Pharmacotherapeutic considerations for late-onset epilepsy

1. Introduction

Late-onset (also termed ‘elderly-onset’) epilepsy is usually defined when seizures first occur at the age of ≥65 years [1], although recent studies have adopted different age cutoffs [2]. Late-onset epilepsy deserves to be regarded as a distinct clinical entity with etiologies and clinical features different from those encountered in younger patients [2].

Together with stroke and dementia, epileptic seizures and epilepsy represent the commonest neurological disorders in the elderly. The incidence of epilepsy rises with increasing age, from 80.8/100 000/year in the general population to 135–175/100 000/year in the group older than 80 years [3]. With the continuing aging of the population, the number of older people presenting with seizures is set to increase further. Cerebrovascular disease and dementia, mainly Alzheimer’s disease, are the most common etiologies of late-onset epilepsy [4], alongside intracranial tumors and traumatic brain injury; acute symptomatic seizures, sometimes presenting as status epilepticus, occur frequently [5]. The risk of recurrence following a single unprovoked late-onset seizure depends on the underlying etiology: for instance, vascular epilepsy can be diagnosed after a first seizure occurring more than 7 days after the stroke [6], but this does not apply for elderly patients with a first unprovoked seizure and concomitant dementia [7].

Effective treatment of epilepsy in elderly patients requires considering the age-related changes that can influence the pharmacokinetics of antiseizure medications (ASMs). They include lower protein binding due to reduced albumin levels, impairment of hepatic metabolism, and decreased renal excretion [4]. Furthermore, the pharmacodynamic properties of ASMs can be affected by age-related changes in receptor density and sensitivity [4].

Polypharmacy and the risk of drug-drug-interactions is also a major cause of concern in this population. Some ASMs can induce (e.g., carbamazepine, phenytoin, phenobarbital) or inhibit (e.g., valproic acid) hepatic cytochrome P450 enzymes, affecting the metabolism of several concomitant drugs, including oral anticoagulants, antiarrhythmics, statins, and antihypertensive agents [4]. Furthermore, enzyme-inducing ASMs can have detrimental metabolic effects by increasing surrogate atherogenic markers and reducing vitamin D levels and calcium uptake, leading to a higher risk of dyslipidemia, atherosclerosis, and osteoporosis [8].

The use of some ASMs is associated with adverse effects that may negatively influence the quality of life and autonomy of elderly patients. The commonest or more clinically relevant adverse effects include sedation (carbamazepine, phenobarbital), dizziness (carbamazepine, phenobarbital), hyponatremia (oxcarbazepine, particularly if concomitantly administered with diuretics), impairment or worsening of cognitive functions (phenobarbital, topiramate, zonisamide), and cardiac conduction abnormalities (carbamazepine, lacosamide, oxcarbazepine phenytoin) [4].

Consequently, the choice of the initial ASM in elderly patients newly diagnosed with epilepsy should be made with caution, considering several pharmacological and clinical issues. In this special patient group, any ASM should be started at a slow dose and up-titrated gradually, paying special attention to comorbidities and comedications, and the occurrence of adverse effects [4].

In regulatory epilepsy trials, elderly patients are often underrepresented due to comorbidities, difficulties in recruitment, and problems in providing informed consent [2]. Furthermore, the number of randomized controlled trials (RCTs) specifically conducted in patients with late-onset epilepsy is extremely scarce because recruitment can be problematic and dropouts are frequent [9–14] (Table 1). Consequently, the evidence supporting the use of specific ASM for this indication is scarce and relies mainly on uncontrolled studies.

Table 1. Randomized controlled trials evaluating the efficacy and tolerability of antiseizure medications used as monotherapy for late-onset epilepsy

Study	Comparisons	Summary of main results
Brodie et al., [9]	LTG vs IR-CBZ	Retention rate higher for LTG No difference in seizure freedom
Rowan et al., [10]	GBP vs LTG vs IR-CBZ	LTG and GBP better tolerated Similar seizure control
Saetre et al., [11]	LTG vs CR-CBZ	LTG better tolerated Similar seizure control
Werhahn et al., [12]	LTG vs LEV vs CR-CBZ	Retention rate higher for LEV Similar seizure control
Trinka et al.; Pohlmann-Eden et al., [13,14]	LEV vs VPA vs CR-CBZ	Retention rate higher for LEV and VPA Similar seizure control

Abbreviations: ASMs: antiseizure medications; CBZ: carbamazepine; CR-CBZ: controlled-release carbamazepine; GBP: gabapentin; IR-CBZ: immediate-release carbamazepine; LEV: levetiracetam; LTG: lamotrigine; VPA: valproic acid

In recent years, several ASMs have been developed and marketed. The efficacy and tolerability of monotherapy treatment of some of these ASMs (lacosamide, lamotrigine, levetiracetam, and gabapentin) have been assessed in RCTs with a non-inferiority design and using immediate- or controlled- release carbamazepine as comparator [9–14]. Overall, these ASMs showed similar seizure control but better tolerability compared to carbamazepine. A recent network meta-analysis of monotherapy RCTs conducted in late-onset epilepsy (total of 1425 patients) did not show statistically significant differences between the treatments in efficacy outcomes, although lacosamide, lamotrigine, and levetiracetam had the greatest likelihood ranking best for seizure freedom [15]. Conversely, this meta-analysis confirmed that carbamazepine (especially the immediate-release formulation) was associated with a higher risk of discontinuation [15]. Similar results were found in a network meta-analysis of RCTs of ASMs used to treat post-stroke epilepsy: lamotrigine and levetiracetam were better tolerated than controlled-release carbamazepine, with no differences in seizure freedom between these three drugs (probably because of the small sample size) [16].

The role of the newest ASMs (brivaracetam, eslicarbazepine acetate, and perampanel) for monotherapy of late-onset epilepsy still needs to be formally evaluated in RCTs [2].

2. Expert opinion

Response to ASMs is generally good and most elderly patients can achieve seizure freedom with a single drug at modest or moderate dosage [17,18]. For the majority of elderly patients, anti‐seizure pharmacotherapy will be required lifelong. As elderly patients often require other long-term medication in addition to antiseizure treatment, drug interactions between ASMs and other drugs can be of major significance. The presence of liver and kidney diseases represents another special feature of pharmacotherapy for epilepsy in the elderly.

Newest ASMs could represent promising options for the treatment of elderly-onset epilepsy. They could reduce the risk of pharmacokinetic drug-drug interactions (brivaracetam, lacosamide, levetiracetam, brivaracetam, eslicarbazepine acetate, and perampanel), increase adherence (perampanel, eslicarbazepine acetate; all available as once-daily formulation), and be used in emergency settings, or for patients with impaired consciousness or temporarily unable to swallow (brivaracetam, lacosamide, and levetiracetam; all available also as intravenous formulation) [2]. However, data on monotherapy with these drugs in elderly patients are limited, as this population is often underrepresented in clinical trials. Hence, further studies (both RCTs and real-world studies) are needed to collect additional data on their efficacy and safety in this special population. These studies should focus not only on traditional outcomes, such as seizure reduction and achievement of seizure freedom, but also on the quality of life, autonomy, cognition, mood, and mobility. Furthermore, more research should focus on individualized predictors of recurrence in elderly patients with dementia and a first seizure, to identify subjects who may benefit from starting treatment with an ASM.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript was not funded.