Reduced chemokine and matrix metalloproteinase expression in patients with rheumatoid arthritis achieving remission.

OBJECTIVE: To quantify the changes in synovial expression of mediators of macrophage chemotaxis, matrix degradation, and macrophage infiltration in the synovial membrane of patients with rheumatoid arthritis (RA) achieving American College of Rheumatology (ACR) defined remission and radiological arrest.
METHODS: Knee synovial biopsies were taken from a selected group of 18 patients with RA before and after treatment and immunostained with antibodies specific for CD68; the chemokines macrophage inflammatory protein (MIP)-1a and monocyte chemoattractant protein (MCP)-1; matrix metalloproteinases (MMP-1 and 3) and their inhibitors, the tissue inhibitors of metalloproteinases (TIMP-1 and 2); as well as isotype-specific negative controls. Immunostaining was quantified using a computer assisted color video image analysis system. Radiographs were performed before and after treatment and the Larsen score determined. Patients were arbitrarily divided into 2 groups: the radiological arrest group (defined as change in Larsen score pound 5 from baseline) and radiological progressors (defined as change in Larsen score > 5). Patients were classified according to ACR response criteria.
RESULTS: In the 8 patients who achieved ACR defined remission, there were tendencies toward reductions in the synovial lining layer (LL) expression of MIP-1a by 36% (p = 0.1) and MCP-1 by 48% (p = 0.1). Significant reductions occurred in the expression of MMP-1, by 53% in the LL (p = 0.008) and 59% in synovial sublining layer (SL) (p = 0.02) and MMP-3, by 76% in LL (p = 0.02), and 72% in SL (p = 0.008), but not in TIMP expression. In this group of patients there were reductions in MMP:TIMP ratios, in particular the MMP-1:TIMP-1 ratio in the LL (p = 0.05), MMP-3:TIMP-1 ratio in the LL (p = 0.05) and SL (p = 0.008), and MMP-3:TIMP-2 ratio in the LL (p = 0.04) and SL (p = 0.08). In this group of patients CD68+ macrophage infiltration was significantly reduced in the LL by 59% (p = 0.008) and in the SL by 52% (p = 0.008), which corresponded with the reductions in chemokine expression. In the remaining 10 patients who did not achieve full remission there were no significant changes in the variables studied. In the group achieving ACR 50% or 70% response there was a reduction in CD68 expression that approached significance (p = 0.06 in LL and SL), but there was no significant change in the other variables. There were no significant changes in the patients with an ACR 20% response. In the radiological arrest group (12 patients) there was a 41% reduction in LL expression of MIP-1a (p = 0.05) and MMP-1 (p = 0.06). Reductions in MMP:TIMP expression were also noted, in particular in MMP-1:TIMP-1 expression in the LL (p = 0.04) and MMP-3:TIMP-1 in the SL (p = 0.01). There were corresponding reductions in CD68 expression by 49% (p = 0.009) in LL and by 42% (p = 0.0005) in SL. In the radiological progressors (6 patients) there were no significant reductions in mediator expression.
CONCLUSION: In RA, ACR defined remission is associated with reductions in MMP-1 and 3 expression, with a corresponding reduction in macrophage infiltration and a tendency to reduction in MIP-1a expression. Radiological arrest is associated with reductions in MMP-1 expression, and significant reductions in macrophage infiltration, MIP-1 expression, and MMP:TIMP ratio.