2413 Does serum neurofilament light chain level contribute to the prediction of treatment response in multiple sclerosis?

• Published in: BMJ neurology open 2022-08, Vol.4 (Suppl 1), p.A56-A57
• Main Authors: Moradi, Nahid, Sharmin, Sifat, Malpas, Charles B, Kuhle, Jens, Benkert, Pascal, Leppert, David, Havrdová, Eva, Horáková, Dana, Kleinová, Pavlina, Uher, Tomas, Hillert, Jan, Olsson, Tomas, Manouchehrinia, Ali, Taylor, Bruce V, Barnett, Michael H, Kilpatrick, Trevor J, Kalincik, Tomas
• Format: Article
• Language: English
• Subjects: Abstracts; Accuracy; Multiple sclerosis
• ISSN: 2632-6140
• DOI: 10.1136/bmjno-2022-ANZAN.151

ObjectivesThe Crystal Ball model is a previously published and externally validated algorithm.1 that helps predict the individual response to Multiple Sclerosis (MS) therapies. In this study, we investigated the contribution of Neurofilament Light (NfL) chain levels to the prediction of treatment response.MethodsData from 3 participating centers were collated to include patients with relapsing MS, recorded Expanded Disability Status Scale and NfL at baseline, while being treated with or commencing a disease modifying therapy. We used principal component analysis to reduce the dimensionality of the demographic and clinical data. Cox proportional hazards models with three principal components and treatment duration with vs. without NfL at baseline were used to model the risk of relapses, 6-month confirmed disability worsening and 9-month confirmed disability improvement. The accuracy of the NfL and non-NfL models was compared using 10-fold cross-validation and Harrell’s C-index.Results1716 patients across 12 MS therapies were included (68% female, mean age 38±11). The accuracy of the prediction of treatment response in the models without NfL was comparable to the original study. Addition of NfL did not further increase the predictive accuracy (C-indices: relapses 0.59, disability worsening 0.64, disability improvement 0.82).ConclusionSerum NfL, an emerging prognostic marker, does not substantially contribute to differentiating individual response to MS therapies among patients with various clinical and demographic characteristics. Clinical and demographic information remains the most useful indicator of future individual response to MS therapies.ReferenceKalincik T, Manouchehrinia A, Sobisek L, et al. Towards personalized therapy for multiple sclerosis: Prediction of individual treatment response. Brain. 2017;140(9):2426–2443. doi:10.1093/brain/awx185