Immunotherapy for Management of Refractory and Super-Refractory Status Epilepticus

Status epilepticus (SE) is a life-threatening condition that affects 18.3 to 41 out of 100,000 people in the United States annually.¹ The International League Against Epilepsy defines SE as a condition resulting from either a failure to terminate seizures or mechanisms that prolong them.² SE can result in long-term consequences secondary to neuronal death, neuronal injury, and alterations of neuronal networks. Seizure activity that continues despite 2 appropriately selected and dosed parenteral medications is called refractory SE (RSE). SE that persists after at least 24 hours of anesthesia, recurs on appropriate anesthetic treatment, or recurs after withdrawal of anesthesia requiring reinitiation is termed super-refractory SE (SRSE).³

Several conditions of interest fall within the RSE classification (Table 1). New-onset refractory SE (NORSE) arises in individuals without known epilepsy or other preexisting neurologic condition and is defined as RSE without a clear structural, toxic, or metabolic cause.³ Whereas some cases of NORSE result from defined immune causes such as autoimmune encephalitis (eg, anti-NMDA receptor [NMDAR]–associated autoimmune encephalitis), most cases remain unexplained (cryptogenic NORSE [cNORSE]).⁴ Febrile infection-related epilepsy syndrome (FIRES), a subcategory of NORSE, occurs in individuals who experienced a fever 2 weeks to 24 hours before onset of RSE.³ There is considerable overlap among these conditions, which is demonstrated in Figure 1.

Progression of SE to RSE is seen in 22% to 43% of cases,^5-9 and to SRSE in 14% to 15% of cases.^10,11 In-hospital mortality rates in RSE and SRSE are variable but typically range between 22% and 31% in RSE^10,12-16 and 30% to 50% in SRSE.¹¹ As such, prompt recognition and therapeutic intervention is crucial.

There is growing interest in the role of neuroinflammation in initiation and propagation of SE, but randomized clinical trial data for the use of immunomodulatory agents in RSE and SRSE remain limited. This review presents the current understanding of neuroinflammation in RSE and the role of therapeutic immunomodulation in its management.

Figure 1. Clinical overlap between syndromes causing status epilepticus.

Neuroinflammation in RSE
A wide array of neuroinflammatory cascades and processes have been implicated in RSE.^17,18 In preclinical models, neuroinflammation has been shown to promote the development and recurrence of seizures. As summarized by Vezzani et al,¹⁹ innate immune signaling pathways involving the interleukin-1 receptor, type 1 (IL-1R1) and Toll-like receptor 4 (TLR4) are central to the initiation of seizure-associated neuroinflammation (Figure 2). Hippocampal injection of their endogenous ligands, interleukin-1β (IL-1β) and high mobility group box 1 (HMGB1), leads to an increase in acute seizure activity in rodents.²⁰ This is further supported by studies demonstrating that inhibiting this pathway results in seizure reduction.^21-24 Collectively, this evidence indicates that activation of IL-1R1 and TLR4 receptors may play a role in initiating seizure activity.

Elevated levels of proinflammatory cytokines (IL-1β, interleukin 6 [IL-6], tumor necrosis factor α [TNF-α], and C-X-C motif chemokine ligand 8 [CXCL8]) are seen in the serum and cerebrospinal fluid (CSF) of individuals with SE or RSE.^4,25-27 This remains true even when compared with individuals with chronic epilepsy with daily seizures.²⁸ Evidence has suggested that elevations in these cytokines or chemokines are mediated by HMGB1 and damage-associated molecular patterns release in the setting of microglial and astrocyte activation as well as neuronal damage.^25,26,29 Seizures may also directly activate microglia and astrocytes, leading to proinflammatory cytokine release.³⁰ These inflammatory mediators may contribute to blood–brain barrier (BBB) breakdown, leading to increased permeability and leukocyte infiltration.³¹ Complement system activation—particularly of the C1q-C3 signaling pathway—and activation of the inflammasome—involved in the innate immune response—have also been observed in SE.^25,32

Figure 2. Pathophysiologic cascade mediated by IL-1R/TLR signaling in status epilepticus that leads to neuroinflammation and the potential for targeted immunotherapy.
Abbreviations: HMGBI, high-mobility group box 1 protein; IL-1R, interleukin-1 receptor; IL-1β, interleukin-1 beta; TLR, toll-like receptor.
This figure was developed using ChatGPT 5 (OpenAI, San Francisco, CA).

Because most NORSE cases remain cryptogenic, there is a growing interest in understanding the neuroinflammatory cascades involved, as this may be shared among cases. It is hypothesized that NORSE results from a postinfectious immunologic process leading to cerebral inflammation.²⁹ Higher cytokine/chemokine levels in individuals with NORSE are associated with worse outcomes.^29,33 In a recent study, individuals with cryptogenic NORSE were shown to have elevated serum levels of CXCL8, macrophage inflammatory protein-1α (MIP-1α), and monocyte chemoattractant protein-1 (MCP-1) compared with individuals with RSE, suggesting a role for the innate immune system in NORSE pathogenesis.²⁹ This was supported by a multiproteomic study demonstrating an upregulation of immune and lymphocyte-mediated pathways in cNORSE vs anti-NMDA encephalitis.³³ Similarly, polymorphisms in cytokine-related genes have been identified in individuals with cryptogenic FIRES.³⁴ Most recently, a study comparing serum cytokine profiles in individuals with cNORSE has suggested 3 distinct subtypes: cluster A, lacking specific inflammatory markers; cluster B, with an innate-immunity cytokine-driven inflammatory response; and cluster C, with dysregulated autoimmune processes.³⁵ This heterogeneity may contribute to the variable response to immunomodulatory treatments reported in individuals with cNORSE.

It remains unclear whether neuroinflammation is predominantly a cause or effect of RSE. Recent data have suggested that neuroinflammation can induce neuronal hyperexcitability and that SE propagates these inflammatory pathways, activating a “feed-forward” cycle. It is hypothesized that the precise relationship varies with underlying etiology, although further investigation is warranted.³⁶

Immunotherapies

First-Line Immunotherapy
Initiation of first-line immunomodulatory agents is recommended within 72 hours of SE onset.³⁷ Typical first-line agents include corticosteroids, intravenous immunoglobulin (IVIg), or plasma exchange (PLEX). There are no randomized controlled clinical trials comparing these therapies in a head-to-head manner. In clinical practice, initiation of corticosteroids and IVIg is common while the diagnostic workup is underway, although efficacy rates vary widely. The broad mechanisms of action of corticosteroids and IVIg in combination with their relative safety profiles are often a primary justification for early use.

Response rates to intravenous corticosteroids have been reported to be between 30% and 40%.^38,39 In systematic reviews focused on the efficacy of IVIg in RSE, response rates of 45% and 16% have been reported in the adult and pediatric literature, respectively.^40,41 PLEX efficacy was reported as 52% and 24% in the adult and pediatric literature, respectively.^40,42 One study focusing on pediatric FIRES demonstrated improved response to IVIg with corticosteroids compared with IVIg alone.⁴³ Overall, there is no clear evidence for choice of one first-line immunotherapy over the others in RSE, nor is there generalizable evidence for combination therapy.

Second-Line Immunotherapy
Anakinra. Anakinra (Kineret; Sobi, Waltham, MA), a recombinant version of the endogenous human interleukin-1 receptor antagonist (IL-1RA), inhibits the actions of IL-1a and IL-1b and has been shown to cross the BBB in areas of breakdown.^44,45 Most published data detailing the use of anakinra in RSE focus on cases of NORSE, including several case studies, case series, and 1 larger international retrospective cohort study. Studies have been predominantly focused on pediatric participants with cryptogenic FIRES, often with objective evidence of neuroinflammation.

The use of anakinra as second-line immunotherapy for NORSE has been increasing.²⁹ In responders, clinical and EEG improvement is typically seen within the first week after administration. Optimal timing for anakinra administration remains unclear, as only some studies have found improved efficacy with early administration. Among published cases, 9 of 12 individuals with NORSE were reported to obtain partial or complete seizure control with the initiation of anakinra in the acute phase.⁴ Anakinra was initiated at a median of 30 days, and doses varied greatly, from 2 to 20 mg/kg/d.4 Notably, response to anakinra in the acute phase did not necessarily lead to better long-term outcomes. The overall impact of anakinra on long-term outcomes remains unclear given the lack of controlled trials. 

In an international retrospective cohort study, 25 children (aged 4 to 16 years) diagnosed with FIRES were examined. All participants received antiseizure medications (ASMs) and anesthetic agents, with 23 out of 25 (92%) given first-line immunotherapy before anakinra administration.¹⁵ Initial and final median dosing of anakinra was 3.8 mg/kg/d and 5 mg/kg/d, respectively, administered a median of 20 days after seizure onset. Electrographic and electroclinical seizure frequency was compared immediately before and 1 week after anakinra administration in 15 children, demonstrating a >50% seizure reduction in 11 participants (73%). Early anakinra initiation was associated with shorter durations of mechanical ventilation, hospital stays, and time in the intensive care unit, although no clear improvement in long-term outcomes was observed. Ten participants developed infections after treatment; however, 9 had infections before anakinra initiation, confounding attribution. Three participants developed drug reactions with eosinophilia and systemic symptoms (DRESS) syndrome, and 2 participants developed cytopenias that resolved without intervention. 

Tocilizumab. Tocilizumab is a recombinant humanized monoclonal antibody against the IL-6 receptor. Although its central nervous system penetration is poor, tocilizumab may play a role in managing systemic inflammation as well as neuroinflammation when the BBB is disrupted.⁴⁶ Most published studies focus on the use of tocilizumab in NORSE, but one prospective randomized controlled study on its use in RSE exists. El-Haggar et al⁴⁷ reported a cohort of 60 adult or teenage participants with RSE who were randomized to receive standard of care treatment vs standard treatment with the addition of 4 mg/kg of tocilizumab administered twice monthly for 3 months. Participants treated with tocilizumab had a statistically significantly greater decrease in modified Status Epilepticus Severity Scores compared with the control group at 12 weeks. They also had a significant decrease in serum IL-6, TNF-α, and nuclear factor κB (NF-κB) levels compared with participants who were not receiving tocilizumab.

In a recent meta-analysis, Hanin et al⁴ reviewed 11 case reports and case series describing the use of tocilizumab in NORSE. A total of 20 individuals, including 14 adults and 6 children (median age, 22.5 years), were included. Among these, 2 individuals were diagnosed with anti-NMDAR encephalitis, 1 with anti-glutamic acid decarboxylase 65 (anti-GAD65) encephalitis, and the remainder with cNORSE. The most frequent dosing regimen was 4 mg/kg weekly for 2 weeks followed by 8 mg/kg at 1 month if needed, although alternative dosing regimens did not show significant differences in response rates. The first administration of tocilizumab in the acute phase of NORSE resulted in rapid partial or complete seizure control in most individuals (14 of 20 [70%]),4 with a median response time of 4 days reported in several studies.^48,49 Most individuals treated had elevated CSF IL-6 levels. Although the administration of tocilizumab did not consistently improve long-term outcomes, 1 study found that those with good or fair functional outcomes received treatment earlier in the disease course.⁴⁹ Safety data were reported in 13 individuals, revealing severe side effects in 9, including neutropenia, leukopenia, pneumonia, sepsis, and multidrug-resistant pathogen infection. 

Intrathecal Dexamethasone. Intrathecal dexamethasone (IT-DEX) was initially proposed as a treatment for FIRES in 2021.⁵⁰ This study involved the administration of IT-DEX to 6 children who had already received first-line immunotherapy. A dosage of between 0.15 and 0.25 mg/kg/d was administered for 4 to 8 cycles. All individuals had improvement in their EEG background and were withdrawn from continuous ASM infusions after a median of 5.5 days. Half of the individuals (3 of 6) demonstrated a >50% reduction in seizure frequency with none reporting significant adverse events.⁵⁰

The more recent international multicenter study enrolled 12 individuals (2 adults, 10 children) with FIRES.⁵¹ Individuals were treated at a median of 20 days after RSE onset with a median dose of 5 mg (0.21 mg/kg/dose). There was a clinician-perceived improvement in seizure control in 83% of cases. Individuals were able to wean off continuous infusions at a median of 5 days from IT-DEX administration. No severe adverse events were reported. 

Individuals in both studies demonstrated continued elevations in proinflammatory cytokines despite clinical improvement. This suggests a possible cytokine-independent mechanism of IT-DEX. Long-term data on IT-DEX for RSE are limited, although most individuals seem to have persistent post-NORSE epilepsy. 

Rituximab. Rituximab, a chimeric anti-CD20 monoclonal antibody, works through a variety of direct and indirect mechanisms including complement-mediated cytotoxicity and antibody-dependent cell-mediated cytotoxicity.⁵² Rituximab has been shown to be an efficacious second-line immunotherapy for RSE, particularly in the setting of autoimmune causes. In one prospective observational study it was found that 20 of 41 individuals with autoimmune encephalitis had inadequate seizure control with first-line immunotherapy.⁵³ A total of 13 of these individuals were treated with rituximab (8 with anti-NMDAR, 3 with LGI-1, and 1 with anti-Ma2/Ta) at a dose of 375 mg/m² weekly for 4 weeks, followed by monthly dosing if needed. Eight of 12 individuals (66.6%) treated gained seizure freedom at 6 months, and 1 had significant seizure reduction.⁵³

Data detailing the use of rituximab for other causes of RSE are limited. A recent study looking at use of second-line immunotherapy in cNORSE showed that 25% of individuals received rituximab as a part of their regimen.⁴ Unlike anakinra and tocilizumab, which have dedicated biomarkers that may prompt utilization, use of rituximab can be justified through the presence of an elevated IgG index or the presence of restricted oligoclonal banding in the CSF, which would indicate intrathecal synthesis of potential unidentified autoantibodies. Safety data for this indication are also limited, although a meta-analysis looking at the rate of adverse events in rituximab use for autoimmune encephalitis reported infusion-related reactions in 15.7% of individuals, pneumonia in 6%, and severe sepsis in 1.1%.⁵⁴

Exploratory Immunotherapies. Investigation into alternative immunotherapeutic agents for RSE is ongoing. Current evidence remains limited to case studies, preclinical models, or select underlying etiologies. Agents that have demonstrated potential in these limited studies include adalimumab, minocycline, natalizumab (Tysabri; Biogen, Cambridge, MA), and N-acetylcysteine.¹⁹ Therapies targeting the mTOR pathway, such as everolimus and sirolimus, may be efficacious in individuals with known pertinent genetic conditions, such as tuberous sclerosis.⁵⁵ Given the recognized contributions of the innate immune system to RSE and NORSE, Janus kinase inhibitors are also emerging as immunotherapeutics of interest, with data limited to preclinical studies.⁵⁶ Additional research is needed to justify the use of these agents in RSE and their consideration is only warranted in extreme cases.

Figure 3. Flow diagram for proposed medical management of status epilepticus, refractory status epilepticus, and super-refractory status epilepticus with immunotherapy.
*Available evidence is based on case reports and small case series (Class IV evidence).
Abbreviations: ASM, antiseizure medication; IT, intrathecal; IV, intravenous; IVIg, intravenous immunoglobulin; PLEX, plasmapheresis.

Clinical Approach
Individuals with RSE are often treated with immunomodulatory agents given presumed inflammatory contributions. After initial ASM and anesthetic trials, first-line immunomodulatory therapy should commence within 72 hours of SE onset, based on international consensus.⁵⁷ These first-line agents include corticosteroids, IVIg, or PLEX. If there is no improvement, a second-line immunotherapeutic agent (Table 2), ketogenic diet, or both should be started within the first week.^57,58 The commonly used second-line immunotherapeutic agents are anakinra, tocilizumab, IT-DEX, and rituximab (Figure 3). Standard of care treatment with ASM and anesthetics should occur in conjunction with immunotherapeutic interventions. 

No randomized clinical trials exist comparing first-line immunotherapies for RSE, and efficacy data are highly variable. The same challenge exists when comparing second-line immunotherapeutics, although rituximab may be more efficacious when an antibody-mediated etiology is suspected. Laboratory indicators of this include the presence of a pathologic autoantibody in the CSF or serum deemed clinically significant, an elevated IgG index in the CSF, or restricted oligoclonal banding in the CSF. Many clinicians use serum and CSF cytokine profiles to guide second-line immunotherapy choice, although the relationship between response rates and elevated target cytokine/chemokine level remains controversial. Elevations of IL-1 and IL-6 may represent targetable avenues for anakinra and tocilizumab, respectively, although further study is needed. Whereas these agents seem to improve seizure control and ability to wean off continuous ASM or anesthetic infusions in the acute period, their impact on long-term outcomes remains unclear. 

There is a growing body of evidence describing the role of inflammation in the initiation and propagation of SE. This raises the question of whether earlier initiation of immunomodulatory therapies in these individuals is warranted. Some studies have suggested that earlier intervention with second-line immunomodulatory agents confers better functional outcomes, although there remains a dearth of data. Multicenter controlled clinical trials are needed to help guide the choice of immunotherapeutic intervention, optimal timing, and dosing, as well as to elucidate the long-term effects of these agents on functional outcomes.