Talk: Resolution of Inflammation Following Treatment of Ankylosing Spondylitis with Anti-TNF Agents Is Associated with New Bone Formation (2009 ACR/ARHP Annual Scientific Meeting)

Monday, October 19, 2009: 4:45 PM

108 B (Pennsylvania Convention Center)

Praveena Chiowchanwisawakit¹, Susanne J. Pedersen², Robert GW Lambert³, Mikkel Ostergaard² and Walter P. Maksymowych³, ¹Siriraj Hospital, Mahidol University, Bangkok, Thailand, ²Copenhagen University Hospitals at Hvidovre and Gentofte, Copenhagen, Denmark, ³University of Alberta, Edmonton, AB

Presentation Number: 1257

Purpose: A previous study of patients with ankylosing spondylitis (AS) receiving anti-TNF agents in clinical trials showed that new bone formation was more likely to occur at those vertebral corners (VC) where inflammation had resolved following institution of anti-TNF therapy. A “release of TNF brake” hypothesis has been proposed whereby decreased TNF downregulates Dickkopf-1 allowing new bone formation through signalling by Wnt and other bone growth factors. However, data supporting the association between development of a new syndesmophyte and resolution of a vertebral corner inflammatory lesion (CIL) remain scarce. We aimed to test the hypothesis that a CIL visible on MRI that resolves following treatment is more likely to develop into a de novo syndesmophyte visible on plain x-ray than a vertebral corner with no prior CIL visible on MRI.

Method: We studied 51 AS patients (28 on standard and 23 on Anti-TNF treatment) followed prospectively who had MRI at baseline and at follow up (mean of 17.9 months), and radiography of the cervical and lumbar spine at baseline and 2 years. A persistent CIL was defined as being present on both MRI scans whilst a resolved CIL was defined as present at baseline MRI and completely disappeared at follow up MRI. The anterior vertebral corners on anonymised MRI were assessed independently and blinded to patient, treatment, and radiography by 2 readers. We conducted the following comparisons:1. proportion of new syndesmophytes developing at each anterior VC from a resolved inflammatory lesion at the corresponding VC; 2. the proportion of new syndesmophytes developing at each anterior VC from a persistent inflammatory lesion at the corresponding VC; 3. proportion of new syndesmophytes developing at each anterior VC where there is no prior inflammatory lesion. Proportions were compared by the Pearson chi-square analysis of the reader concordant data and discordant data was not included in this analysis.

Results: For patients receiving anti-TNF therapy, new syndesmophytes developed significantly more frequently in those vertebral corners with inflammation on MRI at baseline (10.3%) as compared to those that were normal on MRI (1.4%) (p = 0.001). For patients receiving standard treatment, new syndesmophytes developed in 12.5% of vertebral corners with inflammation at baseline as compared to 2.9% of those without (p = 0.01). New syndesmophytes developed in an even higher percentage of CIL that resolved following anti-TNF therapy (27.3%).

New syndesmophyte

Treatment

Resolved CIL

Persistent CIL

Normal VC

P value

Yes

Anti-TNF

3 (27.3)

0

5 (1.7)

<0.0001

No

8 (72.7)

3

282 (98.3)

Yes

Standard

1 (16.7)

0

16 (3.3)

NS

No

5 (83.3)

10

463 (96.7)

Conclusion: This study of AS spines documents that MRI findings predict subsequent damage on X-ray. Further, the demonstration of an increased likelihood of developing new bone following resolution of inflammation offers strong support for the “TNF brake hypothesis”.

Keywords: magnetic resonance imaging (MRI)

Disclosure: P. Chiowchanwisawakit, None; S. J. Pedersen, None; R. G. Lambert, None; M. Ostergaard, None; W. P. Maksymowych, None.

1257 - Resolution of Inflammation Following Treatment of Ankylosing Spondylitis with Anti-TNF Agents Is Associated with New Bone Formation