Outcome and safety of TNFα antagonist therapy in 475 consecutive outpatients (with rheumatoid arthritis or spondyloarthropathies) treated by a single physician according to their eligibility for clinical trials.

OBJECTIVE: To investigate the effectiveness and safety of TNFα antagonists in patients with rheumatoid arthritis (RA) or spondyloarthropathies (SpA) treated by a single physician, according to the presence of the inclusion and non-inclusion criteria used to select patients for pivotal clinical trials. METHODS: Effectiveness was evaluated based on four categories defined by the DAS28-ESR and BASDAI values, from a very good response (mean DAS-28-ESR less than 3.2 and mean BASDAI less than 2.0) to failure (DAS28-ESR unchanged or greater than 5.1 and BASDAI unchanged). Serious adverse events were defined as events that required permanent TNFα antagonist discontinuation or that led to sequelae, hospital admission, or death. RESULTS: The study included 475 patients, 230 with RA, 226 with SpA, 10 with juvenile-onset arthritis, and nine with unclassifiable arthritis. Mean number of TNFα antagonists used per patient was 1.3 and mean duration of TNFα antagonist treatment was 28±23 months. Overall, 41% of patients met the inclusion and non-inclusion criteria used in pivotal trials; the proportion was 43% in the RA group and 40% in the SpA group. These patients had a 3-fold higher rate of very good responses (54 versus 19%) and a 5-fold lower rate of failures (5 versus 25%) compared to the other patients. Of the 15 (3%) patients who died, none met pivotal trial criteria. The group that met pivotal trial criteria had a significantly lower rate of serious adverse events (11 versus 16%; Chi(2), p=0.0001), although age was similar in the two groups (53±16 years versus 57±14 years). CONCLUSION: Patients meeting the selection criteria used in pivotal trials had a higher response rate and significantly fewer serious adverse events.