Objective Several rheumatoid arthritis (RA) susceptibility variants map close to genes involved in the tumor necrosis factor (TNF) signaling pathway, prompting the investigation of RA susceptibility variants in studies of predictors of response to TNF blockade. Based on a previously reported association of RA with the PTPRC genetic locus, the present study was undertaken to test established RA susceptibility variants, including PTPRC, in the prediction of response to TNF blockade in a large cohort of patients from the UK. Methods DNA was extracted from the blood of 1,115 UK patients with RA who were receiving anti-TNF biologic therapy. Samples were analyzed for 29 single-nucleotide polymorphisms (SNPs) previously established as RA susceptibility variants. In the primary analysis, the effect of each SNP on treatment response was assessed by linear regression, using an additive model, in which absolute change in the Disease Activity Score in 28 joints at 6 months of followup was the outcome measure. In a secondary analysis, logistic regression models were used to compare patients with a good treatment response (n = 274) to those with a poor response (n = 195), as defined using the European League Against Rheumatism response criteria. Results were combined with those from previous studies to confirm the findings by meta-analysis. Results The PTPRC rs10919563 SNP was associated with improved treatment response in both the primary analysis (regression coefficient 0.19, 95% confidence interval [95% CI] 0.09, 0.37; P = 0.04) and secondary analysis (odds ratio 0.62, 95% CI 0.40, 0.95; P = 0.03). A meta-analysis combining these data with the results from a previous study strengthened the evidence for association with the PTPRC SNP (P = 5.13 × 10−5). No convincing association of the treatment response with other candidate loci was detected. Conclusion Presence of the rs10919563 RA susceptibility variant at the PTPRC gene locus predicts improved response to anti-TNF biologic therapy. Fine-mapping studies are required to determine whether this SNP or another variant at the locus provides the greatest predictive accuracy for treatment response.