Lupus & antiphospholipid syndrome
Unsupervised machine learning identifies distinct SLE patient endotypes with differential response to belimumab
An erratum to this article was published (Rheumatology (Oxford), 4 Feb 2026 ↗)
Authors' conclusion
Three distinct SLE endotypes were identified based on B cell immunophenotyping and serological profiles, showing differential benefit from belimumab therapy.
Study context
- Population
- Patients with SLE from the phase III trial BLISS-SC
- Participants
- 796 patients
- Comparison
- Belimumab vs placebo within B cell/serological endotypes
- Cluster 1n = 191
- Cluster 2n = 366
- Cluster 3n = 239
- Primary endpoint
- SLE endotypes by B cell immunophenotyping and serological profile; belimumab response (lupus low disease activity state (LLDAS), definitions of remission in SLE (DORIS) remission) within endotypes
Methodological detailsunsupervised machine learning, factor analysis of mixed data (FAMD) and clustering, Cox regression
- Unsupervised machine learning
- Factor analysis of mixed data (FAMD) and clustering
- Cox regression
Primary result
As reported in the abstract
Endotypes identified
- Three endotypes were determined
- 3
- Cluster 1: CD19+CD24b+CD27+ regulatory B cells, mean ± SD
- 35.9%±12.6%
- Cluster 1: CD19+CD20+CD27+ bulk memory B cells
- 32.2%±9.9%
- Cluster 1: CD19+CD20+CD69+ activated B cells
- 0.2%±0.3%
- Cluster 1: CD19+CD20-CD138+ long-lived plasma cells
- 0.7%±1.0%
- Cluster 1: CD19+CD38b+CD27b+ SLE-associated plasma cells
- 6.6%±7.0%
- Cluster 2: CD19+CD24bbrightCD38bbrightCD27- transitional B cells
- 6.3%±9.0%
- Cluster 2: CD19+CD20+CD27- naïve B cells
- 85.4%±7.2%
- Cluster 2: CD19+CD20-CD138+ peripheral long-lived plasma cells (lower)
- 0.2%±0.3%
- Cluster 2: CD19+CD38b+CD27b+ SLE-associated plasma cells (lower)
- 1.6%±2.0%
- Cluster 3: CD19+CD20+CD138+ short-lived plasma cells
- 0.1%±0.1%
Further findings
| Outcome | Estimate | P value | |
|---|---|---|---|
| Sustained LLDAS, belimumabvs placebo in Cluster 2Cox regression within cluster | HR for achieving sustained LLDAS 2.22 (95% CI 1.18–4.17) | P = 0.014 | |
| DORIS remission, belimumabvs placebo in Cluster 2Cox regression within cluster | HR for achieving DORIS remission 3.45 (95% CI 1.2–9.94) | P = 0.022 | |
| 0.10.31310Ratio (log scale) |
Original publication
Unsupervised machine learning identifies distinct SLE patient endotypes with differential response to belimumab. Rheumatology (Oxford). Published 17 April 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →