Anti-tumor necrosis factor (TNF) therapy in rheumatoid arthritis: correlation of TNF-alpha serum level with clinical response and benefit from changing dose or frequency of infliximab infusions

Clin Exp Rheumatol. 2005 Jul-Aug;23(4):469-74.


Objective: To determine the relationship between serum TNF-alpha level and clinical response in rheumatoid arthritis patients treated by infliximab. This could be of value to predict clinical response to infliximab and to determine the optimal dose and interval between dosing of infliximab. RA patients who did not respond adequately to conventional doses (3 mg/kg) of infliximab were studied to see if increasing the dose or frequency of infliximab infusions would be more effective.

Methods: Fifty-five RA patients who fulfilled the American College of Rheumatology criteria and were receiving treatment by anti-TNF-alpha (infliximab 3 mg/kg body weight every 8 weeks) were evaluated by: clinical disease activity using the Richie score index before receiving their scheduled infliximab infusion. Serum level of TNF-alpha, as measured by competitive ELISA assay, was determined immediately before and 9-11 days after receiving infliximab. RA patients who did not respond adequately to treatment with infliximab were given either a larger dose of infliximab or given the infusion at six-week intervals. Their clinical response was then evaluated sixteen months later. RESULTS Patients were divided into 2 groups according to Richie score, active group with score > 10 (score 20.3 +/- 7.7 mean +/- standard deviation, n = 25) and inactive group with scores < or = 10 (score 4.1 +/- 3.2, n = 30). TNF-alpha serum levels pre-infliximab infusion were significantly higher in the active group 76.1 pg/ml than the inactive group 38.0 pg/ml (P < 0.02). Whereas TNF serum level significantly dropped post infliximab in the inactive group (P < 0.05), it did not drop in the active group. The mean level of the post-infusion TNF-alpha serum level was higher (76.6 +/- 93.4 ng/ml) in the-active than the mean level of the post-infusion serum TNF-alpha levels in the inactive group (26.4 ng/ml +/- 7.9) P < 0.01 using the t-test. Increasing the frequency was superior in RA patients' clinical outcome than increasing the dose of infliximab infusions.

Conclusion: RA patients who responded well to infliximab and had inactive disease at the time of the study have lower levels of serum TNF-alpha which could be further suppressed by the recommended doses of infliximab. RA patients with active disease have higher serum levels of TNF-alpha which could not be suppressed after the recommended doses of infliximab infusion. Changing the frequency of infliximab infusions in the active group was more effective than increasing the dose of infliximab in inducing improved clinical outcome. We suggest that the lack of suppression of TNF-alpha in the active group could be due to inadequate dosing of infliximab or to the presence of a neutralizing antibody directed against infliximab. It remains to be seen if serum TNF-alpha levels could be used as a guide in determining the dose and intervals between dosing of anti-TNF therapy in RA in order to achieve the desired clinical response.

Publication types

  • Clinical Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Antibodies, Monoclonal / administration & dosage*
  • Antirheumatic Agents / administration & dosage*
  • Arthritis, Rheumatoid / blood
  • Arthritis, Rheumatoid / drug therapy*
  • Arthritis, Rheumatoid / physiopathology
  • Dose-Response Relationship, Drug
  • Drug Administration Schedule
  • Enzyme-Linked Immunosorbent Assay
  • Female
  • Humans
  • Infliximab
  • Male
  • Severity of Illness Index
  • Treatment Outcome
  • Tumor Necrosis Factor-alpha / metabolism*


  • Antibodies, Monoclonal
  • Antirheumatic Agents
  • Tumor Necrosis Factor-alpha
  • Infliximab