Adverse Events Of TNF-α Inhibitors Application In Patients With Rheumatoid Arthritis:A Network Meta-analysis

Backgrounds:Antitumor necrosis factor(anti-TNF)therapy has been confirmed as an efficacious therapeutic strategy in rheumatoid arthritis(RA).Currently,Five TNF-α inhibitors have been approved for the treatment of RA:adalimumab,golimumab,infliximab,certolizumab pegol,and etanercept.Although there have been some pair-wise meta-analyses and network meta-analyses that evaluate the safety of different TNF-α inhibitors therapies of RA.Nevertheless,most of the trials only focused on total adverse events(AEs)and serious adverse events(SAEs)or just one kind of detailed AEs,and there are inconsistencies between the analysis results of the previous meta-analysis and the later results.In addition,most of the reports on safety of single TNF-α inhibitors combined with or without disease-modifying anti-rheumatic drugs(DMARDs)compared with the evaluation of placebo or DMARDs.There are few studies on direct comparison between different TNF-α inhibitors,and it is difficult to judge their safety,and long-term AEs are also difficult to be detected before drugs are marketed.Objectives:Based on the advantages of direct and indirect comparison of multiple treatment options,the risk of 10 different types of adverse events with TNF-α inhibitors for RA patients was inferred by network meta-analysis.To evaluate the safety of five TNF-α inhibitors in the treatment of RA resulting in AEs,SAEs,infection,serious infection,and malignant tumor,especially the safety of long-term use;To obtain a more comprehensive and accurate judgment and provide guidance for clinical application.Methods:We have deeply searched the records ranging from the earliest dates to August 2021 in PubMed,EMBASE,and the Cochrane Library.Match each "drug" and "RA",limit according to the conditions of "randomized controlled trial".Two researchers screened the literature according to the inclusion and exclusion criteria independently.After excluding repeated studies,review,improper control group and other irrelevant studies are excluded by reading the title and abstract.Finally,by reading the full text,the literatures that can meet the inclusion and exclusion criteria have been further screened.The titles and abstracts are read by using EndNote20 software,and then the remaining articles are browsed in full degree.Data are extracted and the following information is summarized:first author,publication year,country,total number of subjects,type of TNF-α inhibitor,age range,follow-up time,trial duration.The risk of bias is assessed by project scoring.Besides,we use the Jadad scale to assess the quality of randomized to control trials.AEs,SAEs,infections,serious infections,and malignancies are used as primary outcome measures,liver and kidney function indexes and tumor markers were used as secondary outcome indexes.The adverse drug reactions of 12 treatments which include placebo are compared by network meta-analysis.We use STATA16 software for statistical analysis.First,the comparison network diagram is drawn to compare the treatment measures,and the P value is calculated according to the existence of closed loop for consistency test.Then calculate the 95%confidence interval(CI)of the count data to get the OR value.The area under the cumulative ranking probability curve(SUCRA)is analyzed to obtain the possibility of different types of adverse drug reactions for each treatment measure.The higher the SUCRA value is,the higher the risk we may get.Comparison-corrected funnel plots are drawn using STATA16 to assess publication bias for final screening.All tests are two-sided with a significance level of 0.05.Result:72 RCTs involving 28332 subjects are included.In terms of causing the risk of AEs,adalimumab plus DMARDs is higher than etanercept plus DMARDs(OR=1.32,95%CI:1.03,1.67)and DMARDs alone(OR=1.28,95%CI:1.08,1.52);certolizumab pegol combined with DMARDs is higher than etanercept combined with DMARDs(OR=1.34,95%Cl:1.05,1.70)and DMARDs alone(OR=1.30,95%CI:1.10,1.54).In causing the risk of SAEs,compared with golimumab combined DMARDs,etanercept combined with DMARDs(OR=0.39,95%CI:0.15,0.96),adalimumab(OR=0.20,95%CI:0.07,0.59),golimumab(OR=0.19,95%CI:0.05,0.77),infliximab treatment(OR=0.15,95%CI:0.03,0.71)significantly may reduce the risk.In terms of risk of infection,compared with infliximab combined DMARDs,golimumab combined with DMARDs(OR=0.51,95%CI:0.31,0.83),etanercept combined with DMARDs(OR=0.56,95%CI:0.35,0.91),adalimumab combined with DMARDs(OR=0.59,95%CI:0.37,0.95)and etanercept treatment(OR=0.49,95%CI:0.28,0.88)have low risks.The five TNF-α inhibitors increased the risk of severe infection and malignancy equally.Conclusion:The safety of the five TNF-α inhibitor monotherapy is better than the corresponding TNF-α inhibitor combined with DMARDs.Etanercept combined with DMARDs is safer in combination therapy.The five TNF-α inhibitors increased the risk of severe infection and malignancy equally.In clinical application,when the efficacy is similar,we recommend etanercept monotherapy as the best choice for RA patients.Conversely,treatment with certolizumab pegol combined DMARDs is not recommended.