OP0291 How Medication History and Average Disease Duration Modify Treatment Effect in Randomised Trials Using Targeted Therapies for Rheumatoid Arthritis: A Meta-Epidemiological Study

BackgroundRandomised trials that have tested targeted therapies (targeted synthetic disease modifying antirheumatic drugs, tsDMARDs and biological disease modifying antirheumatic drugs, bDMARDs) for rheumatoid arthritis (RA) vary in several of their trial and participant characteristics [1], but little is known about whether these characteristics influence the overall effect of the therapies in the reported trials.ObjectivesTo determine whether trial and participant characteristics modify the treatment response when testing the efficacy of targeted therapies (in contrast to comparator) for RA.MethodsWe conducted a meta-epidemiological study of all trials testing a dosage of a targeted therapy approved by regulatory authorities for the treatment of RA. Included trials reported ACR20 data at months 3-6 and used an add-on design. Anakinra and open-label trials were not included. Odds ratios (ORs) were calculated from the response rates and compared among the trial/participant characteristics of interest. Statistical analyses were performed using SAS v 9.3 (REML based models). Comparisons between strata are presented as the Ratio of Odds Ratios (ROR).ResultsSixty-two trials (19,923 RA patients) were included in the primary analyses using ACR20 as the outcome (table 1). Overall, targeted therapies constituted an effective treatment (OR 3.96 95% confidence interval 3.41 to 4.60). The net benefit of targeted therapies was lower in trials including “DMARD-naïve” patients compared with both “DMARD inadequate responders” (ROR=0.45, 0.31 to 0.66) and “biologic inadequate responders” (0.50, 0.29 to 0.87) trials (test for interaction: p=0.0002). Longer disease duration of the trial population was also statistically significantly associated with a higher likelihood of responding to treatment (β=1.05, 1.00 to 1.11 OR's per year). Analyses conducted using DAS28-remission as the outcome supported the above-mentioned findings.Table 1.Results of the stratified meta-analyses with ACR20 as the dependent variableVariableTrial characteristicsTrialsOR (95% CI)p-interactionOverall623.96 (3.41 to 4.60)N.A.Medication history0.0002DMARD-naïve81.97 (1.39 to 2.79)DMARD-inadequate responders494.34 (3.75 to 5.01)Bio-inadequate responders53.92 (2.58 to 5.97)VariableParticipant characteristicsTrialsCoefficient (95% CI)p-valueBaseline disease duration (years)621.05 (1.00 to 1.11)0.03ConclusionsTrial evidence suggests that using targeted therapies add the greatest net benefit in patients who