The combined use of disease activity and infliximab serum trough concentrations for early prediction of (non-)response to infliximab in rheumatoid arthritis

AIM Early prediction of (non-)response to infliximab therapy can improve therapeutic benefit by avoiding unnecessary periods of high disease activity during ineffective therapy. This prospective cohort study therefore aimed to study the predictive value of (1) disease activity alone and (2) infliximab serum trough concentrations in addition to disease activity 6 weeks after start of treatment for achieving low disease activity after 6 months. METHODS Disease activity and infliximab serum trough concentrations were assessed in all rheumatoid arthritis (RA) patients at 2, 6 and 26 weeks after initiation of infliximab therapy. Receiver operating characteristic (ROC) curves and Youden indices were used to calculate specificity for prediction of good response after 6 months while aiming for maximum sensitivity. RESULTS Fifty-seven consecutive RA patients starting with infliximab therapy were included. After 6 months, 15 (26%, 95 % CI 15, 38%) patients reached good European League against Rheumatism (EULAR) response. A disease activity score <4.2 at 6 weeks after initiation of therapy was a moderate predictor for reaching EULAR response after 6 months (sensitivity 100%, specificity 49%). Infliximab serum trough concentrations (>2.5 mg l(-1)) as predictor complimentary to disease activity (< 4.2) slightly increased the specificity from 49% to 54% without changing the sensitivity (100%). As 39% of the patients did not fulfill at least one of these criteria at week 6, these patients could already be switched to another therapy after 6 weeks. CONCLUSIONS The combination of disease activity and infliximab serum trough concentrations could be a fair predictor to identify early (after 6 weeks) patients who have insufficient response after 6 months of therapy.