Antiepileptic drug therapy in patients with autoimmune epilepsy

DISCUSSION

Although the seizures in AE are often considered to be medication resistant, and immunotherapy is often the focus of management,⁷ this study suggests that in select cases AEDs alone may confer seizure freedom.

Of interest in our study, 9 patients became seizure free after the initiation of AEDs with sodium channel blocking properties⁷ (carbamazepine, n = 3; oxcarbazepine, n = 2; lacosamide alone, n = 3, or with phenytoin, n = 1). The response to AEDs was seen within 2–4 weeks of the initiation of therapy in all patients. In a few patients, AEDs alone were effective in controlling seizures. However, improvement of cognitive symptoms and other manifestations of autoimmune encephalitis in most patients requires immunotherapy.⁷ It is conceivable that these therapeutic responses could be secondary to immunomodulatory properties of these medications.^4,–,6 For example, carbamazepine has been shown to reduce levels of proinflammatory cytokines IL-1β and TNFα in the hippocampus of rats⁴ and inhibits the development of different types of inflammation through dose-dependent reduction of prostaglandin E2 and substance P.⁵ Carbamazepine and oxcarbazepine have also been shown to decrease serum levels of IL-1 and IL-2 in healthy subjects.⁵ Although disruption of blood-brain barrier permeability has been suggested as a possible pathway for cytokines influencing seizures and epilepsy,⁸ the exact role of cytokines in human epilepsy is complex and not yet completely elucidated. Moreover, whether there are unique aspects of the pathophysiology of seizure onset in these patients which make them more susceptible to sodium channel blockage is not known. The observed response could also be due to the effectiveness of sodium channel blockers in focal onset epilepsy which is seen in 66% of patients in our cohort. Of note, while AED responders responded to sodium channel blockers, no patient responded to lamotrigine, which also has sodium channel properties.⁹ Therefore, if there is a unique responsivity of seizures in these patients to this class of medications, the response may not be uniform throughout the entire class. Studies are needed to elucidate whether any differential efficacy is based on their effect on seizure mechanisms or immunomodulatory properties.

Regardless of the type of therapy, VGKC-complex antibody–positive patients were more likely to become seizure free compared with GAD65-positive and autoimmune antibody–negative cases (figure 2). This is consistent with previous observations of superior immunotherapy response in patients with neural surface–related autoantibody-mediated encephalitides as compared to those associated with intracellular antigen autoantibodies.^1,7,10 Of note, among the patients who responded to AEDs in our cohort, the majority were VGKC-complex antibody positive (1 LGI1 positive; 1 CASPR2 positive, and 2 double negative). Two of these responded to oxcarbazepine, one to lacosamide, and the other to carbamazepine (figure 1). Indeed, sodium-channel blocking agents such as carbamazepine and phenytoin have been successfully used to treat autoimmune VGK channelopathies such as neuromyotonia¹¹ and episodic ataxia type 1.¹² These agents act by reducing neuronal repetitive firing through interaction with voltage-gated sodium channels.^11,12 Perhaps, the observed favorable response of these agents in VGKC-complex ab–associated AE is attributable to this mechanism. In our series, one patient with GAD65 ab–associated AE became seizure free with carbamazepine after failing immunotherapy. This is in contrast to a similar report that found lack of AED response in patients with GAD65 antibody–positive AE following which it was postulated that GAD65 ab–associated AEs are intractable to AEDs.¹³ Regardless of the type of therapy, 4/9 (44.4%) autoantibody-negative patients become seizure free, 2 with AEDs (one with lacosamide and the other with carbamazepine after failing immunotherapy), while the other two responded to immunotherapy. Others have noted the potential for response in patients with seronegative AE to antiepileptic medication and immunotherapy.¹⁴

The rationale for selecting one AED over the other was not clarified in the EMR of our cohort. Levetiracetam was the most common used drug (42/50, figure 2); however, none of the patients in our series were rendered seizure free by this medication (0%). This finding is in contrast to previous reports suggesting that levetiracetam may have, at least in part, anti-inflammatory properties.³ The reason for the prevalent use of levetiracetam in our series is likely due to its ease of dosing and lack of drug-drug interactions.⁹ Certainly, AEDs with enzyme induction properties such as carbamazepine and phenytoin could alter the pharmacokinetics of immunosuppressive therapies and other agents used in these patients.⁹ Perhaps, newer AEDs with sodium channel blocking properties and more favorable pharmacokinetic profiles (such as oxcarbazepine and lacosamide) could be considered in this patient population. It is also imperative to monitor for hyponatremia (e.g., oxcarbazepine and carbamazepine) and allergic cutaneous reactions (e.g., carbamazepine and phenytoin), as these have been shown to occur at high rate in patients with VGKC-complex antibody.¹⁵ Prospective studies are needed to clarify AED selection and to elucidate their immunomodulatory properties in AE.

There are several limitations to our study. First, data collection for this retrospective study was based on the extraction of relevant patient information from EMRs. The lack of control group and the possible confounder of concomitant immunotherapy should also be acknowledged. In addition, selection bias could be present due to the tertiary nature of these patients. Finally, seizure quantification is only based on verbal report, thus recall bias could result in the event of amnestic seizures, and related to concurrent cognitive dysfunction that may have been present in a significant proportion of our cohort. Prospective studies are needed to clarify whether there is an optimal AED selection for AE. As the awareness of AE increases, finding effective strategies for seizure control is crucial.