepsy.
Therefore, the authors focused on focal limbic asymmetries in T2 values (using a region of interest approach) because the regional limbic asymmetries correspond to findings in children after FSE (8). When asymmetries were found, lower T2 values were compared among the groups. T2 relaxation time values in the basolateral (and medial) amygdala of FSE rats that developed epilepsy were significantly lower than those in nonepileptic rats after FSE as well as in the controls. After all mathematical and statistical corrections to remove false discoveries, the probability of the effect was 0.03 and the effect size d = 1.77. The differences recorded at 2 hours post-FSE persisted for 4 hours, but the predictability disappeared at 18 hours. Most importantly, these results were reproduced in low magnetic field intensity scanners (4.7T), which are available in clinical settings.
To explain their findings of decrease in T2 relaxation times, the authors proposed and were able to demonstrate a correlation with the levels of unsaturated hemoglobin. This finding is suggestive of increased tissue oxygen extraction; for example, by a process demanding high energy. The authors hypothesized and identified one such process with a plausible relationship to epileptogenesis. They found that in basal amygdala in rats with T2 relaxation time decreases, there is an increase in translocation of high-mobility group box 1 molecule (HMGB1) (a molecular sensor of cellular injury) from the nucleus to cytoplasm, which is a first step preceding its release from the cell. This translocation process is extremely energy demanding. Released HMGB1 then significantly contributes to seizure development in chronic epilepsy, particularly through its interaction with inflammatory markers, such as toll-like receptor 4 (TLR4) (9). Additionally, there is a significant downstream recruitment of the N-methyl-D-aspartate glutamate receptors containing NR2B (Grin2B) subunits, which may further fuel epileptogenesis.
In conclusion, the authors describe a reliable noninvasive technique of early prediction of epileptogenesis after febrile status epilepticus resulting in TLE, which is applicable to and should be tested in the clinical setting. However, there are a couple of additional questions to be clarified: First, is this bio-marker also applicable for TLE resulting from febrile seizures with complex features other than extended duration and for the combinations of complex features? Second, is there a reasonable advance in development of antiepileptogenic treatment that could be properly applied once the high risk for development of TLE is identified?