Interferon gene expression signature in rheumatoid arthritis neutrophils correlates with a good response to TNFi therapy

• Published in: Rheumatology (Oxford, England) England), 2015-01, Vol.54 (1), p.188-193
• Main Authors: Wright, Helen L, Thomas, Huw B, Moots, Robert J, Edwards, Steven W
• Format: Article
• Language: English
• Subjects: Adult; Aged; Antirheumatic Agents - therapeutic use; Arthritis, Rheumatoid - drug therapy; Arthritis, Rheumatoid - metabolism; Arthritis, Rheumatoid - pathology; Biomarkers - metabolism; Case-Control Studies; Female; Gene Expression Profiling; Humans; Interferon Type I - genetics; Interferon Type I - metabolism; Interferon-gamma - genetics; Interferon-gamma - metabolism; Interferons - genetics; Interferons - metabolism; Male; Middle Aged; Neutrophils - metabolism; Neutrophils - pathology; Severity of Illness Index; Signal Transduction - genetics; STAT Transcription Factors - metabolism; Treatment Outcome; Tumor Necrosis Factor-alpha - antagonists & inhibitors
• ISSN: 1462-0324; 1462-0332
• DOI: 10.1093/rheumatology/keu299

The aim of this study was to use whole transcriptome sequencing (RNA-Seq) of RA neutrophils to identify pre-therapy gene expression signatures that correlate with disease activity or response to TNF inhibitor (TNFi) therapy. Neutrophils were isolated from the venous blood of RA patients (n = 20) pre-TNFi therapy and from healthy controls (n = 6). RNA was poly(A) selected and sequenced on the Illumina HiSeq 2000 platform. Reads were mapped to the human genome (hg19) using TopHat and differential expression analysis was carried out using edgeR (5% false discovery rate). Signalling pathway analysis was carried out using Ingenuity Pathway Analysis (IPA) software. IFN signalling was confirmed by western blotting for phosphorylated signal transducer and activator of transcription (STAT) proteins. Response to TNFi was measured at 12 weeks using change in the 28-item DAS (DAS28). Pathway analysis with IPA predicted activation of IFN signalling in RA neutrophils, identifying 178 IFN-response genes regulated by IFN-α, IFN-β or IFN-γ (P < 0.01). IPA also predicted activation of STAT1, STAT2 and STAT3 transcription factors in RA neutrophils (P < 0.01), which was confirmed by western blotting. Expression of IFN-response genes was heterogeneous and patients could be categorized as IFN-high or IFN-low. Patients in the IFN-high group achieved a better response to TNFi therapy [ΔDAS28, P = 0.05, odds ratio (OR) 1.4 (95% CI 1.005, 1.950)] than patients in the IFN-low group. The level of expression of IFN-response genes (IFN score) predicted a good response [European League Against Rheumatism (EULAR) criteria] to TNFi using receiver operating characteristic curve analysis (area under the curve 0.76). IFN-response genes are significantly up-regulated in RA neutrophils compared with healthy controls. Higher IFN-response gene expression in RA neutrophils correlates with a good response to TNFi therapy.