On 16 December, researchers from Germany and the USA published a study in Brain Communications exploring how artificial intelligence (AI) can enhance clinical dementia trials. Dementia due to Alzheimer’s disease (AD) progresses differently among patients, complicating the identification of effective treatments. The study used deep learning to analyse disease trajectories in 283 early dementia patients, clustering them into two subgroups: ‘slow’ and ‘fast’ progressors. These findings were validated in a larger cohort of 2,779 patients. The researchers trained a machine learning model to predict subgroup progression using data from patients’ dementia diagnoses. The classifier demonstrated robust predictive accuracy, with an area under the receiver operating characteristic curve of 0.70 ± 0.01 during external validation. The team simulated a clinical trial enriched with patients predicted to have faster progression. This approach decreased required sample sizes, reduced trial costs by over 13%, and improved trial success rates. The resources saved could expedite drug development and expand efforts to address cognitive impairment. This study highlights the potential of AI to advance precision medicine in AD by streamlining clinical trials and improving their efficiency.
Read the paper here: https://doi.org/10.1093/braincomms/fcae445