Association between the level of circulating bioactive tumor necrosis factor alpha and the tumor necrosis factor alpha gene polymorphism at -308 in patients with rheumatoid arthritis treated with a tumor necrosis factor alpha inhibitor

Arthritis Rheum. 2008 May;58(5):1258-63. doi: 10.1002/art.23430.


Objective: The tumor necrosis factor alpha (TNFalpha) -308A polymorphism has been associated with high production of TNFalpha and poor response to anti-TNFalpha therapy, but these associations remain controversial. The aim of this study was to explore the association between circulating TNFalpha bioactivity, the TNFalpha -308 polymorphism, and the clinical response to infliximab in patients with rheumatoid arthritis (RA).

Methods: One hundred ninety-eight patients with RA were treated with infliximab and methotrexate. Responses at 6 months according to the American College of Rheumatology (ACR) preliminary criteria for improvement in RA were recorded. Genotyping for the TNFalpha -308 polymorphism was performed by enzyme-linked oligosorbent assay. Circulating TNFalpha bioactivity was evaluated in 50 patients with RA by assessing the production of interleukin-6 (IL-6) in synoviocytes induced by a small amount of TNFalpha plus plasma. IL-6 production in 48-hour supernatants and the levels of TNFalpha protein and IL-6 were measured by enzyme-linked immunosorbent assay.

Results: The TNFalpha -308 polymorphism was not associated with the ACR response to infliximab. The level of circulating TNFalpha bioactivity was higher in patients with the TNFalpha -308 A/A or A/G genotype than that in patients with the G/G genotype (median 50.0 ng/ml [interquartile range (IQR) 31.5-62.0] versus 33.0 ng/ml [IQR 16.5-47.5]; P < 0.02). However, no difference was observed for the TNFalpha protein level according to genotype (median 0.62 pg/ml [IQR 0.00-8.85] for G/G versus 3.35 pg/ml [IQR 1.55-4.63] for A/A or A/G; P not significant). The level of circulating TNFalpha bioactivity was higher in good responders (> or =50% improvement) than in poor responders (< or =20% improvement) (median 45.0 ng/ml [IQR 21.0-59.0] versus 28.0 ng/ml [IQR 14.0-39.0]; P = 0.05). However, the level of TNFalpha protein was similar in both groups.

Conclusion: The level of functional circulating TNFalpha is partially genetically determined and is predictive of the clinical response to infliximab. Nonresponders to anti-TNFalpha therapy are likely to have a disease that is not primarily driven by TNFalpha.

MeSH terms

  • Antibodies, Monoclonal / therapeutic use*
  • Arthritis, Rheumatoid / blood*
  • Arthritis, Rheumatoid / drug therapy
  • Arthritis, Rheumatoid / genetics*
  • Female
  • Humans
  • Infliximab
  • Interleukin-6 / biosynthesis
  • Male
  • Middle Aged
  • Polymorphism, Genetic*
  • Synovial Membrane / cytology
  • Synovial Membrane / drug effects
  • Synovial Membrane / metabolism
  • Tumor Necrosis Factor-alpha / antagonists & inhibitors
  • Tumor Necrosis Factor-alpha / blood*
  • Tumor Necrosis Factor-alpha / genetics*


  • Antibodies, Monoclonal
  • Interleukin-6
  • Tumor Necrosis Factor-alpha
  • Infliximab