Integrative proteomic and machine learning strategies identify key biomarkers associated with response to JAK inhibitor in rheumatoid arthritis
Authors' conclusion
Serum proteomic profiling identifies elevated pretreatment IL-12B levels as a key predictor of JAKi non-response and uncovers distinct molecular responses to different JAK inhibitors.
Study context
- Population
- Rheumatoid arthritis (RA) patients treated with tofacitinib, baricitinib, or upadacitinib
- Participants
- 42 patients
- Exposure
- Janus kinase inhibitors (JAKi) responders vs non-responders
- Outcome
- Proteomic biomarkers associated with response to JAKi (ACR/EULAR criteria)
- Follow-up
- 6 months
Methodological detailslongitudinal serum proteomic profiling (Olink® Explore 384), differential expression analysis, ensemble feature selection, machine learning prediction models, SHAP
- Longitudinal serum proteomic profiling (Olink® Explore 384)
- Differential expression analysis
- Ensemble feature selection
- Machine learning prediction models
- SHAP
Main finding
As reported in the abstract
AUC, prediction models using proteomic features
≥ 0.90
AUC, models based on clinical variables alone
0.52 - 0.57
AUC, integrated clinical and proteomic data
0.95
Observational data: associations, not effects.
Original publication
Integrative proteomic and machine learning strategies identify key biomarkers associated with response to JAK inhibitor in rheumatoid arthritis. Arthritis Res Ther. Published 4 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →