Applications for Deep Learning in Epilepsy Genetic Research.

Epilepsy is a group of brain disorders characterised by an enduring predisposition to generate unprovoked seizures. Fuelled by advances in sequencing technologies and computational approaches, more than 900 genes have now been implicated in epilepsy. The development and optimisation of tools and methods for analysing the vast quantity of genomic data is a rapidly evolving area of research. Deep learning (DL) is a subset of machine learning (ML) that brings opportunity for novel investigative strategies that can be harnessed to gain new insights into the genomic risk of people with epilepsy. DL is being harnessed to address limitations in accuracy of long-read sequencing technologies, which improve on short-read methods. Tools that predict the functional consequence of genetic variation can represent breaking ground in addressing critical knowledge gaps, while methods that integrate independent but complimentary data enhance the predictive power of genetic data. We provide an overview of these DL tools and discuss how they may be applied to the analysis of genetic data for epilepsy research.

Dyadic Hair Cortisol Self-Collection Procedure.

BACKGROUND: At-home self-collection of specimens has become more commonplace because of measures taken in response to the coronavirus pandemic. Self-collection of hair cortisol is important because chronic stress is present in many populations, such as older adults living with Alzheimer's disease and their family caregivers. For the evaluation of chronic stress, scalp hair can be used as a predictive biomarker because it examines the cumulative, retrospective stress from previous months. OBJECTIVES: The aim of the paper is to provide a study procedure for at-home, scalp hair self-collection for cortisol concentration analysis from dyads consisting of a person living with Alzheimer's disease and their family caregiver. METHODS: After informed electronic consent is obtained, a package containing the necessary tools for self-collection of hair samples from the dyad is mailed to the participant's home. Participants are provided detailed print and video multimedia guides outlining how to obtain the hair samples. Ideally, the hair samples are obtained during the virtual data collection meeting with research personnel. Participants mail back the hair sample in a prepaid package to the biomedical laboratory for analysis. DISCUSSION: At-home, self-collection of hair provides potential advantages such as reduced participant burden, especially for vulnerable populations where transportation and different environments are challenging. At-home sample collection options may increase research participation and can be applied to multiple research foci. Research considerations for dyads, such as people living with Alzheimer's disease and their caregivers, are discussed.

Multi-omics integration and epilepsy: Towards a better understanding of biological mechanisms.

The epilepsies are a group of complex neurological disorders characterised by recurrent seizures. Approximately 30% of patients fail to respond to anti-seizure medications, despite the recent introduction of many new drugs. The molecular processes underlying epilepsy development are not well understood and this knowledge gap impedes efforts to identify effective targets and develop novel therapies against epilepsy. Omics studies allow a comprehensive characterisation of a class of molecules. Omics-based biomarkers have led to clinically validated diagnostic and prognostic tests for personalised oncology, and more recently for non-cancer diseases. We believe that, in epilepsy, the full potential of multi-omics research is yet to be realised and we envisage that this review will serve as a guide to researchers planning to undertake omics-based mechanistic studies.

Disease-modifying effects of sodium selenate in a model of drug-resistant, temporal lobe epilepsy.

There are no pharmacological disease-modifying treatments with an enduring effect to mitigate the seizures and comorbidities of established chronic temporal lobe epilepsy (TLE). This study aimed to evaluate for disease modifying effects of sodium selenate treatment in the chronically epileptic rat post-status epilepticus (SE) model of drug-resistant TLE. Wistar rats underwent kainic acid-induced SE or sham. Ten-weeks post-SE, animals received sodium selenate, levetiracetam, or vehicle subcutaneousinfusion continuously for 4 weeks. To evaluate the effects of the treatments, one week of continuous video-EEG was acquired before, during, and 4, 8 weeks post-treatment, followed by behavioral tests. Targeted and untargeted proteomics and metabolomics were performed on post-mortem brain tissue to identify potential pathways associated with modified disease outcomes. Telomere length was investigated as a novel surrogate marker of epilepsy disease severity in our current study. The results showed that sodium selenate treatment was associated with mitigation of measures of disease severity at 8 weeks post-treatment cessation; reducing the number of spontaneous seizures (p< 0.05), cognitive dysfunction (p< 0.05), and sensorimotor deficits (p< 0.01). Moreover, selenate treatment was associated with increased protein phosphatase 2A (PP2A) expression, reduced hyperphosphorylated tau, and reversed telomere length shortening (p< 0.05). Network medicine integration of multi-omics/pre-clinical outcomes identified protein-metabolite modules positively correlated with TLE. Our results provide evidence that treatment with sodium selenate results in a sustained disease-modifying effect in chronically epileptic rats in the post-KA SE model of TLE, including improved comorbid learning and memory deficits.

According to the World Health Organization (WHO), there are around 50 million people with epilepsy worldwide. Although drugs are available to control epileptic seizures, these only provide symptomatic relief. They cannot prevent the condition from worsening, and if people with epilepsy stop taking their medication, there is no lasting effect on the severity or frequency of their seizures. Some epilepsy cases are also resistant to these drugs. This is particularly common in adults with temporal epilepsy, with 30% of people continuing to suffer with seizures despite receiving medication. Current treatments also have no effect on problems with learning, memory and mental health that sometimes accompany drug-resistant epilepsy. Previous studies in animals have identified some potential treatments that could slow the progression of temporal epilepsy, but these have only been shown to work when used at a very early stage. Since most individuals with temporal epilepsy have already started having seizures when they are diagnosed (and it is difficult to predict who will develop the condition), these drugs are unlikely to be useful in practice. Here, Casillas-Espinosa et al. set out to find if a novel drug called sodium selenate can stop the progression of epilepsy and reduce the severity of temporal epilepsy when the condition is fully advanced. To do this, they used an animal model of temporal epilepsy, where rats had been modified to develop spontaneous seizures, resistance to normal anti-seizure medications, and problems with learning and memory. Casillas-Espinosa et al. found that sodium selenate not only reduced the number and severity of seizures in these model rats, but also improved their memory and learning ability. Several rats stopped having seizures altogether even after the treatment had stopped, indicating that sodium selenate had a long-lasting protective effect. Genetic analysis of the rats also revealed that shorter telomeres (special DNA sequences at the ends of chromosomes) correlated with increasing severity of the condition, suggesting that telomere length could help predict who might develop temporal epilepsy or respond best to treatment. This study identifies sodium selenate as a potential treatment that could reverse the progression of temporal epilepsy, even in individuals with advanced symptoms. Later this year, sodium selenate will be trialled in people with drug-resistant temporal epilepsy to determine if the drug benefits humans in the same way. Casillas-Espinosa et al. hope that it will improve participants' epilepsy and, ultimately, their quality of life.

Psychophysics of Pain: A Methodological Introduction.

For over 100 years, psychophysics ..÷ the scientific study between physical stimuli and sensation ... has been successfully employed in numerous scientific and healthcare disciplines, as an objective measure of sensory phenomena. This manuscript provides an overview of fundamental psychophysical concepts, emphasizing pain and research application..÷defining common terms, methods, and procedures.Psychophysics can provide systematic and objective measures of sensory perception that can be used by nursing scientists to explore complex, subjective phenomena..÷such as pain perception. While there needs to be improved standardization of terms and techniques, psychophysical approaches are diverse and may be tailored to address or augment current research paradigms. The interdisciplinary nature of psychophysics..÷like nursing..÷provides a unique lens for understanding how our perceptions are influenced by measurable sensations. While the quest to understand human perception is far from complete, nursing science has an opportunity to contribute to pain research by using the techniques and methods available through psychophysical procedures.

Human In Vitro Models of Epilepsy Using Embryonic and Induced Pluripotent Stem Cells.

The challenges in making animal models of complex human epilepsy phenotypes with varied aetiology highlights the need to develop alternative disease models that can address the limitations of animal models by effectively recapitulating human pathophysiology. The advances in stem cell technology provide an opportunity to use human iPSCs to make disease-in-a-dish models. The focus of this review is to report the current information and progress in the generation of epileptic patient-specific iPSCs lines, isogenic control cell lines, and neuronal models. These in vitro models can be used to study the underlying pathological mechanisms of epilepsies, anti-seizure medication resistance, and can also be used for drug testing and drug screening with their isogenic control cell lines.

Study protocol for an integrative theory review of the concept of unidentified pain.

INTRODUCTION: Most definitions of pain require individuals experiencing pain to report their pain. There are three groups who may not always report pain including those who: (1) lack the cognitive ability to verbally communicate their pain and also lack a proxy/surrogate to report pain for them; (2) lack the cognitive ability to verbally communicate their pain but have a proxy to report the pain; and (3) have the cognitive ability to verbally report pain but are unable or unwilling to do so. Clinicians may not be able to determine which patients are at risk for unidentified pain. Therefore, in this study, we present a protocol for an integrative review with the aim of identifying existing theoretical approaches to understanding unidentified pain. METHODS AND ANALYSIS: We propose a systematic overview of the existing theoretical approaches to understanding 'unidentified pain'. We will use Campbell and colleagues' criteria for systematic reviews of theory and the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols guidelines. Our search will be broad to cover theoretical approaches to 'unidentified pain' using MEDLINE, CINAHL, Embase and Google Scholar. Covidence systematic review software will be used for data extraction and analysis. Then, qualitative content analysis will take place. The content analysis will be presented as a narrative. ETHICS AND DISSEMINATION: No human or animal subjects will be involved. The results are to be published in peer-reviewed journals and presented at conferences in the USA and internationally.

Increased pain unpleasantness and pain-related fMRI activation in the periaqueductal gray in Alzheimer's disease.

Background: Pain continues to be underrecognized and undertreated in people with Alzheimer's disease (AD). The periaqueductal gray (PAG) is essential to pain processing and modulation yet is damaged by AD. While evidence exists of altered neural processing of pain in AD, there has not been a focused investigation of the PAG during pain in people with AD. Purpose: To investigate the role of the PAG in sensory and affective pain processing for people living with AD. Methods: Participants from a larger study completed pain psychophysics assessments and then a perceptually-matched heat pain task (warmth, mild, and moderate pain) during a functional MRI scan. In this cross-sectional study, we examined blood oxygenation level-dependent (BOLD) responses in the PAG and other pain-related regions in participants with AD (n = 18) and cognitively intact older adults (age- and sex-matched, n = 18). Associations of BOLD percent signal change and psychophysics were also examined. Results: There were significant main effects of AD status on the temperature needed to reach each perception of warmth or pain, where people with AD reached higher temperatures. Furthermore, participants with AD rated mild and moderate pain as more unpleasant than controls. PAG BOLD activation was greater in AD relative to controls during warmth and mild pain percepts. No significant differences were found for moderate pain or in other regions of interest. Greater PAG activation during mild pain was associated with higher affective/unpleasantness ratings of mild pain in participants with AD but not in controls. Conclusion: Results suggest a role for the PAG in altered pain responses in people with AD. The PAG is the primary source of endogenous opioid pain inhibition in the neuroaxis, thus, altered PAG function in AD suggests possible changes in descending pain inhibitory circuits. People with AD may have a greater risk of suffering from pain compared to cognitively intact older adults.

Promoting successful participation of people living with Alzheimer's disease and related dementias in pain-related neuroimaging research studies.

Recruitment and retention of participants for pain-related neuroimaging research is challenging and becomes increasingly so when research participants have a diagnosis of Alzheimer's disease and related dementias (ADRD). This article shares the authors' recommendations from several years of successful recruitment and completion of pain-related neuroimaging studies of people living with ADRD and includes supportive literature. While not an exhaustive list, this review covers several topics related to recruitment and retention of participants living with ADRD, including community engagement, capacity to consent, dementia diagnostic criteria, pain medication and other study exclusion criteria, participant and caregiver burden, communication concerns, and relationships with neuroimaging facilities. Threaded throughout the paper are important cultural considerations. Additionally, we discuss implications of the coronavirus (COVID-19) pandemic for recruitment. Once tailored to specific research study protocols, these proven strategies may assist researchers with successfully recruiting and retaining participants living with ADRD for pain-related neuroimaging research studies toward improving overall health outcomes.

An Integrated Multi-Omic Network Analysis Identifies Seizure-Associated Dysregulated Pathways in the GAERS Model of Absence Epilepsy.

Absence epilepsy syndromes are part of the genetic generalized epilepsies, the pathogenesis of which remains poorly understood, although a polygenic architecture is presumed. Current focus on single molecule or gene identification to elucidate epileptogenic drivers is unable to fully capture the complex dysfunctional interactions occurring at a genetic/proteomic/metabolomic level. Here, we employ a multi-omic, network-based approach to characterize the molecular signature associated with absence epilepsy-like phenotype seen in a well validated rat model of genetic generalized epilepsy with absence seizures. Electroencephalographic and behavioral data was collected from Genetic Absence Epilepsy Rats from Strasbourg (GAERS, n = 6) and non-epileptic controls (NEC, n = 6), followed by proteomic and metabolomic profiling of the cortical and thalamic tissue of rats from both groups. The general framework of weighted correlation network analysis (WGCNA) was used to identify groups of highly correlated proteins and metabolites, which were then functionally annotated through joint pathway enrichment analysis. In both brain regions a large protein-metabolite module was found to be highly associated with the GAERS strain, absence seizures and associated anxiety and depressive-like phenotype. Quantitative pathway analysis indicated enrichment in oxidative pathways and a downregulation of the lysine degradation pathway in both brain regions. GSTM1 and ALDH2 were identified as central regulatory hubs of the seizure-associated module in the somatosensory cortex and thalamus, respectively. These enzymes are involved in lysine degradation and play important roles in maintaining oxidative balance. We conclude that the dysregulated pathways identified in the seizure-associated module may be involved in the aetiology and maintenance of absence seizure activity. This dysregulated activity could potentially be modulated by targeting one or both central regulatory hubs.

'A platform for goodness, not for badness': The heuristics of hope in patients' evaluations of online health information.

Patient advocates and activists are increasingly relying on online health information that can assist them to manage their health condition. Yet once online, they will confront diverse information whose veracity and utility are difficult to determine. This article offers a sociological analysis of the practical methods, or heuristics, that patient advocates and activists use when making judgements about the credibility and utility of online information. Drawing on the findings from interviews with fifty Australian patient advocates and activists, it is argued that these individuals' use of these heuristics reflects their hopes that information can help them manage their condition which may, in some cases, override fears and uncertainties that arise during searches. The article identifies the common 'rules-of-thumb'-or what we call the 'heuristics of hope'-that patient advocates/activists may use to make judgements and highlights the dangers of over-reliance on them, especially regarding clinically unproven, potentially unsafe treatments. Analyses of the heuristics of hope, we conclude, can assist in understanding the dynamics of decision-making and the role that affect plays in online patient communities which is crucial in an age characterised by the rapid circulation of emotionally charged messages, often based on hope.

Network Preservation Analysis Reveals Dysregulated Synaptic Modules and Regulatory Hubs Shared Between Alzheimer's Disease and Temporal Lobe Epilepsy.

Background: There is increased prevalence of epilepsy in patients with Alzheimer's disease (AD). Although shared pathological and clinical features have been identified, the underlying pathophysiology and cause-effect relationships are poorly understood. We aimed to identify commonly dysregulated groups of genes between these two disorders. Methods: Using publicly available transcriptomic data from hippocampal tissue of patients with temporal lobe epilepsy (TLE), late onset AD and non-AD controls, we constructed gene coexpression networks representing all three states. We then employed network preservation statistics to compare the density and connectivity-based preservation of functional gene modules between TLE, AD and controls and used the difference in significance scores as a surrogate quantifier of module preservation. Results: The majority (>90%) of functional gene modules were highly preserved between all coexpression networks, however several modules identified in the TLE network showed various degrees of preservation in the AD network compared to that of control. Of note, two synaptic signalling-associated modules and two metabolic modules showed substantial gain of preservation, while myelination and immune system-associated modules showed significant loss of preservation. The genes SCN3B and EPHA4 were identified as central regulatory hubs of the highly preserved synaptic signalling-associated module. GABRB3 and SCN2A were identified as central regulatory hubs of a smaller neurogenesis-associated module, which was enriched for multiple epileptic activity and seizure-related human phenotype ontologies. Conclusion: We conclude that these hubs and their downstream signalling pathways are common modulators of synaptic activity in the setting of AD and TLE, and may play a critical role in epileptogenesis in AD.

Potential role of regulatory DNA variants in modifying the risk of severe cutaneous reactions induced by aromatic anti-seizure medications.

OBJECTIVE: Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are severe cutaneous adverse drug reactions. Antiseizure medications (ASMs) with aromatic ring structure, including carbamazepine, are among the most common culprits. Screening for human leukocyte antigen (HLA) allele HLA-B*15:02 is recommended prior to initiating treatment with carbamazepine in Asians, but this allele has low positive predictive value. METHODS: We performed whole genome sequencing and analyzed 6 199 696 common variants among 113 aromatic ASM-induced SJS/TEN cases and 84 tolerant controls of Han Chinese ethnicity. RESULTS: In the primary analysis, nine variants reached genome-wide significance (p < 5e-08), one in the carbamazepine subanalysis (85 cases vs. 77 controls) and a further eight identified in HLA-B*15:02-negative subanalysis (35 cases and 53 controls). Interaction analysis between each novel variant from the primary analysis found that five increased risk irrespective of HLA-B*15:02 status or zygosity. HLA-B*15:02-positive individuals were found to have reduced risk if they also carried a chromosome 12 variant, chr12.9426934 (heterozygotes: relative risk = .71, p = .001; homozygotes: relative risk = .23, p < .001). All significant variants lie within intronic or intergenic regions with poorly understood functional consequence. In silico functional analysis of suggestive variants (p < 5e-6) identified through the primary and subanalyses (stratified by HLA-B*15:02 status and drug exposure) suggests that genetic variation within regulatory DNA may contribute to risk indirectly by disrupting the regulation of pathology-related genes. The genes implicated were specific either to the primary analysis (CD9), HLA-B*15:02 carriers (DOCK10), noncarriers (ABCA1), carbamazepine exposure (HLA-E), or phenytoin exposure (CD24). SIGNIFICANCE: We identified variants that could explain why some carriers of HLA-B*15:02 tolerate treatment, and why some noncarriers develop ASM-induced SJS/TEN. Additionally, this analysis suggests that the mixing of HLA-B*15:02 carrier status in previous studies might have masked variants contributing to susceptibility, and that inheritance of risk for ASM-induced SJS/TEN is complex, likely involving multiple risk variants.

Alpha-lipoic acid analogues in the regulation of redox balance in epilepsy: A molecular docking and simulation study.

OBJECTIVE: Oxidative stress is one of the pathophysiological mechanisms implicated in drug-resistant epilepsy. Recurrent seizures and prolonged treatment with anti-seizure medicines (ASMs) can produce reactive oxygen species (ROS) resulting in neuronal cell damage, cell toxicity, and cell death. This damage may contribute to the loss of efficacy of anti-seizure medicines. Add-on therapy with antioxidants, neuroimmunophilins, and polyphenols may thus be beneficial in drug-resistant epilepsy. In vitro and in vivo studies have shown a significant improvement in drug efficacy and seizure suppression using co-treatment of anti-seizure medication with naturally available antioxidants including alpha-lipoic acid (α-lipoic acid) from walnut; however, the underlying mechanisms of action remain to be fully understood. METHODS: We undertook molecular docking and molecular dynamics simulations to determine whether alpha-lipoic acid and related analogues interacted with the human manganese superoxide dismutase (MnSOD) protein, a member of the oxidative metabolic pathway. The 3D structure of the compounds and the protein were retrieved from protein and chemical databases, binding sites were identified and ligand-protein interactions were performed. RESULTS: Alpha-lipoic acid and various analogues docked within a human MnSOD binding region. Docking results were validated by molecular dynamic simulation. The CMX-2043 analogue showed strong binding with MnSOD compared to alpha-lipoic acid and other analogues. SIGNIFICANCE: Our findings provide new insights into additional mechanisms of action, which may in part, account for the antioxidant properties associated with alpha-lipoic acid and related analogues. The results support further in vitro and in vivo evaluation of these compounds to better understand their potential as add-on therapy for ASM treatment in epilepsy.

ggVolcanoR: A Shiny app for customizable visualization of differential expression datasets.

Volcano and other analytical plots (e.g., correlation plots, upset plots, and heatmaps) serve as important data visualization methods for transcriptomic and proteomic analyses. Customizable generation of these plots is fundamentally important for a better understanding of dysregulated expression data and is therefore instrumental for the ensuing pathway analysis and biomarker identification. Here, we present an R-based Shiny application, termed ggVolcanoR, to allow for customizable generation and visualization of volcano plots, correlation plots, upset plots, and heatmaps for differential expression datasets, via a user-friendly interactive interface in both local executable version and web-based application without requiring programming expertise. Compared to currently existing packages, ggVolcanoR offers more practical options to optimize the generation of publication-quality volcano and other analytical plots for analyzing and comparing dysregulated genes/proteins across multiple differential expression datasets. In addition, ggVolcanoR provides an option to download the customized list of the filtered dysregulated expression data, which can be directly used as input for downstream pathway analysis. The source code of ggVolcanoR is available at https://github.com/KerryAM-R/ggVolcanoR and the webserver of ggVolcanoR 1.0 has been deployed and is freely available for academic purposes at https://ggvolcanor.erc.monash.edu/.

Experimentally evoked pain in Alzheimer's disease.

BACKGROUND: Pain continues to be underrecognized and undertreated in Alzheimer's disease (AD) while existing guidance about pain assessment and management in dementia is not widespread. Brain regions involved in pain processing and modulation are damaged during AD, and the pain experience in AD is not well understood. Experimental pain studies using psychophysics can further our understanding of the pain experience in AD, which may lead to improved assessment and management of pain in people living with AD. OBJECTIVE: A systematic review was conducted to explicate the current understanding of experimentally evoked pain in AD from primary research using psychophysical methods. DATA SOURCES: Peer-reviewed publications were found via PubMed, CINAHL, and PsycINFO. A total of 18 primary research, peer-reviewed full articles that met inclusion criteria were included, representing 929 total participants. CONCLUSIONS: Experimentally evoked pain in people with AD demonstrates that despite cognitive impairment and a reduced ability to effectively communicate, individuals with AD experience pain similar to or more unpleasant than cognitively intact older adults. This may mean amplified pain unpleasantness in people with AD. IMPLICATIONS FOR PRACTICE: Our current best practices need to be widely disseminated and put into clinical practice. Self-report of pain continues to be the gold standard, but it is ineffective for noncommunicative patients and those unable to understand pain scales or instructions because of memory/cognitive impairment. Instead, pain treatment should be ethically initiated based on patient reports and behaviors, caregiver/surrogate reports, review of the medical record for painful conditions, analgesic trials, and regular reassessments.

Prenatal valproate exposure and adverse neurodevelopmental outcomes: Does sex matter?

OBJECTIVE: Prenatal exposure to the antiepileptic drug (AED) valproic acid (VPA) is associated with an increased risk of impaired postnatal neurodevelopment, including autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD). We aimed to evaluate the influence of sex and drug dosage on the association between prenatal VPA exposure and postnatal behavioral outcomes. METHODS: The Australian Pregnancy Register of AEDs was interrogated to identify children aged 4-11 years prenatally exposed to AEDs. Parents reported on their child's behavior using the Autism Spectrum Quotient-Children's Version and the National Institute for Children's Health Quality Vanderbilt Assessment Scale for ADHD. General linear mixed-effects models were used to investigate the relationship between clinicodemographic variables and psychometric scores. RESULTS: A total of 121 children were studied: 54 prenatally exposed to VPA (28 males, 26 females; mean dose ± SD: 644 ± 310 mg/day) and 67 exposed to other AEDs. There was a main effect of sex showing higher ASD scores in males compared to females (p = .006). An interaction between sex and VPA exposure revealed that males had higher ASD symptoms among children exposed to AEDs other than VPA (p = .01); however, this typical sex dynamic was not evident in VPA-exposed children. There was no evidence of any dose-response relationship between VPA exposure and ASD symptoms. Males had higher ADHD scores compared to females, but there was no evidence for a link between ADHD symptoms and VPA exposure. SIGNIFICANCE: Prenatal VPA exposure seems to negate the usual male sex-related predominance in the incidence of ASD. These initial findings deepen the concept of VPA as a "behavioral teratogen" by indicating that its effect might be influenced by sex, with females appearing particularly sensitive to the effects of VPA. No association between VPA doses and adverse postnatal behavioral outcomes was detected, possibly related to the low VPA doses used in this study.

Sex Differences in the Risk of Cutaneous Adverse Drug Reactions Induced by Antiseizure Medications: A Systematic Review and Meta-analysis.

BACKGROUND: Cutaneous adverse drug reactions (cADRs) are one of the most common, severe, and life-threatening types of adverse reactions following treatment with antiseizure medications (ASMs). Some studies have reported a higher incidence of ASM-induced cADRs in females than in males. OBJECTIVE: This study sought to perform a systematic review, meta-analysis, and meta-regression to compare the ASM cADR risks between females and males. METHODS: We searched the literature using three databases (EMBASE, PubMed, and Web of Science) between October 1998 and November 2018, later updated to October 2019. Studies were included in the meta-analysis if they met the following criteria: (1) observational studies that estimated the incidence of cADRs related to ASMs; (2) provided the risk or odds ratio (OR) for cADRs among female and male patients exposed to ASMs; and (3) provided information on patients' characteristics. We assessed the impact of study characteristics, publication bias, and measures to reduce bias, and performed a DerSimonian and Laird random effects meta-analysis. RESULTS: We included 28 studies in this review. Of these, seven studies were eligible for inclusion in the meta-analysis, involving a total of 223,209 patients. Overall, females were more likely to develop cADRs to ASMs than males (OR 1.76, 95% confidence interval [CI] 1.55-1.99). The largest differences were observed in patients prescribed lamotrigine (OR 2.17, 95% CI 1.53-3.08, p < 0.001) and carbamazepine (OR 1.63, 95% CI 1.02-2.60, p = 0.042). Also, the OR trended higher for phenytoin (OR 2.46, 95% CI 0.79-7.65, p = 0.12), followed by oxcarbazepine (OR 1.91, 95% CI 0.75-4.85, p = 0.18) and sodium valproate (OR 0.60, 95% CI 0.12-2.99, p = 0.53), but the difference did not reach statistical significance. In the remaining 21 studies, 13 reported numerically higher risk of cADRs among females compared to male patients, and in five of these, the difference was statistically significant. CONCLUSION: Our findings confirmed that females are more susceptible to cADRs induced by ASMs than males. More research is needed to understand the pathophysiological mechanisms for this difference. PROTOCOL REGISTRATION: PROSPERO (CRD42018111943).