Clinical features and laboratory test results in the 99 patients with early RA (Table ) were not different from those in the entire Mo-Co-To cohort, which have been published elsewhere [12
Baseline characteristics in 99 patients with early rheumatoid arthritis
Sensitivity for RA of a positive anti-CCP2 test in our population of 99 patients with early RA was 55.5% at baseline and 63.6% at any time during the first three years of follow-up. In seven (7%) patients, anti-CCP2 was negative at baseline but converted to positive within the first three years, whereas in 8 (8%) patients anti-CCP2 was positive at baseline but converted to negative within the first three years (Figure , panel 2). Among 19 patients with increasing anti-CCP2 concentrations during the first three years of follow-up, seven were treated with methotrexate (MTX; 37%), seven with sulfasalazine (SSZ; 37%), and five with both drugs in combination (MTX + SSZ; 26%). Of these 19 patients, seven had no anti-CCP antibodies at baseline and converted to positive during follow-up (2 treated with MTX, 3 with SSZ, and 2 with MTX + SSZ). Among 32 patients with decreasing anti-CCP2 concentrations, 14 were treated with MTX (43.75%), 10 with SSZ (31.25%), 6 with MTX + SSZ, and 1 with hydroxychloroquine; 1 patient received no DMARDs. Of the 32 patients who were anti-CCP2 positive at baseline, 8 converted to negative by the end of the first three years of follow-up (5 treated with MTX, 1 with SSZ, 1 with MTX + SSZ, and 1 with hydroxychloroquine). None of the differences in DMARD regimens across these subgroups was statistically significant. Serum IgM RF was detected in 73.7% of patients at baseline and in 83.8% at some time during the first three years. IgM RF titer status changed within the first three years in 45 patients, of whom 34 converted from positive to negative and 11 from negative to positive (data not shown).
Figure 1 Anti-CCP2 concentrations at baseline (month 0 (M0)), 1 year (M12) and 3 years (M36) of follow-up according to (a) decreasing concentrations (panel 1 and 2), (b) increasing concentrations (panel 1 and 2) or (c) steady concentrations. Patients with transition (more ...)
Significant structural damage was present at baseline in 20 patients and after five years in 59 patients (total Sharp score). Presence of anti-CCP2 at the first determination was not significantly associated with radiographic damage at baseline (13/55 (23.6%) patients with anti-CCP and 7/44 (15.2%) patients without anti-CCP; OR, 1.63; 95% CI, 0.59–4.54; P = not significant).
Presence of anti-CCP2 at any time during the first three years was associated with radiographic damage at the hands and feet at baseline (17/63 (27%) patients with anti-CCP and 3/36 (8.3%) patients without anti-CCP; OR, 3.66; 95% CI, 0.99–13.54; P = 0.063).
To investigate the value of a positive anti-CCP2 test for predicting radiographic progression, we computed the ORs for radiographic progression after five years with the serial anti-CCP2 strategy (Table ) and we compared the results to those obtained with anti-CCP2 determination at baseline only. Table reports the ORs for radiographic progression according to serial IgM RF values and to baseline IgM RF status. The 95% CI values showed that presence of anti-CCP2 at any time during the first three years significantly predicted erosions (P = 0.007), joint space narrowing (P = 0.03), and total score deterioration (P = 0.01), whereas the presence of anti-CCP2 detected by a single determination, at baseline, predicted none of these outcomes.
Anti-CCP and IgM RF as predictors of radiographic joint destruction after 5 years
Figure summarizes the anti-CCP2 antibody level variations among patients with increasing concentrations and patients with decreasing concentrations between baseline and month 36. The median antibody level increase (Δ M36 minus M0; 207 U/ml; range, 57 to 1,190) did not differ significantly from the median antibody level decrease (Δ M0 minus M36; 1003 U/ml; range, 27 to 3,200).
Erosions were noted after five years in 35% (22/63) of patients with positive anti-CCP2 at any time during the first three years and in 8.3% (3/36) of patients with negative anti-CCP2 throughout the first three years (two patients had a transiently and weakly positive anti-CCP2 test at 12 months) (Figure ). Among patients with and without anti-CCP2 during the first three years, 54% (35/65) and 29% (10/34) had joint space narrowing, respectively, and 57% (37/65) and 29% (10/34) experienced total score deterioration, respectively. Figure reports the mean (± standard deviation) changes in the erosion score, narrowing score and total score after five years according to anti-CCP2 variations. The mean erosion score and total score were significantly higher in patients with increasing anti-CCP2, compared to those with decreasing antibody levels. Patients with negative anti-CCP2 tests throughout follow-up had less structural damage (erosions, narrowing and total score) than did patients with increasing anti-CCP2 concentrations. Finally, regarding structural deterioration, patients with stable anti-CCP2 concentrations did not differ from those with increasing or decreasing anti-CCP2 titers.
Progression of radiographic Sharp scores (mean ± SD) after five years in patients with or without anti-CCP2 antibodies, according to the change in anti-CCP2 concentrations between baseline and three years.
We evaluated correlations between the baseline anti-CCP2 antibody level, or the mean of the three anti-CCP2 antibody levels, and the Sharp score changes after five years (erosion score, narrowing score and total score) in the 99 patients. Correlation coefficients for the mean of the three anti-CCP2 concentrations are reported in Table . All three radiographic scores were significantly correlated with the mean serial anti-CCP2 concentrations, whereas the correlations with the baseline anti-CCP2 concentration fell slightly short of significance (Figure ).
Correlation between worsening of radiographic Sharp score and mean serial anti-CCP2 concentration or baseline anti-CCP2 concentration
Figure 2 Correlation between (a) anti-cyclic citrullinated peptide (CCP) concentrations at baseline (month 0 (M0); panels 1 to 3) or (b) the mean serial anti-CCP concentrations (M0 + M12 + M36)/3; panels 1 to 3) and progression of the modified Sharp scores (Δ (more ...)
One or two SE alleles (*0401, *0404, *0405, *0101, or *0102) were found in 68% of patients. This proportion was higher in patients with than in patients without anti-CCP2 (61% and 50%, respectively), although the difference was not statistically significant. The mean serial anti-CCP2 concentrations were compared in patients with and without the SE alleles. No significant differences were found between these two subgroups (705 ± 1,408 U/ml versus 450 ± 632 U/ml). The 15 patients with the *0404 allele had a significantly higher mean serial anti-CCP2 concentration (1,163 ± 2,247 U/ml) compared to the other patients (491 ± 776 U/ml) (P = 0.036), whereas no significant difference was found for the baseline anti-CCP2 concentration. We compared five-year total Sharp scores according to the presence or absence of anti-CCP2 during the first 3 years in patients with or without SE alleles. As shown in Table , the mean modified Sharp scores were significantly higher in patients with a positive anti-CCP2 test at any time during the first three years than in patients without anti-CCP2 throughout the first three years, in both the subgroup with and the subgroup without SE alleles.
Mean radiographic Sharp scores after 5 years in patients with or without anti-CCP2 and shared epitope alleles