| 1. INTRODUCTION |
Anticonvulsant medications (ACMs) are the cornerstone of treatment for patients with epilepsy. The arsenal of drugs to choose from helps tailor treatment in epilepsy and also in other circumstances, such as psychiatric disorders and pain management.
The challenges with all of these medications are numerous and warrant a detailed understanding of the pharmacodynamic and pharmacokinetic properties of the drugs. Therapeutic drug monitoring (MTM) can be used to determine and adjust for pharmacokinetic variability and drug interactions, thereby facilitating optimal dosing in each individual patient. However, level A evidence indicates that MTM provides no benefit in the treatment of patients with epilepsy generally, and a Cochrane review found no clear evidence to support its routine use. Therefore, it is important to know how to use MTM correctly in clinically relevant situations.
The purpose of the present review was to develop an educational document addressing three objectives of the International League Against Epilepsy (ILAE) curriculum, based on literature and clinical experience to demonstrate: knowledge of pharmacokinetics and pharmacodynamics; knowledge about proper monitoring of serum MAC levels; and knowledge about drug interactions.
| 2. MAIN BODY |
> 2.1 Basic pharmacology: Pharmacodynamics and pharmacokinetics
2.1.1 Activity spectrum
In general, the spectrum of activity of MACs varies; Most medications first have an approved indication as a complementary medication in focal seizures, and depending on clinical experience, the indications can be expanded, such as for lamotrigine and levetiracetam. Benzodiazepines and valproic acid are generally indicated in generalized epilepsies.
> 2.2 Pharmacodynamics: Mechanisms of action of MACs
Most drugs currently used to treat epilepsy prevent its symptoms (seizures) and not the underlying disease. Hence the use of the term “anti-seizure medications” to describe them.
2.2.1 Inhibition of voltage-gated ion channels
The main mechanism of action of phenytoin, carbamazepine, oxcarbazepine, eslicarbazepine acetate, lamotrigine, and lacosamide is the blockade of voltage-gated sodium channels , which prevents repetitive neuronal activation.
Blockade of sodium channels also contributes to the activity of felbamate, rufinamide, topiramate, zonisamide and cenobamate. Ethosuximide reduces the flow of calcium ions through T-type calcium channels, inhibiting the thalamic rhythm in the spike-and-wave discharges of absence seizures. Gabapentin and pregabalin also exert their effects by binding to voltage-gated calcium channels.
2.2.2 Effects on GABAergic targets
Benzodiazepines (clobazam, clonazepam, diazepam and midazolam) and barbiturates (phenobarbital and its prodrug pidone) are allosteric modulators of GABA A receptors . Its binding to the receptor enhances chloride influx in response to GABA and membrane polarization. Phenobarbital and imientodone in high doses can also act as GABA A receptor agonists , limiting their use (due to risk of overdose/death).
Allosteric modulation is also a mechanism of action of stiripentol, felbamate, topiramate, ganaxolone, and cenobamate. Vigabatrin and tiagabine increase the accumulation of GABA in the brain by irreversible inhibition of its degradation by GABA transaminase or blocking its reuptake in presynaptic neurons and glia, respectively.
2.2.3 Effects on glutamatergic targets
Two types of glutamate receptors play a role in seizure generation and propagation: alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA) receptors are selectively and non-competitively inhibited by perampanel, and blockade of N-methyl-D-aspartate (NMDA) type receptors contributes to the pharmacological activity of felbamate and topiramate. Levetiracetam and brivaracetam bind to the presynaptically located synaptic vesicle protein 2A (SV2A), modulating the fusion of vesicles with the membrane of nerve terminals and the subsequent release of neurotransmitters into the synapse.
2.2.4 Other mechanisms
Cenobamate has a dual mechanism of action, acting as an inhibitor of sodium channels and a weak allosteric modulator of GABA A receptors . Topiramate and zonisamide, in addition to the aforementioned mechanisms of action, are weak inhibitors of carbonic anhydrase in the CNS.
Valproic acid stimulates GABAergic activity, blocking sodium channels and has a weak inhibitory effect on T-type calcium currents. Fenfluramine recently approved in Dravet and Lennox Gastaut syndrome indirectly stimulates serotonergic 5-HT 2C receptors and 5-HT 1D and interacts with sigma-1 receptors.
Cannabidiol exerts its therapeutic activity by antagonizing G protein-coupled receptor-55 (GPR55), desensitizing transient receptor potential vanilloid 1 (TRPV1) channels, decreasing calcium-mediated excitation and improving adenosine-mediated signaling. Everolimus differs from other MACs in that it targets the underlying disease pathology by inhibiting mammalian target of rapamycin (mTOR), a central cell growth regulatory protein kinase.
> 2.3 Principles of pharmacokinetics
In general, absorption is extensive and bioavailability is high for most MACs. The exceptions are gabapentin, which shows dose-related absorption, and cannabidiol, with extensive first-pass metabolism and limited absorption that is increased by fatty foods.
Therefore, for both there is unpredictable variability in their bioavailability. Most MACs are fat-soluble, can easily cross the blood-brain barrier (BBB), and are widely distributed in the body. However, valproic acid, gabapentin, and pregabalin are ionized in serum and their brain distribution is largely mediated by absorption across the BBB.
The degree of protein binding varies. MACs with >90% protein binding are phenytoin, valproic acid, cannabidiol, clobazam, clonazepam, perampanel, stiripentol, and tiagabine. An alteration of this binding and a change in the proportion of free active drug may occur in the event of hypoalbuminemia, chronic liver or kidney disease, pregnancy, and displacement by other drugs or endogenous substances (e.g., uremia).
Most MACs undergo extensive metabolism, primarily through oxidation by cytochrome P450 (CYP) enzymes (phase I reactions) or glucuronidation by UGT (phase II reactions). Exceptions include levetiracetam and rufinamide, which undergo hydrolysis, and gabapentin, pregabalin, and vigabatrin, which are excreted unchanged through the kidneys. Polymorphisms in the CYP2C9/19 genes may affect serum concentrations of phenytoin, cannabidiol, and N-desmethylclobazam. Alteration in renal function is an important determinant of the clearance of drugs that are eliminated through this route, such as levetiracetam, gabapentin and pregabalin.
> 2.4 Pharmacokinetic variability
Most MACs have pronounced pharmacokinetic variability. This variability is much greater between patients than within each patient, and for most MACs, there is a linear and predictable correlation between dose and serum concentration in each individual patient. Pharmacokinetic variability is an important determinant of differences in the response to MACs, and is determined by genetic factors, age, physiological state, pathologies, environmental factors and interactions with other drugs, being difficult to predict individually.
Based on the assumption that clinical effect correlates better with drug concentrations than dose, MTM can be used to tailor treatment to the patient. In MTM, quantification of drug levels in blood or serum is combined with information on pharmaceutical properties, patient characteristics, and a clinical assessment of effects and adverse effects to individualize treatment.
> 2.5 Drug interactions involving MAC
20%-25% of patients with epilepsy and >75% of patients with drug-resistant epilepsy are treated with two or more CAMs. It is also common for MACs to be prescribed together with other medications, mainly in the elderly. Patients may also use over-the-counter medications and dietary supplements that are not always reported. With a greater number of concomitant medications, the risk of adverse effects due to drug-drug interactions (DFI) increases, being a preventable cause of morbidity and mortality.
Many clinically important IFFs involving MAC are pharmacokinetic (one drug alters the concentrations of another), resulting from the induction or inhibition of drug metabolism. This is because many MACs are substrates, inducers and/or inhibitors of drug-metabolizing enzymes. The magnitude of the interaction depends on the fraction of dose that is eliminated by the affected pathway. The elimination of many MACs is largely dependent on a predominant metabolizing enzyme. Therefore, they are prone to IFFs as “victims” of the interaction. Displacement interactions at the protein binding site are relevant in some combinations, leading to an altered balance between the bound and free (active) fraction of the drug.
2.5.1 Drug interactions between MAC
Carbamazepine, phenytoin, phenobarbital, and imientodone are potent inducers of the isoenzymes CYP, UGT, and several drug transporters. Consequently, they may reduce the efficacy of co-administered CAMs such as lamotrigine (UGT substrate), perampanel and everolimus (CYP3A4/5 substrates). Topiramate, oxcarbazepine, eslicarbazepine, cenobamate, felbamate, and rufinamide also reduce serum concentrations of some concomitant MACs.
However, they are generally weak to moderate inducers. Some drugs inhibit drug-metabolizing enzymes. For example, valproic acid may reduce the clearance of lamotrigine, leading to an increased risk of poisoning and hypersensitivity reactions. The combination of cannabidiol and clobazam increases exposure to the main metabolites of both compounds, in particular N-desmethylclobazam.
Interactions based on enzyme inhibition have relatively rapid timelines, with new steady-state concentrations achieved within hours or days. In contrast, the maximum effect of enzyme induction can be observed days or weeks after the start of co-medication due to the time necessary to synthesize more metabolizing enzymes. Caution is also required when discontinuing the inducer, because serum concentrations of affected drugs may return to baseline even weeks after switching.
Pharmacodynamic IFFs between MAC involve additive effects, synergism or antagonism of drug action, without alterations in serum concentrations. These IFFs can be beneficial or dangerous. For example, the combination of lamotrigine and valproic acid may improve therapeutic outcomes, but may also increase adverse effects.
2.5.2 Interactions between MAC and other drug classes
Valproic acid, felbamate and cenobamate, can decrease the clearance of other drugs in addition to MACs, including psychotropics, calcium channel blockers and anticoagulants, causing toxicity. Cannabidiol is emerging as an inhibitor of multiple metabolizing enzymes, which can inhibit the elimination of many drugs. Drugs that require adjustment when prescribed with enzyme-inducing MACs include anticoagulants, calcium channel blockers and statins, immunosuppressants and chemotherapy agents, antibiotics and anti-HIV medications, psychotropics, antidiabetics, and oral contraceptives (OCs).
Proton pump inhibitors may increase the concentration of CYP2C19 substrates such as N-desmethyl-clobazam, carbapenem antibiotics may reduce the serum concentration of valproic acid, and cisplatin may decrease the concentration of phenytoin.
The preferred contraceptive methods in women treated with enzyme-inducing MACs are copper-containing intrauterine devices (IUDs), depot medroxyprogesterone acetate, or levonorgestrel-releasing devices. Combined OCs can decrease lamotrigine concentrations by approximately 50% or more, resulting in seizures in some women. This occurs by induction of glucuronidation of lamotrigine by ethinyl estradiol.
As a group, direct oral anticoagulants (DOACs) are particularly prone to pharmacokinetic interactions with MACs through induction or inhibition of enzymes and transporters. All DOACs are substrates of P-glycoprotein (P-gp), the induction of which would reduce their plasma concentrations.
However, DOACs differ in their metabolic pathways. CYP3A4 metabolizes rivaroxaban and to a lesser extent apixaban. Edoxaban and dabigatran depend on CYP450-mediated metabolism. Co-administration of DOACs and enzyme-inducing MAC may reduce serum concentrations of DOACs and predispose the patient to therapeutic failure.
Additionally, DOAC-MAC pharmacodynamic interactions may result in MAC effects on coagulation; e.g. due to valproic acid-induced thrombocytopenia or hemorrhage with concomitant use of phenytoin, valproic acid, or levetiracetam. The clinical significance of DOAC interactions with mild to moderate CYP3A4/P-gp-inducing MACs such as oxcarbazepine and cenobamate or enzyme-inhibiting MACs (cannabidiol, felbamate) is unknown.
When MAC therapy should be initiated in a patient treated with a DOAC (or vice versa), an interdisciplinary review, and monitoring of serum DOAC concentrations, is crucial. Some adverse effects of CAMs and other medications used in cardiovascular disorders may be additive due to pharmacodynamic interaction. For example, the combination of carbamazepine or oxcarbazepine with diuretics is associated with an increased risk of hyponatremia.
Concern about IFF increased with the emergency authorization of the anti-Covid-19 combination nirmatrelvir/ritonavir. Both compounds are substrates of CYP3A4 and ritonavir is a potent and irreversible inhibitor of CYP3A4, a weak to moderate inhibitor of several other CYP isoenzymes, and an inducer of UGT. Consequently, it is recommended that everolimus not be combined with this preparation. Patients treated with MAC that are CYP3A4 or lamotrigine substrates should be monitored for drug efficacy and adverse reactions.
2.5.3 Other interactions (food, environmental factors)
The best example of a food-CAM interaction is the increased exposure to cannabidiol when taken with a high-fat meal. Use of the antidepressant St. John’s wort (metabolizing enzyme inducer) may result in subtherapeutic levels of several MACs, particularly those that are CYP3A4 substrates, but at high doses. Alcohol consumption is not recommended in patients with epilepsy, due to additive CNS suppression or mood changes.
> 2.6 Principles and clinical use of MTM
2.6.1 Concepts and use of MTM
A key concept for proper implementation of MTM includes appropriate clinical evaluation and a rationale for measuring serum concentration. There are several reasons why MTM has become a commonly used tool to optimize treatment in epilepsy: pharmacokinetic variability and unpredictable drug interactions, as well as adherence and other treatment challenges, where MTM contributes to providing assurance of the quality of the treatment.
2.6.2 Definitions
The ILAE issued guidelines for MTM in 1993, which were updated in 2018 and 2020. The “reference range” is defined as “a range of drug concentrations, which specifies a lower concentration limit below which it is relatively unlikely to occur.” produces a therapeutic response, and an upper limit above which toxicity is relatively likely to occur.
Patients may achieve therapeutic benefit at concentrations outside these ranges and therefore “individual therapeutic concentrations” should be used, defined as “the range of drug concentrations that is associated with the best possible response in a given person.” . Thus, the clinical use of MTM is based on measurements over time in the patient, and the individual therapeutic concentration can then be established and followed when various patient- and drug-related factors vary over time.
2.6.3 What and when to measure
In general, total drug concentrations are measured. However, the pharmacologically active part of a drug is the “free, unbound” proportion. Therefore, serum/plasma is a relevant matrix for MTM. If altered protein binding of certain MACs is suspected, measurement of free and unbound concentrations can be performed and the results interpreted in the clinical context.
The MTM should be used with serum concentration measurements at a standard time point: pharmacological fasting before intake of the morning dose under steady-state conditions. For subsequent measurements and to establish the individual reference concentration, the first measurements could be used as a basis for comparison within that patient and relative to the reference range for the medication in use. If concentration-related adverse effects are suspected, a serum sample can be drawn after a few hours around Cmax to evaluate whether switching to a sustained-release formulation or dividing the daily dose can be recommended.
2.6.4 Special patient groups
During the transition from childhood to old age, physiological changes occur and pathologies may develop that affect the pharmacokinetics of various MACs.
Children are developing rapidly and careful consideration of physiology and organ function and maturation, pharmacokinetic characteristics, and overall drug elimination capacity is required.
In pregnancy, absorption may be affected by physiological changes or vomiting, the volume of distribution may change due to increased water and body fat reserves, and free and unbound concentrations of highly protein-bound medications may increase. In addition, the activity of drug-metabolizing enzymes is altered, and renal blood flow and glomerular filtration rate increase, affecting renal clearance.
Pharmacokinetic changes for CAMs include decreased serum concentrations of lamotrigine, levetiracetam, phenytoin, phenobarbital, licarbazepine, topiramate, and total carbamazepine and VPA, but data are limited or missing for other CAMs. These changes vary considerably between patients and are also influenced by environmental and individual factors. Therefore, regular monthly monitoring is recommended to avoid breakthrough seizures and ensure constant MAC exposure, reducing risks in offspring.
In the elderly, clearance is generally lower, either due to decreased renal function and/or less efficient metabolizing activity, but factors such as frailty, nutritional status, and comorbidities also play an important role.
For better monitoring of vulnerable patients, long-term MTM was introduced as a tool to investigate intra- and inter-patient pharmacokinetic variability over long periods.
2.6.5 Analytical methodologies and recent development of the MTM
MAC measurements are performed using laboratory-developed methods or commercial kits for research and/or clinical analysis purposes. Most use chromatographic methods or immunoassays in serum/plasma or alternative biological fluids such as saliva. Immunoassays are generally easy to perform, specific, and available for most MACs, although they are limited to the determination of a single drug.
Liquid chromatography combined with tandem mass spectrometry (LC–MS/MS) is the most widely used methodology due to its high sensitivity and specificity. Allows alternative techniques to be used, such as dried blood spot analysis and volumetric absorption microsampling. As a general rule, any laboratory providing MTM services should be involved in internal and external quality assessment programs to ensure consistency of results between laboratories.
2.6.6 Proper MTM Implementation
MTM should be used based on a clinical indication and therefore its indiscriminate or routine use without indication in unselected groups of patients is not recommended. The impact on refractory patients with various therapeutic challenges or in situations such as pregnancy is much more useful.
There is no clear correlation between the clinical effects and serum concentrations of all MACs, and therefore the “therapeutic range” is not used to define within which range the patient should remain during maintenance therapy. The term “individual therapeutic range” is therefore more appropriate, as it clarifies where the individual patient has the best treatment outcome.
| 3. FUTURE PROSPECTS |
In the personalized treatment of epilepsy, the use of MTM together with complementary tests can facilitate treatment decisions. This constitutes an important part of future directions. Genetic panels are increasingly used to evaluate genetic epileptic etiologies and enable appropriate therapeutic options, such as certain precision treatments.
Additionally, pharmacogenetic testing may be indicated to better adjust the dose from baseline in those patients who have different drug metabolizing abilities or to avoid exposure to drugs such as carbamazepine in patients with HLA-B*1502 polymorphism, which increases the risk of serious adverse effects. Biochemical markers can also be used to monitor CAM therapy, its adverse effects, and avoid hepatotoxicity seen with valproic acid or cannabidiol.
| 4. CONCLUSIONS |
Knowledge and understanding of basic and clinical pharmacology forms the basis for rational and safe CAM treatment. Pharmacological challenges include pronounced pharmacokinetic variability and numerous interactions both between MAC and between MAC and other drug classes. These are clinical indications for the individualized treatment approach in epilepsy using the MTM.
To provide the best possible control and monitoring of patients with epilepsy, it is essential that research and routine go hand in hand to facilitate safer and more effective CAM treatment in vulnerable patient groups. Future directions point to the combined implementation of MTM with additional tests such as genetic panels for proper diagnosis, pharmacogenetic testing when appropriate, and the use of biochemical markers that will contribute to personalized treatment.
