Abstract
associated with the use of tumor necrosis factor (TNF-) antibody therapies in randomized controlled trials of pa- tients with rheumatoid arthritis (RA). Because each origi- nal study had few events, answering this question requires pooling the findings of many trials, each of which has lim- ited power to examine adverse event rates. The authors did not include the results of the PREMIER trial because at the time of their meta-analysis it was only in abstract form and the data were apparently unavailable. However, this 2-year trial with 799 participants has subse- quently been published. 2 We used exactly the same meth- ods described by Bongartz et al, adding the adverse events of the PREMIER trial, and found that it decreased the pooled odds ratio (OR) for malignancy from 3.29 (95% confidence interval (CI), 1.19-9.08) to 2.02 (95% CI, 0.95- 4.29) using the Mantel-Haenszel method with continuity correction. It decreased the OR from 4.9 (95% CI, 1.6-22.0) to 2.3 (95% CI, 1.1-5.8) using a Bayesian fixed-effects meta-analytic technique. Inclusion of the PREMIER data decreased the pooled ORs for serious infection by approxi- mately 15%, but they remain significantly greater than 1. Themeta-analysisbyBongartzetaladdressesanissuewith important ramifications for persons with rheumatic diseases receivingthesemedications.Theinclusionofthisrecentlarge, long-term trial involving adalimumab has a substantial im- pact on the pooled estimates with respect to the risk of ma- lignancy. We believe that a more precise estimate of the risk awaits the accumulation of yet more longer-term data.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call