Development of a Novel Risk Prediction Model for Rheumatoid Arthritis-Associated Interstitial Lung Disease (RA-ILD): A Longitudinal Study

The study aimed to identify plasma protein biomarkers and develop a prediction model for interstitial lung disease (ILD) in patients with rheumatoid arthritis (RA). Patients with RA have significantly higher risk of developing ILD, with a hazard ratio of 4.425 compared to individuals without RA. Researchers analyzed 2,920 plasma proteins in 781 RA patients and created eight machine learning models. The best-performing CatBoost model achieved an area under the curve (AUC) of 0.884. Analysis identified three most important protein predictors: LAG3, NPC2, and LAMP3. The results suggest that plasma proteomics combined with machine learning may be a promising approach for identifying biomarkers and predicting ILD risk in RA patients.