Long term anti-tumour necrosis factor alpha monotherapy in rheumatoid arthritis: effect on radiological course and prognostic value of markers of cartilage turnover and endothelial activation

Abstract

Objectives: To investigate the effect of prolonged neutralisation of tumour necrosis factor alpha (TNFalpha) on the radiological course in rheumatoid arthritis (RA). To assess whether the radiological course can be predicted by clinical variables or biological markers of cartilage and synovium turnover and of endothelial activation.

Patients and methods: Forty seven patients with active RA enrolled at our centre in monotherapy trials with adalimumab (D2E7), a fully human anti-TNFalpha monoclonal antibody, were studied for two years. Radiographs of hands and feet obtained at baseline and after one and two years were scored in chronological order by a single, blinded observer using the modified Sharp method. Radiological course was classified as stable or progressive using the smallest detectable difference as cut off point. The relation between radiological course and serum markers of cartilage and synovium turnover (metalloproteinases (MMP-1 and MMP-3), cartilage oligomeric matrix protein (COMP), human cartilage glycoprotein-39 (HC gp-39)), endothelial activation (soluble E-selectin and intercellular adhesion molecule (ICAM-1)), and integrated measures of disease activity were assessed using univariate and multivariate analysis.

Results: Radiological evaluation was performed in 36 patients with paired sets of radiographs at baseline and two years. After two years a total of 15/36 (42%) presented no radiological progression. More patients with stable radiological course were still receiving anti-TNFalpha treatment after two years (13/15 (87%) v 11/21 (52%); p=0.03) and had lower baseline COMP and sICAM-1 levels (p=0.01 and 0.04, respectively) than those in the group with progressive disease. In a logistic regression model the combination of sustained TNF neutralisation and baseline COMP and sICAM-1 levels was predictive for radiological outcome (p=0.03). C reactive protein and disease activity score area under the curve were significantly correlated with changes in radiological scores after two years (r=0.40 and 0.37, p<0.05). Long term TNFalpha neutralisation decreased the levels of COMP, sICAM, MMPs, and HC gp-39, but not sE-selectin.

Conclusion: The results suggest that long term monotherapy with anti-TNFalpha has a positive effect on radiological outcome and modulates cartilage and synovium turnover as measured by biological markers. Baseline serum sICAM-1 levels and COMP levels may be helpful to identify patients with progressive or non-progressive radiological outcome.

Figure 1

Scheme of patients receiving treatment, primary reasons for drop out, and treatments used thereafter (box) during the two year follow up. *Patients who did not respond after three administrations of adalimumab were withdrawn according to the study protocol.

Figure 2

Course of the median DAS and CRP during the two year follow up. Intention to treat analysis.

Figure 3

Short and long term changes in markers of synovium turnover (metalloproteinases MMP-1 (A), MMP-3 (B), endothelial activation (sE-selectin (C), sICAM-1 (D)) and cartilage and COMP (E), and HC gp-39 (F)). Short time changes: changes seen within two weeks (D=0 to D=14) after the first dose of anti-TNFα (continuous line) or placebo (dashed line) in the double blind phase. Long term changes: changes occurring between baseline and the two years follow up. The boxes illustrate the mean (horizontal bar), 25th and 75th centiles (box) and 5th and 95th centiles (bars). Significance is shown as p<0.05, tested within groups (*) and between groups (†).

Figure 4

Concentrations of COMP (A) and sICAM-1 (B) in patients with and without radiological progression. The boxes illustrate the mean (horizontal bar), 25th and 75th centiles (box) and 5th and 95th centiles (bars). Significance is shown as p<0.05, tested between groups (†).