BACKGROUND: The added value of neurofilament light chain levels in serum (sNfL) to the concept of no evidence of disease activity-3 (NEDA-3) has not yet been investigated in detail. OBJECTIVE: To assess whether combination of sNfL with NEDA-3 status improves identification of patients at higher risk of disease activity during the following year. METHODS: We analyzed 369 blood samples from 155 early relapsing-remitting MS patients on interferon beta-1a. We compared disease activity, including the rate of brain volume loss in subgroups defined by NEDA-3 status and high or low sNfL (> 90th or < 90th percentile). RESULTS: In patients with disease activity (EDA-3), those with higher sNFL had higher odds of EDA-3 in the following year than those with low sNFL (86.5% vs 57.9%; OR = 4.25, 95% CI: [2.02, 8.95]; p = 0.0001) and greater whole brain volume loss during the following year (β = -0.36%; 95% CI = [-0.60, -0.13]; p = 0.002). Accordingly, NEDA-3 patients with high sNfL showed numerically higher disease activity (EDA-3) in the following year compared with those with low sNfL (57.1% vs 31.1%). CONCLUSION: sNfL improves the ability to identify patients at higher risk of future disease activity, beyond their NEDA-3 status. Measurement of sNfL may assist clinicians in decision-making by providing more sensitive prognostic information.
BACKGROUND AND OBJECTIVES: Research in Multiple Sclerosis (MS) has recently focused on extracting knowledge from real-world clinical data sources. This type of data is more abundant than data produced during clinical trials and potentially more informative about real-world clinical practice. However, this comes at the cost of less curated and controlled data sets. In this work we aim to predict disability progression by optimally extracting information from longitudinal patient data in the real-world setting, with a special focus on the sporadic sampling problem. METHODS: We use machine learning methods suited for patient trajectories modeling, such as recurrent neural networks and tensor factorization. A subset of 6682 patients from the MSBase registry is used. RESULTS: We can predict disability progression of patients in a two-year horizon with an ROC-AUC of 0.85, which represents a 32% decrease in the ranking pair error (1-AUC) compared to reference methods using static clinical features. CONCLUSIONS: Compared to the models available in the literature, this work uses the most complete patient history for MS disease progression prediction and represents a step forward towards AI-assisted precision medicine in MS.
- MeSH
- lidé MeSH
- neuronové sítě (počítačové) MeSH
- roztroušená skleróza * MeSH
- strojové učení * MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- časopisecké články MeSH
