Neurofilament-light chain
Kappa free-light chain
Combining biomarkers

MS biomarker research this year was characterized by further refinements to how current fluid biomarkers may be employed in practice; new/emerging markers of disease; and the use of combined biomarkers that may help to unravel the complex pathobiology of disease progression.

Neurofilament-light chain
A systematic review of the literature reported that higher baseline sNfL levels were associated with an increased risk of relapses and Gd+ lesions, and a higher T2 lesion volume (Freedman et al. J Neurol 2025;272:400). An annual increase in sNfL >10%, a marker of ongoing disease activity, was also associated with greater T1 lesion volume and a higher rate of brain atrophy. The value of sNfL in predicting relapse activity may be greater early on with some studies finding that the association between sNfL and relapses is less robust after 24 months. Baseline sNfL levels are also reportedly predictive of EDSS worsening over the next five years, which may help to identify at-risk patients who would be candidates for higher-efficacy DMTs.

sNfL may also help in differentiating different pathobiological subtypes in MS. A recent study of clinically isolated syndrome/early MS found that patients who presented with optic neuritis (ON) had lower baseline sNfL z-scores than those with a non-ON presentation (Klyscz et al. Eur J Neurol 2025;32:e70375). A higher sNfL z- score was predictive of a subsequent attack only in the non-ON group, which may suggest that neuroaxonal damage is less severe early in the course of ON-onset MS.

It has not yet been determined if absolute sNfL values or z-scores are more useful in practice. One group analysed data from the ASCLEPIOS studies and proposed a cut-off of 12.9 pg/mL for prognosticating disease activity (Ziemssen et al. Mult Scler 2025;31:1543-1556). In contrast, a European study reported that age- and BMI-adjusted z-score (>1.2) outperformed absolute sNfL concentration (> or <10.8 pg/mL) in predicting future relapses (hazard ratio 1.80) (Einsiedler et al. Ann Clin Transl Neurol 2025;12:2214-2225). Upper-quartile z-scores were also associated with a higher incidence of new/enlarging T2 lesions.

Also noteworthy was an analysis of sNfL in patients stopping treatment in the DOT-MS study (Fung et al. J Neurol 2025;272:530). In the group of previously stable patients who stopped treatment, eight experienced significant disease activity (>3 new lesions or >2 Gd+ lesions) that was associated with an increase in sNfL levels and sNfL z-scores. Disease activity was best predicted by a cut-off of >46.4% increase in sNfL, which may help to identify patients who would benefit from re-starting treatment.

Kappa free-light chain
Kappa free-light chain (kFLC), a biomarker of intrathecal antibody production and inflammation, has been less studied but received a promotion this year with the publication of the McDonald 2024 diagnostic criteria (Montalban et al. Lancet Neurol 2025;24:850-865). The McDonald committee, now a supersized 56 members (vs. 16 in 2001), included a positive CSF as one of the additional diagnostic features for typical MS presentations with <3 dissemination in space (DIS) topographies. kFLC and oligoclonal bands can be used interchangeably for the CSF+ criterion. A higher kFLC index is correlated with the number of mononuclear cells, NfL, and chemokine CXCL13 levels in CSF, suggesting that is associated with progressive biology (Duell et al. Eur J Neurol 2025;32:e70291).

Combining biomarkers
A shortcoming of current biomarkers such as NfL and glial fibrillary acidic protein (GFAP) is that they have limited specificity for progression independent of relapse activity (PIRA) (Bsteh et al. Int J Mol Sci 2025;26:4704). One approach to overcome this limitation is to combine biomarkers to give a more complete picture of different underlying disease processes.

One group this year proposed a biomarker panel comprising NfL, GFAP, total-Tau, phosphorylated Tau181/T-Tau ratio and ubiquitin C-terminal hydrolase (UCHL-1) to identify patients at risk of PIRA (Schilke et al. Mult Scler J Exp Transl Clin 2025;11:20552173251372751). The use of Tau and ubiquitin biomarkers suggests that some neurodegenerative processes in MS are similar to those seen in Parkinson’s disease and Alzheimer’s disease. In early MS patients, PIRA was characterized by elevated GFAP and UCHL-1 levels, both due to astrocyte activation; and a lower P-Tau181/T-Tau ratio, an indication of alterations in Tau metabolism. The biomarker pairings of NfL-GFAP, NfL-UCHL-1 and GFAP-UCHL-1 were moderately correlated.

A separate study found that high sNfL z-scores and lipid-specific IgM oligoclonal bands (LS-OMB) in CSF were predictors of active PIRA, while GFAP was predictive of non-active PIRA (Monreal et al. Int J Mol Sci 2025;26:6898). Patients with elevated sNfL + LS-OMB-positivity had the highest risk of active PIRA. The authors concluded that different biomarkers may help to identify distinct mechanisms of disease worsening and enable a more personalized treatment approach.