Background Previous work has suggested subtle infiltration of synovial T cells in the absence of overt synovial inflammation in individuals at risk of developing rheumatoid arthritis (RA).
Objective To study the molecular changes in synovial tissue preceding arthritis development in preclinical RA.
Materials and methods Sixty-one individuals who were IgM rheumatoid factor (RF) and/or anti-citrullinated protein antibody (ACPA) positive and without any evidence of arthritis were included. All individuals underwent mini-arthroscopic synovial biopsy sampling of a knee joint at inclusion and were prospectively followed. An explorative genome-wide transcriptional profiling study was performed on synovial biopsies obtained from 13 individuals using Agilent arrays (test cohort). Survival analysis within the software package Significance Analysis of Microarrays was used to identify transcripts with a significant association with arthritis development. The expression level of differentially expressed genes was validated using quantitative real-time PCR in the total cohort.
Results Six of the 13 individuals in the explorative study developed RA after a median follow up time of 20 months (IQR 2 – 44). The 7 individuals who did not develop RA had a median follow up time of 56 months (IQR 56 – 60). Using a False Discovery Rate of < 5% we found that increased expression of 3,151 transcripts correlated with a higher risk of arthritis development, and increased expression of 2,437 transcripts correlated with a lower risk. Gene Set Enrichment Analysis revealed that synovial biopsies of individuals who developed RA after follow up display higher expression of genes involved in several immune response-related pathways (e.g. T cell and B cell receptor pathways, cytokine and chemokine signalling and antigen processing and presentation) compared with biopsies of individuals who did not develop RA. In contrast, lower expression was observed for genes involved in e.g. extracellular matrix receptor interaction, Wnt-mediated signal transduction and lipid metabolism. Subsequently, the expression level of a selection of 26 differentially expressed genes was validated by quantitative real-time PCR in the total cohort (n = 61). Two-way hierarchical cluster analysis classified the individuals into two groups, where those individuals who developed arthritis (n = 16) showed a preference to cluster together in the left arm of the dendogram (Fisher’s exact test P < 0.05).
Conclusion This study clearly shows molecular changes appearing in synovial tissues before onset of arthritis in the absence of overt synovitis and demonstrates preclinical synovial activation of immune response genes associated with development of arthritis.