EPILEPSY is one of the most common and disabling chronic neurologic conditions, affecting 70 million people globally. Despite this, the selection of antiseizure medications remains a process of trial and error, with clinicians unable to predict which drug a particular patient will respond to. Now, a team led by researchers from Monash University, Melbourne, Australia, has demonstrated the potential of an artificial intelligence model to predict treatment success with the first antiseizure medication for individual patients.
Senior author Patrick Kwan from the Monash Central Clinical School’s Department of Neuroscience summarised the wider relevance of the study: “If the patient doesn’t respond to the first treatment, quite a few will respond to the second or third one, meaning that they might have become seizure-free sooner if the ‘right’ drug was chosen at the outset.” Kwan added: “If they get the wrong medication they still have seizures and may also get side-effects from it.”
The team developed a predictive model on a pooled cohort of 1,798 adults with newly diagnosed epilepsy from five healthcare centres across Australia, Malaysia, China, and the UK.
Kwan noted that the accuracy of the deep learning model in predicting the best medication was “modest.” Indeed, the model that was trained using the pooled cohort had an area under the receiver operating characteristic curve of 0.65. Even so, Kwan emphasised: “That was more than what we expected,” especially since “only very basic clinical factors collected in routine clinical care were used to train this base model.”
Study author Zhibin Chen highlighted the significance of the research: “This is believed to be a world-first model. It assures the predictability of choosing the optimal treatment for patients with newly diagnosed epilepsy. It will open the gate for personalizing the management of epilepsy.”
Going forward, the performance of the model will be improved through the incorporation of genetic and imaging data. The enhanced version will then be tested in the Personalised Selection of Medication for Newly Diagnosed Adult Epilepsy (PERSONAL) trial, a national, multicentre, randomised controlled study. Ultimately, it is hoped that this model will improve the management and treatment of epilepsy and assist practitioners in selecting the appropriate drug at the first trial.
