Predictors, Features and Effects of First Biologic Switch in Rheumatoid Arthritis within GISEA register: Italian 10-Year Experience.

Background/Purpose: To identify predictors for first switch of biologic agents in real-life treatment of moderate to severe rheumatoid arthritis (RA) in Italy. To illustrate switch strategies adopted and to analyze results obtained this way. Methods: Retrospective observational study was performed on RA patients who started biologic agents from 1999 to 2010 because of unresponsive moderate to severe RA within GISEA (“Gruppo italiano per lo studio dell’early arthritis”) register including 14 tertiary rheumatology centers of Italy. Baseline parameters of patients maintaining and patients discontinuing first biologic agent were compared, and stepwise logistic regression model was calculated to find baseline parameters predictive for first switch. Switch strategies between biologics, time to and reason of first switch were analyzed. Efficacy of first switch was determined by EULAR response achieved at 6 and 12 months, and by rate of failure occurring with second line biologic divided for type of agent used and reason of switch. For statistical analysis the Kruskal-Wallis and chi-squared test were used as appropriate. P < 0.05 was considered statistically significant. Results: 3,702 RA patients starting biologic treatment were included. Patients continuing first biologic had better EULAR responses at 6 and 12 months after starting biologic agent than those switching later on. Predictors for first switch were disease activity, glucocorticoids, lower age and disease duration, whereas concomitant DMARDS and the use of adalimumab and etanercept instead of infliximab resulted protective. Switching to second TNF_ blocker was the strategy most applied, and showed greatest EULAR responses at 6 and 12 months. Etanercept performed better for response and failure rate than adalimumab and infliximab. As second line agents biologics with novel mechanisms of action presented lowest discontinuation rates, but were not significantly different from anakinra and etanercept. Switch because of inefficacy resulted in worse response and higher probability to fail again due to inefficacy. Conclusion: Behavior in managing biologics is changing because more agents are available including those working on alternative pathways. Nevertheless, TNF_ blockers confirmed their efficacy in first and second line, but possible superiority of new agents in long-term disease control has to be considered in further studies.