Osteoarthritis (OA) is a disease commonly diagnosed at advanced stages, when the damage to the joint is irreversible. In OA pathogenesis, changes involving the joint could lead to the production of autoantibodies (AAbs) through a humoral response, even at asymptomatic stages. Therefore, the discovery of an AAb signature might facilitate the identification of those patients who will develop the disorder.
In this study we analyzed sera obtained at baseline from subjects belonging to the Osteoarthritis Initiative (OAI) cohort that were subsequently followed for 96-months. Ten pools of sera per study group (incident and not-incident) were blindly analyzed by Nucleic Acid-Programmable Protein Arrays (NAPPA) to screen immunoreactivity against 2,125 human proteins. Quantitative data were normalized following the Biodesign Institute criteria and a cut-off level was calculated by the median intensity absolute deviation rule from all the spots through all the pools to determine AAb reactivities. A differential spot analysis was carried out with the antigens over the cut-off by Wilcoxson Rank-Sum test. Among the final panel of candidate AAbs, the association of the reactivity levels against methionine adenosyltransferase 2 subunit beta (MAT2ß) with OA onset was verified in 327 sera from the OAI cohort by NAPPA-ELISA assays.
Among the 2125 proteins in the array, a panel of 6 AAbs showed significant (p<0.05) different baseline reactivity levels between the incident and not-incident subjects. The NAPPA-ELISA analysis verified the presence of significant higher baseline reactivity levels of MAT2ß–AAb in those patients who did develop OA at some point of the 96-months follow-up period (0.58 ± 0.22 vs 0.49 ± 0.23 a.u., p= 3.140E-04). The odds ratio of incident OA was 5.99 (2.16–16.63) times higher per one-unit increase of MAT2ß-AAb.
In conclusion, we have identified an autoantibody signature in serum that could be useful for the diagnosis of osteoarthritis at early, asymptomatic stages.