Neurofilament light as a prognostic marker in multiple sclerosis

Abstract

Relapsing-remitting multiple sclerosis has a variable prognosis and lacks a reliable laboratory prognostic marker. Our aim in this study was to investigate the association between neurofilament light levels in cerebrospinal fluid in early multiple sclerosis and disease severity at long-term follow-up. Neurofilament light levels in cerebrospinal fluid collected at diagnostic lumbar puncture were measured in 99 multiple sclerosis cases. Clinical data were obtained from 95 out of those at follow-up visits made 14 years (range 8—20 years) after disease onset. Significant correlations between neurofilament light levels and the multiple sclerosis severity score were found for all cases (r = 0.30, p = 0.005), for relapsing-remitting multiple sclerosis cases (r = 0.47, p < 0.001) and for cases with a recent relapse (r = 0.60, p < 0.001). In the multivariate logistic regression analysis, neurofilament light levels >386 ng/L (median value of cases with detectable levels) increased the risk for severe multiple sclerosis fivefold (odds ratio 5.2, 95% confidence interval 1.8—15). Kaplan—Meier analysis showed that conversion to secondary-progressive multiple sclerosis was more likely in cases with neurofilament light levels >386 ng/L than in those with neurofilament light levels <60 ng/L (p = 0.01) or 60—386 ng/L (p = 0.03). We conclude that elevated levels of neurofilament light in cerebrospinal fluid collected at diagnostic lumbar puncture were associated with unfavourable prognosis. These data suggest that the neurofilament light level could be used as a prognostic marker in early relapsing-remitting multiple sclerosis.