Background/Purpose :

TNF-Related Apoptosis Inducing Ligand (TRAIL) is a member of the TNF familly. Its soluble receptor Osteoprotegerin (OPG) also inhibits Receptor activator of nuclear factor kappa-B ligand (RANKL). We previously reported that, in a cohort of early arthritis (< 2 years), a high OPG/TRAIL ratio at baseline was associated with remission (DAS28<2.6) at 6 months, suggesting that OPG and TRAIL could be used to predict outcomes in early arthritis.

Methods :

The aim of this study was to confirm these results in the larger French cohort ESPOIR. To be included, patients should have an early arthritis (< 6 months). We studied patients clinical status at 12 months (M12) and radiographic progression (Sharp’s score) at 24 months (M24). TRAIL and OPG concentrations in serum were measured at baseline (M0). Values of TRAIL and OPG were log transformed to be normalized. We first compared TRAIL and OPG between patients with arthritis responding to ACR/EULAR 2010 criteria (RA) and undifferentiated arthritis (UA). We then compared logOPG, logTRAIL and logOPG/logTRAIL using general linear model adjusted for DAS28, rheumatoid factor and anti-citrullinated protein antibodies positivity, and total Sharp score at M0. Logistic regression was used to determine predictive value for remission (DAS28 ≤2.6) and radiographic progression (DSharp score>0).

Results :

TRAIL, OPG and TRAIL/OPG at M0 were not different between RA (n = 641) and UA patients (n = 53). Among RA patients, patients in remission at M12 had a significantly lower concentration of M0 logOPG than those with DAS28>2.6 (2.93±0.18 (n=204) vs 2.98±0.16 pg/ml (n=341), respectively, p=0.002 and p=0.026 after adjustment). M0 TRAIL and M0 OPG/TRAIL were not significantly different between patients in remission at M12 and others. Logistic regression adjusted for DAS28, rheumatoid factor and anti-citrullinated protein antibodies positivity, and total Sharp score at M0 showed that logOPG≥3.1 (=1259 pg/ml) was predictive of absence of M12 remission (B=0.55, CI: 0.33-0.90, p=0.018), with a sensibility of 21%, a specificity of 87%, a positive predictive value of 75% and a negative predictive value of 38%.

Patients with progression of Sharp score erosion at M24 had significant lower M0 logTRAIL (2.96±0.22 vs 3.00±0.19, p=0.009 and p=0.002 after adjustment). M0 OPG and M0 OPG/TRAIL were not significantly different between patients with or without radiographic progression at M24. Logistic regression adjusted for DAS28, rheumatoid factor and anti-citrullinated protein antibodies positivity, and total Sharp score at M0 showed that logTRAIL≤2.95 (891 pg/ml) was predictive of erosion progression (B=0.47, CI: 0.30-0.72, p=0.001) with a sensibility of 52%, a specificity of 63%, a VPP of 39% and a VPN of 74%.

Conclusion :

Concentrations of TRAIL and OPG could not help in distinguish UA and RA. Low M0 OPG is associated with remission at M12 and low TRAIL is associated with erosion progression at M24 after multiple adjustments for usual prognosis factors. This suggests a promoting effect of OPG on disease activity and a protective role of TRAIL on bone erosion. However, no relevant threshold could be determined to use OPG and TRAIL in daily practice to identify early RA patients at risk of poor outcomes.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/could-osteoprotegerin-and-tnf-related-apoptosis-inducing-ligand-assessments-help-us-to-manage-early-rheumatoid-arthritis-results-from-the-espoir-cohort/