Epilepsy research in 2021: multidisciplinary achievements

2021 was steeped in challenges associated with the COVID-19 pandemic, but many advances in epilepsy research have nevertheless been accomplished. Many of these endeavours were initiated and developed before the pandemic, but completed despite the disruptive lock down of social as well as academic activity from 2020 to 2021.

A highlight of the year was the increased understanding of the role of astrocytes in epileptogenesis.¹ The role of neurons in epileptogenesis has been extensively investigated. Once the detailed role of astrocytes in regulating excitatory glutamate release and in maintaining homoeostasis of extracellular potassium concentrations in the synaptic cleft were delineated, the significance of astrocytes in the development of epileptogenesis has now been analysed. Astrocytes were shown to have strong control over neuronal excitability and their mode of information processing.¹ The network of astrocytes connected via gap junctions allows either a wide or a more confined distribution of these ions, depending on gap junctions. In transgenic mice with astrocyte-specific expression of a pH sensor, astrocytes react to epileptiform activity with intracellular alkalisation. This finding suggests that alkaline pH shifts in astrocytes can lead to gap junction uncoupling, hampering potassium clearance, and therefore to exacerbation of epilepsy. This notion is consistent with the finding of ictal slow shifts (ictal direct current shifts) generated by astrocyte functional syncytium occurring before the onset of a conventional ictal EEG pattern in patients with focal epilepsy, assessed by invasive EEG recording before epilepsy surgery.²

Other interesting developments took place with the use of mathematical modelling of invasive ictal EEG and virtual brain models.3, 4 Non-linear dynamic mathematical modelling was applied to invasive ictal EEG and a new proposal for seizure classification was introduced.³ Following the analysis of more than 2000 focal-onset seizures in patients from multiple centres, 16 dynamotypes were proposed from the viewpoint of principle of operation by considering both the fast and infraslow EEG components. The mathematical models were further developed, ranging from a single neuron to a neural population, and combined with structural information from non-invasive brain imaging of diffusion-weighted MRI, thus building large-scale brain network models of patients.⁴ Such modelling enables clinicians to explain changes in spatiotemporal organisation of brain dynamics, such as epileptic focus resection, by means of the integration of the clinical hypotheses on the epileptogenic zone before surgery.

Another interesting finding was reported in people with benign adult familial myoclonus epilepsy (BAFME).5, 6 This rare disorder is autosomal dominantly inherited and is characterised by myoclonic tremor and epilepsy. Abnormal expansions of TTTCA and TTTTA repeats were noted in intron 4 of SAMD12 in people with BAFME type 1 (mainly seen in Japan) and other introns of STARD7 in people with BAFME type 2 (mainly seen in Europe). These non-coding repeat expansions cause common clinical features (cortical tremor, rare generalised tonic-clonic seizure, absent cerebellar symptoms, and intellectual impairment). Other types of BAFME (3, 4, 6, and 7) had different gene abnormalities but the intron of each gene commonly had the expansion of TTTTA and TTTCA.⁵ Furthermore, TTTCA repeat insertions in SAMD12 were proven to be pathogenic, whereas TTTTA repeat expansions in TNRC6A were non-pathogenic, which suggests there might be unknown factors in the ancestry of patients with BAFME because the TTTTA expansion in TNRC6A responsible for BAFME 6 (a rare type of BAFME in Japan) often occurs together with mutations in SAMD12, but clinically symptoms are not modified by this additional expansion in TNRC6A.⁶

Progress was also made in diagnosis of autoimmune encephalitis by use of a signs and symptoms score.⁷ The antibodies contributing focal epilepsy signs and symptoms (ACES) score encompasses six factors: temporal MRI hyperintensities, autoimmune diseases, behavioural changes, autonomic symptoms, cognitive symptoms, and speech problems. It was developed on the basis of prospectively collected data from patients with focal epilepsy of unknown cause; the score can be used to guide screening for autoimmune causes of seizures. The ACES score has subsequently been validated in a second, external cohort. An ACES score of at least 2 had a sensitivity of 100% to diagnose seizures of autoimmune origin, and a specificity of 84·9%.

Finally, there were also achievements in the field of big data analysis and epileptogenic networks.8, 9 In epilepsy surgery, subdural electrode implantations were very popular, but the procedure has been gradually replaced by stereotactic EEG (SEEG). This transition occurred because cortical malformation, one of the most important pharmacological-resistant epileptogenic pathologies, is often buried in the subcortical area, which is beyond the extent of subdural electrode recording. Additionally, an epileptogenic network as the ictal propagation has been emphasised to control even focal epilepsy, which could be better delineated by multiple SEEG than by subdural electrode. However, a direct comparison of clinical utility between the two methods and with big data remains to be investigated in a real-world setting.

The surgical therapies commission of the International League Against Epilepsy did an international registry of patients with invasive EEG implanted between 2005 and 2019, comprising 2012 patients, including 1468 (72·9%) eligible for analysis (526 by subdural electrode and 942 by SEEG).⁸ Subdural electrode evaluations rather than SEEG are more likely to lead to brain surgery in patients with drug-resistant epilepsy, but have more surgical complications and lower probability of seizure freedom. In relation to the concept of epileptogenic networks, closed-loop medial septum electrical stimulation can quickly terminate intrahippocampal seizures and suppress secondary generalisation in a rat kindling model, by means of time-targeted proxy stimulation for intervening pathological oscillations.⁹

Most epilepsy research questions have the fundamental aim of providing benefits for people with epilepsy and associated neurological disorders, which is synergistic with the aim of the ongoing global campaign against epilepsy.¹⁰

Department of Epilepsy, Movement Disorders and Physiology, to which I am currently affiliated, is the Industry-Academia Collaboration Courses, supported by Eisai, Nihon Kohden, Otsuka Pharmaceutical, and UCB Japan, outside of the submitted work.