Mechanisms of idiosyncratic drug reactions: the case of felbamate

https://doi.org/10.1016/S0009-2797(02)00057-1Get rights and content

Abstract

Idiosyncratic drug reactions (IDR) are a specific type of drug toxicity characterized by their delayed onset, low incidence and reactive metabolite formation with little, if any, correlation between pharmacokinetics or pharmacodynamics and the toxicological outcome. As the name implies, IDR are unpredictable and ofter result in the post marketing failure of otherwise useful therapies. Examples of drugs, which have failed as a result of IDR in recent years, include trovafloxacin, zileuton, troglitazone, tolcapone and felbamate. To date there exists no pre-clinical model to predict these adverse drug reactions and a mechanistic understanding of these toxicities remains limited. In an attempt to better understand this class of drug toxicities and gain mechanistic insight, we have studied the IDR associated with a model compound, felbamate. Our studies with felbamate are consistent with the theory that compounds which cause IDR undergo bioactivation to a highly reactive electrophilic metabolite that is capable of forming covalent protein adducts in vivo. In additon, our data suggest that under normal physiological conditions glutathione plays a protective role in preventing IDR during felbamate therapy, further emphasizing a correlation between reactive metabolite formation and a toxic outcome. Clinical studies with felbamate have been able to demonstrate an association between reactive metabolite formation and a clinically relevant toxicity; however, additonal research is required to more fully understand the link between reactive metabolite formation and the events which elicit toxicity. Going forward, it seems reasonable that screening for reactive metabolite formation in early drug discovery may be an important tool in eliminating the post-marketing failure of otherwise useful therapies.

Introduction

The synthesis of felbamate (FBM), 2-phenyl-1,3-propanediol dicarbamate, was first achieved in the late 1950s in an effort to explore the structure activity relationship (SAR) of the known anxiolytic compound, meprobamate, 2-methyl-2-propyl-1,3-propanediol dicarbamate [1]. The SAR study of the 1,3-propanediol dicarbamate analogs proved that the 2-phenyl-1,3-propanediol dicarbamate analogue lacked sedative activity [1]. As a result, felbamate did not receive much attention until approximately thirty years later when Wallace Laboratories submitted felbamate to the Epilepsy Branch of the National Institute of Neurological Disorders and Stroke (NINDS) to be screened as an anti-convulsant therapy [2]. These collaborative efforts identified the therapeutic potential of felbamate as an anti-epileptic drug (AED) and prompted a series of clinical trials, ultimately leading to FDA approval in 1993 [2], [4], [5], [6].

At the time of felbamate's release in 1993, it was the first new AED in over fifteen years [7]. The need for new and better AEDs was highlighted by the fact that over 25% of epileptic patients were refractory to the available medications [8]. In addition, the older AEDs produced significant side effects, including marked sedation affecting the patient's quality of life. Felbamate debuted as a promising new therapy because it proved effective in the treatment of refractory partial seizures and in particular, Lennox-Gastaut syndrome [3]. Lennox-Gastaut Syndrome is a severe form of childhood epileptic encephalopathy, whose treatment is refractory to other AEDs [7]. Children with Lennox-Gastaut Syndrome can experience as many as 200 seizures a day, resulting in mental retardation [9]. In patients refractory to other AEDs, treatment with felbamate resulted in a 46% lower seizure frequency and 39% fewer drop attacks [9] Another study demonstrated that treatment with felbamate caused 5% of refractory patients to become seizure free, 11% to improve 75 – 100% by a reduction in their number of seizures, and 23% to improve 50 – 75% [10]. In addition to a reduction in seizure frequency and severity, felbamate demonstrated relatively few side effects and was notably less sedative than earlier therapies [11].

In its first year on the market, felbamate reached an estimated patient population of 120 000, which is significant in a patient population resistant to changes in therapeutic regimens [7]. However, within that first year, an unexpected association between felbamate therapy and reported cases of aplastic anemia and hepatotoxicity were realized [12], [13], [14]. In all, 34 cases of aplastic anemia, resulting in 13 deaths, and 23 cases of hepatic failure, resulting in 5 deaths, were reported to the manufacturer [7], [15]. Evaluation of the case reports of aplastic anemia considers a conservative risk assessment to be 1 in 4,800 and a liberal risk assessment to be 1 in 37,000 [16]. The risk for hepatotoxicity is estimated to be 1 in 18,500 to 25,000 [7]. The toxicities associated with felbamate resulted in a black box label warning by the FDA, which severely curtailed its use. Since felbamate offers relief to refractory patients, the FDA did not withdraw its use, but rather limited its use to cases where the benefit of therapy outweighs the risks of untreated seizures. An estimated 12,000 patients remain on felbamate therapy within the United States today [7].

The toxicities associated with felbamate therapy have in common several characteristic features of a particular class of adverse drug reactions referred to as idiosyncratic drug reactions. Idiosyncratic drug reactions, as the name implies, refer to drug toxicities that are unpredictable based on the pharmacological mechanism(s) of action of the drug. Thus, the reactions of felbamate, an antiseizure medication, are idiosyncratic in that they cannot be explained through the pharmacology of the compound. Other examples of drugs that cause idiosyncratic reactions include halothane, carbamazepine, phenytoin, acetaminophen, tienillic acid, troglitazone, trovafloxacin, zileuton, diclofenac and tolcapone [17], [18], [19], [20], [21], [22], [23], [24], [25], [26]. Given their unpredictable nature, their low incidence, and rare occurrence in animal species, idiosyncratic reactions are not recognized during pre-clinical or clinical evaluation and generally first appear during the widespread distribution of the drug. The hallmarks of idiosyncratic reactions include: 1) a delayed onset, usually 14 days to several months from the time of initiating drug treatment; 2) low incidence, occurring at a rate of 1 in 100 to 1 in 100 000; 3) reactive metabolite formation, usually at the site of toxicity; 4) little correlation between dosage and risk, although most cases occur at a dosage>10 mg/day [27]. Often, idiosyncratic reactions occur in the liver and blood, resulting in hepatic failure and blood dyscrasias. Although idiosyncratic reactions generally involve the formation of reactive species, the resultant toxicity in most cases is not thought to occur directly through modification of tissue nucleophiles. Instead the toxicity associated with most idiosyncratic reactions is thought to occur indirectly through the response of the immune system to the presence of modified proteins or possibily to the drug or its metabolites—a kind of drug-induced autoimmune response [27].

Two theories offer explanations of immune-mediated idiosyncratic drug toxicities. Dr. Jack Utrecht and Dr. Polly Matzinger posit the most widely accepted theory, which combines the hapten hypothesis and danger hypothesis [27], [28]. Evidence suggests that small molecular weight molecules (<1000 Da) are incapable of eliciting an immune response. The hapten hypothesis suggests that a drug or a reactive metabolite alkylates an endogenous protein(s) forming a hapten-protein complex. The hapten-protein undergoes normal protein processing, producing peptides of 8–15 amino acids. The resultant peptides are capable of eliciting two types of an immune response based on the specific antigen [27]. If the antigen is the hapten-peptide, the antibodies will recognize the hapten-protein. However, hapten-protein formation may result in altered protein proteolysis and the presentation of cryptic antigens by the major histocompatibility complex (MHC), thereby inducing a response to a self-protein and causing an autoimmune response or irreversible toxicity. Often, the site of hapten-protein formation is the site of toxicity [29], [30]. For instance, if drug-metabolizing enzymes in the liver, such as CYP450s, generate the reactive metabolite, the toxicity may be hepatotoxic (e.g. acetaminophen). If the reactive metabolite results from macrophage oxidation, a blood dyscrasia may develop (e.g. clozapine).

The danger hypothesis suggests that MHC-II peptide presentation by the antigen presenting cell (APC) alone does not cause a physiological T-cell response, but requires an additional or danger signal, thought to involve cytokine stimulated expression of B7 for recognition by CD28 on the surface of the T-cell [27]. The hypothesis also suggests that peptide presentation without the danger signal leads to tolerance, while peptide presentation with the danger signal leads to an immune reaction. A host of events may lead to the expression of cytokines required to stimulate the danger signal. They include normal infection, surgery, and disease states, such as HIV infection [27]. Another mechanism by which cytokines could be induced is through the production of reactive intermediates that deplete GSH. This could lead to the activation of transcription factors sensitive to the cellular redox status, such as NF-κB. In those terms, a sufficient loading of reactive intermediate alone would be enough to induce the danger signal.

The alternative Pichler hypothesis of idiosyncratic reactions, termed direct metabolism-independent T-cell stimulation, suggests that hapten-protein formation is not required for adverse drug toxicities [31], [32], [33]. Rather, Pichler holds that drugs (<1000 Da) can directly bind to peptide-loaded MHC on APC for T-cell receptor recognition. Evidence for this hypothesis includes: 1) direct T-cell activation by tucaresol; 2) glutaraldehyde-fixed APC were able to present drugs to specific CD4+ and CD8+ T-cell clones; 3) T-cell receptor expression kinetics after incubation with APC; 4) washing APC that had been incubated with drug nullified the stimulatory capacity [33]. While the Pichler hypothesis may offer insight into some idiosyncratic reactions, the hypothesis does not account for irreversible drug toxicity or autoimmune-type drug reactions that persist upon therapy cessation. As noted below, we have produced evidence that the idiosyncratic reactions of felbamate appear to be initiated by a reactive intermediate that is capable of reacting with tissue nucleophiles. At present, we have not produced sufficient data to conclude that tissue modification induces the idiosyncratic reactions through a directly toxic (apoptotic or necrotic) mechanism or through an immune-mediated indirectly toxic mechanism. However because the balance of evidence points toward an indirect mechanism, we have chosen to adopt the hapten hypothesis and danger signal hypothesis in our discussion of the molecular toxicology of felbamate.

Section snippets

Felbamate asorption, distribution, metabolism and excretion (ADME)

In man, felbamate is a relatively insoluble compound with good (> 90%) oral absorption, a volume of distribution of 0.81 liters/kg, and less than 25% binding to plasma protein [34], [35]. Of interest for an AED, felbamate has excellent blood brain barrier penetration with a plasma to brain ratio of 0.6 [36]. Felbamate undergoes extensive metabolism with a human renal clearance of greater than 95%. At the time of its approval, the metabolism of felbamate had been studied in several species

Felbamate metabolic activation–in vitro

Upon the reports of felbamate-associated toxicity in 1994 and given the previously characterized felbamate metabolites, a reactive felbamate metabolite was proposed [40]. At the time that the reported toxicities were recognized, four felbamate metabolites had been characterized, p-OHF, 2-OHF, MCF and CPPA. Given that a one-step four-electron oxidation from the alcohol (MCF) to the acid (CPPA) was unlikely, an intermediate aldehyde was proposed to exist [40]. If formed, the proposed aldehyde

Felbamate metabolic activation–in vivo

Since approximately 10,000 patients remain on felbamate therapy within the United States, further studies to characterize the potential for atropaldehyde formation in vivo were warranted. If formed in vivo, one could expect that atropaldehyde would react with endogenous GSH and, after subsequent processing, would ultimately be excreted as the corresponding mercapturates, N-acetyl-S-(2-phenylpropan-3-ol)-L-cysteine and N-acetyl-S-(2-phenylpropanoic acid)-L-cysteine. The first evidence for the in

In vitro enzymatic characterization

The observation of an individual with a dramatic shift in the ratio of CPPA to mercapturates raised the obvious question of the potential role for polymorphisms in felbamate metabolism. Given the low-incidence of felbamate toxicity, we thought that if polymorphisms did play a role in the formation or detoxification of atropaldehyde, that toxicity would only be observed in a patient that had a two or more polymorphisms, given the relative high incidence of polymorphisms to low incidence of

In vitro covalent protein binding

One element that remains to be explored in understanding felbamate-associated toxicities is the hapten-protein conjugate that may mediate the observed toxicities. Given the danger hapten-hypothesis of idiosyncratic reactions, an important step in activation of the immune system involves the formation of a hapten-protein conjugate. To date, the covalent modification of three proteins, GSTM1-1, ALDH and HSA, by atropaldehyde have been identified [46], [58], [62]. Irreversible inhibition of

Future directions and conclusions

All of our data to date support the hypothesis that atropaldehyde is the toxic felbamate metabolite that may lead to idiosyncratic toxicity. The research reviewed here demonstrates the identification of a reactive metabolite in vitro and in vivo and correlates a patient who may have developed felbamate-related toxicity as a result of the identified reactive metabolite. The research on felbamate has also demonstrate that the reactive felbamate metabolite, atropaldehyde, is capable of covalently

References (64)

  • B.J. Ludwig et al.

    Carbamate derivitives related to meprobamate

    J. Med. Chem.

    (1969)
  • E.A. Swinyard et al.

    Comparative anticonvulsant activity and neurotoxicity of felbamate and four prototype antiepileptic drugs in mice and rats

    Epilepsia

    (1986)
  • The Felbamate Study Group in Lennox-Gastaut Syndrome. (1993) Efficacy of felbamate in childhood epileptic...
  • E. Faught et al.

    Felbamate monotherapy for partial onset seizures: an active-control trial

    Neurology

    (1993)
  • R. Sachdeo et al.

    Felbamate monotherapy: controlled trial in patients with partial onset seizures

    Ann. Neurol.

    (1992)
  • I.E. Leppik et al.

    Felbamate for partial seizures: results of a controlled clinical trial

    Neurology

    (1991)
  • J.M. Pellock

    Felbamate

    Epilepsia

    (1999)
  • J.M. Pellock et al.

    Use of new antiepileptic drugs in the treatment of childhood epilepsy

    Epilepsia

    (1999)
  • I.E. Leppik et al.

    The place of felbamate in the treatment of epilepsies

    CNS Drugs

    (1995)
  • P.B. Pennell et al.

    Aplastic anemia in a patient receiving felbamate for complex partial seizures

    Neurology

    (1995)
  • M.G. O'Neil et al.

    Felbamate-associated fatal acute hepatic necrosis

    Neurology

    (1996)
  • S.L. Nightingale

    From the food and drug administration. Recommendation to immediately withdraw patients from treatment with felbamate

    J. Am. Med. Assoc.

    (1994)
  • J.M. Pellock et al.

    Felbamate: 1997 Update

    Epilepsia

    (1997)
  • D.W. Kaufman et al.

    Evaluation of case reports of aplastic anemia among patients treated with felbamate

    Epilepsia

    (1997)
  • C.S. Davidson et al.

    Concerning hepatotoxicity of halothane

    New Eng. J. Med.

    (1966)
  • N.C. Moore et al.

    Three cases of carbamazepine toxicity

    Am. J. Psychiatry

    (1985)
  • M.J. Eadie

    Formation of active metabolites of anticonvulsant drugs. A review of their pharmacokinetic and therapeutic significance

    Clin. Pharmacokinet.

    (1991)
  • J.C. Homberg et al.

    A new anti-liver microsome antibody (anti-LKM2) in tienilic acid-induced hepatitis

    Clinica. Chimica. Acta

    (1984)
  • S. Subramaniam

    The emerging role of thiazolidinediones in the treatment of diabetes-mellitus and related disorders

    Clin. Exp. Hypertension

    (1999)
  • C.A. Sorkness

    The use of 5-lipoxygenase inhibitors and leukotriene receptor antagonists in the treatment of chronic asthma

    Pharmacotherapy

    (1997)
  • D. Schapira et al.

    Diclofenac-induced hepatotoxicity

    Postgraduate Medical Journal.

    (1986)
  • J. Uetrecht

    New concepts in immunology relevant to idiosyncratic reactions: the Danger Hypothesis and innate immune system

    Chem. Res. Toxicol.

    (1999)
  • Cited by (89)

    • Reactive metabolites of the anticonvulsant drugs and approaches to minimize the adverse drug reaction

      2021, European Journal of Medicinal Chemistry
      Citation Excerpt :

      Therefore, it is not considered among the first-line treatment drugs [282,283]. FBM metabolized to form a major oxidized metabolite 3-carbamoyl-2-phenyl propionic acid [277,283]. It is highly unstable and believed to be the RM, responsible for the toxicities associated with FBM [18,284].

    • Metabolic profiling of norepinephrine reuptake inhibitor atomoxetine

      2020, European Journal of Pharmaceutical Sciences
    • Drug-Induced Liver Injury

      2017, Zakim and Boyer's Hepatology: A Textbook of Liver Disease
    View all citing articles on Scopus
    View full text