Serum proteomic networks associate with pre-clinical rheumatoid arthritis autoantibodies and longitudinal outcomes

Abstract

Read online

ObjectivesThe development of autoantibody directed towards citrullinated proteins (ACPA) are predictive of RA in at-risk individuals. The biological events that underpin loss of immune tolerance and progression into inflammatory arthritis are not known. We sought to identify serum proteomic alterations that drive autoantibody formation, persistence and progression into inflammatory arthritis in a cohort of first-degree relatives (FDR) of RA patients.MethodsWe studied baseline serum samples from a cohort of Indigenous FDR (n = 147) and quantified serum proteins using a 48-plex platform. Longitudinal outcomes were defined on the basis of ACPA status and progression into inflammatory arthritis (IA). K-means clustering, differential expression, and principal components analyze group differences. A co-expression module analysis was used to identify enriched networks. Random forest was used to classify ACPA positive samples, while network analysis was used to understand underlying biological processes based on protein expression.ResultsWe defined 6 proteomic clusters, with enrichment of ACPA positive samples in one of the clusters. 23 of 24 differentially expressed proteins in ACPA positive samples were upregulated. A co-expression network was enriched in ACPA positive sera and individuals who progressed into IA. Random Forest achieved an area under the curve of 0.767 to classify ACPA positive sera in a test dataset. Network analysis revealed upregulation of JAK-STAT signalling as being activated in those at highest risk to develop future IA.ConclusionsThe serum proteome provides a rich dataset to understand biological processes in ACPA seropositive individuals. A combination of serum biomarkers, including ACPA, may predict future arthritis onset in at-risk individuals.

Keywords

• rheumatoid arthritis
• proteomics
• pre-clinical
• jak-stat
• prevention
• machine learning