Vasculitis & polymyalgia rheumatica
Data-driven subclassification of antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis: model-based clustering of a federated international cohort
Authors' conclusion
Our study reinforces the view that ANCA-associated vasculitis is not merely a binary construct. Data-driven subclassification of ANCA-associated vasculitis exhibits higher predictive value than current approaches for key outcomes.
Study context
- Population
- Patients diagnosed with AAV (granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA)) between Nov 1, 1966, and March 1, 2023, in six European vasculitis registries (FAIRVASC)
- Participants
- 3,868 patients6 registries
- Exposure
- Data-driven cluster subclassification vs classification by clinical diagnosis (GPA/MPA) or ANCA specificity
- SKn = 555
- MPO-Kn = 782
- PR3-Kn = 683
- IMSn = 1,202
- YRn = 646
- Outcome
- Identification of phenotypically distinct subgroups for a data-driven subclassification of AAV
Methodological detailsmulticentre (six European vasculitis registries), federated international cohort, model-based clustering of 17 mixed-type clinical variables, People with lived experience were involved throughout the project
- Multicentre (six European vasculitis registries)
- Federated international cohort
- Model-based clustering of 17 mixed-type clinical variables
- People with lived experience were involved throughout the project
Main finding
As reported in the abstract
Clusters identified (share of 3868 patients)
- Five clusters with distinct phenotype, biochemical presentation and disease outcome
- 5
Observational data: associations, not effects.
Further findings
- Cluster-assigned models showed improved predictive power for both patient and kidney survival vs models fitted with clinical diagnosis or ANCA status
Original publication
Data-driven subclassification of ANCA-associated vasculitis: model-based clustering of a federated international cohort. Lancet Rheumatol. Published 22 August 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →