RESUMEN
WHAT IS KNOWN AND OBJECTIVE: Psoriatic arthritis is an autoimmune disease characterized by chronic inflammation of the skin and joints. Anti-TNF drugs reduce the severity of the disease in the long term. This study compares the efficacy and safety of adalimumab, etanercept, infliximab and golimumab in patients with psoriatic arthritis. METHODS: Direct comparison was based on a literature search of drug comparison studies, whereas indirect treatment comparison was based on phase III clinical trials with biological agents, involving similar populations and durations, and with the same outcome. ACR50 was taken as primary outcome for comparison, whereas ACR20 and ACR70 were used as secondary outcomes. Indirect comparisons were made using infliximab as the reference drug and the Bucher method. In calculating δ (the maximum acceptable difference as a clinical criterion of equivalence), use was made of half of the absolute risk reduction obtained in the meta-analysis of the clinical trials included in the indirect comparison (ARR 32%; δ: 16%). The four anti-TNF drugs were also compared in relation to the secondary outcomes and adverse effects. RESULTS AND DISCUSSION: Reported direct and indirect comparisons of the four drugs did not include golimumab, and did not yield conclusive results. Four clinical trials - one for each drug studied - were identified. The estimated differences for the primary outcome, ACR50, between infliximab and the other drugs were adalimumab (ARR 4%, 95% CI -9·5 to 17·5), etanercept (ARR 4%, 95% CI -10·5 to 18·5) and golimumab (ARR 9%, 95% CI -5·4 to 23·4). Likewise, there were no relevant differences between the drugs in relation to the secondary efficacy outcomes, except for etanercept, which was less effective in ACR70 response. For adverse reactions, there were also no significant differences except for injection site, reactions which were more frequent with etanercept, with a mean difference of 26% relative to infliximab. WHAT IS NEW AND CONCLUSION: No significant differences were found in ACR50 responses to the four drugs after 24 weeks. Injection-site reactions were more common with etanercept, but this was insufficient to invalidate the inference that clinically the four drugs can be regarded as clinically equivalent for the treatment of psoriatic arthritis.