Neurofilament heavy chain in CSF correlates with relapses and disability in multiple sclerosis

Objective: Neurodegeneration is now accepted as a pathologic hallmark of multiple sclerosis (MS). We sought to discover whether CSF levels of neurofilament heavy chain protein (NfH(SMI35)) correlate with disability, disease activity, or specific stages of MS. Methods: An electrochemiluminescence immunoassay was used to retrospectively measure NfHSMI35 in CSF of patients with clinically isolated syndrome (CIS) (n = 63), relapsing-remitting multiple sclerosis (RRMS) (n = 39), secondary progressive multiple sclerosis (SPMS) (n = 25), primary progressive multiple sclerosis (PPMS) (n = 23), or controls (n = 73). Cell count and CSF levels of immunoglobulin and albumin were also measured. Results: CSF levels of NfH(SMI35) increased with age in controls (r(s) = 0.50, p < 0.0001) and CIS (r(s) = 0.50, p < 0.0001); this effect was less pronounced in RRMS (r(s) = 0.35, p = 0.027) and absent in SPMS/PPMS. After age correction, NfH(SMI35) levels were found to be higher in all disease stages compared to control. Relapses were associated with higher CSF NfH(SMI35) values compared with stable disease. NfH(SMI35) levels correlated with EDSS scores in patients with CIS and RRMS (r(s) = 0.33, p = 0.001), and during relapse (r(s) = 0.35, p = 0.01); the correlation was most prominent in RRMS during relapse (r(s) = 0.54, p = 0.01). This was not the case for any of the other CSF markers examined. Conclusions: Neuronal loss is a feature of aging, and the age-dependent increase of CSF NfH(SMI35) suggests that this loss accelerates over time. For MS, increased NfH(SMI35) levels reflect the superimposed presence of further neurodegenerative processes. Evaluation of NfH(SMI35) levels is likely to provide a useful surrogate for measuring the rate of neurodegeneration in MS. Furthermore, the dissociation of NfH(SMI35) levels with biomarkers of inflammation suggests that the mechanisms responsible for their production are at least partly independent. Neurology (R) 2011;76:1206-1213