Article Text

Download PDFPDF
02.18 Can rankl serum levels predict future progression to rheumatoid arthritis in early arthritis clinic patients?
  1. Agata N Burska1,
  2. Jehan J El-Jawhari1,
  3. Ai L Tan1,
  4. Richard J Wakefield1,
  5. Helena Marzo-Ortega1,
  6. Philip G Conaghan1,
  7. Paul Emery1,
  8. Frederique Ponchel1,
  9. Jane E Freeston1,2
  1. 1Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, UK and NIHR Leeds Musculoskeletal Biomedical Research Centre, Chapel Allerton Hospital, Leeds, UK
  2. 2Leeds Teaching Hospitals NHS Trust, Chapel Allerton Hospital, Leeds, UK


Background/objectives The earliest diagnosis of rheumatoid arthritis (RA) is crucial to initiate treatment and prevent further disease progression and the accumulation of irreversible damages to the joints. Despite recent advances with the discovery and integration of anti–citrullinated protein antibody (ACPA) in diagnostic criteria, there is still an unmet need for new diagnostic biomarkers, notably for ACPA-negative disease. The receptor-activator-nuclear-factor-κB axis (RANK/RANKL) is known to regulate bone homeostasis. The aim of this pilot study is to establish whether serum RANKL levels in people with early inflammatory arthritis, are associated with RA diagnosis at follow-up and to evaluate the added value of RANKL for early RA diagnosis.

Materials/methods Serum from 298 subjects (95/204 Male/Female) was collected at the baseline participant visit to the Leeds early arthritis clinic. Demographic (age, symptom duration) and clinical data (joint counts swollen and tender (SJC, TJC), CRP, DAS28, rheumatoid factor (RF) and ACPA, shared epitope (SE)) were collected. A commercial ELISA (BioVENDOR) was used to measure RANKL. Ultrasound of 26 joints (bilateral elbows, wrists, MCP 2–3, PIP 2–3, knees, ankles and MTP 1–5) was performed at baseline recording summative scores for power Doppler (PD), grey scale hypertrophy (GS) and synovitis (SYN).

Results At follow-up, 151 patients had a confirmed diagnosis of RA (EULAR-2010 criteria) and 147 were classified as non-RA (undifferentiated arthritis, other inflammatory or non-inflammatory diagnosis). All routinely-used biomarkers were associated with RA diagnosis (ACPA, RF, SE, TJC, SJC, CRP, DAS28, p<0.0001), similarly to imaging biomarkers (PD, GS, SYN, p<0.001). RANKL levels were significantly higher in RA (RA 1002.4±1053.2 pmol/L, non-RA 339.2±451.5 pmol/L, p<0.0001). A regression analysis suggested that four parameters were sufficient to account for all associations with RA: RANKL, age, SJC, PD with 75.3% accurate prediction. An AUROC analysis suggested cut-off for each parameter and a score was calculated adding 1 point for each of factors (RANKL >700, age >62, TPD >3, SJC>4). This score predicted RA with an AUROC of 0.782 ((0.23–0.840), p<0.0001).

Conclusions A score including RANKL, age, SJC and PD showed good predictive value for non-RA when low and for RA when high. Furthermore, the regression redone in ACPA-negative only patients performed particularly well for (77.4% accurate).

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.