Univariate and multivariate analysis of predicting factors of efficacy
| Patients with EULAR (either good or moderate) response (n=330) | Patients without EULAR response (n=228) | OR 95% CI and p value (univariate analysis) | OR 95% CI and p value, for multivariate analysis with multiple imputation | |
|---|---|---|---|---|
| Age (years), median (IQR) | 59.0 (49.0–68.0) (0.3%) | 58.0 (49.0–68.0) (0.4%) | OR per 10 years: 1.1 (0.9–1.2), p=0.2 | |
| Female (%) | 80.9 (0.0%) | 77.9 (0.9%) | OR: 1.2 (0.8 to 2.3), p=0.4 | |
| Disease duration (years), median (IQR) | 13.0 (8.0–20.0) (1.5%) | 14.0 (9.0–20.0) (0.9%) | OR per 5 years: 0.9 (0.9–1.1), p=0.5 | |
| RF-positivity (%) | 75.6 (8.2%) | 66.7 (9.2%) | OR: 1.6 (1.1–2.3), p=0.03 | OR: 1.03 (0.7 to 1.6), p=0.90 |
| Anti-CCP positivity (%) | 75.9 (17.0%) | 62.2 (15.4%) | OR: 1.9 (1.3–2.9), p=0.001 | OR: 1.9 (1.2 to 2.9), p=0.007 |
| Baseline IgG>12g/l | 40.3% (63.9%) | 36.7% (57.0%) | OR: 1.2 (0.7–2.0), p=0.6 | |
| Number of previous synthetic DMARDs, median (IQR) | 3.0 (2.0–4.0) (3.9%) | 3.0 (2.0–4.0) (7.9%) | OR: 1.0 (0.9–1.2), p=0.5 | |
| % of 0/1/2/3 previous anti-TNF | OR: 1.0 (0.9–1.2), p=0.7 | |||
| None | 11.5 | 10.5 | ||
| 1 | 20.3 | 23.7 | ||
| 2 | 38.5 | 35.5 | ||
| 3 | 29.7 (0.0%) | 30.3 (0.0%) | ||
| % of patients previously treated with RTX | 31.2 (0.0%) | 29.8 (0.0%) | OR: 1.1 (0.7–1.5), p=0.7 | |
| Concomitant treatment with corticosteroids* (%) | 77.8 (1.6%) | 72.2 (0.4%) | OR: 1.4 (0.9–2.0), p=0.1 | |
| Median dosage (mg/d) | 8.0 (5.0–10.0) (2.7%) | 7.5 (0.0–10.0) (1.3%) | OR per 10 units: 1.1 (0.9–1.3), p=0.2 | |
| Concomitant treatment with a synthetic DMARD (%)* | 65.2 (1.5%) | 67.0 (0.4%) | OR: 0.9 (0.6–1.3) | |
| Methotrexate** | 76.0 | 77.6 | OR: 0.8 (0.4–1.5) | |
| Leflunomide** | 13.2 | 15.1 | OR: 1.4 (0.6–3.0), p=0.7 | |
| Combination/others** | 10.8 | 7.4 | ||
| Number of swollen articulations, median (IQR)* | 6 (3–11) (23.0%) | 4 (1–8) (22.4%) | OR: 1.1 (1.0–1.1), p=0.004 | |
| DAS28 ESR at initiation of ABA, median (IQR)* | 5.4 (4.7–6.5) (0.0%) | 4.9 (4.0–6.0) (0.0%) | OR: 1.4 (1.2–1.6), p<0.0001 | OR: 1.3 (1.1 to 1.6), p=0.002 |
| ESR, median (IQR)* | 28.0 (15.0–52.0) (23.0%) | 27.0 (15.0–42.0) (22.4%) | OR per 10 units: 1.1 (0.9–1.1), p=0.2 | |
| CRP, median (IQR)* | 15.7 (5.6–40.1) (16.4%) | 12.6 (4.1–28.0) (15.4%) | OR per 10 units: 1.1 (1.0–1.2), p=0.03 | OR per 10 units 1.02; (0.9 to 1.10), p=0.5 |
Results are median (IQR) or %. Proportions of missing values are in brackets ().
The imputation model contained 6 variables (RF, anti-CCP, concomitant treatment with corticosteroids, DAS28 vs ESR at initiation of ABA, CRP and number of swollen articulations before initiation of ABA). Missing data among these variables were as follows: RF-positivity: 8.6%;anti-CCP positivity: 16.3%; concomitant treatment with corticosteroids: 1.1%; DAS28 vs ESR at initiation of ABA: 0%; CRP: 16.0%; number of swollen articulations before initiation of ABA: 22.8%.
↵** Proportion of treated patients among those treated with a synthetic DMARD
↵* Defined as the day of ABA first infusion.
ABA, abatacept; anti-CCP, anti-cyclic citrullinated peptide; CRP, C reactive protein; DAS, disease activity score; DMARD, disease-modifying anti-rheumatic drugs; ESR, erythrocyte sedimentation rate; EULAR, European League Against Rheumatism; RF, rheumatoid factor; RTX, rituximab; TNF, tumour necrosis factor.