abstract
- Although tumor necrosis factor-alpha (TNF-alpha) blockade is a very effective therapy for rheumatoid arthritis (RA), not all patients achieve a favorable outcome. The objective of this study was to determine if the common TNF-alpha variant -308(A) predicts poor response to TNF-alpha inhibitors in RA patients using meta-analysis. Studies were identified using MEDLINE and EMBASE. Data were extracted based on DAS28 or achieving at least American College of Rheumatology 20 response. A total of nine studies met the inclusion criteria representing a total of 692 RA patients. There was no significant heterogeneity among study effect sizes (P=0.36). The frequency of the A allele was 22% (119/531) in responders and 37% (60/161) in non-responders. The odds of having the A allele was lower in responders versus non-responders (odds ratio (OR)=0.43, 95% confidence intervals (CI): 0.28-0.68, P=0.000245), irrespective of the TNF-alpha inhibitor prescribed, indicating that the -308(A) variant predicts poor response to TNF-alpha inhibitors. The clinical utility of prospectively genotyping for this variant when initiating anti-TNF-alpha therapy for RA should now be formally assessed.