A first unprovoked seizure is a common presentation, 10% of the population will have at least one seizure and approximately 50% will have a recurrence. It remains a major challenge for neurologists to reliably identify those that will have a recurrent seizure, creating uncertainty for both patients and clinicians. This uncertainty is associated with serious physical, psychological and social consequences for patients, with significant impacts on their driving and future employment prospects.

This fellowship will be the first to combine and explore the utility of using serum, quantitative EEG and quantitative MRI biomarkers to predict seizure recurrence. Patients with a first unprovoked seizure will be recruited and serum samples will be collected for measurement of circulating biomarkers, a resting-state EEG will be performed for computational analysis and we will acquire advanced quantitative structural MR imaging sequences to investigate brain connectivity and network dynamics. Patients will then be followed up at various time points to assess for seizure recurrence. The primary outcome event is a seizure recurrence, and the primary analysis will be of time to seizure recurrence using a multivariable regression model.

The identification of reliable biomarkers of seizure recurrence following a first unprovoked seizure will identify potential mechanisms associated with seizure predisposition and epileptogenesis. It will also better inform patient stratification for counselling and treatment decisions as well as optimising the recruitment of high-risk patients to future clinical trials of novel disease modifying agents, with the ultimate goal of improving the natural history of epilepsy.