Contributions
Abstract: P1181
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Biomarkers
Objective: NfL is a marker for axonal damage while Chi3l1 is related to chronic inflammation. Both have proven value as potential MS biomarkers for disease activity, progression, and prognosis. Their measure in the CSF is widely available in most centers.
Aims: The aim of this work was to investigate the variables that summed to isolated measure of CSF NfL and Chi3l1 could predict disease course of MS patients, and therefore we propose a multivariable logistic regression predictive model.
Methods: NfL and CHI3L1 levels were measured in CSF samples (n=156) with ELISA assay in a cross-sectional MS cohort. Relapse was considered when lumbar puncture (LP) was within 90 days of a clinical attack. Prior to statistical analysis, NfL and CHI3L1 levels were log-transformed. We assessed the association between both variables and in relation to clinical, imaging, treatments, and disability outcomes with parametric comparison tests. Finally, we developed a multivariable logistic regression analysis for predicting disability progression.
Results: A cross-sectional cohort of 156 patients (70% females) was included in this study. Clinical forms distributed in 93 RRMS, 40 SPMS, and 23 PPMS at LP. NfL and Chi3l1 were measured in the CSF of 144 and 156 patients respectively. Median NfL in patients with activity (clinical attack or gadolinium-enhanced lesions) was higher 664,8 pg/ml [125,3-8290] compared to patients without active inflammation, 352,2 pg/ml [45,28-5770] (p< 0,001). Median Chi3l1 was not influenced by inflammation, although a significant increase was found in progressive forms (171,45 [34,84-802,7]) compared to relapsing (115,35 [33,60-507,4]) (p< 0,000). Progressive forms with clinical or radiological inflammatory activity (SPMS) had higher NF concentrations (486,80 pg/ml vs. 1108 pg/ml, p< 0,04). Current therapies at time of LP did not influence significantly NfL or Chi3l1. Other demographic and clinical variables affecting NfL and Chi3l1 were investigated. A multivariable logistic regression model was adjusted with NfL and Chi3l1, influencing variables (age, IgM-OCB, disease duration) and EDSS longitudinal data, resulting an accurate tool to predict disability progression at 10 to 20 years from LP, with an adjusted R-squared of 0.81.
Conclusion: Isolated simultaneous measure of NfL and Chi3L in CSF informs us about disease activity in relapsing and progressive forms of disease, but it is also a valuable tool to predict disability progression.
Disclosure: SGP is funded by Health Institute Carlos III Rio Hortega CM12/00014. LCN, JCV, JGM, SC, ME DHM: nothing to disclose. BCE serves on the scientific advisory board for Novartis, Genzyme, TEVA and Merck Pharmaceuticals, and is funded by Health Institute Carlos III, grant number PS13/00663 (BC) and a local grant from Hospital La Fe VlcBiomed (UV/IIS la Fe #33-2015-A). FPM serves on the scientific advisory board for Novartis, Genzyme, TEVA, Almirall, and Roche Pharmaceuticals. CAV serves on the scientific advisory board for Genzyme. FG serves on the scientific advisory board for Genzyme and Roche Pharmaceuticals.
Abstract: P1181
Type: Poster Sessions
Abstract Category: Pathology and pathogenesis of MS - Biomarkers
Objective: NfL is a marker for axonal damage while Chi3l1 is related to chronic inflammation. Both have proven value as potential MS biomarkers for disease activity, progression, and prognosis. Their measure in the CSF is widely available in most centers.
Aims: The aim of this work was to investigate the variables that summed to isolated measure of CSF NfL and Chi3l1 could predict disease course of MS patients, and therefore we propose a multivariable logistic regression predictive model.
Methods: NfL and CHI3L1 levels were measured in CSF samples (n=156) with ELISA assay in a cross-sectional MS cohort. Relapse was considered when lumbar puncture (LP) was within 90 days of a clinical attack. Prior to statistical analysis, NfL and CHI3L1 levels were log-transformed. We assessed the association between both variables and in relation to clinical, imaging, treatments, and disability outcomes with parametric comparison tests. Finally, we developed a multivariable logistic regression analysis for predicting disability progression.
Results: A cross-sectional cohort of 156 patients (70% females) was included in this study. Clinical forms distributed in 93 RRMS, 40 SPMS, and 23 PPMS at LP. NfL and Chi3l1 were measured in the CSF of 144 and 156 patients respectively. Median NfL in patients with activity (clinical attack or gadolinium-enhanced lesions) was higher 664,8 pg/ml [125,3-8290] compared to patients without active inflammation, 352,2 pg/ml [45,28-5770] (p< 0,001). Median Chi3l1 was not influenced by inflammation, although a significant increase was found in progressive forms (171,45 [34,84-802,7]) compared to relapsing (115,35 [33,60-507,4]) (p< 0,000). Progressive forms with clinical or radiological inflammatory activity (SPMS) had higher NF concentrations (486,80 pg/ml vs. 1108 pg/ml, p< 0,04). Current therapies at time of LP did not influence significantly NfL or Chi3l1. Other demographic and clinical variables affecting NfL and Chi3l1 were investigated. A multivariable logistic regression model was adjusted with NfL and Chi3l1, influencing variables (age, IgM-OCB, disease duration) and EDSS longitudinal data, resulting an accurate tool to predict disability progression at 10 to 20 years from LP, with an adjusted R-squared of 0.81.
Conclusion: Isolated simultaneous measure of NfL and Chi3L in CSF informs us about disease activity in relapsing and progressive forms of disease, but it is also a valuable tool to predict disability progression.
Disclosure: SGP is funded by Health Institute Carlos III Rio Hortega CM12/00014. LCN, JCV, JGM, SC, ME DHM: nothing to disclose. BCE serves on the scientific advisory board for Novartis, Genzyme, TEVA and Merck Pharmaceuticals, and is funded by Health Institute Carlos III, grant number PS13/00663 (BC) and a local grant from Hospital La Fe VlcBiomed (UV/IIS la Fe #33-2015-A). FPM serves on the scientific advisory board for Novartis, Genzyme, TEVA, Almirall, and Roche Pharmaceuticals. CAV serves on the scientific advisory board for Genzyme. FG serves on the scientific advisory board for Genzyme and Roche Pharmaceuticals.