Background Osteoarthritis (OA) is the most common rheumatic disease. To date, the diagnostic methods of OA are very limited and low accurate to detect changes in the joint, what means to wait for a long time to get reliable information about the disease progression. Moreover, there are no available medications capable of halting its characteristic cartilage degradation. Therefore, there is a significant interest to find new biomarkers potentially useful for early diagnosis, prognosis and therapeutic monitoring.
Objectives In the present study, we utilize different affinity proteomics approaches for profiling autoantibodies and proteins in a large number of serum samples from OA patients and compare this profile with that observed in control individuals and other rheumatic diseases. Our overall objective is to develop a novel diagnosis method for the high-throughput analysis of OA patients, applicable in clinical practice.
Methods We analyzed a total of 960 serum samples from control individuals (Ctrl) and patients with osteoarthritis (OA), rheumatoid arthritis (RA) and psoriatic arthritis (PSA). On one hand, to characterize the autoantibody repertoire from each group of study, we used antigen arrays and NAPPA (Nucleic Acid Programmable Protein Arrays). In the first platform, a total of 3840 protein fragments were analysed and 373 were selected for further validation analysis. Using NAPPA, we studied autoantibodies towards 80 different proteins selected by their relevance in OA pathology. On the other hand, to characterize the different protein profile of these samples, 174 antibodies towards 78 different proteins included in Human Protein Atlas (HPA) were coupled to bead array and screened with the 960-sample set.
Results Using antigen array and NAPPA, 15 autoantibodies were identified as potential candidates to distinguish between OA patients, RA patients and controls. In complementary efforts, using antibody arrays, we identified a panel of 12 serum proteins whose levels distinguish between OA patients and controls. Moreover, we identified 34 proteins significantly different between PSA and Ctrl, and a panel of 45 proteins, which distinguish between OA patients and PSA patients.
Conclusions Broad-scaled protein profiling of autoantibody repertoires and protein levels in serum enables discovery of potential novel biomarkers in osteoarthritis. Using different protein array formats we have defined interesting marker candidates, which allow distinguishing serum samples from control individuals, OA patients and other rheumatic diseases. These proteins are now brought into the next level of verification where a new cohort of new serum samples will be utilized.
Disclosure of Interest None declared