IL-1β is one of the most extensively studied pro-inflammatory cytokines in the central nervous system. The expression of IL-1β and its receptor, IL-1R1, is significantly upregulated in postoperative focal brain tissues of epilepsy patients and in various animal models of epilepsy (Lee et al., 2020b; Rana and Musto, 2018). IL-1β is predominantly released by activated microglia, astrocytes, and damaged neurons, and it activates the downstream MyD88/NF-κB signaling pathway through the IL-1R1 receptor, inducing the expression of numerous inflammatory genes (Jiang et al., 2023; Lee et al., 2020b; Rana and Musto, 2018). In the context of epilepsy, IL-1β exerts several pro-seizure effects:

Reduction of Inhibitory GABAergic Transmission: Electrophysiological studies have shown that IL-1β exposure reduces the amplitude of GABA_A receptor-mediated currents in hippocampal neurons by approximately 30%, suggesting that IL-1β attenuates postsynaptic inhibitory potentiation. Roseti et al. observed that in tissue slices of human temporal lobe epileptic foci, the addition of IL-1β markedly reduced GABA_A receptor-mediated currents, and the use of an IL-1 receptor antagonist (IL-1Ra) blocked this effect. A reduction in inhibitory transmission disrupts the excitatory/inhibitory balance, increasing the likelihood of overexcitation of the neural network (Horn et al., 2022; Sanz and Garcia-Gimeno, 2020; Kamali et al., 2021).

Enhancement of Excitatory Glutamatergic Transmission: IL-1β promotes the inactivation of the intracellular glutamine synthetase (GS) in astrocytes, impairing the clearance and recycling of synaptic glutamate and leading to its accumulation in the extracellular space. Additionally, IL-1β acts directly on presynaptic terminals, increasing the probability of glutamate release. The resulting excess glutamate causes continuous neuronal depolarization, making neurons more susceptible to synchronous discharges (Sanz and Garcia-Gimeno, 2020; Kamali et al., 2021; Phan et al., 2022).

Mediation of Other Epileptogenic Factors: IL-1β stimulates glial cells to express cyclooxygenase-2 (COX-2) and mPGES-1 enzymes, which upregulate the synthesis of prostaglandin E2 (PGE2). PGE2 acts on EP receptors located on neuronal membranes, inhibiting Na⁺/K⁺-ATPase activity and thereby causing membrane depolarization and increased neuronal excitability. IL-1β also promotes the release of “danger signals” such as HMGB1 from damaged cells, further amplifying the inflammatory response via the TLR4 pathway (discussed later in the mechanism of HMGB1) (Phan et al., 2022; Radu et al., 2017;Vezzani et al., 2019).

Inducing Neuronal Changes: In animal studies, direct injection of IL-1β into the rat brain induced epileptiform discharges, whereas injection of IL-1Ra increased the convulsion threshold and reduced seizure frequency. Vezzani et al. demonstrated that sustained osmotic pumping of human recombinant IL-1Ra protein into the hippocampus, or overexpression of IL-1Ra in transgenic mice, significantly inhibited spontaneous seizures in epileptogenic states. In contrast, IL-1R1 knockout mice exhibited elevated seizure thresholds and reduced seizure severity in experimental epilepsy models. These findings strongly suggest that IL-1β/IL-1R1 signaling contributes to seizure onset, establishing IL-1β as one of the key mediators of epilepsy-associated neuroinflammation (Radu et al., 2017; Vezzani et al., 2019; Soltani Khaboushan et al., 2022; Yue et al., 2022).

IL-6 is another pro-inflammatory cytokine closely related to epilepsy and is widely present in central and peripheral immune responses. Under normal physiological conditions, IL-6 expression in the brain is extremely low; however, IL-6 levels are significantly elevated in cerebrospinal fluid and serum of patients with epilepsy. Similar increases are observed in the hippocampus, cortex, and other brain regions of animals modelels, where levels can remain elevated for up to 24 h after status epilepticus. IL-6 is secreted predominantly by activated astrocytes and microglial cells, and its production is regulated by other cytokines, such as TNF-α, IL-1β, and IL-17, which can induce glial cells to synthesize IL-6 (Phan et al., 2022; Radu et al., 2017; Vezzani et al., 2019).

The role of IL-6 in epilepsy is more complex, but overall its pro-convulsant effect is dominant:

IL-17 is an epilepsy-related cytokine that has gained attention in recent years. It belongs to the pro-inflammatory mediators secreted by Th17 cells. The IL-17 family consists of isoforms ranging from IL-17A to F, with IL-17A (often abbreviated as IL-17) being the most active. Th17 cells, which play an important role in intestinal and autoimmune diseases, produce IL-17 that has been found to be significantly elevated in both epilepsy patients and experimental models. Interestingly, peripherally infiltrating Th17 cells are not the only source of IL-17 in the CNS; some glial cells and neurons also express IL-17 in epileptic pathology (Sharma et al., 2025; Soltani Khaboushan et al., 2022; Choi et al., 2021).

The effects of IL-17 on epilepsy are primarily mediated by disrupting the blood-brain barrier and triggering persistent inflammation:

TNF-α is a classical pro-inflammatory factor secreted by macrophages, microglia, and is equally important in epilepsy pathology. In patient brain tissue and epilepsy models, upregulation of TNF-α expression often coexists with epileptic activity and neuronal injury. TNF-α mediates its effects through its receptors TNFR1 and TNFR2. TNFR1 is widely expressed and associated with pro-inflammatory effects, whereas TNFR2 is expressed mainly on immune cells and is linked to pro-survival and regenerative functions (Kamali et al., 2021; Shan et al., 2022; Soltani Khaboushan et al., 2022). The mechanisms of action of TNF-α in epileptic conditions are complex and varied.

HMGB1 is an intranuclear DNA-binding protein that is not normally secreted; however, in response to inflammation or cellular injury, HMGB1 can be released passively into the extracellular space or actively secreted by activated microglia, functioning as a typical Danger-Associated Molecular Patterns (DAMP) (Binabaj et al., 2019). Vezzani et al. found that HMGB1 is present in abundantly present in human and rodent epileptic focal brain tissue, and that its activation by binding to the TLR4 receptor plays a critical role in seizure generation. HMGB1-TLR4 signaling induces microglia and astrocytes to produce a cascade of inflammatory mediators, including IL-1β, TNF-α, and upregulate the production of inflammatory factors on neuronal membranes. And upregulates NMDA receptor function on neuronal membranes, increasing Ca²⁺ inward flow, leading to neuronal hyperexcitability and delayed neuronal death (Pasheva et al., 1998; Phan et al., 2022). HMGB1 has also been shown to disrupt the tight junctions of the blood-brain barrier and to induce epileptiform discharges. Moreover, experimental epilepsy models demonstrated that blockade of either HMGB1 or genetic deletion of TLR4 using neutralizing antibodies significantly reduced seizure incidence, thereby validating the pathogenic role of the HMGB1/TLR4 pathway in epilepsy (Zwilling et al., 1995; Schoenmakers and Van de Ven, 1997; Xiao et al., 2011; Ramstein et al., 1999).

Another pathway of interest is transforming growth factor-β (TGF-β) signaling. In the context of blood–brain barrier (BBB) damage, plasma albumin that enters the brain can activate TGF-β receptor I (ALK5) on astrocytes, triggering the SMAD cascade and leading to functional changes in astrocytes (Kamali et al., 2021). Specifically, under the TGF-β/ALK5 signaling pathway, the expression of Kir4.1, an inwardly rectifying potassium channel, was downregulated in the astrocyte membrane, and the polarized distribution of aquaporin 4 (AQP4) was impaired. Reduced Kir4.1 expression implies poorer potassium buffering capacity and easy accumulation of extracellular K⁺, which results in sustained excitability; and dysfunction of AQP4 affects the neurovascular unit homeostasis. Both are thought to be part of the epileptogenic effect of albumin. Thus, glial cell pathologic alterations (e.g., reactive proliferation but dysfunction of astrocytes) commonly seen in chronic epileptic foci are also associated with the involvement of inflammatory mediators (Yan et al., 2025; Hou et al., 2024).

In addition to the major inflammatory factors mentioned above, several other mediators are also implicated in epilepsy. For example, nitric oxide (NO) is a gaseous signaling molecule synthesized in response to inflammation and produced by inducible nitric oxide synthase (iNOS) (Langeh and Singh, 2021; Garg et al., 2023). After seizures, activated microglia produce large amounts of NO, which activates the cyclic guanosine monophosphate (cGMP) pathway in bystander neurons and exacerbates neuronal excitability (Kawakami et al., 2021; Kalati et al., 2022; Akyuz et al., 2021). Chemokines such as CCL2 and CXCL10 are significantly elevated in the epileptic focal brain, driving more peripheral immune cells (e.g., CCR2⁺ monocytes) into the center and amplifying the inflammatory cascade. Prostaglandin E2 (PGE2), previously mentioned, synthesized via COX-2 and mPGES-1, is also elevated during seizures and inhibits neuronal membrane pumps thereby triggering neuronal hyperexcitability. In addition, activation of the complement system has also been reported in epilepsy: C1q, C3, and other complement components accumulate in the brain of epileptic patients, and microglia express the complement receptor CR3, suggesting that complement may mediate synaptic phagocytosis and contribute to neuronal injury (Akyuz et al., 2021; Javed et al., 2022; Rahimi et al., 2022). However, the role of the complement system appears to be dual, removing necrotic debris on the one hand and damaging healthy tissue on the other. In summary, multiple inflammatory mediators collectively create a complex pro-inflammatory microenvironment within epileptic foci (as shown in Figure 1). These mediators interact and promote each other, leading to a sustained increase in neural network excitability and a heightened propensity for dysregulated activity (Javed et al., 2022; Rahimi et al., 2022; Vega Rasgado et al., 2023).

Building on the growing understanding of the role of pro-inflammatory mediators, attempts have been made in recent years to target these inflammatory pathways for intervention with a view to achieving antiepileptic effects. For example, IL-1 receptor antagonists that block IL-1β/IL-1R signaling (e.g., Anakinra) have achieved efficacy in refractory epilepsy with febrile infection-related status epilepticus (FIRES), suggesting the feasibility of this strategy. As another example, neutralizing antibodies targeting HMGB1 effectively reduced spontaneous seizures in animal models (Chao et al., 1994; Almostafa et al., 2024). Collectively, these findings underscore both the pathogenic importance and therapeutic potential of inflammatory mediators in epilepsy. Nevertheless, further clinical investigations are required to establish the safety and efficacy of such anti-inflammatory interventions in broader patient populations (Kawakami et al., 2021; Kalati et al., 2022; Akyuz et al., 2021; Javed et al., 2022).

Microglia, the resident macrophages of the central nervous system (CNS), account for approximately 5%–20% of the glial cell population in the brain (Hu et al., 2024). Originating from yolk sac hematopoietic progenitors, they perform essential myeloid functions, including phagocytic clearance, antigen presentation, and the secretion of inflammatory mediators (Tagliatti et al., 2024; Tröscher et al., 2023). In their resting state, microglia continuously survey the surrounding microenvironment through their dendritic extensions and can rapidly shift to an activated state in response to stimuli such as injury or abnormal discharges (Abbott et al., 2010; Kumar et al., 2022; Su et al., 2022). Seizure activity is a potent activator of microglia: in both human epileptic foci and animal models, they undergo a morphological transformation from a ramified, surveillant form to a hypertrophic, amoeboid “phagocytic” phenotype, accompanied by increased expression of activation markers such as Iba-1 and CD68.

Microglia activation has a dual effect on epilepsy. On one hand, excessive and sustained M1-type microglial responses exacerbate neuroinflammation and induce neuronal hyperexcitability. Studies have shown that activated microglia release large amounts of IL-1β, TNF-α, complement components, and other mediators, directly lowering seizure thresholds and contribute to neuronal damage (Kumar et al., 2022; Su et al., 2022; Wang et al., 2021a; Sanz and Garcia-Gimeno, 2020). Additionally, microglia interact with astrocytes, where inflammatory factors released by microglia stimulate the release of glutamate from surrounding astrocytes, increasing neuronal excitability and leading to excitotoxic cell death. On the one hand, microglial activation may also exert protective functions. For example, microglial proliferation during the early phase of status epilepticus facilitates clearance of cellular debris and helps limit acute injury. However, persistent microglial proliferation in the chronic phase may perpetuate neuroinflammation and promote recurrent seizures.

On the other hand, microglia also exert beneficial effects (Kiani Shabestari et al., 2022; Li et al., 2024; Massey et al., 2023). For example, they phagocytose and remove apoptotic neurons, thereby preventing the formation of aberrant neural circuits. In addition, microglia have been reported to suppress the excessive proliferation of hippocampal dentate gyrus granule cells through the TLR9–TNF pathway during the remodeling phase following epileptogenesis (Meng et al., 2023; Kiani Shabestari et al., 2022). This process reduces the generation of ectopic neurons and limits aberrant circuit formation, contributing to a reduction in seizures. Conversely, some findings suggest that inhibiting microglial activation increases the number of ectopic newborn neurons in the hippocampus after sustained epilepsy, suggesting that microglia may also contribute to pathological neurogenesis under certain conditions. These paradoxical findings may be explained by differences in epilepsy models, seizure stages, and microenvironmental factors (Zhong et al., 2022; Zhang et al., 2024b; Chen et al., 2023). The current consensus is that microglia act as “inflammatory regulators” in epilepsy. Excessive M1-type pro-inflammatory responses increasing seizure susceptibility, whereas moderate M2-type responses or clearance functions promote tissue repair and seizure suppression. Thus, microglial dysfunction—whether due to excessive inflammatory activation or impaired regulatory activity—may contribute to epileptogenesis. Modulation microglial function, for example by promoting a shift toward an anti-inflammatory phenotype or blocking specific pro-inflammatory receptors (e.g., P2X7 receptors, TLRs), has shown promise in reducing seizures in animal models. A deeper understanding of microglial roles at different stages of epilepsy will be essential for the development of relevant immunomodulatory therapies in the future (Su et al., 2022; Wang et al., 2021a; Sanz and Garcia-Gimeno, 2020).

Astrocytes are the most abundant glial cells in the central nervous system (CNS) and play essential roles in maintaining brain homeostasis, including supporting the blood-brain barrier, regulation extracellular ion concentrations, clearing excess neurotransmitters, and providing metabolic substrates (Langeh and Singh, 2021; Su et al., 2022). In epilepsy, astrocytes undergo significantly pathological alterations. Proliferation and hypertrophy of astrocytes (i.e., gliosis, gliosis) are commonly observed in epileptic foci, representing a reactive response of brain tissue to injury or abnormal excitation (Su et al., 2022; Sanz and Garcia-Gimeno, 2020; Miller et al., 2023). Activated astrocytes can release a variety of cytokines and neurotrophic factors, and they also regulate neuronal excitability through phagocytosis and uptake of glutamate and K+. Dysregulated astrocyte function in epilepsy is primarily manifested in the following aspects:

Under normal physiological conditions, the blood - brain barrier (BBB) restricts the entry of peripheral immune cells into the brain parenchyma. However, in epileptic pathological states, particularly during status epilepticus or after severe trauma, the disruption of the BBB allows the migration of numerous peripheral inflammatory cells into the brain. Monocytes-macrophages and neutrophils are the most common types of infiltrating cells, and there have also been reports of T - cell infiltration (Hu et al., 2024; Ding et al., 2022; Tröscher et al., 2023).

In status epilepticus models, numerous CCR2⁺ monocytes have been observed to exit the bloodstream and infiltrate injured brain regions such as the hippocampus, where they differentiate into macrophages, with phagocytic activity and elevated expression of pro - inflammatory genes. Demonstrated that in status epilepticus models, CCR2 receptor knockout mice—lacking the receptor essential for monocyte recruitment—exhibited significantly reduced neuronal damage and cognitive deficits (Hendrix et al., 2024; Kamali et al., 2021; Phan et al., 2022). This demonstrates that brain - infiltrating monocytes/macrophages aggravate epilepsy - related brain injury. These peripherally derived macrophages share many functional similarities with central microglia and can release pro - inflammatory factors such as IL - 1β and TNF-α, exacerbating the inflammatory environment within the brain (Huppert et al., 2010; Sun et al., 2022; Vezzani et al., 2022). Moreover, they can engulf dendrites and synapses of surviving neurons, disrupting the normal architecture of neural circuits. In human epileptic foci (e.g., cortical dysplasia), a significant accumulation of CD68 - positive macrophages has been detected. These cells are likely peripherally derived macrophages that have infiltrated and, together with resident microglia, participate in the inflammatory response of the epileptic focus. Therefore, preventing abnormal infiltration of peripheral monocytes into the central nervous system or limiting their detrimental effects may provide neuroprotective and anti - epileptic benefits. For instance, studies are currently exploring the use of CCR2 antagonists or strategies to block monocyte entry into the central nervous system to alleviate post - epileptic sequelae and seizure susceptibility (Choi et al., 2021; Que et al., 2024; Spitzer et al., 2023; Chen et al., 2026).

Neutrophils are another type of immune cell that are rapidly mobilized during acute epilepsy - related inflammation. Although neutrophils are primarily active in the peripheral innate immune system, they can also infiltrate the brain during severe epileptic seizures. A research has shown that epileptic seizures can cause adhesion and accumulation of neutrophils in cerebral capillaries, leading to impaired local microcirculation perfusion. This mechanism may help explain the post-ictal hypoperfusion commonly observed in patients following seizures (Daneman et al., 2010; Abbott and Friedman, 2012; Sanz et al., 2024). Additionally, clinical studies have demonstrated that the neutrophil - to - lymphocyte ratio (NLR) in the peripheral blood of epilepsy patients is frequently elevated, reflecting the systemic inflammatory response and stress levels associated with seizures. A systematic review further confirmed that independent studies consistently reported significantly higher NLR values in epilepsy patients compared to healthy controls, with the degree of elevation correlating with seizure frequency and poor prognosis (Yue et al., 2022; Hu et al., 2023; Lagarde et al., 2022). This indirectly supports the involvement of neutrophils in epilepsy - related systemic inflammation.

Neutrophils may contribute to cerebralvascular and tissue injury through mechanisms such as the release of reactive oxygen species (ROS), secretion of proteases, and the formation of extracellular traps (NETs), thereby exacerbating epilepsy seizure susceptibility. Although the precise role of neutrophils in epilepsy has not yet been fully elucidated their status as key indicators of systemic inflammation highlights the need for further investigation (Lee et al., 2020a; Hu et al., 2024; Hermann et al., 2017). Evidence for colchicine in epilepsy is limited to case reports in familial Mediterranean fever (FMF), where seizure control improved after initiating colchicine; epilepsy-specific randomized or prospective trials are lacking (Parvez et al., 2015). Mechanistically, colchicine can reduce neutrophil recruitment/NET formation, providing a rationale for anti-inflammatory targeting (Leung et al., 2015; Vaidya et al., 2020). If proven effective, this would suggest a potential approach to modulating epilepsy - related inflammatory responses from another perspective.

T lymphocytes play a significant role in certain epileptic pathologies, such as autoimmune - related epilepsy and Rasmussen’s encephalitis. In the rare condition of Rasmussen’s encephalitis, a large number of CD8⁺ T cells can be seen infiltrating the cerebral cortex and attacking neurons. Which is thought to directly mediate epileptic seizures and progressive neurological deterioration. In more common forms of epilepsy, the role of T cells is less pronounced, but there is still evidence indicating that functional abnormalities of T - cell subsets are associated with epilepsy (Tröscher et al., 2023; Hendrix et al., 2024; Yue et al., 2022).

As mentioned earlier, IL - 17 produced by Th17 cells has a significant impact on epilepsy. A reduction in the number or function of regulatory T cells (Tregs) may weaken the suppression of abnormal inflammation and increase epilepsy susceptibility. Studies have applied IL - 2/anti - IL - 2 complexes in epileptic mouse models to expand Treg cells, resulting in reduced seizure frequency. This suggests that enhancing the immunomodulatory role of T cells may have the potential to suppress epilepsy. Overall, although peripheral T - cell infiltration into the brain is not common in most epilepsy patients, the systemic T - cell immune status (e.g., the presence of autoimmunity) can still influence the course and treatment response of epilepsy. For example, epilepsy patients with autoimmunity often respond well to immunotherapies such as steroids, IVIG, and plasma exchange. Therefore, while considering epilepsy as a disorder of local brain circuits, it is also crucial to take into account the contribution of peripheral immunity (Lai et al., 2024; Huppert et al., 2010; Geis et al., 2019). This is key to identifying certain epilepsies that may be treatable with immunotherapeutic approaches.

In summary, various types of inflammatory cells contribute to the inflammatory microenvironment of epilepsy. Microglia and astrocytes are the core “internal players.” Their continuous activation and interaction generate a large number of pro - seizure mediators (Lee et al., 2020a; Yan et al., 2025; Abbott et al., 2010). When the BBB is compromised, “external players” such as monocytes, granulocytes, and lymphocytes enter the central nervous system, further amplifying inflammation and tissue damage (Figure 2). The combined actions of these inflammatory cells not only lower the seizure threshold and promote seizure occurrence but also contribute to the chronic progression of epilepsy and drug resistance (as sustained inflammation can alter BBB permeability and neuronal responses to drugs). This suggests that in the comprehensive management of epilepsy, in addition to traditional strategies focused on inhibiting neuronal excitability, immunomodulatory strategies targeting inflammatory cells and their products may become the next research and therapeutic focus (Chen et al., 2024b; Tang et al., 2024; Zhong et al., 2022; Wang et al., 2017).

Numerous clinical studies have reported significantly elevated levels of multiple inflammatory factors in the peripheral blood and cerebrospinal fluid (CSF) of patients with active epilepsy. For instance, as early as 1998, Peltola et al. found that IL-6 levels in patients’ CSF rapidly increased shortly after a seizure, positively correlated with seizure duration. Subsequently, Mao et al. analyzed the cytokine profiles in the peripheral blood and CSF of epilepsy patients and found that levels of inflammatory factors such as IL-1β, IL-6, IFN-γ, and IL-17A were significantly higher than in healthy controls. Among these, IL-17A levels showed a positive correlation with annual seizure frequency and severity. These findings established that epilepsy patients exhibit systemic and central inflammatory activation (Kanemura, 2024; Vinti et al., 2021; Chang et al., 2021).

Additional inflammatory markers have since been proposed as potential biomarkers for epilepsy. For example, the neutrophil-to-lymphocyte ratio (NLR), mentioned earlier, has consistently been reported to be elevated across multiple independent cohorts in epilepsy patients and shows a stronger association with refractory epilepsy. Non-specific inflammatory indicators such as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) may also be elevated in some epilepsy patients, particularly those with systemic infections or autoimmune comorbidities. Additionally, the positivity rate of anti-neuronal autoantibodies (e.g., anti-glutamate receptor antibodies, GAD antibodies, etc.) is higher in epilepsy patients of unknown etiology compared to the general population, supporting the notion that a subset of epilepsies may have an autoimmune-inflammatory basis.

Although inflammatory markers do not possess sufficient specificity to function as stand-alone diagnostic tools for epilepsy, they offer additional insights into patients’ pathophysiological states. In the future, integrating multiple markers into a composite ‘epilepsy inflammation score could help guide personalized immunotherapeutic strategies (Wan et al., 2021; Khedpande and Barve, 2025; Auvin et al., 2025; Tanaka et al., 2024).

A subset of epilepsies is closely linked to immune-mediated encephalitis. A typical example is autoimmune limbic encephalitis, which frequently presents as anti-NMDA receptor encephalitis or anti-LGI1 encephalitis. In many cases, epileptic seizures constitute the initial symptom, followed by cognitive decline and psychiatric manifestations. These cases share several common features: detection of autoantibodies against neuronal antigens in cerebrospinal fluid (CSF) and serum, increased inflammatory cells and protein in CSF, and brain imaging showing encephalitis changes. Immunosuppressive therapies (e.g., glucocorticoids, IVIG, rituximab, etc.) usually significantly alleviate seizures and symptoms, indicating that autoimmune-mediated inflammation is a key pathogenic factor (Tanaka et al., 2024; Riva et al., 2024a; Ravizza et al., 2024).

Beyond definitive cases of autoimmune encephalitis, some patients with refractory epilepsy may have occult immune mechanisms. For example, Rasmussen encephalitis, which typically onsets in childhood, is characterized by progressive unilateral cerebral hemisphere atrophy and intractable seizures. Pathologically, it is marked by T lymphocyte-mediated inflammation and neuronal injury. Such patients sometimes respond to high-dose glucocorticoids or immunosuppressive therapy, but often require surgical resection of the affected area for a definitive treatment. These cases highlight the importance of targeting immune-mediated inflammation in certain epilepsies (Tang et al., 2024; Zhong et al., 2022; Vezzani et al., 2022).

Some scholars have proposed the concept of “autoimmune-related epilepsy syndromes” to describe epilepsy patients who have autoimmune diseases (e.g., systemic lupus erythematosus, gluten intolerance, etc.) or neuro-autoantibodies. For these patients, adding immunomodulatory therapy to standard anti-epileptic drugs may improve prognosis (Symonds et al., 2021; Scott, 2021; Tang et al., 2024). Indeed, some refractory epilepsy patients who failed to respond to conventional anti-epileptic medications have shown success with immunotherapy. For example, in adult refractory epilepsy patients with autoimmune thyroiditis, the addition of immunosuppressants significantly reduced seizure frequency. Although such evidence is currently mostly anecdotal, it suggests the need for immunological evaluation in refractory epilepsy patients to identify potential treatable immune triggers (Tanaka et al., 2024; Riva et al., 2024a; Ravizza et al., 2024).

Based on the aforementioned mechanisms, clinical exploration of various anti-inflammatory and immunomodulatory therapies for refractory epilepsy has gained momentum in recent years. One notable attempt is the use of the IL-1 receptor antagonist Anakinra in children with FIRES (febrile infection-related epilepsy syndrome). FIRES is a catastrophic pediatric epilepsy syndrome characterized by intractable epileptic status epilepticus following a fever, often leaving severe sequelae. Several recent clinical reports indicate that adding intravenous Anakinra to conventional sedation and anti-epileptic regimens can significantly reduce seizure frequency and improve consciousness in children. These observations support the critical role of IL-1-mediated inflammatory storms in FIRES and suggest that blocking IL-1 signaling may suppress abnormal discharges (Ihezie et al., 2021; Chen et al., 2024c; Symonds et al., 2021).

Similarly, the IL-6 neutralizing antibody Tocilizumab has been administered in the chronic phase treatment of children with FIRES. A study involving several chronic-phase FIRES patients reported that Tocilizumab reduced seizure frequency and severity in some cases (Aledo-Serrano et al., 2022; Donnelly et al., 2021; Wadayama et al., 2021; Girardin et al., 2023). In post-encephalitic epilepsy or autoimmune-related epilepsy, immunotherapies such as glucocorticoids and intravenous immunoglobulin (IVIG) have also been effective in controlling seizures and, in some cases, achieving remission. For example, in patients with anti-LGI1 encephalitis, long-term low-dose corticosteroid maintenance following acute-phase seizures has been reported to prevent seizure recurrence.

In addition to systemic medications, local anti-inflammatory strategies are also under investigation (Nangia et al., 2024; Chen et al., 2026; Gotra et al., 2021). For instance, Dutch researchers injected inflammation-suppressing neuropeptides delivered via lentiviral vectors into the hippocampus of mice and found that it reduced seizure frequency. Given the potential side effects of systemic anti-inflammatory drug use, such as increased infection and tumor risks, local drug delivery or controlled-release carrier methods may represent one of the future research directions (Wirrell et al., 2022; Lorigados Pedre et al., 2013; Mula et al., 2022).

Considering the role of the gut-immune axis in epilepsy, gut microbiota-targeted therapies are gaining increasing attention. The most well-known approach is the ketogenic diet (KD), a high-fat, low-carbohydrate dietary intervention with significant efficacy in refractory epilepsy, particularly in children. The exact mechanism of KD has long been unclear, but it is now widely accepted that KD may exert its effects by reshaping gut microbiota and metabolites (Ding et al., 2022; Yan et al., 2025; Zhang et al., 2024a).

Research by Olson et al. revealed that KD treatment increases the proportion of two specific gut bacterial taxa in mice: Akkermansia muciniphila (phylum Verrucomicrobia) and Parabacteroides (phylum Bacteroidetes). When these two bacteria coexist, they can enhance the gut’s capacity to produce gamma-aminobutyric acid (GABA), thereby reducing seizure frequency. Importantly, the anti-seizure effect of KD was abolished when to deplete gut microbiota or when KD was administered to germ-free mice, the anti-seizure effects of KD disappeared, highlighting the indispensability of gut microbiota (Li et al., 2025; Qi-Xiang et al., 2022; Chen et al., 2024a).

Beyond KD, probiotics and prebiotics (substrates that promote the growth of beneficial bacteria) are being investigated as adjuvant therapies for epilepsy. A small-scale randomized controlled trial administered a probiotic formulation containing Lactobacillus acidophilus, Bifidobacterium, and other beneficial bacteria to drug-resistant epilepsy patients. Approximately 28% of patients in the treatment group experienced a >50% reduction in seizure frequency, whereas no significant improvement was observed in the control group. Although the sample size was limited, these finding suggests that gut microbiota modulation may lower seizure susceptibility in some patients (Riva et al., 2024b; Mejía-Granados et al., 2021; Mazarati, 2024).

Additionally, several case reports have documented the potential benefit of fecal microbiota transplantation (FMT) in patients with epilepsy and coexisting conditions. For example, a case report described a patient with refractory epilepsy whose seizure frequency dramatically decreased and remained reduced for over 6 months following FMT for Clostridium difficile infection. While it is unclear whether this outcome was coincidental or causally related, it has sparked interest in FMT as a potential epilepsy therapy (Chen et al., 2024a; Diaz-Marugan et al., 2024; Riva et al., 2024b).

Overall, gut microbiota–modulating therapies for epilepsy are still in the exploratory phase, with their mechanisms of action, target populations, and long-term safety requiring further investigation. However, as a manipulable entity, gut microbiota has the potential to become an emerging field in future epilepsy management (Wang et al., 2024; Liang et al., 2022).

In addition to the aforementioned therapies directly targeting cytokines or gut microbiota, some commonly used medications have also demonstrated dual anti-inflammatory and anti-epileptic effects (Hermann et al., 2017; Arredondo et al., 2024). For example, metformin, an antidiabetic drug, has been found to activate the adenosine pathway and attenuate microglial inflammation. In animal epilepsy models, it reduced seizure frequency. Statins, lipid-lowering drugs with anti-inflammatory and endothelial function improvement properties, have been reported to alleviate post-epileptic cognitive impairments in when administered after status epilepticus. These “drug repurposing” findings highlight promising avenues for anti-inflammatory epilepsy treatment. However, these strategies require rigorous clinical trial validation (Gotra et al., 2021).

Finally, for some definitive inflammation-related epilepsies, surgical treatment remains a crucial option. For instance, in advanced Rasmussen encephalitis, hemispherectomy is often necessary to radically cure seizures due to irreversible inflammatory damage. Early identification and intervention in inflammatory mechanisms are thus critical to avoiding surgical outcomes.