Lupus & antiphospholipid syndrome
Unsupervised clustering of clinical and laboratory features of systemic lupus erythematosus: insights from a multicentre cohort
Authors' conclusion
We identified five SLE clusters with distinct clinical, immunological features and treatment outcomes, demonstrating an intrinsic link between clinical involvement and immune-inflammatory alterations.
Study context
- Population
- Systemic lupus erythematosus (SLE) patients from five teaching hospitals899 followed >1 year
- Participants
- 1,065 patients
- Exposure
- Cluster 1 vs Cluster 2 vs Cluster 3 vs Cluster 4 vs Cluster 5
- Outcome
- Cluster size and defining feature
Methodological detailsunsupervised machine learning, multi-omics integration, prospective follow-up
- Unsupervised machine learning
- Multi-omics integration
- Prospective follow-up
Main finding
As reported in the abstract
| Group | Value | Renal involvement | Cutaneous manifestations | Neuropsychiatric manifestations | anti-U1 RNP positivity | Cytopenia | Direct Coombs positivity | Cardiopulmonary involvement |
|---|---|---|---|---|---|---|---|---|
| Cluster 1 (renal) | 14.27% | 87.5% | ||||||
| Cluster 2 (mucocutaneous) | 39.53% | 72.45% | ||||||
| Cluster 3 (neuropsychiatric) | 14.74% | 96.18% | 42.04% | |||||
| Cluster 4 (haematologic) | 22.63% | 95.85% | 87.55% | |||||
| Cluster 5 (cardiopulmonary) | 8.83% | 97.87% |
Value
Cluster 1 (renal)14.27%
Cluster 2 (mucocutaneous)39.53%
Cluster 3 (neuropsychiatric)14.74%
Cluster 4 (haematologic)22.63%
Cluster 5 (cardiopulmonary)8.83%
Renal involvement
Cluster 1 (renal)87.5%
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)— not stated in the abstract
Cluster 4 (haematologic)— not stated in the abstract
Cluster 5 (cardiopulmonary)— not stated in the abstract
Cutaneous manifestations
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)72.45%
Cluster 3 (neuropsychiatric)— not stated in the abstract
Cluster 4 (haematologic)— not stated in the abstract
Cluster 5 (cardiopulmonary)— not stated in the abstract
Neuropsychiatric manifestations
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)96.18%
Cluster 4 (haematologic)— not stated in the abstract
Cluster 5 (cardiopulmonary)— not stated in the abstract
anti-U1 RNP positivity
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)42.04%
Cluster 4 (haematologic)— not stated in the abstract
Cluster 5 (cardiopulmonary)— not stated in the abstract
Cytopenia
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)— not stated in the abstract
Cluster 4 (haematologic)95.85%
Cluster 5 (cardiopulmonary)— not stated in the abstract
Direct Coombs positivity
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)— not stated in the abstract
Cluster 4 (haematologic)87.55%
Cluster 5 (cardiopulmonary)— not stated in the abstract
Cardiopulmonary involvement
Cluster 1 (renal)— not stated in the abstract
Cluster 2 (mucocutaneous)— not stated in the abstract
Cluster 3 (neuropsychiatric)— not stated in the abstract
Cluster 4 (haematologic)— not stated in the abstract
Cluster 5 (cardiopulmonary)97.87%
Observational data: associations, not effects.
Further findings
| Outcome | Values |
|---|---|
| Lupus low disease activity state · 1 year | Cluster 2 (mucocutaneous) 46.42% |
Lupus low disease activity state · 1 year
- Cluster 2 had the highest proportions achieving low disease activity and tapering GC
CYC use and clinical improvement, Cluster 3 · 1 year
- R
- 0.284
- p
- P = 0.023
RTX use and favourable response, Cluster 5 · 1 year
- R
- 0.286
- p
- P = 0.019
Original publication
Unsupervised clustering of clinical and laboratory features of systemic lupus erythematosus: insights from a multicentre cohort. Lupus Sci Med. Published 8 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →