Blood markers flag drug-resistant epilepsy in children, study finds

Children with epilepsy whose seizures resist multiple medications show a distinctive pattern of inflammatory markers in their blood, according to a cross-sectional study of pediatric patients in Jordan. Researchers measured serum levels of three adipokines—leptin, adiponectin, and resistin—along with an extended panel of inflammatory cytokines in 67 children with epilepsy and 13 age- and sex-matched healthy controls. All of the children with epilepsy had been seizure-free for at least 72 hours before blood sampling, an interval chosen to minimize the influence of acute seizures on cytokine release. The findings, published in the journal Pharmacia, suggest that two markers in particular—resistin and tumor necrosis factor-alpha (TNF-α)—may help distinguish children whose epilepsy is drug-resistant from those whose seizures are controlled or simply not yet meeting the formal criteria for resistance.

The study enrolled children under 18 years of age who presented to the neurology clinic at the University of Jordan Hospital between June 2020 and June 2022. The participants were divided according to seizure control status. Twenty children, or 29.85 percent of the cohort, had controlled epilepsy, meaning their seizures were well managed with antiseizure medications and they had been seizure-free for at least three months. The remaining 47 children, or 70.15 percent, had uncontrolled epilepsy, which the researchers split into two subgroups: 24 children with drug-resistant epilepsy and 23 with non-drug-resistant uncontrolled epilepsy. Drug resistance was defined according to the International League Against Epilepsy criteria, as the failure of adequate trials of two tolerated, appropriately chosen antiseizure medications, used alone or in combination at maximum tolerable doses, to achieve sustained seizure freedom. Children with non-drug-resistant uncontrolled epilepsy had not yet met these criteria, either because only one medication had been tried or because two drugs had been used at doses below the maximum tolerated level.

The patients had an average age of 9.02 years, ranging from 0.1 to 17 years, and an average body mass index of 20.24 kilograms per square meter. Importantly, there were no significant differences between the patients and the healthy controls in age, weight, height, body mass index, or sex, which strengthens the comparability of the groups. The researchers also excluded children with any other neurological conditions or any condition that could alter inflammatory status, including active infection, thyroid dysfunction, chronic inflammatory diseases, autoimmune diseases, cancer, kidney or liver impairment, recent surgery or major trauma, or the use of immunosuppressive or immunomodulatory medications in the six weeks before the study.

Blood samples were processed into serum and analyzed with a human ProcartaPlex multiplex immunoassay, a bead-based technique that can quantify many proteins simultaneously in a small volume. The investigators measured leptin, resistin, adiponectin, mannose-binding lectin, TNF-α, BAFF, APRIL, IL-1β, IL-1RA, IL-2, IL-6, IL-17A, interferon-alpha, and interferon-gamma. Samples were run on a Bio-Plex 200 System and analyzed with the corresponding software. Statistical comparisons relied on non-parametric tests, including the Mann–Whitney U and Kruskal–Wallis tests, with Spearman correlations for continuous associations and logistic regression for identifying predictors of drug resistance.

The headline result concerned the adipokines. Across all children with epilepsy, leptin and adiponectin levels were significantly higher than in healthy controls, with p-values of 0.022 and 0.045 respectively, while resistin levels were significantly lower, at p = 0.009. When the patient groups were compared with each other, a gradient emerged with worsening control. Leptin levels did not differ significantly between the drug-resistant and non-drug-resistant uncontrolled groups, or between the controlled and uncontrolled epilepsy groups, although leptin was significantly elevated in the drug-resistant group compared with healthy controls. Adiponectin, by contrast, was significantly higher in the uncontrolled epilepsy group than in the controlled group, and within the uncontrolled group it was the drug-resistant subgroup—rather than the non-drug-resistant one—that showed significantly higher adiponectin than both the controlled group and the healthy controls. Resistin followed a similar pattern: levels were significantly higher in uncontrolled than in controlled epilepsy, with both subgroups exceeding the controlled group, and resistin was lowest of all in the controlled epilepsy group, falling significantly below even the healthy controls.

Among the cytokines, most were simply undetectable in the serum of these children. BAFF, APRIL, IL-1β, IL-1RA, IL-2, IL-6, IL-17A, interferon-alpha, and interferon-gamma all fell below the detection limits of the assay. Mannose-binding lectin, a component of the complement system, did not differ significantly between patients and controls or among the patient groups. TNF-α was the exception: it did not differ overall between children with epilepsy and healthy controls, but it was significantly higher in the drug-resistant group than in either the non-drug-resistant uncontrolled group (p = 0.040) or the controlled group (p = 0.046). Notably, TNF-α levels were also correlated with the use of combination antiseizure therapy, averaging 1.71 picograms per milliliter in children on combination therapy versus 0.99 picograms per milliliter in those on monotherapy, and all children with drug-resistant epilepsy in this study were receiving combination therapy.

To test which markers actually predicted drug resistance, the researchers compared the drug-resistant group with all other children with epilepsy combined. Levels of resistin, adiponectin, and TNF-α were all significantly higher in the drug-resistant group, while the difference in leptin was not statistically significant. These candidate markers were then entered into a multiple binary logistic regression model. Only resistin and TNF-α retained significant predictive value. Each unit increase in resistin carried an odds ratio of 1.003 for drug resistance, while higher TNF-α carried an odds ratio of approximately 7.9, with a 95 percent confidence interval of 1.578 to 39.800. In practical terms, children with higher TNF-α levels were roughly eight times more likely to have drug-resistant epilepsy than other children with epilepsy in this cohort, and resistin added independent predictive information.

The study also explored which clinical characteristics tracked with the levels of each mediator. Leptin and resistin levels were positively correlated with the children’s age, weight, and height, and leptin was additionally positively correlated with body mass index, consistent with the role of these adipokines in energy metabolism. Adiponectin moved in the opposite direction, showing significant negative correlations with age, weight, height, and body mass index. In stepwise multiple linear regression, body mass index, mannose-binding lectin levels, and resistin levels emerged as the strongest predictors of leptin levels, with the model explaining about 34 percent of the variance. For resistin, the strongest predictors were age, leptin levels, and a shorter time since the last seizure, explaining about 26 percent of the variance. For adiponectin, the presence of developmental delay was the only positive predictor and lower body weight the only negative predictor, with the model explaining about 36 percent of the variance. Adiponectin averaged 2.4 × 10⁸ picograms per milliliter in children with developmental delay versus 1.00 × 10⁸ in those without.

Two findings stood out as novel. Adiponectin and resistin levels were both negatively correlated with the time elapsed since the last seizure, meaning their levels were higher in children with more recent seizures and lower in those with better seizure control—an observation the authors report for the first time. Adiponectin levels were also significantly higher in children with developmental delay, another first. The authors caution, however, that the results should be generalized carefully because of the small sample size and the single-center design. Moreover, the cross-sectional design cannot establish whether the elevated markers drive drug resistance or merely accompany it. The authors suggest that resistin and TNF-α might represent potential targets for disease-modifying, anti-inflammatory approaches in pediatric epilepsy, particularly drug-resistant disease, but such therapeutic implications remain hypothetical pending interventional research. What the study firmly establishes is a measurable serum signature—higher adiponectin and resistin, and higher TNF-α relative to other patient groups—that distinguishes children whose seizures defy multiple medications from those whose epilepsy is controlled.

Subject of Research: Serum adipokine and cytokine profiles in pediatric epilepsy and their association with seizure control and drug resistance

Article Title: Adipokine and cytokine profiles in pediatric epilepsy: a comparative analysis of epilepsy level of control and drug resistance

Article References: Abu-rish, E. Y., Masri, A., Alnajjar, R., Bustanji, Y., Tawaha, K., & Mohamed, Y. S. (2026). Adipokine and cytokine profiles in pediatric epilepsy: a comparative analysis of epilepsy level of control and drug resistance. Pharmacia, 73, Article e207848. https://doi.org/10.3897/pharmacia.73.e207848

Image Credits: AI Generated

DOI: 10.3897/pharmacia.73.e207848

Keywords: pediatric epilepsy, drug-resistant epilepsy, adipokines, leptin, adiponectin, resistin, TNF-alpha, inflammatory cytokines, multiplex immunoassay, seizure control, cross-sectional study, biomarkers