Serum Antigenome Profiling Reveals Diagnostic Models for Rheumatoid Arthritis

Abstract

ObjectiveThe study aimed to investigate the serum antigenomic profiling in rheumatoid arthritis (RA) and determine potential diagnostic biomarkers using label-free proteomic technology implemented with machine-learning algorithm.MethodSerum antigens were captured from a cohort consisting of 60 RA patients (45 ACPA-positive RA patients and 15 ACPA-negative RA patients), together with sex- and age-matched 30 osteoarthritis (OA) patients and 30 healthy controls. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was then performed. The significantly upregulated and downregulated proteins with fold change > 1.5 (p < 0.05) were selected. Based on these differentially expressed proteins (DEPs), a machine learning model was trained and validated to classify RA, ACPA-positive RA, and ACPA-negative RA.ResultsWe identified 62, 71, and 49 DEPs in RA, ACPA-positive RA, and ACPA-negative RA, respectively, as compared to OA and healthy controls. Typical pathway enrichment and protein–protein interaction networks were shown among these DEPs. Three panels were constructed to classify RA, ACPA-positive RA, and ACPA-negative RA using random forest models algorithm based on the molecular signature of DEPs, whose area under curve (AUC) were calculated as 0.9949 (95% CI = 0.9792–1), 0.9913 (95% CI = 0.9653–1), and 1.0 (95% CI = 1–1).ConclusionThis study illustrated the serum auto-antigen profiling of RA. Among them, three panels of antigens were identified as diagnostic biomarkers to classify RA, ACPA-positive, and ACPA-negative RA patients.

Advancing precision rheumatology: applications of machine learning for rheumatoid arthritis management

“…Subsequently, distinct biomarker sets were identified for the differentiation of RA, ACPA-positive RA, and ACPA-negative RA using feature selection through the Random Forest algorithm. The model demonstrated exceptional performance with AUC values of 0.9949, 0.9913, and 1.0, respectively, establishing a proteomics-based diagnostic model for RA ( 39 ). Furthermore, leveraging metagenomic data to predict the microbiomic characteristics of the gut in autoimmune diseases has been demonstrated to discriminate between various types of autoimmune disorders ( 40 ).…”