Biomarkers have gained increasing interest in recent years. Intrathecal IgM synthesis (ITMS) and serum neurofilament light chain (sNfL) levels have been associated with inflammation and poor prognosis. Serum glial fibrillary acidic protein (sGFAP) has been linked to progression independent of inflammation. However, their combined predictive value has not yet been studied.

To assess whether CSF and serum biomarkers at disease onset can predict the risk of relapse-associated worsening (RAW) and progression independent of relapse activity (PIRA). We also analyzed differences in lymphocyte subpopulations to explore distinct mechanisms of disability worsening.

A multicenter observational study including patients with multiple sclerosis (MS) treated at six Spanish hospitals, with CSF and serum samples collected within 12 months of disease onset. Multivariable Cox regression models were used to estimate risks.

Of the 529 patients included, the median age was 34.1 years (IQR, 27.6–42.6), and 367 (69.4%) were female. The median follow-up was 7.04 years (IQR, 4.92–10.5). Higher sNfL z-scores were associated with an increased risk of RAW (HR 1.30, 95% CI 1.07–1.56; P=0.007) and PIRA (HR 3.93, 95% CI 1.31–11.8; P=0.02), whereas sGFAP levels were only associated with PIRA risk (HR 2.18, 95% CI 1.25–3.81; P=0.006). Other clinical and demographic variables did not predict risk in the multivariable model, except for age (associated with PIRA) and T2 lesion load (associated with RAW). Lipid-specific IgM oligoclonal bands (LS-OCMB) were linked to a higher risk of RAW (HR 1.91, 95% CI 1.18–3.11; P=0.009). Significant differences in lymphocyte subpopulations were observed between cohorts, particularly regarding regulatory cells.

The role of biomarkers surpasses the utility of baseline clinical and radiological findings. ITMS and sNfL levels are associated with inflammation-related disability worsening, while sGFAP values indicate a more compartmentalized inflammatory process.