BACKGROUND The cytokine osteopontin (OPN) is a potential key player in the immunopathogenesis of multiple sclerosis (MS) and a candidate biomarker for disease activity. OBJECTIVE The objective of this study was to examine concentrations of OPN in the cerebrospinal fluid (CSF) across the clinical spectrum of MS. METHODS Our research consisted of a cross-sectional study of patients from two randomized, placebo-controlled trials. Concentrations of OPN and other blood and CSF markers were determined using an enzyme-linked immunosorbent assay (ELISA). Samples were obtained from untreated patients with exacerbation of clinically isolated syndrome (CIS) (n = 25) and relapsing-remitting MS (RRMS) (n = 41) of whom 48 participated in clinical trials, randomly allocated to treatment with placebo or methylprednisolone (MP) and undergoing repeated sampling after 3 weeks. Furthermore, we obtained CSF and blood samples from patients with primary progressive MS (PPMS, n = 9), secondary progressive MS (SPMS, n = 28) and other neurological disorders (OND, n = 44), and blood samples from 24 healthy subjects. RESULTS OPN concentrations were significantly increased in the CSF of patients with CIS (p = 0.02) and RRMS (p < 0.001) in exacerbation compared to patients with OND, and increased levels of OPN were associated with high values of other biomarkers of inflammation. At 3-week follow-up CSF OPN concentrations had decreased significantly from baseline regardless treatment with placebo or MP. Patients with PPMS had increased OPN levels in the CSF (p = 0.004) and high CSF levels of OPN were associated with high degrees of disability. CONCLUSIONS OPN concentration in the CSF is a dynamic indicator of disease activity in RRMS, presumably reflecting ongoing inflammation. Increased CSF OPN concentrations in PPMS may indicate ongoing inflammation even in these patients.