Background/Purpose: Efficacy of TNF alpha inhibitors (TNFi) has been clearly established in patients with axial SpondyloArthritis (ax-SpA), but its effectiveness and adherence with regard to specific patient characteristics remains to be evaluated in patients with early disease. The aim of the study was to evaluate the effectiveness of TNFi in a cohort of patients with inflammatory back pain (IBP) suggestive of ax-SpA. Secondary objectives were to estimate the retention rate of TNFi after 2 years of follow-up and the baseline predictors of such retention rate.

Methods:

Observational prospective multi-centre study, with 708 patients with early IBP (< 3 years duration) suggestive of ax-SpA. Data included: demographics and disease characteristics of the first two years of follow-up.  Statistical analysis: Effectiveness: patients receiving TNFi were matched with patients not receiving TNFi (but receiving usual care) at a 1:1 ratio according to a propensity-score. The primary endpoint of the study was ASAS40, assessed here at the last available visit still on treatment. Exploratory subgroup analyses evaluated the potential interactions existing between baseline characteristics and the treatment effect (p<0.10). Retention rate was estimated for all TNFi using a survival-data analysis by Kaplan-Meier. Cox analyses were performed to identify potential predisposing factors of such retention rate.

Results:

1. Treatment effect: 203 patients (28.7%) received at least one TNFi during follow-up; data for the assessment of the primary endpoint was available in 197 patients (97.0%) receiving TNFi, that were matched according to a propensity score to 197 patients receiving any other usual care. An ASAS40 response at the last available visit (74±30.9 weeks after initiation of TNFi) was achieved in 62 (31.8%) patients receiving TNFi treatment vs. 31 (16.0%) patients receiving usual care (OR = 2.45 [1.50 – 3.99], p=0,004).
2. Exploratory subgroup analyses detected that patients with X-ray sacroiliitis (OR=9.12 [3.22–25.80] and OR=1.32[0.73–2.41], for X-ray sacroiliitis yes/no, p=0.002), MRI sacroiliitis (OR=5.10 [2.28–11.41] and OR=1.32[0.68–2.56] for MRI sacroiliitis yes/no, p=0.011) and fulfilling the imaging arm (OR=5.34 [2.51–11.33] and OR=0.78 [0.32–1.89] for the imaging and clinical arms, respectively, p=0.001) were more likely to achieve an ASAS40 response with TNFi, compared to usual care. No significant interaction was detected between CRP abnormality and treatment effect.
3. Retention rate: Of the 203 patients included, Kaplan-Meier estimates of the proportion of patients still on TNFi over time were 75.2% [69.0–81.3] and 41.2% [33.2–49.1], at 6 and 24 months after initiation, respectively.  HLA B27 presence (HR=0.66[0.44-0.99], p=0.044) and X-ray sacroiliitis at baseline (HR=0.48 [0.30-0.78], p=0.003) were associated with a prolonged continuation of the TNFi over time.

Conclusion: In daily practice, use of TNFi was frequent in active early ax-SpA, whatever the subgroup of the ASAS criteria the patients were fulfilling. However patients with imaging abnormalities were more likely to achieve a TNFi response compared to usual care

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/patients-with-imaging-abnormalities-of-the-sacroiliac-joints-are-more-likely-to-respond-to-tnf-alpha-inhibitors-in-early-axial-spondyloarthritis-data-from-the-desir-cohort/