Short- and long-term outcomes of immediate and delayed treatment in epilepsy diagnosed after one or multiple seizures

Clinical prediction models for treatment outcomes in newly diagnosed epilepsy: A systematic review

Up to 35% of individuals diagnosed with epilepsy continue to have seizures despite treatment, commonly referred to as drug-resistant epilepsy. Uncontrolled seizures can directly, or indirectly, negatively impact an individual's quality of life. To inform clinical management and life decisions, it is important to be able to predict the likelihood of seizure control. Those likely to achieve seizure control will be able to return sooner to their usual work and leisure activities and require less follow-up, whereas those with a poor prognosis will need more frequent clinical attendance and earlier consideration of epilepsy surgery. This is a systematic review aimed at identifying demographic, clinical, physiological (e.g., electroencephalographic), and imaging (e.g., magnetic resonance imaging) factors that may be predictive of treatment outcomes in patients with newly diagnosed epilepsy (NDE). MEDLINE and Embase were searched for prediction models of treatment outcomes in patients with NDE. Study characteristics were extracted and subjected to assessment of risk of bias (and applicability concerns) using the PROBAST (Prediction Model Risk of Bias Assessment Tool) tool. Baseline variables associated with treatment outcomes are reported as prognostic factors. After screening, 48 models were identified in 32 studies, which generally scored low for concerns of applicability, but universally scored high for susceptibility to bias. Outcomes reported fit broadly into four categories: drug resistance, short-term treatment response, seizure remission, and mortality. Prognostic factors were also heterogenous, but the predictors that were commonly significantly associated with outcomes were those related to seizure characteristics/types, epilepsy history, and age at onset. Antiseizure medication response was often included as a baseline variable, potentially obscuring other factor relationships at baseline. Currently, outcome prediction models for NDE demonstrate a high risk of bias. Model development could be improved with a stronger adherence to recommended TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) practices. Furthermore, we outline actionable changes to common practices that are intended to improve the overall quality of prediction model development in NDE.

Long-term physical and psychiatric morbidities and mortality of untreated, deferred, and immediately treated epilepsy

OBJECTIVE

In Australia, 30% of newly diagnosed epilepsy patients were not immediately treated at diagnosis. We explored health outcomes between patients receiving immediate, deferred, or no treatment, and compared them to the general population.

METHODS

Adults with newly diagnosed epilepsy in Western Australia between 1999 and 2016 were linked with statewide health care data collections. Health care utilization, comorbidity, and mortality at up to 10 years postdiagnosis were compared between patients receiving immediate, deferred, and no treatment, as well as with age- and sex-matched population controls.

RESULTS

Of 603 epilepsy patients (61% male, median age = 40 years) were included, 422 (70%) were treated immediately at diagnosis, 110 (18%) received deferred treatment, and 71 (12%) were untreated at the end of follow-up (median = 6.8 years). Immediately treated patients had a higher 10-year rate of all-cause admissions or emergency department presentations than the untreated (incidence rate ratio [IRR] = 2.0, 95% confidence interval [CI] = 1.4-2.9) and deferred treatment groups (IRR = 1.7, 95% CI = 1.0-2.8). They had similar 10-year risks of mortality and developing new physical and psychiatric comorbidities compared with the deferred and untreated groups. Compared to population controls, epilepsy patients had higher 10-year mortality (hazard ratio = 2.6, 95% CI = 2.1-3.3), hospital admissions (IRR = 2.3, 95% CI = 1.6-3.3), and psychiatric outpatient visits (IRR = 3.2, 95% CI = 1.6-6.3). Patients with epilepsy were also 2.5 (95% CI = 2.1-3.1) and 3.9 (95% CI = 2.6-5.8) times more likely to develop a new physical and psychiatric comorbidity, respectively.

SIGNIFICANCE

Newly diagnosed epilepsy patients with deferred or no treatment did not have worse outcomes than those immediately treated. Instead, immediately treated patients had greater health care utilization, likely reflecting more severe underlying epilepsy etiology. Our findings emphasize the importance of individualizing epilepsy treatment and recognition and management of the significant comorbidities, particularly psychiatric, that ensue following a diagnosis of epilepsy.

Treatment Outcomes of Newly Diagnosed Epilepsy: A Systematic Review and Meta-analysis

BACKGROUND AND OBJECTIVES

Understanding the multi-faceted treatment outcomes of newly diagnosed epilepsy is critical for developing rational therapeutic strategies. A meta-analysis was conducted to derive pooled estimates of a range of seizure outcomes in children and adults with newly diagnosed epilepsy commenced on antiseizure medication treatment, and to identify factors associated with different outcomes.

METHODS

PubMed/EMBASE were screened for eligible articles between 1 January, 1995 and 1 May, 2021 to include unselected cohort studies with a ≥ 12-month follow-up of seizure outcomes. Proportions of patients seizure free at different follow-up timepoints and their characteristics at the study population level were extracted. The patients were group-wise aggregated using a random-effects model. Primary outcomes were proportions of patients with cumulative 1-year seizure freedom (C1YSF), and 1-year and 5-year terminal seizure freedom (T1YSF and T5YSF). Secondary outcomes included the proportions of patients with early sustained seizure freedom, drug-resistant epilepsy and seizure-free off antiseizure medication at the last follow-up (off antiseizure medications). A separate random-effects meta-analysis was performed for nine predictors of importance.

RESULTS

In total, 39 cohorts (total n = 21,139) met eligibility criteria. They included 15 predominantly adult cohorts (n = 12,024), 19 children (n = 6569), and 5 of mixed-age groups (n = 2546). The pooled C1YSF was 79% (95% confidence interval [CI] 74-83). T1YSF was 68% (95% CI 63-72) and T5YSF was 69% (95% CI 62-75). Children had higher C1YSF (85% vs 68%, p < 0.001) and T1YSF than adult cohorts (74% vs 61%, p = 0.007). For secondary outcomes, 33% (95% CI 27-39) of patients achieved early sustained seizure freedom, 17% (95% CI 13-21) developed drug resistance, and 39% (95% CI 30-50) were off antiseizure medications at the last follow-up. Studies with a longer follow-up duration correlated with higher C1YSF (p < 0.001) and being off antiseizure medications (p = 0.045). Outcomes were not associated with study design (prospective vs retrospective), cohort size, publication year, or the earliest date of recruitment. Predictors of importance in newly diagnosed epilepsy include etiology, epilepsy type, abnormal diagnostics (neuroimaging, examination, and electroencephalogram findings), number of seizure types, and pre-treatment seizure burden.

CONCLUSIONS

Seizure freedom is achieved with currently available antiseizure medications in most patients with newly diagnosed epilepsy, yet this is often not immediate, may not be sustainable, and has not improved over recent decades. Symptomatic etiology, abnormal neuro-diagnostics, and increased pre-treatment seizure burden and seizure types are important predictors for unfavorable outcomes in newly diagnosed epilepsy. The study findings may be used as a quantitative benchmark on the efficacy of future antiseizure medication therapy for this patient population.

Development and Validation of a Deep Learning Model for Predicting Treatment Response in Patients With Newly Diagnosed Epilepsy

Importance

Selection of antiseizure medications (ASMs) for epilepsy remains largely a trial-and-error approach. Under this approach, many patients have to endure sequential trials of ineffective treatments until the "right drugs" are prescribed.

Objective

To develop and validate a deep learning model using readily available clinical information to predict treatment success with the first ASM for individual patients.

Design, Setting, and Participants

This cohort study developed and validated a prognostic model. Patients were treated between 1982 and 2020. All patients were followed up for a minimum of 1 year or until failure of the first ASM. A total of 2404 adults with epilepsy newly treated at specialist clinics in Scotland, Malaysia, Australia, and China between 1982 and 2020 were considered for inclusion, of whom 606 (25.2%) were excluded from the final cohort because of missing information in 1 or more variables.

Exposures

One of 7 antiseizure medications.

Main Outcomes and Measures

With the use of the transformer model architecture on 16 clinical factors and ASM information, this cohort study first pooled all cohorts for model training and testing. The model was trained again using the largest cohort and externally validated on the other 4 cohorts. The area under the receiver operating characteristic curve (AUROC), weighted balanced accuracy, sensitivity, and specificity of the model were all assessed for predicting treatment success based on the optimal probability cutoff. Treatment success was defined as complete seizure freedom for the first year of treatment while taking the first ASM. Performance of the transformer model was compared with other machine learning models.

Results

The final pooled cohort included 1798 adults (54.5% female; median age, 34 years [IQR, 24-50 years]). The transformer model that was trained using the pooled cohort had an AUROC of 0.65 (95% CI, 0.63-0.67) and a weighted balanced accuracy of 0.62 (95% CI, 0.60-0.64) on the test set. The model that was trained using the largest cohort only had AUROCs ranging from 0.52 to 0.60 and a weighted balanced accuracy ranging from 0.51 to 0.62 in the external validation cohorts. Number of pretreatment seizures, presence of psychiatric disorders, electroencephalography, and brain imaging findings were the most important clinical variables for predicted outcomes in both models. The transformer model that was developed using the pooled cohort outperformed 2 of the 5 other models tested in terms of AUROC.

Conclusions and Relevance

In this cohort study, a deep learning model showed the feasibility of personalized prediction of response to ASMs based on clinical information. With improvement of performance, such as by incorporating genetic and imaging data, this model may potentially assist clinicians in selecting the right drug at the first trial.

Risk factors and consequences of self-discontinuation of treatment by patients with newly diagnosed epilepsy

OBJECTIVE

Patients with epilepsy not uncommonly self-discontinue treatment with antiseizure medications (ASM). The rate, reasons for this, and consequences have not been well studied.

METHODS

We analyzed self-discontinuation of ASM treatment in patients with recently diagnosed epilepsy via review of clinic letters and hospital correspondence in a prospective cohort of first seizure patients.

RESULTS

We studied 489 patients with newly diagnosed and treated epilepsy (median age 41, range 14-88, 62% male), followed up for a median duration of 3.0 years (interquartile range [IQR]: 1.2-6.0). Seventy eight (16.0%) self-discontinued ASM therapy after a median treatment duration of 1.4 years (IQR: 0.4-2.9), and after a median duration of seizure freedom of 11.8 months (IQR: 4.6-31.8). Patients commonly self-discontinued treatment due to adverse effects (41%), perception that treatment was no longer required (35%), and planned or current pregnancy (12%). Patients who self-discontinued were less likely to have epileptogenic lesions on neuroimaging (hazard ratio [HR] = 0.44, 95% confidence interval [CI]: 0.23-0.83), presentation with seizure clusters (HR = 0.32, 95% CI: 0.14-0.69) and presentation with tonic-clonic seizures (HR = 0.36, 95% CI: 0.19-0.70). Patients with shorter interval since the last seizure (HR = 0.76, 95% CI: 0.66-0.86) were more likely to self-discontinue treatment. Sleep deprivation prior to seizures before diagnosis (HR = 1.80, 95% CI: 1.05-3.09) and significant alcohol or illicit drug use (HR = 2.35, 95% CI: 1.20-4.59) were also associated with higher rates of discontinuation. After discontinuation, 51 patients (65%) experienced seizure recurrence, and 43 (84%) restarted treatment. Twenty two patients (28%) experienced a seizure-related injury after treatment discontinuation.

SIGNIFICANCE

Self-initiated discontinuation of ASM treatment was not uncommon in patients with newly treated epilepsy. Reasons for discontinuation highlight areas for improved discussion with patients, including the chronicity of epilepsy and management strategies for current or potential adverse effects.

Markov modelling of treatment response in a 30-year cohort study of newly diagnosed epilepsy

People with epilepsy have variable and dynamic trajectories in response to antiseizure medications. Accurately modelling long-term treatment response will aid prognostication at the individual level and health resource planning at the societal level. Unfortunately, a robust model is lacking. We aimed to develop a Markov model to predict the probability of future seizure-freedom based on current seizure state and number of antiseizure medication regimens trialled. We included 1,795 people with newly diagnosed epilepsy who attended a specialist clinic in Glasgow, Scotland, between July 1982 and October 2012. They were followed up until October 2014 or death. We developed a simple Markov model, based on current seizure state only, and a more detailed model, based on both current seizure state and number of antiseizure medication regimens trialled. Sensitivity analyses were performed for the regimen-based states model to examine the effect of regimen changes due to adverse effects. The model was externally validated in a separate cohort of 455 newly diagnosis epilepsy patients seen in Perth, Australia, between May 1999 and May 2016. Our models suggested that once seizure-freedom was achieved, it was likely to persist, regardless of the number of antiseizure medications trialled to reach that point. The likelihood of achieving long-term seizure-freedom was highest with the first antiseizure medication regimen, at approximately 50%. The chance of achieving seizure-freedom fell with subsequent regimens. Fluctuations between seizure-free and not seizure-free states were highest earlier on, but decreased with chronicity of epilepsy. Seizure-freedom/recurrence risk tables were constructed with these probability data, similar to cardiovascular risk tables. Sensitivity analyses showed that the general trends and conclusions from the base model were maintained despite perturbing the model and input data with regimen changes due to adverse effects. Quantitative comparison with the external validation cohort showed excellent consistency at year 1, good at year 3 and moderate at year 5. Quantitative models, as used in this study, can provide pertinent clinical insights that are not apparent from simple statistical analysis alone. Attaining seizure freedom at any time in a patient's epilepsy journey will confer durable benefit. Seizure-freedom risk tables may be used to individualise the prediction of future seizure control trajectory.