Revealed heterogeneity in rheumatoid arthritis based on multivariate innate signature analysis

Abstract

Objective A growing body of evidence highlights the persistent activation of the innate immune system and type I interferon (IFN) signature in the pathogenesis of rheumatoid arthritis (RA) and its association with disease activity. Since the recent study revealed heterogeneity in the IFN signature in RA, we investigated for the first time the heterogeneity in innate signature in RA. Methods The innate gene expression signature (10 TLRs, 7 IL1/IL1R family members, and CXCL8/ IL8) was assessed in peripheral blood mononuclear cells from RA patients (n=67), both with active (DAS28 >= 3.2, n=32) and inactive disease (DAS28<3.2, n=35), and in healthy control subjects (n=55). Results Of the 13 deregulated innate genes (TLR2, TLR3, TLR4, TLR5, TLR8, TLR10, IL1B, IL1RN, IL18, IL18R1, IL1RAP, and SIGIRR/IL1R8) associated with RA, TLR10 and IL1RAP are being reported for the first time. Multivariate analysis based on utilising patient similarity networks revealed the existence of four patient's subsets (clusters) based on different TLR8 and IL1RN expression profiles, two in active and two in inactive RA. Moreover, neural network analysis identified two main gene sets describing active RA within an activity- related innate signature (TLR1, TLR2, TLR3, TLR7, TLR8, CXCL8/IL8, IL1RN, IL18R1). When comparing active and inactive RA, upregulated TLR2, TLR4, TLR6, and TLR8 and downregulated TLR10 (P<0.04) expression was associated with the disease activity. Conclusion Our study on the comprehensive innate gene profiling together with multivariate analysis revealed a certain heterogeneity in innate signature within RA patients. Whether the heterogeneity of RA elucidated from diversity in innate signatures may impact the disease course and treatment response deserves future investigations.