Background/Purpose: An interferon-α gene signature (IGS) has been demonstrated in a number of rheumatological conditions, including rheumatoid arthritis (RA) where it is present in about 20-30% of patients. In established RA the IGS does not correlate with any disease activity parameters however ongoing immunomodulatory treatment may mask any underlying associations. We therefore wished to investigate any association between the IGS and disease activity in early RA patients prior to administration of any disease modifying therapy.

Methods: Disease modifying drug naïve patients fulfilling the ACR/EULAR criteria for a new diagnosis of RA were recruited from the Early Arthritis Clinic at the Freeman Hospital, Newcastle-upon-Tyne, UK. Clinical parameters included sex, age, swollen joint count (SJC), tender joint count (TJC), patient reported visual analogue scale of disease severity (VAS), disease activity score (DAS-28), inflammatory markers (CRP,  ESR), autoantibody status (seropositive defined as either rheumatoid factor and/or anti CCP antibody positive) and symptom duration. In addition whole blood RNA was collected in Tempus Blood RNA tubes at diagnosis and following RNA extraction the expression of 5 interferon-α response genes (MXA, IFI6, OAS-1, ISG-15 and IFI-44L) was quantified using TAQMAN real-time PCR. Using the whole blood expression of these genes in an independent healthy control population an interferon gene score was calculated as previously published (Feng et al., Arthritis and rheumatism. 2006; 54(9):2951-62). Statistical tests were performed using GraphPad prism v5 (GraphPad Software, Inc) and included linear regression and Mann-Whitney U tests. Significance was taken when p<0.05.

Results:

We recruited 36 early RA patients. Median age was 59 years [range 30-91] and male:female ratio was 1:2. The healthy control population (n=17) had a median age 33 [range 23-57] and a male:female ratio 1:1. Notably age and sex had no significant association with the IGS within our early RA cohort (p=0.549 and p=0.513 respectively, linear regression). We found no significant difference in IGS score between our early seropositive RA and seronegative RA (p=0.925, Mann=Whitney U). Furthermore there was no association between any of the disease activity parameters measured (SJC, TJC, VAS, DAS-28, symptom duration, ESR and CRP) and the interferon gene score or individual gene expression (p>0.05 for all, linear regression).

Conclusion:

We demonstrate the novel finding that in early, drug naïve RA the IGS does not associate with markers of disease activity. This is similar to what has been shown in established RA. We know however that a high IGS score may predict poor response to certain therapies therefore, rather than reflecting inflammation/disease activity, interferon-α exposure may represent a distinct pathological process within a subset of RA patients. We propose that understanding this heterogeneity further is likely to assist in the development of novel therapeutic pathways for a subset of RA patients.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-interferon-gene-signature-in-early-rheumatoid-arthritis-demonstrates-no-significant-association-with-disease-activity/