SPECIAL REPORT
This year saw significant advances in understanding how fluid biomarkers may elucidate the underlying pathophysiology of MS and how these measures may be employed in clinical practice. One novel fluid biomarker – kappa free light chain (KFLC), a marker of intrathecal inflammation – was admitted into the revised McDonald diagnostic criteria along with two imaging biomarkers (central vein sign [CVS] and paramagnetic rim lesions [PRL] (Montalban et al. ECTRIMS 2024; Scientific session 1).
However, most biomarkers denote a CNS pathology that is not specific to MS, so their diagnostic utility is limited. Rather, their application in practice will be to predict which patients are at risk of disease progression, identify the extent and severity of the compartmentalized inflammation that is associated with neurodegeneration, and as a measure of response to a disease-modifying therapy.
Neurofilament-light chain (NfL)
Serum levels of this marker of neuroaxonal injury are elevated prior to the onset of MS (Bjornevik et al. JAMA Neurol 2020;77:58-64) and higher sNfL levels at baseline are prognostic of poorer outcomes. For example, a post-hoc analysis of the ASCLEPIOS trials found that higher baseline levels (>9.3 pg/mL) were associated with an increased risk of new MS lesions and brain atrophy (Ziemssen et al. Front Immunol 2022;13:852563). A follow-up analysis presented at ECTRIMS reported that a higher baseline sNfL level was also associated with a greater number of new T2 lesions both in previously-treated patients (relative risk 2.25) and in treatment-naïve patients (RR 2.16) (Cross et al. ECTRIMS 2024;P258).
In the RESTORE treatment interruption study of natalizumab, sNfL was not significantly higher in 71% of patients with Gd+ lesions compared to those without Gd+ lesions, a finding that may be due in part to the study population (Fox et al. Neurology 2024;102:e209357). sNfL levels peaked at a median of eight weeks after a Gd+ lesion was observed in the subgroup with elevated sNfL. A Swiss group previously reported sNfL elevations in the three months after the appearance of a Gd+ lesion (Kuhle et al. Mult Scler 2020;13:1691-1699).
Also noteworthy this year was the publication of the Consortium of MS Clinics’ guidance on the use of NfL in practice (Freedman et al. eBioMedicine 2024;101:104970). The group recommended that NfL be evaluated at diagnosis and after 3-6 months of treatment. If sNfL levels do not decrease with treatment, escalation to a higher-efficacy agent may be considered. Clinical and MRI findings would need to be factored into the decision to switch therapies. This recommendation was supported by a study by the Barcelona group presented at ECTRIMS this year (Fernandez et al. ECTRIMS 2024;O022). In that analysis, patients who did not demonstrate a reduction in sNfL z-score in the first year of treatment had a three-fold higher risk of disease activity in year two.
Glial fibrillary acidic protein (GFAP)
This marker of astrocytosis and astrocyte damage may be more closely related to the neurodegenerative processes underlying progression than to inflammatory disease activity. However, this NfL/inflammation versus GFAP/neurodegeneration dichotomy is likely overstated.
In the EmBioProMS (Explorative Study of Emerging Blood Biomarkers in Progressive Multiple Sclerosis) published this year, both NfL and GFAP z-scores were elevated in PMS patients (Abdelhak et al. Ann Clin Transl Neurol 2024;11:477-485). A GFAP z-score >3 was associated with a higher risk of disability progression in PPMS (hazard ratio 2.88). An elevated GFAP z-score was more strongly associated with disability progression in PPMS when serum NfL was low (HR 4.31). This may reflect the emerging predominance of neurodegeneration as MS progresses, although other factors (e.g. patient age) may also be relevant.
A CLIMB analysis presented at ECTRIMS reported that GFAP levels were strongly associated with progression independent of relapse activity (PIRA), progression to EDSS 6.0 and SPMS onset over a 10-year follow-up period (Madill et al. ECTRIMS 2024;O060).
An examination of serum biomarkers as predictors of disease progression examined NfL, GFAP and chitinase 3-like 1 (CHI3L1), a marker of glial activation, in untreated PPMS patients (5.7% were on treatment) (Fissolo et al. J Neurol Neurosurg Psychiatry 2024;95:410-418). All three markers were associated with EDSS worsening. However, the three measures were only weakly correlated with each other, suggesting that they are indicating somewhat different aspects of the disease processes. In the subset analysis of non-inflammatory patients, only CHI3L1 was associated with EDSS change, which may have been due to more diffuse gliosis at this stage of MS.
GFAP levels may be reduced with treatment, although there may be varying effects depending on the DMT used. An analysis of the Swiss MS cohort looked at long-term PIRA risk in patients treated with fingolimod or a B cell-depleting therapy (Abdelhak et al. ECTRIMS 2024;O131). GFAP z-scores were reduced in the first two years of treatment in 64% of the fingolimod group and 48% of the B cell-depleting group. Every z-score unit reduction was associated with a lower risk of PIRA with fingolimod (HR 0.53 at 6.8 years) and B cell-depleting therapy (HR 0.62 at 3.1 years). An increasing GFAP z-score during treatment was a predictor of poorer outcomes in both groups. Interestingly, a stable GFAP z-score was also predictive of PIRA in the fingolimod group. The authors concluded that monitoring GFAP dynamics during treatment may help to identify patients at higher risk of early PIRA.
Also reported this year were the results of a meta-analysis, which found that GFAP is predictive of EDSS worsening in patients with neuromyelitis optica spectrum disorder (NMOSD) (Shaygannejad et al. Medicina (Kaunas) 2024;60:1050).
Chitinase-1 (CHIT1)
This biomarker of microglial activation can distinguish between chronic active and chronic inactive lesions (Oldoni Eet al. Ann Neurol 2020;87:633-645). In the OBOE (Ocrelizumab Biomarker Outcome Evaluation) study, CHIT1 levels were correlated with measures of inflammatory disease activity (e.g. Gd+ lesions, NfL) as well as of progressive biology (slowly expanding lesion volume) (Bar-Or et al. ECTRIMS 2024;O026). CHIT1 levels were significantly reduced in RMS and PPMS patients receiving ocrelizumab, suggesting that anti-CD20 therapy may reduce the microglial activation that drives the ongoing tissue destruction seen in chronic active lesions.