Gray Matter-Based Models Predict No Evidence of Disease Activity Status in Patients with Relapsing-Remitting Multiple Sclerosis

Motivation: Prediction of achieving no evidence of disease activity (NEDA) status in relapsing-remitting multiple sclerosis (RRMS) patients on oral disease-modifying therapies (DMTs) is crucial for clinical decision-making and patient management. Goal(s): To develop a clinical and regional gray matter atrophy variables-based machine learning model to predict achieving NEDA status in RRMS patients on oral DMTs. Approach: Composite performance scores were calculated based on eight standard performance metrics ranking to screen out the optimal model among five models, and risk stratification analysis was performed. Results: The logistic regression (LR) model demonstrated the best performance and can distinguish well between high- and low-risk RRMS individuals. Impact: Baseline clinical and regional gray matter atrophy variables-based machine learning models can help physicians predict NEDA status in RRMS patients on oral DMTs for clinical decision-making and patient management.