A Bayesian mixed treatment comparison of efficacy of biologics and small molecules in early rheumatoid arthritis

摘要

The current paradigm in the management of rheumatoid arthritis (RA) is to treat patients in the early stage of the disease (ERA). Previous meta-analysis-based mixed treatment comparisons (MTCs), aimed to identify the most effective drugs in ERA, are biased by the wide window of early definition, ranging from 6months to 2years. The aim of this study was to estimate through a Bayesian Network Meta-Analysis which biologics or small molecules are more likely to achieve a 1-year good clinical response in ERA patients with disease duration <1year. According to the Preferred Reporting Items for Systematic Reviews and Meta-analyses statement, randomized controlled trials (RCTs) of biologic agents and small molecules in combination with MTX to treat patients affected with ERA lasting <1year were searched through MEDLINE, EMBASE, Cochrane Library, and Clinicaltrials.gov between 1990 and September 2017. The outcome of interest was the achievement of American College of Rheumatology (ACR) 50 and ACR 70 response at 1year. WinBUGS 1.4 software (MRC Biostatistics Unit, Cambridge, UK) was used to perform the analyses, using a fixed effect model. Fourteen studies were included in the analysis. Tofacitinib (64.83%) followed by Etanercept (23.26%) were the drugs with the highest probability of achieving ACR50 response. Rituximab showed the highest probability of inducing ACR70 response (52.81%) followed by Etanercept (26.85%). This is the first MTC involving only RCTs on ERA patients with disease duration <1year. Tofacitinib and rituximab were the drugs ranked first in inducing 1-year ACR50 and ACR70 response, respectively.