Epilepsia最新文献

Objective: This systematic review synthesizes evidence on multimodal machine learning (ML) decision support systems for epilepsy surgery focusing on postsurgical outcome prediction, with emphasis on methodological quality and implications for clinical practice.

Methods: Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we searched PubMed, Scopus, and Web of Science using predefined keywords. Seventy records were screened; 10 studies met inclusion/exclusion criteria, reporting ML-based prediction of surgical outcomes in drug-resistant epilepsy (DRE) and using ≥2 data modalities. Extracted items included study design, population, data sources, algorithms, validation strategy, performance metrics, and outcome definitions. Two reviewers independently screened records.

Results: Nine studies were retrospective and one prospective; seven were single-center and two multicenter. Most integrated neuroimaging (9/10), electroencephalography (8/10), and clinical variables (7/10); two included neuropsychology, and one added ablation parameters for magnetic resonance-guided laser interstitial thermal therapy. Sample sizes ranged from 15 to 11 067. Performance varied; best results (area under the curve [AUC] ≈ .95) were reported with multimodal gradient boosting, whereas ablation-based models achieved lower discrimination (AUC ≈ .67). The oldest neural-network study reported 98% accuracy on a small, nonstandard dataset. Cross-validation predominated; only two studies assumed prospective validation. Outcome definitions were heterogeneous, and time points were inconsistently specified. Despite variability, several clinically relevant findings emerged; multimodal ML improved, but not universally, prediction of seizure freedom, supported epileptogenic-zone localization, and in one large multicenter study, enabled earlier identification of surgical candidates compared with routine referral pathways.

Significance: ML shows promise for outcome prediction and presurgical decision support in DRE, particularly when integrating multimodal data. Translation is currently constrained by limited external and prospective validation, inconsistent outcome frameworks, and insufficient interpretability. Future research should prioritize harmonized endpoints, multicenter external validation (including federated approaches), and explainable models capable of informing both patient selection and surgical strategy.

The recent International League Against Epilepsy (ILAE) official and updated classification of focal cortical dysplasia (FCD) includes a third type-FCD type 3-characterized by architectural abnormalities (cortical dyslamination) associated with another "principal" lesion: hippocampal sclerosis (HS), developmental tumors, vascular malformations, or gliotic scars. We posit that the clinical relevance of FCD type 3 is not established, as dyslamination cannot be reliably identified preoperatively and (although unproven, because unidentifiable) persistence of cortical tissue with putative dyslamination after resective surgery does not preclude seizure control. Here we discuss these issues from the epileptologist's perspective and stimulate the debate on whether the FCD type 3 construct impacts decision-making in the epilepsy surgery field-or else is of little practical significance.

Objective: To examine whether initiation of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) is associated with seizure recurrence and related outcomes in adults with epilepsy and type 2 diabetes.

Methods: We conducted a retrospective cohort study using de-identified electronic health records from the TriNetX Research Network (January 2003-August 2025), including adults ≥18 years with ≥3 epilepsy or recurrent seizure diagnoses. Patients initiating a GLP-1 RA (exenatide, liraglutide, dulaglutide, lixisenatide, semaglutide, or tirzepatide) without prior comparator therapy were compared with those initiating other glucose-lowering agents (sodium-glucose cotransporter 2 inhibitors, dipeptidyl peptidase 4 inhibitors, sulfonylureas, or insulin) without GLP-1 RA exposure. Propensity score matching (1:1) was performed on 82 covariates, yielding 8688 matched pairs. Outcomes were assessed using Cox proportional hazards models.

Results: After matching, the mean age was 52.6 years, and 67.6% were female. Median follow-up was 514 days (interquartile range [IQR] 671) for GLP-1 RA initiators and 415 days (IQR 769) for comparators. GLP-1 RA initiation was associated with lower risk of seizure recurrence (HR .82, 95% confidence interval [CI] .78-.86; RD -2.1%), hospitalization (HR .35, 95% CI .29-.43; RD -2.6%), and all-cause mortality (HR .40, 95% CI .34-.47; RD -4.8%). Associations with status epilepticus (HR .75, 95% CI .66-.85; RD -.7%) and ICU admission (HR .82, 95% CI .69-.96; RD -.3%) were smaller; the latter was not statistically significant.

Significance: In this large multinational cohort, GLP-1 RA initiation was associated with reduced risks of seizure recurrence, hospitalization, and mortality compared with other glucose-lowering therapies. These hypothesis-generating findings warrant confirmation in prospective studies before translation into clinical practice.

Objective: To examine early responses to cenobamate therapy using prospective data from a dose-response study in Asian patients with focal seizures (YKP3089C035, C035) that employed a titration regimen starting at 12.5 mg/day.

Methods: In Study C035, adults 18-70 years of age with uncontrolled focal seizures despite treatment with 1-3 antiseizure medications were randomized 1:1:1:1 to receive placebo or adjunctive cenobamate 100, 200, or 400 mg/day. The 24-week double-blind study included an 18-week titration and 6-week maintenance phase. During the first 8 weeks of titration ("early titration phase"), all cenobamate patients received the same dosing regimen: 12.5 mg/day for 2 weeks, 25 mg/day for 2 weeks, 50 mg/day for 2 weeks, and 100 mg/day for 2 weeks. Change in seizure frequency from baseline and responder rates were assessed at these 2-week intervals for combined cenobamate dose groups vs placebo. Analyses were performed on the modified intent-to-treat maintenance (MITT-M) population (≥1 study drug dose and seizure data in the maintenance phase); all patients completed early titration.

Results: Of 519 patients randomized, 446 were included in the MITT-M population (placebo n = 117, cenobamate n = 329). During Weeks 1-2, 3-4, 5-6, and 7-8 of titration, cenobamate patients experienced a median reduction in 28-day seizure frequency of 16.0% (vs 20.0% placebo, p = .81), 27.3% (vs 22.2% placebo, p = .42), 42.9% (vs 15.4% placebo, p = .002), and 55.6% (vs 20.0% placebo, p < .001), respectively. During Weeks 5-6 and 7-8, the 100% responder rates for cenobamate 50 and 100 mg/day were 17.0% (vs 12.8% placebo, p = .29) and 26.7% (vs 8.5% placebo, p < .001), respectively.

Significance: Statistically significant responses to cenobamate treatment occurred within the first 8 weeks of titration, including a 42.9% median reduction in 28-day seizure frequency (Weeks 5-6) and a seizure-free rate of 26.7% (Weeks 7-8). These data show that substantial seizure reductions occurred in many patients early during cenobamate titration.

Objective: Lamotrigine is one of the most widely prescribed antiseizure medication (ASM) and mood stabilizer in the United States due to its favorable side-effect profile, lower risk of teratogenicity, and minimal drug-drug interactions. This study aimed to examine age- and sex-associated variability in prescribing and pharmacokinetics, focusing on postmenopausal women.

Methods: Data were from electronic health records. Individuals were included if ≥18 years and received an ASM between January 1, 2015 and December 31, 2021. Lamotrigine prescriptions were compared based on age, sex, epilepsy diagnosis, and monotherapy/polytherapy. Statistical comparisons of proportions were conducted using two-proportion tests. To characterize age- and sex-related differences in LTG apparent oral clearance and assess the impact of covariates, linear mixed-effects modeling was employed.

Results: Records were available for 314 890 individuals, with 23 906 patients being prescribed lamotrigine at least once (as monotherapy or polytherapy) for both epilepsy and non-epilepsy diagnoses. The lamotrigine prescription rate was lower in postmenopausal women compared to younger women but higher than in older men, irrespective of diagnosis. Notably, lamotrigine was prescribed as monotherapy more frequently to patients without epilepsy than those with epilepsy, regardless of sex and age. The clearance of lamotrigine was 22% lower in postmenopausal women compared to younger women and 9% in older men. Lamotrigine clearance increased by 49% and 11% with co-administration of inducers or the presence of smoking, respectively. Lamotrigine clearance decreased by 51% in the presence of an inhibiting medication.

Significance: Prescription rates for lamotrigine varied between patients with epilepsy and those with non-epilepsy conditions. Age and sex differences in pharmacokinetics suggest the need for lamotrigine dose adjustments, highlighting the importance of therapeutic drug monitoring in personalized epilepsy care. Lamotrigine use was less frequent in postmenopausal women compared to younger women but higher compared to older men. Postmenopausal women were prescribed lamotrigine as monotherapy to a lesser extent than younger women and older men.

Background and objective: Lafora disease (LD) is a rare progressive disorder caused by mutations in the EPM2A or EPM2B genes, characterized by the accumulation of Lafora bodies, drug-resistant epilepsy, and cognitive decline. To investigate the early molecular mechanisms of LD, we studied electrophysiological changes in the dentate gyrus (DG) of the Epm2a^R240X knock-in mouse model at various ages.

Methods: Electrophysiological recordings measured neuronal membrane properties, epileptic-like activity, epileptic thresholds, and synaptic plasticity in Epm2a^R240X mice at 1, 3, and 12 months. We also employed Periodic Acid-Schiff (PAS) diastase staining, immunofluorescence, and Western blotting to detect Lafora bodies, amyloid beta deposition, and the expression of glutamate receptor subunits.

Results: Epileptic-like activity began at 1 month and intensified with age. Aberrant long-term potentiation (LTP) appeared at 3 months and worsened by 12 months. Notably, cannabidiol treatment reduced excitability and restored LTP in older mice, suggesting its potential therapeutic value.

Significance: The reversibility of synaptopathy, even at advanced stages, reinforces the importance of early detection of hyperexcitability and the development of effective therapeutic approaches.

Objective: This study was undertaken to identify novel blood-based biomarkers associated with the early diagnosis of status epilepticus (SE) by exploring proteomic alterations in plasma extracellular vesicles (EVs).

Methods: We conducted an observational and prospective study in a two-phase design: a discovery proteomic analysis of plasma EVs from 16 patients (eight with SE and eight matched non-SE), followed by a validation phase using basic immunodetection techniques in a separate cohort of 160 patients (80 SE, 80 non-SE) for selected proteins in plasma. Proteomic profiling was performed with liquid chromatography-tandem mass spectrometry, and protein interaction networks were explored using STRING and Cytoscape software platforms.

Results: In the discovery phase, 2046 proteins were identified, and 977 met inclusion criteria. CDH1, APOC4, and IGHG2 showed the highest differential expression in SE patients. Network analysis revealed strong interactions involving SLC4A1 and SRC. In the validation cohort, levels of APOC4, CDH1, SLC4A1, and S100B were significantly higher in SE patients (p < .05). In the multivariate logistic regression model, only APOC4, S100B, and SLC4A1 remained independently associated with SE diagnosis. A combined model including these proteins predicted SE with 84.6% accuracy (95% confidence interval = 64.3%-94.9%). Other proteins, including NSE and HMGB1, showed no significant differences.

Significance: This study identifies new plasma biomarkers that may support the early diagnosis of SE, particularly in emergency settings where access to electroencephalography is limited. These findings suggest potential applicability to clinical practice; nevertheless, prospective validation in independent and larger cohorts is required.