"Classic" neurodegenerative disorders, such as Alzheimer's disease and amyotrophic lateral sclerosis share common pathophysiological features and involve progressive loss of specific neuronal populations, axonal or synaptic loss and dysfunction, reactive astrogliosis, and reduction in myelin. Furthermore, despite the absence of astrogliosis, impaired expression of astrocyte- and oligodendrocyte-related genes has been observed in patients with major psychiatric disorders, including schizophrenia and mood disorders. Because S100B is expressed in astrocytes and oligodendrocytes, its concentration in cerebrospinal fluid (CSF) or serum has been considered a suitable surrogate marker for the diagnostic or prognostic assessment of neurodegeneration. This review summarizes previous postmortem, CSF and serum studies regarding the role of S100B in this context. A general drawback is that only small single-center studies have been performed. Many potential confounding factors exist because of the wide extra-astrocytic and extracerebral expression of S100B. Due to lack of disease specificity, reliance on S100B concentrations for differential diagnostic purposes in cases of suspected neurodegenerative disorders is not recommended. Moreover, there is no consistent evidence for a correlation between disease severity and concentrations of S100B in CSF or serum. Therefore, S100B has limited usefulness for monitoring disease progression.