Vasculitis & polymyalgia rheumatica
Immunophenotyping and machine learning-based stratification reveal prognostic immune signatures in Behçet’s disease
Authors' conclusion
This study provided a comprehensive immunophenotypic profile of treatment-naive BD patients.
Study context
- Population
- Treatment-naive patients with BD and age- and sex-matched healthy controls
- Participants
- 71 vs 71
- Exposure
- Treatment-naive BD patients vs age- and sex-matched healthy controls (HCs); BD clusters compared for relapse-free survival
- BDn = 71
- HCsn = 71reference
- Outcome
- Immune profiles and relapse in BD: lymphocyte subsets vs HCs, ML-derived subgroups and predictors of relapse
Methodological detailsage- and sex-matched healthy controls, flow cytometry immunophenotyping, machine learning (XGBoost, PLS-DA, k-means clustering), Cox regression
- Age- and sex-matched healthy controls
- Flow cytometry immunophenotyping
- Machine learning (XGBoost, PLS-DA, k-means clustering)
- Cox regression
Main finding
As reported in the abstract
HR-weighted combination of CD8+DR+T%, hsCRP and disease severity scores, relapse prediction AUC
0.815
Observational data: associations, not effects.
Original publication
Immunophenotyping and machine learning-based stratification reveal prognostic immune signatures in Behçet’s disease. Rheumatology (Oxford). Published 22 November 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →