Background Using musculoskeletal ultrasound (US) to assess joint erosions and disease activity in patients with rheumatoid arthritis (RA) an atypical subgroup was identified with active disease demonstrated by significant Power Doppler, but normal C-reactive protein (CRP) levels. We questioned whether this presentation was associated with delayed diagnosis or relative under treatment, risking worse disease outcome and/or disability. We hypothesized that understanding the underlying immune pathology in this atypical subset of patients could directly influence therapeutic targeting in patients whose needs are not currently met.
Objectives Define the immunological and clinical phenotype of RA patients with active disease and either normal or high CRP levels. Investigate whether lack of CRP response is associated with altered inflammatory cytokine signaling and exploit the immunological and clinical phenotype of this patient subgroup to propose a more appropriate treatment algorithm.
Methods 44 RA patients with active synovitis were recruited, defined by ≥1 joint with Power Doppler detected by US, 29 had normal (n)CRP (≤5mg/L) and 15 had high (h)CRP (>5mg/L) levels. Peripheral blood mononuclear cells (PBMCs) and serum as well as detailed clinical and disease activity scores were collected. Blood from 18 age and sex matched healthy donors was obtained. Multicolour flow cytometry was used to perform in depth PBMC immunophenotyping and serum cytokines were assessed using a Cytometric Bead Array.
Results The erosion accrual rate was elevated in patients with nCRP compared to hCRP suggesting more disease-associated joint damage, while no significant differences were detected between patient groups in terms of autoantibody levels, ESR and disease activity scores. Inflammatory cytokine levels were elevated in both the nCRP and hCRP patients including IL1β (p=0.0364; 0.0233) and IL6 (p=0.0009; 0.0007) which are known to trigger CRP production. This suggested that nCRP patients could have defects in downstream IL6 signaling, or alternatively, the disease mechanism may not be IL6-dependent in the nCRP group. Since IL6 is known to support T-cell and T-follicular helper-cell (Tfh) activation and differentiation we compared the T-cell phenotype in the two patient groups. As expected, CD4+ T-cells from hCRP patients with RA were activated compared to healthy donors (p=0.0054) and had increased central memory (p=0.0380, T-CM) and Th17 populations. Similarly, only Tfh-cells in hCRP patients were activated compared to nCRP (p=0.0026). In contrast, nCRP patients had a less inflammatory phenotype characterized by the highest level of regulatory T-cells (T-reg) compared to healthy donors (p=0.0036) and hCRP (p=0.0303) and raised serum IL10 suggesting increased immune modulation in these patients. This was supported by a significant correlation between Treg and TCM populations in the nCRP (p=0.05) but not the hCRP patients.
Conclusions Overall, this supports altered immunological mechanisms in nCRP compared to hCRP patients which could have therapeutic implications.
Disclosure of Interest None declared