Background It is known that serum IL-6 (sIL-6) levels are associated with disease activity and joint destruction in rheumatoid arthritis (RA). Tocilizumab (TCZ) blocks binding of interleukin 6 (IL-6) to both membrane and soluble receptors, thereby preventing the signal transmission and its efficacy in the treatment of RA and other inflammatory diseases has been well documented. A better knowledge on IL-6 modulation in our patients may help us to design strategies for optimal use of TCZ in RA patients.
Objectives To analyse the genetic, sociodemographic and clinical factors that could be involved in the regulation of sIL-6 levels in a population of patients with early arthritis (EA).
Methods We studied 176 patients from our prospective EA register, 75% women, age at disease-onset 53 [43-67] (mean [p25-p75]) years, disease duration at inclusion in the register 5.5 [3.1-8.6] months and 2 to 5 years of follow-up. A total of 693 visits were analyzed (3.7 visits/patient). 71% met the 1987 ACR criteria for classification of RA at two years of follow-up. Rheumatoid factor (RF) was positive in 44% and anti-cyclic citrullinated peptide antibodies (ACPA) in 45%. Sociodemographic data were collected at the first visit and clinical and laboratory data, at each visit, according to a protocol. sIL-6 levels were measured using an ELISA kit from R&D Systems. Genotyping of single-nucleotide polymorphisms rs7574865 on STAT4, rs2476601 on PTPN22, rs231775 on CTLA-4, rs2228145 on the IL-6 receptor and rs1800795 on IL-6 was performed using real-time PCR and specific TaqMan probes (Applied Biosystems). To determine the effect of the independent variables on sIL-6 levels, a generalised linear model, nested by patient and visit, was fitted using the xtgee command of Stata 10.1 for Windows (StataCorp LP, College Station, TX, USA).
Results sIL-6 levels diminished along the follow-up from 3.4 pg/ml [1.5-8] at basal visit to 1.64 pg/ml [0.75-3.4] at the two-year follow-up. The variables which reached statistically significant association with increased sIL-6 levels were: moderate or high activity versus remission (p<0.001 in both cases), age (>65 versus <45 y; p<0.001) and the presence of ACPA (p=0.004). Once the statistical model was adjusted by these variables, methotrexate (MTX) was the only drug associated with a reduction in sIL-6 levels (p<0.001) after fitting the statistical model by the level of disease activity. In addition, the alleles from the STAT4, PTPN22 and CTLA4 polymorphisms associated with the greatest risk for RA correlated with lower sIL-6 levels although without statistical significance. The genetic variants studied on IL-6 gene and its receptor did not have any effect on serum cytokine levels.
Conclusions Our data suggest that elderly patients with EA, those with positive ACPA and with greater disease activity show the highest sIL-6 levels and confirm the association between these levels and disease activity. Therefore, our results would explain, at least in part, the lack of synergy in the MTX/TCZ combination that has been revealed in clinical trials, likely as both these drugs exert their action on the same target.
This work is partially funded by the RETICS program, RD08/0075 (RIER) and FIS 080754 from the Carlos III Health Institute (ISCIII) and a Roche Farma unrestricted grant.
Disclosure of Interest None Declared