This study investigated the use of machine learning (ML) methods to identify a genetic signature for predicting response to fingolimod (FTY) treatment in relapsing-remitting multiple sclerosis (RRMS) patients. Two cohorts of RRMS patients were analyzed, one from Italy (OSR) and one from France (CHUT). ML models were trained using clinical and genetic data, with a focus on Random Forests. The study identified a 123-SNP genetic signature that predicted FTY response (AUROC = 0.65), with improved prediction when clinical data was added (AUROC = 0.71). While the predictive accuracy wasn't sufficient for clinical practice, the findings suggest that combining clinical and genetic data via ML can aid in predicting FTY response.
Publisher
Journal of Personalized Medicine
Published On
Jan 06, 2023
Authors
Ferrè, L, Clarelli, F, Pignolet, B, Mascia, E, Frasca, M, Santoro, S, Sorosina, M, Bucciarelli, F, Moiola, L, Martinelli, V, Comi, G, Liberatore, G, Filippi, M, Esposito, F
Tags
machine learning
genetic signature
fingolimod
multiple sclerosis
predictive modeling
relapsing-remitting
clinical data
Related Publications
Explore these studies to deepen your understanding of the subject.
