Pan-disease blood protein profiles of rheumatic autoimmune diseases.

内容摘要

Systemic autoimmune rheumatic diseases (SARDs) are a heterogeneous group of autoimmune conditions characterized by immune system dysregulation leading to chronic inflammation and tissue damage. The overlapping clinical manifestations make differential diagnosis challenging, highlighting the need for novel biomarkers to facilitate early diagnosis, stratification, and personalized treatment. Five SARDs including idiopathic inflammatory myopathies (n = 210), rheumatoid arthritis (n = 84), systemic sclerosis (n = 100), Sjögren disease (n = 99), and systemic lupus erythematosus (n = 99), as well as healthy controls (n = 400) and controls with acute infectious diseases (n = 218) were selected for plasma protein profiling using Olink Explore 1536. Differential abundance analysis and machine learning were used to identify proteins with both known and novel association to SARDs. The five SARDs share hundreds of proteins with consistently altered abundance compared to both healthy and infectious controls, reflecting common underlying molecular dysregulation. Despite the overlap, we identify multiple proteins with higher abundance specific to individual SARDs. Machine learning further enables accurate classification of the five SARDs, identifying a panel of 48 proteins with high discriminatory performance, several of which are also supported by differential abundance analysis. Altogether, this explorative cross-sectional study demonstrates the importance of a pan-disease approach, including also infectious and healthy controls, to identify robust and disease-informative protein panels for improved classification of SARDs. Protein levels from this study are available open access through the Human Protein Atlas, facilitating further plasma proteome research on autoimmune disease. Systemic autoimmune rheumatic diseases (SARDs) are a group of diseases caused by a malfunctioning immune system that attacks and damages the body´s own tissues and organs. They can be difficult to distinguish and diagnose correctly because they share similar symptoms; therefore, additional molecular indicators could be helpful to aid in diagnosis. In this study, blood samples were collected from patients representing five SARDs. These samples were measured using a technology named Olink Explore, which measures around 1500 proteins in the blood at the same time. Using protein measurements from 592 patients, as well as from control groups, we identified proteins that may help distinguish the five diseases from one another. The proteins, if confirmed in future studies, could help clinicians in the diagnosis of these diseases with higher precision.