Pharmacokinetics and Pharmacodynamics of Levetiracetam in Neonatal Seizures: What We Still Need to Know

Introduction

Seizures are the most common neurological dysfunction in neonates, affecting 1 to 4 neonates per 1000 live births.^1–4 Neonatal seizures are often an indication of some underlying disorders including hypoxic-ischemic encephalopathy, metabolic disturbances, and infection of the central nervous system.¹ While identifying and treating the cause of neonatal seizures are critical, seizures often require immediate attention because high seizure burden in neonates can lead to increased neurologic impairment and mortality.^5,6 As such, antiseizure medications (ASM) are initiated promptly to prevent further brain injury. Currently, phenobarbital is recommended as a first-line agent in neonatal seizures.⁷ However, phenobarbital provides seizure control in fewer than half of the neonates.⁸ Phenobarbital has been associated with major cognitive impairments in early developmental periods.^9,10 Due to the limited efficacy and serious safety concerns, another ASM, levetiracetam, is commonly used in neonates.

Levetiracetam improves seizure control and has an acceptable tolerability profile in children as young as 1 month old.^11,12 The dosing regimens of levetiracetam in children >1 month old were determined to achieve drug exposure that is similar to adults, largely by accounting for the pharmacokinetic differences due to well-described developmental changes.^11–13 However, the optimal dosing regimen of levetiracetam in neonates remain unknown. A recent randomized controlled trial concluded that levetiracetam was less efficacious than phenobarbital as a first-line agent for neonatal seizures but has better tolerability.¹⁴ This study used a maximal levetiracetam dose of 60 mg/kg/day because it is the highest US Food and Drug Administration (FDA)-approved dose in any age group. Several retrospective studies reported that levetiracetam was at least as effective as phenobarbital as a first-line agent and had lower rates of adverse effects.^15–19 Though the evidence in the retrospective studies are not as rigorous as a randomized controlled trial, the wide range of levetiracetam dosing (10–150 mg/kg/day) used in the retrospective studies may be a critical factor for the favorable levetiracetam efficacy, as a larger dose of levetiracetam has been associated with increased seizure control.²⁰ As the optimal use of levetiracetam in neonates remains elusive, this narrative review 1) evaluates the pharmacology, pharmacokinetics, and pharmacodynamics of levetiracetam, 2) discusses the evidence of levetiracetam use in neonatal seizures, and 3) discusses the information still needed to assess the efficacy and safety of levetiracetam in neonatal seizures.

Neonatal Seizures and Pharmacology of Levetiracetam

Neonates are more susceptible to seizures due to several age-related pathophysiologic factors.⁹ For example, neonates have an increased number of N-methyl-D-aspartate (NMDA) receptors, resulting in a higher amount of neuronal glutamatergic excitatory signals.²¹ Another type of glutamate receptor, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor, is also overexpressed in neonates. The immature AMPA receptors in neonates have enhanced calcium permeability, contributing to the increased neuronal excitability and other downstream signaling.²² Additionally, neonates have an immature gamma-aminobutyric acid (GABA) neurotransmitter system. The activation of GABA receptors typically results in neuronal inhibition; however, the activation of GABA receptors of immature neurons may cause a neuronal membrane depolarization, rather than a hyperpolarization, leading to neuronal excitation.^23,24 Due to excess excitation and reduced inhibition of the immature neurons, neonates with insults to the central nervous system (CNS), such as hypoxic-ischemic encephalopathy and meningitis, have higher chances of seizures.²⁵

Levetiracetam exerts its action primarily by binding to the ​​synaptic vesicle glycoprotein 2A (SV2A), interfering with the ability of the vesicle to release excitatory neurotransmitters in presynaptic neurons.²⁶ Many other antiseizure mechanisms of levetiracetam have also been reported.²⁷ Levetiracetam may block the potassium and voltage-dependent calcium channels, decreasing the synaptic transmission. It may also modulate the glutamatergic AMPA receptors on postsynaptic neurons, diminishing the excitatory current. The SV2A protein was found to be expressed in the postmortem human neocortex of neonates with a gestational age (GA) of 20 to 37 weeks, providing the evidence that the target of levetiracetam is present in neonates.²⁸ However, the expression of SV2A was reduced in premature neonates (62% of adult values at 20 weeks GA and 81% at 27–32 weeks GA) and approached adult values near term birth (94% at 37 weeks GA).²⁸ The differential expression of SV2A may suggest a reduced efficacy or a different exposure-response relationship of levetiracetam in premature neonates. In addition to targeting SV2A, levetiracetam may exert its antiseizure effect by inhibiting AMPA receptors, which are overexpressed during early postnatal periods. The activation of AMPA receptors has been found to play a crucial role in rodent model of neonatal hypoxia-induced seizures.²⁸ A higher concentration of levetiracetam may be needed to inhibit the overexpressed AMPA receptors, thereby increasing the antiseizure effect of levetiracetam. Although the exact mechanisms are unknown, levetiracetam likely exerts its effects in neonatal seizures by acting on multiple targets. Due to the several pathophysiologic factors unique to seizures in neonates, the target levetiracetam systemic exposure may be different in neonates compared to infants and children with seizures.

Pharmacokinetics of Levetiracetam in Children.

A summary of pediatric levetiracetam pharmacokinetic studies is provided in Table 1. As children grow and mature, the changes in organ functions and body composition can lead to alterations in pharmacokinetics.³⁷ For levetiracetam, absorption in children appears to be similar to that in adults, where it is rapid and complete.³¹ Consistently, the reported mean oral clearance values^38–44 (i.e., clearance divided by bioavailability; range: 0.69–2.18 mL/min/kg) from pediatric studies using the oral formulation are generally comparable with mean clearance values^45,46 (range: 1.33–1.49 mL/min/kg) determined from pediatric studies using the intravenous formulation (Table 1). This suggests that the bioavailability of levetiracetam in children is close to 100%. The weight-normalized Vd was found to be similar in adults and children (0.5–0.7 L/kg; Table 1). This suggests that the Vd of levetiracetam changes proportionally with the body weight in infants (>1 month) and children. The similarity in weight-normalized Vd in adults and children may imply that a comparable weight-based loading dose of levetiracetam will result in similar initial plasma concentrations in both adults and children. The weight-normalized clearance of levetiracetam was found to be higher in children. Specifically, the mean clearance values (1.4–1.5 mL/min/kg; SD: 0.4 mL/min/kg)^39,44 in children 2 months to 12 years of age are about 50% higher than that of an adult (mean [SD]: 0.96 [0.14] mL/min/kg).³¹ This observation is likely because the drug-elimination organs mature at a faster rate than the rate of weight gain in younger children. The higher weight-normalized clearance in children suggests that larger doses of levetiracetam are needed in children to achieve steady-state plasma concentrations comparable with those in adults. The differences in weight-normalized clearance observed in children and adults are reflected in the dosing regimens in the levetiracetam product label: 25 to 30 mg/kg twice daily is recommended for partial onset seizures in children 6 months to 16 years, whereas 1500 mg twice daily is recommended for adults (i.e., 21 mg/kg twice daily for a 70-kg person).

Table 1.Summary of Levetiracetam Pharmacokinetics in Neonates and Pediatrics

View Fullscreen

Pharmacodynamics of Levetiracetam in Adults and Children

The dose-response and exposure-response relationship of levetiracetam have been reported in adults and children with partial onset seizures. In 4 double-blind placebo-controlled trials, Snoeck and Stockis⁵² found that both the placebo and levetiracetam reduced seizure frequency over a period of 16 to 24 weeks in adult refractory epileptic patients with partial onset seizures who received levetiracetam therapy as an add-on treatment.⁵² They found that 59% patients who received placebo had a lower number of seizures compared with baseline. The proportion of improving patients was approximately 75% among those who received levetiracetam treatment. The proportion of improving patients increased slightly (from 72% to 77%) with increasing daily levetiracetam dose (from 1 g/day to 3 g/day). Rhee et al⁵³ evaluated the effect of levetiracetam dose and concentration on seizure occurrence among adult patients who received levetiracetam with or without other ASM. They reported that among patients receiving 500 to 3000 mg of daily levetiracetam dose, 34% to 71% of patients were seizure free. However, a higher total daily dose and plasma concentration were not associated with a higher proportion of seizure-free patients. They suggested that the seizure response to levetiracetam may be all-or-none, where patients with seizures that are not controlled after titrating to 2 g/day may require a change of ASM with a different mechanism of action. Based on the dose-response relationship evaluated in the abovementioned studies by Snoeck and Stockis⁵² and Rhee et al,⁵³ larger doses of levetiracetam did not provide a clear benefit in adult patients with partial onset seizures.

Dose-response relationship is valuable for determining optimal dose of drugs with relatively low pharmacokinetic variability. For drugs with large interpatient pharmacokinetic variability, the same dose may result in different exposures (trough concentration, area under the concentration-time curve [AUC], or average concentration) in different patients. In such cases, exposure-response relationship is more informative to help individualize dosing regimen. Similarly, exposure-response relationship is more helpful to determine optimal dosing in patients who have different physiological variables, such as children and patients with organ dysfunction. In a recent study, Mehrotra et al⁵⁴ performed exposure-response analyses of add-on ASM including levetiracetam for treatment of partial onset seizures, using adult and pediatric phase 3 clinical trial datasets available at the FDA. They found that in adults and children >4 years of age, the relationship between the levetiracetam average plasma concentration and the percentage change from baseline (%CFB) in seizure frequency could be described using a log-linear model. The baseline seizure frequency was determined over an 8- to 12-week period prior to levetiracetam use, and the %CFB was calculated based on the weekly seizure frequency during levetiracetam treatment period. In this model, the %CFB was −45%, −52%, −58%, −64%, and −67% when the average concentration was 0 mg/L (placebo), 10 mg/L, 20 mg/L, 30 mg/L, and 40 mg/L, respectively. They found that the effect of levetiracetam on the %CFB did not differ between adults and children (p = 0.25).

Schoemaker et al⁵⁵ performed PK/PD modeling on levetiracetam in adults and children using datasets from placebo-controlled phase 3 clinical trials, where levetiracetam was used as add-on treatment for refractory partial onset seizures. They also found that the levetiracetam concentration-effect relationship was similar in adults and children >4 years of age. In this study, they reported that adult and pediatric patients who receive levetiracetam were generally divided into 2 populations: placebo-like population (66%) and responder population (34%). The placebo-like population had a 15% reduction in daily seizure frequency following levetiracetam treatment. Such reduction in seizure frequency was similar to patients who were randomized to the placebo group. In the responder population, a higher levetiracetam concentration was associated with greater reduction in seizure frequency (up to a maximum of 96% reduction), and a levetiracetam concentration of 31.4 mg/L would reduce the seizure frequency by 48% (half of the maximum levetiracetam effect). The authors reported that there was no significant difference in the relationship between levetiracetam concentration and the reduction of seizure frequency in adults and children.

The analyses of clinical trial data from both Mehrotra et al⁵⁴ and Schoemaker et al⁵⁵ showed that the concentration-effect relationship of levetiracetam is similar in adults and children >4 years old. To achieve a similar antiseizure effect, dosing of levetiracetam in children can be derived by accounting for the pharmacokinetic differences between children and adults. This is consistent with the findings reported by Pellock et al¹³ that show similarity in pathophysiology of partial onset seizures in adults and children down to 2 years of age. In May 2020, at the Epilepsy Foundation Research Roundtable for Epilepsy, researchers further showed that in patients 1 month to <2 years of age, the partial onset seizure characteristics and symptoms, the features of electroencephalogram, the progression of disease, and the response to ASM were similar to children >2 years of age.⁵⁴ Consistently, a randomized, placebo-controlled trial demonstrated the efficacy of levetiracetam in children with partial onset seizures as young as 1 month old.¹¹ Extrapolating the efficacy of ASM for partial onset seizures from adults to children and infants can help avoid additional dose-finding studies in young children. In general, studies to assess pharmacokinetics and safety in the extrapolated population are still needed. Although the concentration-effect relationship of levetiracetam has been determined in adults and children with partial onset seizures, the therapeutic range is not well-defined due to large patient-to-patient variability.^56,57 Also, as reported by Schoemaker et al,⁵⁵ the concentration-effect relationship may only be observed among responders and not in nonresponders. As such, therapeutic drug monitoring for levetiracetam is not performed routinely in clinical practice.

Efficacy of Levetiracetam in Neonatal Seizures

The evidence of efficacy and pharmacodynamics of levetiracetam in neonatal seizures remain scarce. A nice review of levetiracetam in neonatal seizures by Mruk et al⁵⁸ has been previously published in this journal. More recently, the Neonatal Task Force of the International League Against Epilepsy (ILAE) conducted a systematic literature review (up to June 2020) of neonatal seizure management.⁷ It is recommended that phenobarbital should be used as the first-line ASM, and levetiracetam may be considered as a second-line agent.⁷ Levetiracetam may be the preferred second-line agent in neonates with cardiac disorders.⁷ Table 2 includes a summary of the efficacy, effectiveness, and safety of levetiracetam in neonatal seizures. The recommendation by the ILAE Task Force regarding the levetiracetam use was primarily supported by a recent randomized controlled, phase 2b clinical trial. In the trial (NEOLEV2), levetiracetam and phenobarbital were compared as first-line agents for neonatal seizures confirmed by continuous electroencephalogram (EEG).¹⁴ Study neonates were either treated with an initial dose of 40 to 60 mg/kg of levetiracetam or 20 to 40 mg/kg of phenobarbital. Phenobarbital was found to be more efficacious in 24-hour seizure freedom when compared with levetiracetam (80% vs 28%, p < 0.001). However, a slight improvement (7.5%) in seizure control was observed when levetiracetam was increased from 40 mg/kg to 60 mg/kg. The initial loading dose of 40 mg/kg of levetiracetam was determined to achieve a trough of >20 mg/L in majority of neonates.⁴⁸ This target concentration was based on therapeutic concentration of levetiracetam determined from adult and pediatric populations. However, the target concentration remains unknown for complete seizure control in neonates. A dose-ranging study may be needed to determine whether levetiracetam is truly ineffective in neonatal seizures or if larger doses may produce favorable outcomes. In fact, the investigators of NEOLEV2 are conducting a dose escalation study (NCT05610085) that will examine the efficacy and safety of levetiracetam in doses from 60 mg/kg to a maximum dose of 150 mg/kg. Furthermore, given the large variability in neonatal pharmacokinetics, determining the concentration-response relationship, rather than dose-response relationship, would be most helpful in finding optimal dosing regimens for levetiracetam in neonatal seizures.

Table 2.Efficacy, Effectiveness, and Safety of Levetiracetam Used as First- or Non-First-Line Agent for Neonatal Seizures

View Fullscreen

Unlike the NEOLEV2 trial, other studies that evaluated the effectiveness of levetiracetam in neonatal seizures were observational trials (Table 2). These studies included varying study outcome measures and dosing regimens, with most loading doses being less than 60 mg/kg.^{14,17–19,59–65} The achievement of seizure freedom ranged from 26% to 57% among neonates who received levetiracetam as the first-line ASM for varying duration (Table 2). Rao et al⁶² reported that levetiracetam, as opposed to phenobarbital, as the first-line ASM may shorten the time to reach seizure freedom (adjusted hazard ratio: 2.57, p = 0.01). Due to the nature of the retrospective observational study design, these studies are limited by the risk of bias.

Some evidence has suggested that larger doses of levetiracetam (>60 mg/kg) may be needed in neonatal seizures. A retrospective cohort study by Verwoerd et al⁶⁶ examined the use of levetiracetam (50–100 mg/kg loading dose) and phenobarbital (20 mg/kg) as a first-line treatment for neonatal seizures. They found that levetiracetam and phenobarbital resulted in similar percentage of patients (63%–65%) having a sustained seizure burden <10% (<6 minutes of electrographic seizure per 1-hour period). The use of phenobarbital as the first-line agent resulted in a larger absolute reduction in average seizure burden (−24.3 vs −14.2 min/hr) an hour before and after the treatment. Conversely, the use of levetiracetam as the first-line agent led to a shorter average time to achieve seizure freedom (15 vs 21 hours). It should be noted that this study only included 25 neonates and the results were not statistically significant. However, initial cumulative bolus doses of levetiracetam up to 100 mg/kg were found to be well tolerated. This is consistent with the safety profile reported by Venkatesan et al,²⁰ who reported no adverse side effects in cumulative bolus doses of levetiracetam up to 120 mg/kg for the treatment of seizures in neonatal hypoxic-ischemic encephalopathy. In addition to the dosing of levetiracetam, the effectiveness of levetiracetam for neonatal seizures may be affected by the prior use of benzodiazepine. A retrospective cohort study by Wagner et al⁶⁷ found that, as compared with neonates without prior benzodiazepine exposure, neonates with prior benzodiazepine exposure had a higher seizure resolution rate when treated with phenobarbital but a lower seizure resolution rate when treated with levetiracetam. This finding may be a result of pharmacodynamic interactions; however, the differential response may also stem from differences in disease severity between those who did and did not receive benzodiazepine treatment. It should also be noted that, like phenobarbital, prolonged benzodiazepine exposure (>7 days) is associated with negative neurodevelopment outcomes.⁶⁸

Discussion and Future Directions

More effective and safe treatments of neonatal seizures are desperately needed. Despite the unfavorable levetiracetam treatment outcomes in the NEOLEV2 trial, critical information such as the pharmacodynamics (concentration-effect relationship) of levetiracetam in larger doses is still needed to confirm its efficacy or lack thereof. Levetiracetam has been shown to be safe and effective in children; however, most pharmacokinetic and pharmacodynamic data of levetiracetam in pediatrics are from children with partial onset seizures. Furthermore, the pathophysiology of seizures in neonates are different from that in children and adults. The treatment goals for neonatal seizures are also different from those for seizures in children and adults. In children with partial onset seizures, the desired clinical outcome for ASM treatment is typically a reduction in seizure frequency over a period of time. ASM treatment is usually long-term and may be discontinued if the child remains seizure free for two years on monotherapy. In neonates, most neonatal seizures are acute provoked seizures, and the primary goal is to achieve complete control of seizures within 1 to 2 days of starting ASM treatment. A loading dose of the first line ASM is given initially, and an additional loading dose may be used if seizure activity persists after 30 minutes. Maintenance doses follow the loading doses, and a second-line agent may be added if the first-line agent fails.⁵ The ASM may be discontinued after 72 hours of seizure freedom, prior to hospital discharge.^7,73

To achieve complete seizure control within a short amount of time (1–2 days) in neonatal seizures, the target levetiracetam concentration may differ from that targeted for children with partial onset seizures. In children 4 to 15 years of age with partial onset seizures, the therapeutic dose of 60 mg/kg/day resulted in the average levetiracetam plasma concentration of 33 mg/L (IQR: 29–36 mg/L).⁵⁴ This plasma levetiracetam concentration range observed in children is similar to that found in adult patients with partial onset seizures (mean: 32.3 mg/L; IQR: 30–35 mg/L) and those with status epilepticus (mean: 25.8 mg/L; range: 19.4–49.2 mg/L) who responded to levetiracetam therapy.^54,57 This plasma concentration range may not be enough for complete seizure control in neonatal seizures. The effective concentration of levetiracetam may also differ when other ASM are used concomitantly. Additionally, the differential expression of the molecular target of levetiracetam SV2A in neonates with different gestational age may further contribute to the variability in levetiracetam treatment response.²⁸

Further pharmacokinetic and pharmacodynamic information are needed for levetiracetam in neonates to individualize the optimal dosing regimen. As the pharmacokinetics of neonates undergo dramatic changes during the first few weeks of life, optimal dosing of levetiracetam may vary significantly for neonates with different ages and/or body weights. For example, the clearance of levetiracetam in neonates nearly doubles during the first week of life, which may result in significantly different plasma concentrations if the same dosing regimen is used. Approaches like physiologically based pharmacokinetic modeling may offer invaluable insights on how the intertwined relationship among age (gestational and postnatal), weight, and organ growth and maturation may affect the distribution and elimination of levetiracetam. Furthermore, the optimal dosing regimen for levetiracetam in neonatal seizures is determined by the target plasma concentrations, which could differ from those used in levetiracetam therapy for seizures in children and adults. As such, it is necessary to determine the relationship between the levetiracetam plasma concentration and its antiseizure effect in neonatal seizures. The use of video EEG monitoring to measure seizure burden and qualify seizure freedom may further allow the quantification of the concentration-effect relationship. Long-term neurodevelopmental effects of levetiracetam vs phenobarbital should also be evaluated. This may further demonstrate the significance of drug selection in the early neonatal period. In conclusion, the additional information on levetiracetam may help determine whether it is a safer first-line alternative for treating neonatal seizures.