AB0023 Gene-gene interaction and early rheumatoid arthritis: effects on the response to therapy

Abstract

Objectives To evaluate the relationship between the response to therapy in early rheumatoid arthritis (ERA) and polymorphisms in the genes of molecules involved in inflammatory processes (TNF-α, TNF-RII, IL6, IL1b, IL1Ra, BAFF) or in the homeostasis of T and B cells (PTPN22, HS1,2 enhancer). Methods Two hundreds and eighty six ERA patients were enrolled in the study (females: 75.2%; mean age: 55.1±13.6 years; anti-CCP positive patients: 63.3%; IgA RF positive patients: 34.5%; IgM RF positive patients: 51.0%). All patients were treated with Methotrexate for three months, then with a combination with Tumor Necrosis Factor (TNF) blockers if they had an incomplete response after 3-month treatment period. DNA from patients was genotyped for the TNF-α (-308G>A and -238G>A), TNF-RII +196T>G, IL6 -174G>C, IL1b (-511C>T and +3953C>T), IL1Ra VNTR, BAFF -871C>T, PTPN22 +1858C>T, HS1,2A polymorphisms by polymerase chain reaction restriction fragment length polymorphism method. Clinical and laboratory data were recorded at baseline and every 3 months. The EULAR response criteria based on DAS were used to assess the disease activity after 3 and 6 months of follow-up (good response: DAS <2.4 and a fall in this score from baseline by >1.2; sustained remission: DAS value of <1.6 for at least two consecutive visits 3 months apart). Exact Hardy Weinberg Equilibrium (HWE) tests were performed for each polymorphism independently. Generalized multifactor dimensionality reduction (MDR) was used to verify interaction results. Models were evaluated on the testing balanced accuracy statistic (TBA), the cross-validation consistency (CVC) and the statistical significance of the model. Results All the polymorphisms were in Hardy-Weinberg equilibrium. According to the multi-locus GMDR approach, TNF-308G>A, TNFRII+196T>G, IL1RA VNTR, IL1b-511, IL1b+3956, IL6-174G>C and BAFF-871C>T polymorphisms predicted the EULAR good response at 3 months correctly 60.9% of the time (Sign Test p=0.001). The EULAR sustained remission at 6 months was predicted 55.9% of the time by an ensemble of 5 polymorphisms (TNF-α-308, TNF-RII+196, IL1Ra VNTR, IL1β+3953, BAFF-871; Sign Test p=0.055). Conclusions The genetic background is a possible predictor of the response to therapy of patients with early rheumatoid arthritis. The analysis of gene-gene interactions, especially in genes of molecules related to the major inflammatory cytokines, will lead to a better understanding of the biological pathways underlying the therapeutic outcome in the earliest phases of the disease. Disclosure of Interest None Declared