Neurodegeneration is the correlate of disease progression in multiple sclerosis (MS) and thus biological biomarkers that sensitively reflect this process are much needed. Neurofilament protein subunits are potential cerebrospinal fluid (CSF) biomarkers for disease progression in MS. We argue that the neurofilament light subunit can reflect acute axonal damage mediated by inflammatory mechanisms and can imply prognostic value for conversion from clinically isolated syndrome (CIS) to definite MS. The neurofilament heavy subunit may rather reflect chronic irreversible damage and has prognostic value for disease progression or disability. The neurofilament intermediate subunit has not yet been studied. Recent studies showing higher neurofilament light or heavy subunit levels to be altered upon treatment regimes indicate their potential clinical value in monitoring treatment or side effects. Future studies should be aimed at the optimisation, standardisation and interlaboratory implementation of the assays and address the predictive value of these biomarkers.