Sjögren's disease & overlap diseases
Machine learning–driven immunophenotypic stratification of mixed connective tissue disease, corroborating the clinical heterogeneity
Authors' conclusion
This study describes the stratification of MCTD patients, based on immunophenotyping, suggesting the presence of distinct immunological processes behind the clinical subtypes of MCTD.
Study context
- Population
- Patients with MCTD, SLE, idiopathic inflammatory myopathy (IIM) and SSc from the ImmuNexUT functional genome database; 24 immune cell subsets analysed22 with MCTD
- Participants
- 215 patients
- Exposure
- MCTD with SLE immunophenotype vs MCTD with non-SLE immunophenotype
- SLE immunophenotypen = 16
- Non-SLE immunophenotypen = 6reference
- Outcome
- Immunophenotype-based stratification of MCTD patients
Methodological detailsmachine-learning classification (including Random Forest), immunophenotyping and transcriptome analysis, gene set variation analysis (GSVA)
- Machine-learning classification (including Random Forest)
- Immunophenotyping and transcriptome analysis
- GSVA
Main finding
As reported in the abstract
Classification of MCTD patients by immunophenotype
SLE immunophenotype
16
events
Non-SLE immunophenotype
6
events
Observational data: associations, not effects.
Further findings
| Outcome | SLE immunophenotype | Non-SLE immunophenotype | P value |
|---|---|---|---|
| Proportion of Th1 cells | 2.85% (IQR 1.54–3.91) | 1.33% (IQR 0.99–1.74) | P = 0.027 |
| Proportion of plasmablasts | 6.35% (IQR 4.17–17.49) | 2.00% (IQR 1.20–2.80) | P = 0.010 |
| Number of SLE-related symptoms | 2.0 (IQR 1.0–2.0) | 1.0 (IQR 1.0–1.0) | P = 0.038 |
GSVA scores (transcriptome)
- IFN-α and -γ responses significantly higher with SLE immunophenotype in central memory CD8+ T cells
- Hedgehog signalling higher with non-SLE immunophenotype in five cell subsets
Original publication
Machine learning–driven immunophenotypic stratification of mixed connective tissue disease, corroborating the clinical heterogeneity. Rheumatology (Oxford). Published 13 March 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →