Article Text

AB0653 Retention Rate and Predictive Factors of Tnf-α Inhibitor Discontinuation in Patients with Ankylosing Spondylitis - Results from the Rheumatic Diseases Portuguese Register Reuma.Pt
  1. A.R. Sepriano1,2,
  2. F. Araujo2,
  3. R. Aguiar3,
  4. R. Vieira4,
  5. E. Sousa5,
  6. F. Pimentel-Santos2,6,
  7. G. Sequeira7,
  8. H. Canhão5,
  9. H. Santos8,
  10. J. Garcia9,
  11. J.A. Pereira Silva5,
  12. J. Canas Silva10,
  13. L. Miranda8,
  14. M. Oliveira11,
  15. M.J. Salvador12,
  16. M. Bernardes4,
  17. P. Monteiro13,
  18. T. Nόvoa14,
  19. J. Branco1,2
  1. 1CEDOC, Faculdade Ciências Médicas, Universidade Nova de Lisboa
  2. 2Rheumatology, Hospital Egas Moniz, CHLO, Lisbon
  3. 3Rheumatology, Centro Hospitalar Baixo Vouga, Aveiro
  4. 4Rheumatology, Centro Hospitalar São João, Porto
  5. 5Rheumatology, Hospital Santa Maria, CHLN
  6. 6Rheumatology, CEDOC, Faculdade de Ciência Médicas, Universidade Nova de Lisboa, Lisbon
  7. 7Rheumatology, Hospital Faro, Faro
  8. 8Rheumatology, Instituto Português Reumatologia, Lisbon
  9. 9Rheumatology, Centro Hospitalar Médio Tejo, Torres Novas
  10. 10Rheumatology, Hospital Garcia Orta, Almada
  11. 11Rheumatology, Centro Hospitalar da Cova da Beira, Covilhã
  12. 12Rheumatology, Centro Hospitalar Universitário Coimbra, Coimbra
  13. 13Rheumatology, Centro Hospitalar Tondela-Viseu, Viseu
  14. 14Rheumatology, Hospital do Divino Espírito Santo, Ponta Delgada, Portugal


Background Tumor necrosis factor-alpha inhibitors (TNFi) are currently the only therapeutic option when conventional treatment fails in ankylosing spondylitis (AS) patients.

Objectives To assess the retention rate and investigate predictive factors and reasons for drug discontinuation in patients with AS starting their first TNFi.

Methods We included all new biological starters AS patients fulfilling the 1984 modified New York classification criteria, registered at the Rheumatic Diseases Portuguese Register, from June 2008 until October 2011. Retention rate at 2 years was evaluated using survival-data analysis methods with discontinuation of the drug, regardless the reason, as the primary outcome. Potential predictive factors of drug discontinuation (demographic, clinical and laboratorial) were assessed using log-rank tests and a Cox proportional-hazards regression model.

Results Of the 334 AS patients starting a TNFi, 265 (79.34%) maintain treatment after 2 years of follow-up. Median drug survival among patients who discontinued treatment was 10.9 months (95% CI: 7.7 – 13.0 months) with a discontinuation rate of 11.6% per year. The main reason for treatment cessation was adverse events (34.4%) followed by lack of response at 12 weeks (23.2%). Drug survivals were similar regardless of the reason for discontinuation or TNFi used. Compared with patients who retain their first TNFi those who discontinue treatment, were more frequently women (p=0.04), were older at disease beginning (p=0.01) and at TNFi beginning (p=0.03), had higher BMI (p=0.01) and higher Bath Ankylosing Spondylitis Functional Index (BASFI, p=0.02) at baseline. In multivariable Cox regression, older age at TNFi beginning was the only baseline predictor of drug discontinuation (HR 1.06, 95% CI 1.02 – 1.10, p=0.006).

Conclusions Retention rate was high among AS patients starting their first TNFi. Adverse events were the main reason for drug discontinuation and older age at treatment onset was a predictor of shorter drug survival.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.3916

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.