Statistics from Altmetric.com
Acute symptomatic seizures occur in 10–20% of patients with acute intermittent porphyria in relapse. There are rarer reports of porphyria in people with chronic epilepsy who are mostly drug resistant. The association between epilepsy and porphyria is unclear. Porphyria might be the cause of chronic symptomatic epilepsy (if so, this would be rare or frequently undiagnosed) or there might be a chance association, given that epilepsy is common. Nevertheless, in drug-resistant epilepsy, clinicians should consider metabolic causes such as the porphyrias, especially when the seizure frequency increases on higher doses of certain antiepileptic drugs. Acute porphyric attacks can be potentially fatal and are usually precipitated, in susceptible individuals, by exposure to commonly used drugs, including certain antiepileptic drugs. It is, therefore, important to determine the correct dose and safety of use of certain drugs in people with porphyria. Having diagnosed porphyria-related seizures, the patient will also need treatment of the porphyria itself; its management includes appropriate selection of a non-porphyrinogenic antiepileptic drug. Enzyme-inducing medications induce hepatic heme synthesis, which can exacerbate porphyria symptoms, or provoke acute attacks.
Clinical experience and findings from experimental systems using whole animal or cell culture models have been used to determine porphyrogenicity (the potential of a drug to induce an acute porphyric attack) and to classify drugs as safe or unsafe in freely available drug lists (eg, http://www.drugs-porphyria.org; http://www.wmic.wales.nhs.uk/pdfs/porphyria/2015%20Porphyria%20safe%20list.pdf).
In acute porphyric attacks, seizures can be treated with intravenous diazepam, levetiracetam or propofol if status epilepticus develops; a single or few seizures may not require antiepileptic drug treatment in the long term, provided the porphyria itself is properly managed (box 1).
Antiepileptic drugs and porphyria
Antiepileptic drugs considered safe for use in the acute porphyrias (not porphyrinogenic or probably not porphyrinogenic)
Antiepileptic drugs that should not be used in the acute porphyrias (porphyrinogenic or probably porphyrinogenic)
Uncertain (possibly porphyrinogenic or not yet classified)
The porphyrias form a heterogeneous group of inherited metabolic disorders, each of which results from deficiency of a specific enzyme in the multistep heme biosynthetic pathway.1
Variegate porphyria is an autosomal-dominant form of hepatic porphyria associated with disease-predisposing mutations in the gene encoding protoporphyrinogen oxidase (PPOX). Its phenotypic expression results from deficiency of this enzyme, which converts protoporphyrinogen to protoporphyrin in the penultimate step of heme biosynthesis.
In PPOX deficiency, the gene carrier state is generally clinically quiescent, but exposure to a variety of precipitating factors, including a wide range of commonly prescribed medications, can trigger symptoms of variegate porphyria.
Attacks in acute porphyria mainly feature gastrointestinal and neuropsychiatric symptoms. Accumulation of phototoxic porphyrins in the skin may lead to solar hypersensitivity and bullous dermal lesions. Treating epilepsy in acute porphyrias is challenging because many commonly used antiepileptic drugs strongly induce heme-dependent cytochromes P450 in the liver. In the course of holoenzyme assembly, the liver meets demands for heme by accelerating de novo heme biosynthesis. In PPOX deficiency, the pathway may become overloaded on porphyrogenic challenge, with resulting accumulation of phototoxic porphyrins, as well as presumably neurotoxic pre-porphyrin intermediates.
Perampanel is a structurally novel, selective non-competitive AMPA receptor antagonist recently approved for treating partial-onset and secondarily generalised seizures. We describe a patient with epilepsy who was a carrier of variegate porphyria, in which we tried perampanel after safety assessment.
A 44-year-old woman had her first tonic-clonic seizure aged 9. She subsequently tried many antiepileptic drugs without full seizure control. At age 29, she was found to have variegate porphyria after developing a right hemiparesis with ataxia and dysphasia. DNA analysis showed a deletion [IVS5-(24-16) del CTTAGTCCT] in intron 5 of the PPOX gene, likely to be the cause of her variegate porphyria. She was also diagnosed with primary hypothyroidism. At aged 31, she had a vagus nerve stimulator implanted, without benefit. She was also treated with GnRH analogues and low doses of oestrogens, given some correlation between her menstrual cycle and seizure frequency.
She was referred to our centre aged 42. At that time, her antiepileptic medication was gabapentin, levetiracetam, pregabalin and clobazam, all previously assessed not to be porphyrogenic. She was experiencing multiple seizures daily of various types. None of her previous or current antiepileptic medication had precipitated any porphyric crisis. Prolonged EEG-videotelemetry showed interictal left frontotemporal abnormalities, also with some right-sided epileptiform changes. We captured multiple seizures and electroclinical evidence suggested left frontotemporal origin. Her ECG was normal. Neuropsychometry showed widespread cerebral dysfunction, with verbal and visuospatial skills below the average range and verbal memory, naming and fluency all within the borderline-to-impaired range.
We introduced lacosamide, considered safe in porphyria. Unfortunately, she developed a skin rash, but no symptoms of a porphyric crisis. We could not exclude an allergy to lacosamide and therefore stopped it.
Perampanel is among the newest antiepileptic drugs, with no previous report of its use in acute porphyrias and no patient exposure data at the National Acute Porphyria Service (http://www.cardiffandvaleuhb.wales.nhs.uk/national-acute-porphyria-service-naps, accessed 15 March 2013).
The Porphyria Centre in Sweden categorises perampanel as ‘probably not porphyrogenic’, the lowest risk category after ‘safe’. This is based upon in silico analysis, using pharmacokinetic data to assess drug cytochrome (CYP)-inductive power. In addition, there were evaluations of pharmacodynamic mechanisms, physiological actions and side effects of the drug for potential capacity to activate or co-activate nuclear receptors responsible for CYP induction. The only relevant finding was evidence of its weak capacity to induce CYP3A4. There were no reports of clinical observations against in silico safety assessment results.
After discussing this with the patient and her family, we started perampanel. Her urinary porphobilinogen concentration pretreatment and at 1 and 2 weeks on treatment were normal. The National Acute Porphyria Service did not recommend further routine urinary monitoring. The perampanel dose was initially 2 mg nocte, gradually titrated to 4 mg daily.
Her seizure frequency initially improved, with up to 15 days without seizures, compared with daily seizures before perampanel. After 3 months, she developed unsteadiness and had difficulty transferring from her wheelchair. This adverse effect required admission to hospital and physiotherapy. We reduced the perampanel dose to 2 mg daily. She had no symptoms of acute porphyria while on perampanel during 23 months of treatment.
While acute symptomatic generalised seizures are recognised features of acute attacks of porphyria, the association between porphyria and drug-resistant epilepsy is less clear, with few reported cases.2
In treating epilepsy in patients with acute porphyrias, the choice of drugs should be from among the non-CYP-inducing antiepileptic drugs (see box 1) to avoid porphyrogenic acceleration of hepatic heme biosynthesis. Treatment with carbamazepine, phenobarbital, phenytoin, primidone, topiramate and sodium valproate should be avoided as far as possible. Tiagabine shows porphyrinogenicity in in vitro studies using cultured liver cells and may be hazardous.
In this case of variegate porphyria and drug-resistant epilepsy, use of perampanel was not followed by any clinical or biochemical signs of activation of the disorder. A single observation of tolerance to a drug in a carrier of acute porphyria, however, cannot be taken as proof of non-porphyrogenicity because of the great variability between carriers—as well as in one carrier over time—in susceptibility to the action of porphyria precipitating agents. Women are more prone than men to attacks of acute porphyria: in the present case, the carrier exposed to perampanel is a female with a history of clinically manifest acute porphyria, indicating that she is not among the group of people seemingly constitutionally resistant to the phenotypic manifestations of acute porphyria. The circumstance that our patient is potentially vulnerable to porphyrogenic challenge serves to enhance the significance of the observation of her tolerance to perampanel and helps to validate the in silico assessment of probable non-porphyrogenicity of the drug.
There have been two recently reported cases of drug-resistant epilepsy in non-carriers of acute porphyria. Both presented in convulsive status epilepticus and were on treatment with perampanel. Both had significantly low serum concentrations of concurrent antiepileptic medication (phenytoin, phenobarbital, rufinamide) in comparison with concentrations before perampanel introduction. In one of the cases, further increasing the perampanel dose led to a further drop of the serum phenytoin concentration. As demonstrated in cultured human hepatocytes and through drug interaction studies, perampanel weakly induces CYP2B6 and CYP3A4/5, as well as uridine 5′-diphospho-glucuronosyltransferase, while not affecting drug transporters. It is, therefore, conceivable that the increased rate of elimination of the CYP-metabolised co-administered drugs in the reported cases is an effect of the weak CYP-inductive capacity of perampanel. Until we acquire more experience, initial monitoring of urinary porphobilinogen excretion and subsequent clinical vigilance should be routine when using perampanel in acute porphyrias, especially at higher doses. Monitoring of concomitant antiepileptic drug levels is also indicated. The common side effects, nausea and disturbed appetite, necessitate attention to nutrition of the patient, to reduce risk for potentially porphyrogenic decrease of caloric intake.
Non-CYP-inducing antiepileptic drug should be used to treat epilepsy associated with acute porphyrias.
Perampanel is a weak inducer of cytochrome P450 enzymes.
Perampanel can be tolerated in long-term therapy in carriers of acute porphyrias.
Vigilance should be maintained for clinical and biochemical signs of activation of acute porphyria when used in a carrier of acute porphyria.
Management of epilepsy associated with porphyrias-what to use and what not to use
Two useful websites:
http://www.drugs-porphyria.org (accessed 20 October 2015).
http://www.wmic.wales.nhs.uk/porphyria_info.php#resources_healthcare (accessed 20 October 2015).
Contributors Study concept: SMS. Clinical data acquisition: SB and YH. Drafting of the manuscript: SB. Critical revision of the manuscript for important intellectual content: ST and SMS.
Funding Part of this work was undertaken at University College London Hospitals, who received a proportion of funding from the NIHR Biomedical Research Centres funding scheme. SB was supported by the Polytechnic University of Marche, Italy, for a 1-year research fellowship.
Disclaimer In the report, the classifications given for drugs in terms of porphyrogenicity/non-porphyrogenicity are from the literature and generally based on clinical observations, and experimental or in vitro findings. In some cases, they are results of pharmacological considerations applied to a genometabolic model of acute porphyria. There are, however, potential sources of error in all presently available techniques for drug porphyrogenicity assessment. Even with care taken to eliminate them errors lege artis, it is not possible to take legal responsibility for the drug classifications provided and the data should not be taken as advice.
Competing interests None declared.
Patient consent Obtained.
Provenance and peer review Not commissioned; externally peer reviewed. This paper was reviewed by Mark Manford, Cambridge, UK.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.