Background TNF antagonists have been equally effective in the treatment of immunoinflammatory diseases. However, differences in their characteristics and the introduction of biosimilars may be associated with different safety profiles.

Objectives The main goal of this study is to compare the safety of infliximab (IFX), etanercept (ETN), and adalimumab (ADA), as well as of IFX-biosimilar.

Methods We assessed the adverse events reported in the EudraVigilance database between 2004 and 2016. The MEdDRA® system was used to classify the adverse events according to the primary system organ class. For a direct comparison of the data obtained, the adverse events reported for each drug were normalized using the number of treatments for the same period. The reporting odds ratio (ROR) and its 95% confidence intervals (CI) were calculated regarding the different categories of adverse events. The incidence of serious adverse events, serious infections, withdrawals due to adverse events and deaths were also calculated.

Results The EudraVigilance database contains 851 882 adverse events reported for IFX, ETN, and ADA. During this period, the different TNF antagonists have shown almost the same safety profile. The reported adverse events were classified by systems organ class (SOC) and the most frequent were administration site conditions (28.8%) and infections and infestations (11.2%), Safety was not statistically different. The comparison between IFX originator and its biosimilar did not show statistically significant differences in safety (ROR 1.08 [0.80, 1.46]) during the initial 3-years after launch for both drugs. However, a small non-significant increase in immune reactions during administration was reported for IFX-biosimilar, which might reflect increasing attention for this class of drugs.

Conclusions The comparison of reference IFX and IFX-biosimilar did not demonstrate statistically significant differences in safety. This pharmacovigilance study provides the first analysis of TNF antagonists from the EudraVigilance database and offers a framework for safety comparison between originators and biosimilar TNF antagonists.

Acknowledgements We would like to acknowledge Eudravigilance for providing access to the database.

Disclosure of Interest None declared