Guidance for interactions between antiseizure medications

KEYWORDS:

• antiepileptic drugs
• clinical relevance
• drug-drug interactions
• prevention

1. Introduction

History of antiseizure medications (ASMs) starts in 1857, with introduction of bromides in clinical practice, and nowadays about 30 drugs with anticonvulsant properties are available [1,2]. ASMs are frequently prescribed in clinical practice since prevalence of epilepsy in general population is about 1%, and there are other approved indications for ASMs, like bipolar disorder, neuralgia, or neuropathy [3]. Co-prescribing of two or more ASMs occurs in about 25% of children and 59.6% of adolescents and adults, typically with more severe types of epilepsy [4,5]. If at least one of co-administered ASMs has narrow therapeutic window [6], pharmacodynamic or pharmacokinetic interactions between them are more likely to be clinically relevant, resulting either with changes in therapeutic effect (augmentation or diminution), or with potentiation of adverse effects. Sometimes ASMs are deliberately combined, taking advantage of the pharmacodynamic interaction and augmentation of antiseizure effect, but this may require dose adjustment due to simultaneous pharmacokinetic interaction. Knowledge of the main principles of avoiding (or utilizing) drug–drug interactions (DDIs) between the ASMs should be of practical help when prescribing combinations of these drugs.

2. Types and mechanisms of DDIs

DDIs could be roughly divided into chemical, pharmacodynamic, and pharmacokinetic interactions. Pharmacodynamic interactions could be physiological and pharmacological [7]. Chemical interactions happen when two drugs bind each other in a syringe, infusion bottle, or in the blood; such interactions are usually called incompatibilities, they are mostly well known, and are avoided if drugs are administered according to instructions from summary of product characteristics (SmPC). Physiological interactions happen when drugs have same or opposite action on a tissue or organ, but are acting on different receptors and use different intracellular machinery. Pharmacological interactions are consequence of binding of drugs for the same receptors, and acting there as orthosteric agonists or antagonists, or allosteric receptor modulators, with resultant increase or decrease in efficacy. On the other hand, pharmacokinetic DDIs occur when drugs affect each other’s absorption, distribution, metabolism, or elimination; they are mostly caused by inducing or inhibiting metabolizing enzymes, cell membrane transporters, or efflux pumps [7]. If a drug’s pharmacokinetics or action on target tissue is changed by another drug that is simultaneously present in the human body, it is called ‘victim,’ while the other drug is designated as ‘perpetrator.’ However, sometimes a drug may take both roles, depending on precise mechanism of DDI.

3. DDIs among antiseizure medications

Although DDIs among ASMs are investigated on in vitro models and in animal studies during the drug development process, only after direct observation in several clinical trials or cohort studies a DDI becomes recognized as clinically relevant and is described and interpreted in official SmPC. Results of in vitro and animal studies mostly serve to explain mechanism of a DDI, but it is rarely completely understood [8]. However, clinical studies of a DDI sometimes have conflicting results, primarily due to large interindividual variations in drug metabolism and transport [8,9]. What we can observe in clinical studies as a proof of an DDI among ASMs is also rather limited: changes in control of epilepsy, adverse events, and concentration of ASMs and their metabolites in serum. Routine therapeutic drug monitoring (TDM) of ASMs in clinical practice may also direct our attention to possible pharmacokinetic DDIs if plasma concentrations of co-administered ASMs are significantly increased or decreased in comparison to values obtained during monotherapy. The guidelines for TDM in neuropsychopharmacology issued by working group for neuropsychopharmacology and pharmacopsychiatry are of great help in this sense, as they offer not only practical instructions for TDM but also insight into mechanisms and interpretation of possible DDIs [10]. Taking all together, interpretations of DDIs in SmPCs of ASMs are frequently not sufficiently explicit to warrant unequivocal action of a prescriber and prevention of certain DDI in routine clinical practice [11]. The DDIs described in SmPCs of currently marketed ASMs [12] are shown in Table 1, in a symbolic manner.

Table 1. Interactions between antiseizure medications with clinical significance.

	Phenobarbital	Phenytoin	Primidone	Ethosuximide	Carbamazepine	Valproate	Clonazepam	Clobazam	Vigabatrin	Zonisamide	Lamotrigine	Oxcarbazepine	Felbamate	Gabapentin	Topiramate	Tiagabine	Levetiracetam	Pregabalin	Rufinamide	Stiripentol	Lacosamide	Eslicarbazepine	Perampanel	Brivaracetam	Cannabidiol	Cenobamate	Fenfluramine
Phenobarbital		↑±	+	-	-	↑±	↑-		-	-	-	±	±			-				+			-	-		+
Phenytoin			↑-	+	±	↑+	↑-	±	-	-	-	±	±		±	-				+		-	-	±	+	±
Primidone				-	↑-	↑±	±			-	-	-	-			-				±			-
Ethosuximide					-	±					-									+
Carbamazepine						↑-	-	-	+	-	↑-	-	-		-	-	↑			+		↑-	-	↑±		-
Valproate							↑-	±			↑+		+		↑				+	+					↑
Clonazepam
Clobazam																				↑+					↑+	+
Vigabatrin
Zonisamide
Lamotrigine												↑			+							-			+	-
Oxcarbazepine																						↑	±
Felbamate
Gabapentin
Topiramate																						-	-
Tiagabine																				+
Levetiracetam
Pregabalin
Rufinamide
Stiripentol																									+		+
Lacosamide
Eslicarbazepine
Perampanel
Brivaracetam																									+		+
Cannabidiol
Cenobamate
Fenfluramine

(↑) Increase in adverse effects intensity; (↓) decrease in adverse effects intensity; (+) increase in serum concentration of one or both antiseizure medications and possibly intensification of anticonvulsant activity; (-) decrease in serum concentration of one or both antiseizure medications and possibly diminution of anticonvulsant activity; (±) increase in serum concentration of one and decrease in serum concentration of another antiseizure medication given in combination. Fosphenytoin is not mentioned in the table due to similarity of its interaction potential with phenytoin.

While there are nowadays tools available for rapid checkup of potential DDIs among the ASMs, like Lexicomp or Micromedex DDI checkers, their recommendations are sometimes not explicit enough, too; especially, choice of alternative ASMs in case of serious potential DDIs that require replacement of primarily intended prescription (e.g. ‘contraindicated’ concomitant administration) is not tackled in the interaction checkers [13]. Therefore, rapid orientation, and use of some simple yet sufficiently directional principles concerning DDIs among ASMs are necessary. The first principle is predominance of pharmacokinetic over pharmacodynamic interactions among ASMs, mostly because combining drugs with the same known adverse effects is avoided in the first place. Second, all ASMs could be classified into three groups according to their propensity to engage in pharmacokinetic DDIs among ASMs: highly interacting group (phenobarbital, phenytoin, primidone, carbamazepine, valproate, lamotrigine, fosphenytoin, and stiripentol), moderately interacting group (ethosuximide, clobazam, clonazepam, zonisamide, topiramate, tiagabine, eslicarbazepine, oxcarbazepine, perampanel, brivaracetam, cannabidiol, and cenobamate), and low-interacting group (gabapentin, levetiracetam, vigabatrin, pregabalin, rufinamide, lacosamide, and fenfluramine) [14]. This classification corresponds to extent of lipophilicity of the ASMs: members of the highly interacting group are highly lipophilic substances that according to the Biopharmaceutics Drug Disposition Classification System (BDDCS) belong to class 1 (highly soluble, highly permeable, and extensively metabolized in the liver in order to be transformed to sufficiently hidrosoluble metabolites capable of being excreted in the kidneys). The low-interacting group consists of drugs that are low lipophilic and belong to BDDCS class 3 (high solubility, low permeability, low metabolism), which means that they are minimally or not at all metabolized in the liver, but mostly excreted unchanged in the urine [15]. Since main pharmacokinetic DDIs of ASMs happen on liver cytochromes, it is not surprising that drugs that are extensively metabolized at cytochromes are also subject to DDIs. The moderately interacting group consists either of low-lipophilic drugs (which are excreted unchanged in urine) or of moderately lipophilic drugs that are metabolized in liver, but only partially or not at all on cytochromes (they are either acetylated or directly conjugated). The third principle holds that prescription of two or more ASMs from highly or moderately interacting groups requires use of an interaction checker or insight in SmPCs of the prescribed drugs to establish whether dose adjustment or some kind of monitoring adverse effects are mandatory. Therefore, rational approach to problem of DDIs among ASMs is to pay special attention to cases of prescribing highly or moderately interacting drugs, explore each such case thoroughly in regard to interaction risk, use modern informatic tools, and, if necessary and feasible, prescribe an ASMs from low-interacting group. However, classification to highly, moderately, or low-interacting group may not be fully applicable to ASMs when DDIs with drugs from other groups are in question, but this issue is out of scope of this article.

4. Expert opinion

Although the possibility of DDIs between ASMs is first investigated in in vitro or animal studies, clinical studies are necessary for definitive conclusions. However, the investigation of interactions between ASMs in clinical studies faces two problems: (1) it is necessary to plan and carry out the measurement of serum concentrations of ASMs and their active metabolites, and (2) it is necessary to evaluate the clinical significance of the interactions that are observed. Although newer ASMs are being investigated for their potential to affect metabolism of other drugs in healthy volunteers if an in vitro signal for the potential of such interactions was observed, complete insight into clinical consequences of such interactions is possible only in studies involving patients with epilepsy who take ASMs doses titrated to optimal effect. Paying more attention to registration, follow-up, and causality assessment of adverse events when designing both clinical trials and observational studies would help with identification of clinically relevant DDIs, which may further be analyzed in terms of pathophysiology.

A more detailed investigation of the interactions between ASMs in the future, and especially assessment of their clinical significance are needed in future. The final goal of research on DDIs would be the formation of precise but complex algorithms for the application of combinations of ASMs for each type of epilepsy, which would avoid the occurrence of interactions, or would use them to achieve a better therapeutic effect. In order to implement such complex algorithms in a busy clinical practice, it will be necessary to translate them to an easy-to-use smartphone application functional both online and offline. In the meantime, clinicians should pay special attention to highly interacting ASMs: phenobarbital, phenytoin, primidone, carbamazepine, valproate, lamotrigine, fosphenytoin, and stiripentol.

Of great importance in future research on DDIs between ASMs will be new technologies for measuring the concentration of ASMs and their metabolites in the serum of subjects, such as liquid chromatography with double mass detection (LC-MS/MS), which allow rapid and reliable measurement of a large number of substances from only one blood sample. By introducing the measurement of the concentration of ASMs and their metabolites in all clinical studies dealing with ASMs, the conditions will be created to monitor and detect pharmacokinetic interactions, and by comparing the measurement results with the clinical picture, their clinical significance will be better assessed.

In the coming years, we can expect more clinical studies investigating DDIs among ASMs that relatively recently received marketing authorization (fenfluramine, cannabidiol, brivaracetam), especially those with mechanism of action completely different from the other ASMs (e.g. soticlestat). With the advent of causal therapy of Dravet syndrome and other childhood epileptic syndromes with clear genetic etiology, there will be more innovative drugs administered in combination with classic ASMs, and more potential DDIs that will have to be investigated.

Special area of research involving DDIs among ASMs that will gain more attention is question how genetic variation of drug-metabolizing enzymes or membrane transporters/efflux pumps may influence DDIs [16]. Starting from the present time, when we can determine genetic variants of key enzymes relevant for pharmacokinetics of ASMs in each patient, carefully designed clinical trials will explain differences in clinical consequences of DDIs among patients who participated in earlier studies.

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 study was partially funded by Serbian Ministry of Science, Technological Development and Innovations, grant no. 175007, contract no. 451-03-47/2023-01/200111.