Article Text

AB0694 Rheumatologists' attribution of patient-reported symptoms in axial spondyloarthritis (AXSPA): impact on response to TNF-inhibitors (TNFI) in 508 patients
  1. L Gossec,
  2. A Moltό,
  3. A Etcheto,
  4. N Boudersa,
  5. P Claudepierre,
  6. N Roux,
  7. F Berenbaum,
  8. A Martin,
  9. L Sparsa,
  10. P Coquerelle,
  11. M Soubrier,
  12. S Perrot,
  13. M Dougados
  1. Predict-SpA Study Group, Paris, France


Background In axSpA, treatment decisions are mainly based on patient-reported symptoms: recommendations are to initiate TNFi in patients with active disease and with physician conviction that treatment is needed (ref). However, the attribution of symptoms to inflammation is difficult to establish in axSpA.

Objectives The objective of the present analysis was to explore the link between physician-attributed causality for symptoms and treatment response to TNFi.

Methods The PredictSpA study ( NCT03039088) was a longitudinal observational multicenter study in France in 2015. Patients with physician-defined definite axSpA and starting a TNFi treatment were included, a TNFi was prescribed according to usual practice and efficacy was assessed at 12 weeks by BASDAI50 response. At baseline, symptoms levels including BASDAI and ASDAS were collected and the physician evaluated the causality of symptoms by answering the following 3 questions: how convinced are you that the symptoms of this patient are due to (A) inflammatory axSpA activity (B) to axSpA severity (eg syndesmophytes, kyphosis) and NOT to disease activity and (C) to other diseases and NOT axSpA. Each question was assessed 0–10 (not convinced at all to absolutely convinced). The link between a score ≥4/10 on each of the 3 questions and BASDAI50 response was assessed by univariate logistic regression. Patients interrupting the TNFi before 3 months were considered as non-responders and missing data were imputed using non-responder imputation.

Results In all, 519 patients were included and 508 had data over 3 months: mean age 41.3 (SD 11.6) years, mean disease duration 6.1 (SD 8.4) years, 237 (46.7%) were women, 424 (83.5%) satisfied the ASAS criteria for axSpA of whom 379 (74.6%) were in the imaging arm and 45 (8.9%) in the clinical arm. Symptom levels were high: mean BASDAI was 5.7 (SD 1.8) and mean ASDAS-CRP was 3.3 (SD 0.9) with only 6 (1.2%) patients in inactive disease state according to ASDAS. The physician-attributed causality of symptoms was mostly related to inflammatory activity: mean scores for (A), (B) and (C) were respectively, 7.4 (SD 2.0), 2.3 (SD 2.5) and 2.1 (SD 2.2). When physicians attributed causality to non-axSpA (score (C) ≥4/10), BASDAI50 response was less frequent: 45/118 (38.4%) vs 213/390 (54.6%), odds ratio 0.5 [95% CI 0.3, 0.8].

Conclusions In axSpA patients starting a TNFi with high symptom levels, physician-attributed causality of symptoms was mainly related to inflammatory activity of the axSpA. When physician-attributed causality was more oriented towards non-axSpA causes, BASDAI50 response after 3 months of a TNFi was lower. This confirms the validity of the ASAS-EULAR recommendations for starting a TNFi which rest on a level of symptoms but associated to physician conviction of the indication.


  1. van der Heijde D, et al. 2016 update of the ASAS-EULAR management recommendations for axial spondyloarthritis. Ann Rheum Dis. 2017 [Epub ahead of print].


Acknowledgements This study was conducted thanks to an unrestricted grant from MSD.

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.