Background Progranulin autoantibodies (PGRN) have been previously described in sera of patients with different autoimmune diseases such as ANCA associated vasculitis, large vessel vasculitis, Crohn's disease, and psoriatic arthritis. Basically, progranulin seems to interact with TNF receptors 1 and 2 causing subsequently antiinflammatory effects. However, the presence of autoantibodies against progranulin was proved to result in lower serum progranulin levels, which is hypothesized to entertain proinflammatory environment due to lack of inhibited TNF alpha activity. Here we conducted a study evaluation the presence of PGRN in patients with rheumatoid arthritis, which is an autoimmune disorder with proven TNF alpha mediated joint inflammation.
Objectives Prevalence of serum progranulin autoantibodies in RA with stratification into DAS28 subgroups with prognostic relevance.
Methods From Nov 2011 till Dec 2013 sera of 307 RA patients (Saarland, Berlin) were collected and analyized for PGRN with ELISA (as previously published ) using a retrospective, blinded, non-randomized, two-center study design. All patients were on treatment with conventional DMARDs (n=193) or TNF inhibitors (n=114) at least for three months. All patients fullfill the ACR criteria from 1987 and the revised diagnostic EULAR/ACR criteria from 2011. RA patients characteristics concerning age, gender, treatment history, radiological joint imaging, and RF-/ACPA-serum status were availiable. DAS28 and HAQ representing RA activity were documented at study inclusion visit completed by the previous DAS28 scores as written in the patient's chart under the same present antirheumatic treatment as availiable (retrospective analysis of data from three months till two years in the past).
Results The overall prevalence of PGRN-Abs in RA patients was 26.6%. The RA patients were further stratifyied according to the lowest DAS 28 score during the observation period under the present treatment into patients with scores higher versus lower than 3.0 representing “good treatment response” or not, respectively. Table 1 demonstrates that the presence of PGRN-Abs is independent risk factor for failing “good treatment response” beyond erosive diesease, RF- and ACPA-positivity, HAQ score, and Larsen score in this setting.
Conclusions PGRN-Abs in sera of RA patients seems to be predictive for poor response of conventional DMARDs and biological TNF-alpha blocking agents. However, further prospective studies with larger cohorts of RA patients are required to confirm the results.
Thurner L & Assmann G, Arth Res Ther, 2013
Disclosure of Interest None declared