Antibodies against type II collagen in rheumatoid arthritis. Extended investigations in a large case-control study.

Abstract Introduction Failure in the mechanism of self-tolerance in T or B cells can lead to autoimmunity. One of the autoimmune diseases is rheumatoid arthritis (RA), which is a chronic inflammatory disease of unknown cause and is characterized by systemic inflammation, autoantibodies and joint destruction. Serology is crucial for the classification of this disease. The first autoantibody found in RA patients was Rheumatoid factor (RF). However, anti-citrullinated peptide antibodies (ACPAs), a relatively new group of autoantibodies found in 70-90% of RA patients, are diagnostically more specific than RF. Type II collagen (CII) is the most abundant protein in human cartilage. In RA patients, immunity against CII leads to cartilage degradation and loss of joint function. Already from the 1970s, antibodies to CII (anti-CII) were found in RA sera, suggesting that CII autoimmunity might be pathogenetically important. Previous studies from our group show that a subgroup of patients with high levels of anti-CII at the time of diagnosis at the same time have high levels of inflammation in the joints. This is probably caused by anti-CII immune complexes (IC) inducing pro-inflammatory cytokines from macrophages. Although anti-CII positive patients have high inflammatory activity early on, as anti-CII levels decrease during the first year, the associated inflammation also diminishes. Thus, anti-CII positive patients have a rather good prognosis. Moreover, it is assumed that since anti-CII positive patients have a better prognosis than ACPA positive, patients with elevated anti-CII at the time of diagnosis might benefit from different and milder treatment. Previous studies from the group were performed on stored patient samples from the time before modern treatments with biologic agents (1995-2005). In this study, we aimed to investigate patients belonging to a more recent RA cohort, diagnosed between 2005-2014, with the aim to investigate whether patients with the anti-CII-associated RA phenotype would respond differently depending on the use of different modern RA therapies. Patients and Methods The primary cohort consisted of 2335 RA patients and 480 non-RA controls from the Epidermiological Investigations in Rheumatoid Arthritis (EIRA) case-control study. As we run into methodological problems two subgroups with 62 and 40 RA patients from the previous anti-CII studies were investigated when modifying the ELISA procedure, as well as a group of earlier investigated patients with non-specific ELISA reactivity. Totally 2776 RA patients were investigated. All investigated patients fulfilled the American College of Rheumatology classification criteria. To measure the anti-CII levels in RA patients and healthy controls anti-CII ELISA was performed. During the experiments, several different sources of CII from human, rat and bovine origin, and two different alternative coating buffers were used. The optical density (OD) was measured at 450 nm and anti-CII concentrations were calculated against the standard curve from an RA patient with high anti-CII levels. Results My first analysis of the EIRA cohort showed that anti-CII are higher in RA patients than in controls, but could not confirm the association with acute onset RA. This was an unexpected finding and changed the focus of this master thesis project, to modify the measurement of anti-CII. Re-investigation of EIRA I showed that a proprietary coating buffer is important in the assay. Moreover, when different samples from RA patients were tested with bovine, rat and three different lots of human CII, correlation tests with clinical measures showed that bovine collagen and a new lot of human CII- prepared by the supplying company solely for this project- showed the strongest associations. Thereafter the EIRA cohort was re-investigated with two ELISAs, using bovine and human CII coated with the proprietary buffer. At the time of thesis writing almost all of the EIRA samples have been re-analysed, and results from both the modified ELISAs show the awaited clinical associations to early inflammation. Conclusion Keeping the integrity of triple helical collagen is very important for the identification anti-native CII in RA patients. Our results show that the use of the proprietary coating buffer appears to be instrumental in this assay, irrespective of what source of CII was used. The new lot of human CII shows significant associations with the clinical measures, but associations are somewhat stronger with bovine CII. After finalising the re-investigations, we will be able to conclude which of the two analyses is most appropriate, and the corresponding dataset will then be merged with data from the first part of the EIRA study investigated previously by other group members. As anti-CII analysis shows the association to disease activity and prognosis, it can be used for predicting prognosis of RA and choosing the appropriate therapy in newly diagnosed RA patients, which might be clinically useful for rheumatologists. Our hypothesis is that as anti-CII positive patients have strong early inflammatory response but good long-term prognosis, they might benefit from other and perhaps short-term treatment compared to other RA patients. If this is correct, our finding can have impact on the economy as it can define the patients who will not need expensive long-term medications. As modern anti-rheumatic therapies carry the risk of infections, such individualized therapies might also benefit anti-CII positive patients