Identifying patients who will not reachieve remission after breakthrough seizures

Predicting efficacy of antiseizure medication treatment with machine learning algorithms in North Indian population

PURPOSE

This study aimed to develop a classifier using supervised machine learning to effectively assess the impact of clinical, demographical, and biochemical factors in accurately predicting the antiseizure medications (ASMs) treatment response in people with epilepsy (PWE).

METHODS

Data was collected from 786 PWE at the Outpatient Department of Neurology, Institute of Human Behavior and Allied Sciences (IHBAS), New Delhi, India from 2005 to 2015. Patients were followed up at the 2nd, 4th, 8th, and 12th month over the span of 1 year for the drugs being administered and their dosage, the serum drug levels, the frequency of seizure control, drug efficacy, the adverse drug reactions (ADRs), and their compliance to ASMs. Several features, including demographic details, medical history, and auxiliary examinations electroencephalogram (EEG) or Computed Tomography (CT) were chosen to discern between patients with distinct remission outcomes. Remission outcomes were categorized into 'good responder (GR)' and 'poor responder (PR)' based on the number of seizures experienced by the patients over the study duration. Our dataset was utilized to train seven classical machine learning algorithms i.e Extreme Gradient Boost (XGB), K-Nearest Neighbor (KNN), Support Vector Classifier (SVC), Decision Tree (DT), Random Forest (RF), Naïve Bayes (NB) and Logistic Regression (LR) to construct classification models.

RESULTS

Our research findings indicate that 1) among the seven algorithms examined, XGB and SVC demonstrated superior predictive performances of ASM treatment outcomes with an accuracy of 0.66 each and ROC-AUC scores of 0.67 (XGB) and 0.66 (SVC) in distinguishing between PR and GR patients. 2) The most influential factor in discerning PR to GR patients is a family history of seizures (no), education (literate) and multitherapy with Chi-square (χ2) values of 12.1539, 8.7232 and 13.620 respectively and odds ratio (OR) of 2.2671, 0.4467, and 1.9453 each. 3). Furthermore, our surrogate analysis revealed that the null hypothesis for both XGB and SVC was rejected at a 100 % confidence level, underscoring the significance of their predictive performance. These findings underscore the robustness and reliability of XGB and SVC in our predictive modelling framework.

SIGNIFICANCE

Utilizing XG Boost and SVC-based machine learning classifier, we successfully forecasted the likelihood of a patient's response to ASM treatment, categorizing them as either PR or GR, post-completion of standard epilepsy examinations. The classifier's predictions were found to be statistically significant, suggesting their potential utility in improving treatment strategies, particularly in the personalized selection of ASM regimens for individual epilepsy patients.

Risk of breakthrough seizures depends on type and etiology of epilepsy

OBJECTIVE

This study was undertaken to analyze whether the rate of breakthrough seizures in patients taking antiseizure medication (ASM) who have been seizure-free for at least 12 months varies among different types and etiologies of epilepsy. Given the relative ease of achieving seizure freedom with ASM in patients with post-ischemic stroke epilepsy, we hypothesized that this etiology is associated with a reduced risk of breakthrough seizures.

METHODS

We defined a breakthrough seizure as an unprovoked seizure occurring while the patient was taking ASM after a period of at least 12 months without seizures. Data were analyzed retrospectively from a tertiary epilepsy outpatient clinic. Patients were eligible for inclusion if they either had a breakthrough seizure at any time or a seizure-free interval of at least 2 years. Our primary endpoint was rate of breakthrough seizures. We conducted univariable and multivariable analyses to identify variables associated with breakthrough seizures.

RESULTS

Of 521 patients (53% females, median age = 49 years) included, 29% had a breakthrough seizure, which occurred after a median seizure-free interval of 34 months (quartiles = 22, 62). When controlling for clinically relevant covariates, breakthrough seizures were associated with post-ischemic stroke epilepsy (odds ratio [OR] = .267, 95% confidence interval [CI] = .075-.946), genetic generalized epilepsy (OR = .559; 95% CI = .319-.978), intellectual disability (OR = 2.768, 95% CI = 1.271-6.031), and the number of ASMs previously and currently tried (OR = 1.203, 95% CI = 1.056-1.371). Of the 151 patients with breakthrough seizures, 34.3% did not reachieve terminal 12-month seizure freedom at the last visit.

SIGNIFICANCE

This is the first study to show an association between type and etiology of epilepsy and risk of breakthrough seizures. Our data suggest that epilepsies in which seizure freedom can be obtained more easily also exhibit a lower risk of breakthrough seizures. These findings may help to better counsel seizure-free patients on their further seizure prognosis.

Molecularly imprinted polymer-based extended-gate field-effect transistor chemosensors for selective determination of antiepileptic drug

Ultrathin molecularly imprinted polymer (MIP) films were deposited on the surfaces of ZnO nanorods (ZNRs) and nanosheets (ZNSs) by electropolymerization to afford extended-gate field-effect transistor sensors for detecting phenytoin (PHT) in plasma. Molecular imprinting efficiency was optimized by controlling the contents of functional monomers and the template in the precursor solution. PHT sensing was performed in plasma solutions with various concentrations by monitoring the drain current as a function of drain voltage under an applied gate voltage of 1.5 V. The reliability and reproducibility of the fabricated sensors were evaluated through a solution treatment process for complete PHT removal and PHT adsorption-removal cycling, while selectivity was examined by analyzing responses to chemicals with structures analogous to that of PHT. Compared with the ZNS/extracted-MIP sensor and sensors with non-imprinted polymer (NIP) films, the ZNR/extracted-MIP sensor showed superior responses to PHT-containing plasma due to selective PHT adsorption, achieving an imprinting factor of 4.23, detection limit of 12.9 ng/mL, quantitation limit of 53.0 ng/mL, and selectivity coefficients of 3-4 (against tramadol) and ~ 5 (against diphenhydramine). Therefore, we believe that the MIP-based ZNR sensing platform is promising for the practical detection of PHT and other drugs and evaluation of their proper dosages.

Treatment Outcome and Risk Factors of Adult Newly Diagnosed Epilepsy: A Prospective Hospital-Based Study in Northeast China

Objective: The study was conducted to summarize the treatment outcomes of newly diagnosed epilepsy (NDE) and analyse the risk factors for refractory epilepsy (RE) in Northeast China.

Methods: A total of 466 adult patients with NDE were consecutively enrolled in this programme. Clinical data were collected at baseline and each follow-up. Several scales concerning recognition and mood were also completed at the first visit.

Results: Seizure-free status was achieved by 52% (n = 244) of the patients; however, 15% (n = 68) manifested RE. A total of 286 (61%) patients continued with the first ASM as monotherapy, among which 186 (40%) patients became seizure-free. Fifteen (22%) patients with RE became seizure-free following ASM adjustment and 34 patients (14%) had breakthrough seizures after being classified as seizure-free. One patient developed RE after attaining seizure-free status. Breakthrough seizures during the first expected interictal interval [Odds ratio (OR) = 5.81, 95% CI: 2.70–12.50], high seizure frequency at baseline (OR = 1.24, 95% CI: 1.04–1.49), younger age of onset (OR = 1.42, 95% CI: 1.12–1.79), and male sex (OR = 2.64, 95% CI: 1.26–5.53) were risk factors for RE.

Significance: Treatment outcomes of the majority of NDE cases are good. New risk factors could help physicians more promptly and accurately identify patients who are likely to develop RE. Seizure-free state is not long enough to commence the withdrawal of ASMs. RE is not permanent and seizure-free may be achieved subsequently by appropriate drug adjustment.

Course and prognosis of adult-onset epilepsy in Brazil: A cohort study

BACKGROUND

Most of the epilepsy longitudinal studies have analyzed children. However, in endemic regions, such as Brazil, neurocysticercosis accounts for many adult-onset epilepsy cases. So, the main objective of this study was to identify the clinical predictors associated with drug-resistant adult-onset epilepsy in Brazil during a long-term follow-up.

METHODS

We followed 302 individuals with adult-onset epilepsy for 9.8 years in our University Hospital. Structured questionnaires about drug-resistant epilepsy were applied. The presence of drug-resistant epilepsy was the primary outcome. We used multilevel linear modeling in our data analysis.

RESULTS

Overall 47 (15.6%) individuals presented drug-resistant epilepsy and the etiology was structural in 70.2% of them, while infectious etiology was present in 8.5% of this group. Infectious etiology occurred in 25.9% (n = 66) of the patients from the nondrug-resistant group. Those with developmental delay were two times more likely to present seizures. Structural epilepsy etiology was associated with an increased chance of relapsing. Poor school performance and abnormal electroencephalogram were also associated with an increased chance of seizures.

CONCLUSION

The course of epilepsy was favorable in the majority of our patients, and drug-resistant epilepsy rates were similar to those found in other studies, although we evaluated older individuals with higher levels of infectious etiology. Also, we found that neurocysticercosis was associated with well-controlled epilepsy, while structural epilepsy was directly related to the occurrence of seizures. We also hypothesized that the smaller size of lesions found in neurocysticercosis could contribute to better treatment response.

Identifying patients who will not reachieve remission after breakthrough seizures

Objective

We aim to identify people with epilepsy who are unlikely to reachieve a 12‐month remission within 2 years after experiencing a breakthrough seizure following an initial 12‐month remission.

Methods

We apply a novel longitudinal discriminant approach to data from the Standard and New Antiepileptic Drugs study to dynamically predict the risk of a patient not achieving a second remission after a breakthrough seizure by combining both baseline covariates (collected at the time of breakthrough seizure) and follow‐up data.

Results

The model classifies 83% of patients. Of these, 73% of patients (95% confidence interval [CI] = 58%‐88%) who did not achieve a second remission were correctly identified (sensitivity), and 84% of patients (95% CI = 69%‐96%) who achieved a second remission were correctly identified (specificity). The area under the curve from our model was 87% (95% CI = 80%‐94%). Patients who did not achieve a second remission were correctly identified on average after 10 months of observation postbreakthrough. Occurrence of seizures after breakthrough and the number of seizures experienced were the most informative longitudinal variables. These longitudinal profiles were influenced by the following baseline covariates: age at breakthrough seizure, presence of neurological insult, and number of antiepileptic drugs required to achieve first remission.

Significance

Using longitudinal data gathered during patient follow‐up allows more accurate predictions than using baseline covariates in a standard Cox model. The model developed in this paper is a useful first step in developing a tool for identifying patients who develop drug resistance after an initial remission.