Proteomics: VSIG4 Flags RA Flare Risk in Biologic Tapering
Of the rheumatoid arthritis (RA) patients who cut their biologic dose in the OPTIBIO trial, 46% flared during up to three years of follow-up, and routine clinical scores gave only a weak hint of who would. A rheumatology and proteomics group in A Coruña, Spain, went back to the trial's banked baseline serum. Label-free mass spectrometry on 44 samples nominated candidates for ELISA in all 194 patients, alongside a 15-plex anti-cytokine autoantibody array. One protein held up. Reported in RMD Open, it is VSIG4, a complement receptor of tissue-resident macrophages.
Key Takeaways
Serum VSIG4 was higher at baseline in rheumatoid arthritis patients who flared after biologic tapering; no such difference appeared in patients kept on full-dose therapy.
Discovery proteomics on 44 banked trial serum samples produced a flare marker that replicated by ELISA across the full 194-patient rheumatoid arthritis cohort.
Adding VSIG4 and anti-interferon-gamma autoantibodies to the DAS28-CRP disease activity score raised RA flare prediction during tapering from an AUC of 0.67 to 0.76.
The serum flare-risk model for RA biologic tapering has been tested only inside one Spanish trial and needs external validation before it guides treatment.
Serum Proteomics Inside a Tapering Trial
The analysis drew on 194 OPTIBIO patients from five Spanish hospitals, all in sustained remission on a biologic and randomized to dose tapering or unchanged therapy. Flare, defined as losing every remission criterion at once, hit 45 of 97 tapered patients and 30 of 97 controls. Discovery sera (n = 44, balanced for flare, arm and anti-citrullinated protein antibody status) were stripped of the 14 most abundant proteins and run by data-independent acquisition (DIA-PASEF) on a timsTOF Pro, giving 806 proteins. A q-value below 0.1 with a 1.5-fold change counted as differential, a lenient bar that suits hypothesis generation at this size. The hand-off from DIA discovery to serum ELISA mirrors a lupus nephritis proteomics study we reviewed recently.
Key Findings
87 proteins tracked flare only under tapering: In the optimisation arm, 87 of 806 proteins separated future flares from sustained remission, about 35 of them tied to immune pathways such as neutrophil degranulation.
VSIG4 replicated in the full cohort: Baseline VSIG4 was higher in tapered patients who flared (p = 0.018), with no difference in the control arm; above-median levels gave an odds ratio of 5.07 (95% CI 2.13 to 12.06) and a hazard ratio of 3.25.
An autoantibody adds its own signal: Of 15 anti-cytokine autoantibodies, IgG against interferon gamma (anti-IFN-gamma) stood out; the top tertile flared sooner than the bottom (HR 2.11, p = 0.044).
Markers plus disease activity: VSIG4, anti-IFN-gamma and DAS28-CRP together reached an AUC of 0.760 versus 0.666 for DAS28-CRP alone (p = 0.003), and 0.839 when flares were compared with patients still in remission after two years.
Figure 1. VSIG4 and flare after biologic tapering. Panel A plots serum VSIG4 (ng/mL) at tapering for patients who flared (red) or stayed in remission (blue), separately for the control arm (p = 0.36) and the optimisation arm (p = 0.018), with medians and interquartile ranges marked. Panel B gives Kaplan-Meier flare-free survival over about 130 weeks, split at the median VSIG4 level, with the high group flaring earlier (log-rank p = 0.000096). Adapted from Blanco et al. (2026), RMD Open.
A Macrophage Receptor, Not an Acute-Phase Protein
VSIG4, also called CRIg, is a B7-family receptor on tissue-resident macrophages that binds complement fragments and dampens immune responses, and its expression is reduced in refractory RA. The authors read high serum levels as subclinical immune activity that remission scores miss. Plausible, but untested: nothing here shows the circulating protein comes from the joint. Calprotectin separated neither arm, a mild surprise after de Moel and colleagues (2019) reported higher calprotectin in patients relapsing within a year of tapering. VSIG4 levels did not differ by biologic class, reassuring when almost a third of patients took tocilizumab, which flattens IL-6-driven acute-phase proteins.
What the Combined Model Buys a Clinic
The biomarker pair did most of the work: alone it reached an AUC of 0.722, and adding DAS28-CRP gave no significant improvement (p = 0.562). At its Youden cutoff the combined model caught 86.7% of flares yet cleared only 55.8% of stable patients. That trade is tolerable here, since a false alarm just means keeping the current dose. Decision curves put the gain at about 16 avoided inappropriate tapering attempts per 100 patients at a 30% risk threshold. One number needs a second look. The anti-IFN-gamma odds ratio of 1.889 carries a 95% CI of 1.888 to 1.890, far too tight for 97 patients and more likely a scaling quirk than real precision.
Widening the Validation Funnel
Only two of the 87 discovery hits reached the full cohort, and the authors list commercial ELISA availability among their selection criteria. A scheduled parallel reaction monitoring (PRM) assay on the same depleted-serum digest could have carried 30 or 40 candidates through all 194 samples at once, which is how we typically extend MS-based proteomics discovery at Dalton. Trial serum with outcomes attached can't be collected twice, so it's better spent on a panel than on a kit-catalogue shortlist, a principle behind our approach to multi-omics biomarker discovery.
Frequently Asked Questions
What is serum proteomics used for in rheumatoid arthritis?
Serum proteomics measures hundreds of circulating proteins at once, usually by mass spectrometry, to find markers of disease activity, drug response or relapse. In RA it is mainly a discovery tool; candidates still need targeted assays and an independent cohort before clinical use.
Can a blood test predict flare when tapering biologics in rheumatoid arthritis?
Not reliably yet. In the OPTIBIO trial, a model combining serum VSIG4, anti-IFN-gamma autoantibodies and DAS28-CRP reached an AUC of 0.76 for flare, against 0.67 for DAS28-CRP alone. It has not been tested outside that trial.
How many samples does a proteomics biomarker discovery study need?
Discovery can start smaller than many teams expect; this study profiled 44 depleted serum samples at 10 micrograms of protein each. The expensive part is validation, which here meant 194 trial patients with up to three years of outcome data.
Conclusion
Elevated baseline VSIG4 is now a credible flag for RA patients whose remission depends on full-dose biologic therapy. It isn't a decision tool yet: there is no external cohort, and a specificity near 56% would keep many stable patients on drug. Sponsors running tapering or withdrawal trials should bank outcome-linked baseline serum from the first visit, because studies like this one can't be done any other way.
Related Reading
See how we run these analyses in one lab: Dalton's multi-omics CRO services.
Citation
Blanco, F. J., Quaranta, P., Domínguez-Guerrero, P., Calamia, V., Fernández-Puente, P., Paz-González, R., Balboa-Barreiro, V., Noriega, D., Galindo, L., Acasuso, B., Oreiro, N., Rojo, R., Lourido, L., & Ruiz-Romero, C. (2026). Integrated serum proteomics and autoantibody analyses reveal a biomarker signature predictive of flare during biologic tapering in rheumatoid arthritis. RMD Open, 12(3), e007198. https://doi.org/10.1136/rmdopen-2026-007198
Note
This blog post summarizes findings from the above-cited research. Figures are adapted from the original publication. For full details, please refer to the source article.
By Seungjun Yeo, CEO at Dalton Bioanalytics. Specializing in multi-omics mass spectrometry for drug discovery and biomarker research.