Is epilepsy a curable neurodegenerative disease?

This scientific commentary refers to ‘Hyperphosphorylated tau in patients with refractory epilepsy correlates with cognitive decline: a study of temporal lobe resections’, by Tai et al. (doi:10.1093/brain/aww187). Temporal lobe epilepsy (TLE) is the most frequent focal epilepsy in adults. The epileptogenic network responsible for seizures encompasses multiple limbic structures including the hippocampus and neighbouring cortices. The histomorphological hallmark of TLE is mesial temporal sclerosis, described originally in post-mortem studies as neuronal loss and gliosis in the hippocampus, entorhinal cortex and amygdala. About a third of patients do not respond to anti-epileptic medication despite continued development of new molecules. Drug-resistant TLE is a serious chronic condition associated with high risk for psychosocial impairment, cognitive decline and mortality. In these patients, surgery is currently the only therapeutic solution to reduce the seizure burden and, in many instances, cure the epilepsy. MRI has been instrumental in TLE surgery by allowing the in vivo detection of neuronal loss and gliosis as indexed by hippocampal atrophy on T1-weighted MRI and increased signal in T2-weighted images. In this issue of Brain, Tai and co-workers provide compelling evidence that long-standing epilepsy is associated with tauopathy, providing a basis for conceptualizing TLE, and possibly all epilepsies, as neurodegenerative disorders (Tai et al., 2016). In the strictest sense, neurodegeneration refers to pathology affecting neuronal structure and consequently function. In practice, neurodegenerative diseases form an ensemble of conditions with heterogeneous pathological and clinical expressions defined by the affected brain regions (Spillantini and Goedert, 2013). The main characteristics common to these disorders are relentless progression and cognitive decline. Rather than being exclusive to a given disorder, the processes contributing to neurodegeneration also share commonalities, both at cellular and molecular levels. In particular, the pathway leading from soluble to hyperphosphorylated insoluble tau protein is central to many neurodegenerative disorders, collectively called tauopathies, with Alzheimer’s disease as the archetype. In Alzheimer’s disease, tau pathology spreads from the disease epicentre, i.e. the transentorhinal area (Brodmann area 35 or perirhinal cortex), to other limbic regions at first, and later to neocortical areas in a pattern that suggests propagation along anatomical connections. Alzheimer’s disease is increasingly recognized as a disease of synaptic dysfunction; not surprisingly, the incidence of unprovoked seizures is up to 10-fold greater in sporadic Alzheimer’s disease than in the general population, and as much as 87-fold greater in patients with ‘early’ disease onset (before 60 years of age). Notably, a number of studies have shown strong association between neuronal activity and tauopathy. In particular, mouse models of Alzheimer’s disease have demonstrated that tau may affect excitability and seizure threshold; enhanced neuronal activity, in turn, not only stimulates tau release and transfer, but also exacerbates tauopathy (Wu et al., 2016). Moreover, tau pathology may contribute to neurotoxicity and neuronal loss. Interestingly, reduction of tau reduces seizures and improves cognitive function (Roberson et al., 2007). Analysing tissue obtained from 33 patients aged 50 to 65 during surgery for the treatment of drug-resistant TLE related to mesial temporal sclerosis, Tai et al. identified hyperphosphorylated tau in all but two. In addition to showing similarities to both Alzheimer’s disease and chronic traumatic encephalopathy, the authors described through an elegant modification of the tau score (necessary because of the lack of extra-temporal tissue) TLE-specific subpial tau deposition sparing the hippocampus. Modified scores were weakly related to age at onset of seizures and at surgery, but not to preoperative memory scores. Conversely, there was a negative correlation between tau burden and decline in verbal memory and recall (as well as word retrieval) 1 year after temporal lobectomy. Though the mechanism by which tau is released is yet to be defined and may differ among disorders, the discovery of tau pathology provides new evidence for disease progression in TLE. So far the most convincing support has come from neuroimaging data showing progressive grey matter atrophy in the disease epicentre and at a distance from it (Bernhardt et al., 2009, 2013). Arguably, additional longitudinal studies are needed to dissociate disease progression from ageing and to establish its relation to cognitive decline. Current hypotheses posit a multifactorial phenomenon combining effects of seizures, limbic disconnection and antiepileptic medication. The study by Tai et al. shows similarities in the distribution of tau between TLE and chronic traumatic encephalopathies often expressing as frontotemporal dementia. In the latter, tauopathy is thought to result from concussions or repetitive subconcussive blows sustained by athletes. In light of the association between tau load and a history of generalized convulsions, Tai et al. formulate an appealing hypothesis stipulating that subtle head injuries resulting from secondarily generalized seizures may contribute to tau release in TLE. Together with the recently established reciprocal relationship between tau accumulation and neuronal excitability in animal models, these results suggest new mechanisms for seizure-related neurotoxicity and damage in epilepsy, in which aggregated tau might play a mediating role. While mechanisms underlying the onset and progression of tau-related neurotoxicity remain elusive, tauopathy is recognized as the key driver of disease progression in Alzheimer’s disease and other neurodegenerative disorders. Timely diagnosis is, however, hampered by the difficulty in obtaining biopsies. On the other hand, epilepsy surgery offers a unique opportunity to validate the diagnosis and increase our understanding of tau biology. The availability of such data may inform not only the development of novel imaging agents to map epilepsy-specific tau deposits, but also help research in neurodegeneration at large, with the ultimate goal to develop mechanism-based treatments. Longitudinal studies, ideally with accelerated design (i.e. enrolling patients at different time points in the disease course, particularly early stages) would clarify the trajectory of tauopathy, and its link to clinical, cognitive and imaging markers of structure and function. It is now widely recognized that the longer the duration of drug-resistant epilepsy, the higher the likelihood of exposure to factors leading to neuronal damage: in other words, time means neurons (Cascino, 2009). Despite evidence in favour of the efficacy and superiority of surgical procedures to reduce or eliminate seizures compared to medical treatment, as documented by randomized controlled trials and observational studies (Jobst and Cascino, 2015), referrals to specialized centres for evaluation tend to occur several years after medication failure (Haneef et al., 2010). To date, long-term benefits of surgery with respect to cognitive, psychiatric and quality-of-life measures have not been examined with the same scrutiny as seizure control. With respect to the former, younger age at surgery and high presurgical levels of cognitive reserve may positively influence recovery after temporal lobe resection (Baxendale et al., 2006); postsurgical seizure cessation and reduction of anti-epileptic drug load are other important contributors to cognitive improvement. In their work, Tai et al. convincingly demonstrate that long-standing epilepsy is associated with neurodegeneration and has a negative impact on cognitive function after surgery. Admittedly, the retrospective nature of the study did not allow accounting for key parameters, such as seizure frequency, the evolution of psychometric tests, degree of neuronal loss, and imaging biomarkers. Validation in larger, possibly multicentric correlative ex vivo–in vivo studies has the potential to establish specificity and generalizability, fuelling the ongoing efforts to implement novel screening tools, such as tau imaging, and ultimately accelerating referrals for epilepsy surgery before irreversible brain damage occurs.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.