Maxime Verhoeven

83 TCZ vs. TCZ+MTX; prevention of radiographic progression in RA baseline joint damage, disease duration or DAS28 were possible treatment effect modifiers (i.e., p≤0.20 for interaction term), stratified analyses were performed for these factors to better interpret the interaction using log-binomial regression (to obtain relative chances) and forest plots. Stratification was based on median scores of effect modifiers. Patients with a score at or below this cut-off were classified in the low-level subgroup. Patients with scores above this cut-off were classified in the high-level subgroup. Relative chances (relative risk; RR) of preventing radiographic progression were calculated, and graphically illustrated per (sub)group. Furthermore, the absolute difference in the risk of preventing radiographic progression was calculated per subgroup. This was calculated using the RR and the rate of no-progression in the TCZ+MTX group as reference. By multiplying these, the rate of no-progression in the TCZ group is calculated. The difference between these rates of no-progression is the absolute risk difference. 1 All analyses were performed with SAS v9.4. All tests were two-sided and p≤0.05 was considered statistically significant. RESULTS Table 1 shows characteristics of all patients included in the analyses (n=1506). In total 1089 patients were classified as having early RA and 417 patients as having established RA. Baseline DAS28 was generally high (reflecting active disease), although slightly lower in U-Act-Early, most likely due to the inclusion criteria being less strict. The median change over two years in total SvdH score, as well as scores for erosions and JSN, were 0 in all trials and treatment arms, Table 1. The maximum change in total SvdH score was 33 in U-Act-Early, 31 in FUNCTION, and 23 in ACT-RAY, respectively (data not shown). 1 An example, the relative chance of preventing radiographic progression of 0.91 for TCZ vs. TCZ+MTX can be translated into an absolute risk difference using the average percentage of patients treated with TCZ+MTX who have no-progression in this group (reference rate: 91%). Using the RR and this reference rate, the percentage of patients treated with TCZ with no-progression would be 0.91*91%= 83%, and the absolute risk difference (91%-83%=) 8%. 5