Epilepsy and Alzheimer’s Disease: Current Concepts and Treatment Perspective on Two Closely Related Pathologies

Abstract

The literature on epileptic seizures in Alzheimer's disease has significantly increased over the past decades. Remarkably, several studies suggest a bi-directional link between these two common neurological diseases, with either condition carrying a nearly 2-fold risk of contracting the other in comparison to healthy subjects. In this respect, evidence from both clinical and preclinical studies indicates that epileptogenesis and neurodegeneration possibly share common underlying mechanisms. However, the precise association between epileptogenesis and neurodegeneration still needs to be fully elucidated. Targeted intervention to reduce abnormal network hyperexcitability might constitute a therapeutic strategy to postpone the onset of later neurodegenerative changes and consequent cognitive decline by many years in patients. By virtue of this, an early diagnosis and treatment of seizures in patients with Alzheimer’s disease should be pursued. To date, no guidelines are available for treating epileptic activity in this context, largely due to the paucity of studies sufficient to answer the related questions. Accordingly, clinical trials are mandatory, not only to inform clinicians about symptomatic management of seizures in Alzheimer’s disease patients but also to detect if treatment with antiseizure medications could have disease-modifying effects. Moreover, it will be fundamental to expand the application of animal models of Alzheimer’s disease to comorbid conditions, such as epilepsy both to reveal the mechanisms underlying seizure onset and to better define their role in cognitive decline. Such models could also be useful to identify pharmacological compounds having therapeutically effectiveness as well as reliable early biomarkers for seizures in Alzheimer’s disease.

1. INTRODUCTION

Alzheimer’s disease (AD) is the most prevalent type of dementia; according to Global Burden of Disease systematic reports, more than 50 million people worldwide suffer from AD. The economic impact is huge; in fact, AD represents a public health financial burden of ~1 trillion US dollars annually. Due to the rapid ageing of the population, it is expected that these numbers will double over the next two decades [1, 2]. Unfortunately, none of the currently available treatments has shown clinical efficacy in delaying the onset and/or curing already established AD. In fact, currently licensed drugs show only modest symptomatic effects at best [3]. It is well established, since Alois Alzheimer’s earliest case reports, that patients with AD have an augmented risk, mainly in the early disease stage, of spontaneous seizures and epilepsy in comparison to healthy age-matched controls. However, despite a relatively lower risk, even late-onset AD patients have an increased risk of spontaneous recurrent seizures (SRSs) in comparison to healthy age-matched controls [4]. Particularly, seizures are 3 times more frequent in patients with AD in comparison to healthy controls and the absolute risk of spontaneous seizures in patients with AD is about 10 to 20%. The predominant epileptic seizure subtypes diagnosed in AD patients are non-motor focal onset seizures with impaired awareness, the symptoms of which may often overlap with behavioral features of AD, such as amnestic spells, déjà vu, unexplained emotions, and/or sensory phenomena (e.g., epigastric sensations). A broad range of seizures have recently been detected in patients with AD, most of which are subtle and easily unrecognized [5]. Notably, several lines of evidence indicate that 22% to 54% of patients with AD have subclinical epileptiform activity. These epileptiform discharges, much more common during sleep in patients with AD, may contribute to hasten cognitive impairment and reduce the quality of life in these patients [6]. Furthermore, a case-control analysis reported that regular use of antiseizure medications (ASMs) was more frequent in patients with dementia compared to healthy controls. Moreover, authors also reported that the use of ASMs was linked to a significantly greater risk of dementia and AD compared to no ASMs use [7]. However, this study was widely criticized, due to several biases [7-9]. Remarkably, also people with epilepsy (PWE) are more likely to develop AD/dementia in comparison to the general population. Accordingly, a bi-directional link has been suggested between epilepsy and AD/dementia, with either condition carrying a nearly 2-fold risk of contracting the other in comparison to healthy subjects [10, 11]. In fact, the ancient unidirectional viewpoint that epilepsy and seizures are the major risk factor for cognitive and behavioral comorbidity was rejected. Cognitive and behavioral disorders often precede epilepsy and may represent potential biomarkers for its development [11, 12]. Moreover, it should be underlined that in different illnesses both epileptic seizures and cognitive and behavioral deficits are symptoms of a common underlying pathological condition, such as synaptic loss, temporal lobe atrophy, disruption of the blood-brain barrier, neurodegeneration, and neuroinflammation. Various of these conditions, implicated also in AD pathogenesis, have been shown to cause changes in neuronal excitability [13]. In this respect, studies indicate that epileptogenesis and neurodegeneration possibly share common underlying mechanisms. Indeed, similar molecular alterations have been observed in the hippocampus of patients with AD [14], epilepsy [15], and in animal models with a chronic aberrant increase of excitatory neuronal activity [16]. Moreover, although seizure occurrence in AD patients is generally considered a secondary process resulting from advanced stages of neurodegeneration [17], studies performed in experimental models of AD suggest that aberrant excitatory neuronal activity also represents an early mechanism contributing to cognitive deficits [16, 18]. Recently, an observational, prospective study has reported that patients with late-onset epilepsy had pathological CSF Aβ1-42 and t-tau levels compared to healthy individuals [19, 20], suggesting that Aβ pathology might lead to both epileptogenic alterations and cognitive impairment. Moreover, we also reported that acute exposure of mouse hippocampal slices to Aβ1-42 oligomers reduces the epileptic threshold and impairs synaptic plasticity in the dentate gyrus (DG) of the hippocampus with a mechanism dependent on the stimulation of D1 dopamine receptor signaling, involving AMPAR subunits rearrangements [19]. The presence of early network hyperexcitability may have profound effects on the progression of neurodegeneration because the abnormal aggregation of neurodegenerative proteins is known to be activity-dependent. Moreover, seizures promote alterations in molecular signaling, accelerating neurodegeneration and thus, leading to the anticipation of cognitive symptoms [21]. This could generate a “vicious cycle” where seizures increase neurodegenerative protein deposition, leading to an increase in neuronal excitability and a further predisposition to develop epilepsy [22, 23]. While the potential association of neurodegenerative proteins and epilepsy has been widely explored for Aβ, limited studies exist on the impact of α-synuclein-mediated neurodegeneration in epileptogenesis. Indeed, although the role of α-synuclein (α-syn) in neurodegeneration is well established, it is still unclear whether hyperexcitability can be considered a key feature of α-synucleinopathies and Lewy body dementias (LBDs). LBDs are associated with progressive cognitive impairments, occurring prior to the onset of motor symptoms, and are the second most common cause of neurodegenerative dementia after AD. Clinical studies report upregulated α-syn expression in epileptic patients, highlighting the possibility of α-syn-mediated neurodegeneration in epilepsy [24, 25]. In addition to evidence from clinical studies, emerging preclinical data highlight a possible correlation between α-syn and epileptic seizures [26, 27], but whether hyperexcitability is a consistent feature of early α-syn pathology is still unclear to date [28-30]. All these findings have increased the attention on neurodegenerative proteins, as possible targets to develop a promising therapeutic approach against epilepsy. Although the precise association between epileptogenesis and neurodegeneration still needs to be fully elucidated, targeted intervention to reduce abnormal network hyperexcitability might constitute a therapeutic strategy to postpone the onset of later neurodegenerative changes and consequent cognitive decline by many years in patients. Since several lines of evidence indicate the role of seizures and epileptiform activity in cognitive decline, an early diagnosis and treatment of seizures in AD patients should be pursued. Despite the pharmacological armamentarium against epilepsy has significantly increased with newer antiseizure medications (ASMs), to date, the decision on whether to treat seizures and the choice of ASMs for an AD patient may be arduous. Even more complex is the decision of whether to treat subclinical epileptiform discharges in AD patients. In this regard, drug treatment should only be initiated whether these epileptiform discharges can promote cognitive deficits. On the contrary, it is a common opinion that the empirical ASM treatment, in AD patients without clinical or subclinical epileptiform activity, should be avoided [31]. The use of ASM in AD patients must take into account several factors, such as age, comorbid conditions, pharmacokinetic profile, and adverse effects of ASMs (e.g. sedation, mood alterations, motor coordination disturbances, and cognitive alterations are the most common) [12, 32]. Furthermore, one of the main issues regarding seizures in AD patients is linked to the probability of seizure threshold decreased by drugs commonly prescribed to treat AD and associated behavioral disorders [33, 34]. By virtue of the few randomized clinical trials (RCTs) investigating the efficacy and the tolerability of ASMs in AD patients, pharmacological choices must be based on data obtained from studies of ASMs in elderly patients with or without AD [33]. Basically, regarding cognitive side effects of ASMs, it has been described that these are worst for the first-generation ASMs compared to the newer ASMs, which have a better profile of effectiveness and tolerability [12, 35]. Levetiracetam, brivaracetam, lamotrigine, lacosamide, gabapentin, eslicarbazepine, and perampanel appear to have fewer effects on cognitive performance than older ASMs [36, 37]. Interestingly, several randomized double-blind studies performed on new-onset geriatric epilepsy reported that cognitive side effects are less prominent with gabapentin, lamotrigine, and levetiracetam compared to carbamazepine [38-40]. Similarly, a retrospective, uncontrolled study on geriatric PWE suggested that among 10 ASMs (including lamotrigine, levetiracetam, carbamazepine, gabapentin, oxcarbazepine, phenytoin, and topiramate), lamotrigine and levetiracetam were more effective compared to other ASMs in geriatric PWE and they did not seem to influence cognitive performance [41]. ASMs with intermediate effects on cognitive function include carbamazepine, oxcarbazepine, and valproate, the use of which, as first-line treatment, should be avoided in patients with AD with or without a seizure. Despite cognitive deterioration being documented with phenytoin, the drug is still prescribed in the elderly and AD patients. The wide use of phenytoin in geriatric PWE with cognitive deterioration was supported by a retrospective study [42]. It is now known that benzodiazepines, topiramate, and zonisamide, due to their cognitive adverse events, should be avoided in AD patients with or without a history of seizures [36, 43, 44], although data on zonisamide remains controversial.To date, few randomized clinical trials have evaluated whether an ASM may offer beneficial effects and good tolerability in patients with AD. Recently, a Cochrane review, including one RCT on pharmacological interventions in 95 people with AD, has been performed to study the efficacy and tolerability of treatment of epilepsy in AD patients. Overall, this study does not provide sufficient evidence to support levetiracetam, phenobarbital, or lamotrigine for the treatment of seizures in people with AD. Moreover, no significant changes in efficacy and tolerability have been discovered among these ASMs. Authors concluded that large RCT with a double-blind, parallel-group design is necessary to establish the efficacy and tolerability of seizures treatment in AD patients [45]. Remarkably, a phase 2 randomized double-blinded placebo-controlled crossover clinical trial enrolling 34 patients with AD was recently conducted to investigate whether AD patients with epileptiform discharges responded better to levetiracetam than those without this chronic network hyperexcitability. Unfortunately, in this study levetiracetam did not significantly improve cognitive function, however, it was able to recover executive function and spatial memory among AD patients with seizures or subclinical epileptiform activity.The relationship between epilepsy and AD raises different pertinent questions that deserve to be answered. In fact, a major goal for future studies will be both understanding the mechanisms underlying epileptic seizure onset in AD patients and if epilepsy drives dementia or vice versa. Toward this end, the experimental models of AD may represent, albeit with some limitations, powerful tools to better define the links between AD and epilepsy [4, 46]. Furthermore, up to now, the treatment of seizures in AD patients remains unclear, mainly due to the paucity of RCTs necessary to address this question. Essentially, RCTs are mandatory, not only to inform clinicians about symptomatic management of seizures, in AD patients but also to detect whether treatment with ASMs could have disease-modifying effects, rescuing cognitive symptoms and/or slowing decline. Similarly, future research should also be focused on the identification of reliable biomarkers for epilepsy in AD.

CONFLICT OF INTEREST

The authors declare no conflict of interest, financial or otherwise.