Article Text

THU0162 Patterns of Clinical Response Predict Failure of TNF Inhibitor Treatment in Rheumatoid Arthritis: Data from a Longitudinal Observational Study in Romania
  1. I. Ancuta1,
  2. C. Codreanu2,
  3. R. Ionescu3,
  4. M. Parvu4,
  5. M. Bojinca1
  1. 1Rheumatology, “Dr. I. Cantacuzino” Hospital
  2. 2Rheumatology, “Dr. I. Stoia” Center for Rheumatic Diseases
  3. 3Rheumatology, “Sfanta Maria” Clinical Hospital
  4. 4Rheumatology, “N. Gh. Lupu” Clinical Hospital, Bucharest, Romania


Background In recent years, emphasis has shifted in rheumatoid arthritis (RA) to early intervention in the disease course. The introduction of tumour necrosis factor (TNF) inhibitors significantly changed the overall treatment goals and guidelines. Best practice use still needs to be determined in patients with oscillating or increased disease activity treated with TNF inhibitors over long periods of time.

Objectives To determine patterns of clinical response that can predict failure of anti-TNF treatment over time, indicating the need for change to second line therapy in patients suffering from severe RA with high disease activity.

Methods In this longitudinal, observational, population-based, cohort study we included a total of 400 patients treated with anti-TNF therapy. Data were retrieved from the National Health Insurance House (NHIH) database (2002-2011). Over time, 195(51.5%) patients received a single TNF inhibitor and 184(48.5%) had 1-2 anti-TNF therapies. The 28-joint disease activity score (DAS28) response rates and the difference between values found at two consecutive evaluations (ΔDAS28) were calculated and tested. Patients were classified as low (LDA), moderate (MDA) and high (HDA) disease activity. Statistical analysis was performed using STATA SE 11.0 software

Results The MDA group was divided into MDA1, MDA2 and MDA3 to determine the factors predicting treatment failure. In our clinical practice, we identified five patterns of treatment response: 1) Responder (R) DAS28<3.2 or ΔDAS28>1.2 at 6 months; 2) MDA1 (Stable MDA) - ΔDAS28<1.2 and 3.2<DAS28<3.5, with final evolution towards breakthrough response; 3) MDA2 (Breakthrough Response) - 3.5<DAS28<3.8 usually stationary around 3.8 or increase in DAS28>0.4 per year, predictor of therapeutic failure with oscillating clinical response, secondary loss of treatment efficacy and shift towards HDA; 4) MDA3 (Unstable MDA) - 3.8<DAS28<5.1 with response similar to HDA with most rapid loss of treatment efficacy and immediate evolution towards HDA; 5) Primary non-responder (NR) identical to HDA DAS28>5.1 at 6 months or ΔDAS28<0.6 ideally diagnosed at 3 months.

Conclusions MDA3 (Unstable MDA) response showed similar response to HDA in our statistical analysis and therefore second line therapy should be considered at first evaluation (6 weeks). Likewise, change to second line therapy is indicated after two consecutive MDA2 evaluations. MDA1 response requires monitoring and re-evaluation of treatment options if patients reach MDA2 breakthrough stage. In current clinical practice in Romania, in all MDA patients, we recommend strict patient evaluation and tight control in the first 1-3 months after anti-TNF treatment initiation, according to ACR/EULAR criteria and ACR 2012 treat to target guidelines.

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.