Long follow-up links baseline biomarkers
Baseline samples from untreated RRMS patients are connected to a median nine years of follow-up, including SPMS conversion and sustained PIRMA outcomes.
↳ Methods, Patients and Samples; Results, Fig. 4
Abstract
Progressive multiple sclerosis is characterized by gradual neurological decline, often occurring independently of relapses or MRI activity-a phenomenon known as progression independent of relapse and MRI activity (PIRMA). Despite the effectiveness of disease-modifying therapies in controlling inflammatory activity, identifying individuals at risk of PIRMA remains an unmet clinical need. The objective of this exploratory study was to identify biomarkers and underlying molecular pathways associated with multiple sclerosis progression and especially PIRMA. Using the NUcleic acid Linked Immuno-Sandwich Assay (NULISA) inflammatory panel, we quantified 250 immune-related proteins in CSF and plasma from 49 controls, 49 patients with early active relapsing-remitting multiple sclerosis (RRMS) and 33 patients with inactive progressive multiple sclerosis (iPMS). Longitudinal clinical data were used to define PIRMA and conversion to secondary progressive multiple sclerosis (SPMS). We identified distinct proteomic signatures in CSF of both RRMS and iPMS patients compared with controls, with no significant differences in plasma. Both were associated with elevated markers of adaptive immunity, while iPMS showed a shift towards innate immune markers. Among RRMS patients, low baseline CSF concentrations of KIT ligand (KITLG) predicted both conversion to SPMS and future PIRMA events. Receiver operating characteristic analysis demonstrated KITLG's potential as a prognostic biomarker. Additionally, plasma concentrations of interleukin 1 beta and 36 gamma were elevated in RRMS patients who later developed SPMS. A model selection analysis identified a two-protein logistic regression model including interleukin 1 beta and interleukin 36 gamma as the best-performing combination (area under the curve = 0.990). Our findings reveal distinct immunological profiles across multiple sclerosis subtypes and identify KITLG as a promising biomarker for predicting disease progression and PIRMA. These results highlight potential targets for therapeutic intervention and demonstrate the utility of NULISA in uncovering novel molecular signatures in multiple sclerosis.
Key Claims
Primary findingTentative
Among RRMS patients, low baseline CSF concentrations of KIT ligand (KITLG) predicted both conversion to SPMS and future PIRMA events.
Fragile — the claim's weakest requirement is that the effect exists at all. Experimental. Every point below is traceable to a quoted passage, but this forecast has not yet been validated against known replication outcomes.
small overlapping outcome groups, an internally derived cut-off, event-poor adjusted models, and no external validation
Abstract; Results, Fig. 4
SecondarySupported
We identified distinct proteomic signatures in CSF of both RRMS and iPMS patients compared with controls, with no significant differences in plasma.
FDR-controlled adjusted CSF comparisons support the direction, while the small plasma sample limits confidence in the null result
Abstract; Results, Fig. 1–2
SecondaryTentative
A model selection analysis identified a two-protein logistic regression model including interleukin 1 beta and interleukin 36 gamma as the best-performing combination (area under the curve = 0.990).
the apparent AUC follows selection in a very small cohort without cross-validation and duplicates another model’s reported statistics
Abstract; Results, Fig. 5C
Derived from the full evaluation — not a separate score.
Strengths · Limitations
Strengths
Standardized sampling and multiplicity control
Samples followed the BioMS-EU handling protocol, and differential-expression analyses used age- and sex-adjusted models with FDR correction.
↳ Methods, NULISA Proteomic Analysis and Statistical Analysis
Prior proteomic evidence is engaged
The Discussion compares the findings with prior CSF signatures and a large plasma proteomic study that reported broader alterations, rather than presenting the results in isolation.
↳ Discussion, first two paragraphs
Limitations
Potentially duplicated headline model statistics
The IL1B–IL36G SPMS model and the CCL27–NGF–SPP1 PIRMA model report exactly the same AUC, confidence interval, and P-value, despite using different proteins and outcomes.
↳ Results, Fig. 5C and Fig. 5E
Model selection precedes performance evaluation
Proteins were filtered by outcome correlation and models were dredged by AICc, while validation was absent for plasma models and was not shown to encompass feature selection for the CSF models.
↳ Methods, Statistical Analysis; Results, Fig. 3 and Fig. 5
Progression evidence lacks independent validation
The prognostic analyses use few events, internally selected cut-offs, and no strong clinical-only comparator or external cohort. PIRMA and SPMS conversion also overlap substantially.
↳ Results, Fig. 4–5; Discussion, limitations
Nabu Commentary
The study's strongest foundation is its treatment-naïve baseline cohort, standardized biofluid handling, and long follow-up for clinically relevant progression outcomes. Its cross-sectional CSF findings are supported by adjusted, FDR-controlled analyses, but much of that profile confirms existing biomarker literature. Confidence falls sharply for the prognostic models because proteins and cut-offs were selected internally, plasma models were not validated, and no clinical-only prognostic comparator was presented. The identical AUC, confidence interval, and P-value reported for two different plasma models, together with other Results–Discussion inconsistencies, requires correction before the headline performance claims can be relied upon.
Check another paper
Author Note / Field Feedback
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Dimension breakdown
Study Quality
Limited2.5
Confidence mediumContribution
3.1
The study combines baseline CSF and plasma proteomics with long clinical follow-up and nominates KITLG as a potentially novel progression-associated marker. The advance remains incremental because much of the CSF profile is confirmatory and the prognostic findings have not been externally validated.
Abstract; Results, Fig. 1–5; Discussion, opening two paragraphs
These results are largely confirmatory
Methodological Rigour
1.9
Standardized sample handling, FDR-controlled expression analysis, covariate-adjusted Cox models, and internal cross-validation support the exploratory design. However, outcome-informed feature selection was not nested within validation, the plasma models were evaluated in the selection cohort, and no clinical prognostic baseline was tested.
Methods, NULISA Proteomic Analysis and Statistical Analysis; Results, Fig. 3–5
the single best-performing model was selected
Reporting
3.1
The article follows a coherent progression from cohort definition through proteomic comparison and prognostic analysis. Predictive language in the Abstract and Discussion overstates small internally selected models, while references to IKBKG, PTX3, and an unreported log-rank test interrupt the evidence chain.
Abstract; Discussion, KITLG paragraph, intrathecal-indices paragraph, and summary
KITLG as a promising biomarker for predicting disease progression and PIRMA
Positioning
2.8
The Discussion engages prior CSF and plasma proteomic work, including a contrasting large plasma study, and acknowledges sample size and external-validation needs. It does not directly address selection-induced optimism in the reported AUCs or the substantial overlap between PIRMA and SPMS conversion.
Discussion, opening literature comparison and limitations paragraphs; Results, PIRMA section
external validation in independent cohorts is needed
Reliability
Concerns4 of 4 checks
Internal CoherenceConcern
Several numerical and narrative inconsistencies affect central prognostic claims, including potentially duplicated plasma-model statistics. These issues require correction or reconciliation before the reported performance can be interpreted confidently.
Research ConductNoted
Ethics approval, written consent, and a data-availability statement are declared. The supplied article does not include funding or conflict-of-interest disclosures and does not provide analysis code.
Flags: 2 declared / 5 total
Reference IntegrityClean
50 references in manuscript 50 of 50 checkable references found in an index
Post-Publication RecordClean
No retraction notice found in Retraction Watch.
Last checked: 08 Oct 2026
Sources: Retraction Watch ✓
Readiness
Where this paper’s evidence sits on the path from initial observation to real-world use.
Mechanistic
KITLG and the plasma cytokine combinations remain discovery-stage candidates from one cohort. There is no external prognostic validation, calibration, decision-curve analysis, or evidence of added value over routine clinical predictors.
Results, Fig. 4–5; Discussion, limitations and summary
validation in independent cohorts and integration with imaging and clinical data are essential
AI-generated, human-governed. Something look off? Contact us to request a review.