European Journal of Neurology · 2019 · 93 citations · 17 references
Life SciencesNeurologyNeuroscienceClinical NeuroscienceNeurodegenerationSurvival AnalysisAmyotrophic Lateral SclerosisBiomarkers Of NeurodegenerationMolecular Basis Of NeurodegenerationNeurodegenerative DiseasesNeurodegenerative DisordersPrognostic AssessmentBiomarker-based PrognosticationCox RegressionDisease PrognosticationFunctional Rating ScalesNeurofilament BiomarkersNeurological Disease BiomarkersPrognostic BiomarkersPrognostic Factors Analysis
Background and purpose The prognostic value of serum neurofilament light chain ( sNfL ), a biomarker of neurodegeneration, compared to other prognostic factors of amyotrophic lateral sclerosis ( ALS ) at the time of diagnosis, remains unclear. Methods Sera from ALS patients were prospectively collected at the first diagnostic visit in our centre. sNfL levels were determined by single molecule array in 207 ALS patients and in 21 healthy controls. The prognostic value of sNfL was compared with that of other known clinical prognostic factors using a Cox regression model and multivariate analysis. Results Serum neurofilament light chain levels were higher in ALS patients than in controls ( P < 0.0001). Seven parameters were predictive of death in ALS : older age, bulbar onset, higher ALS Functional Rating Scale revised ( ALSFRS ‐R) score, greater weight loss, lower maximal inspiratory pressure, forced vital capacity and higher sNfL levels. A Cox regression model showed that sNfL ( P < 0.0001), weight loss ( P = 0.040) and site at onset ( P = 0.048) were independent predictive factors of death. In a sub‐cohort restricted to 139 patients with complete spirometry data, sNfL level ( P < 0.005) and forced vital capacity ( P = 0.022) were independent factors predictive of death. In a subgroup of 142 patients in whom ALSFRS ‐R score was available at several time points, sNfL levels positively correlated with ALSFRS ‐R rate of decline ( r = 0.571, P < 10 −12 ). Conclusions Higher sNfL concentration is a strong and independent prognostic factor of death in ALS as early as the time of diagnosis.
17
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10
Jens Kühle, Christian Barro, Ulf Andréasson et al. · Clinical Chemistry and Laboratory Medicine (CCLM) · 2016 · 739 citations
Ching‐Hua Lu, Corrie Macdonald‐Wallis, Elizabeth Gray et al. · Neurology · 2015 · 551 citations · Full text