OP0090: Impact of Etanercept on Radiographic Progression
OP0038: Impact of Adaimumab on Spinal Joint Inflammation
OP0090 TWO-YEAR ETANERCEPT THERAPY DOES NOT INHIBIT RADIOGRAPHIC PROGRESSION IN PATIENTS WITH ANKYLOSING SPONDYLITIS
D.M. van der Heijde1, R.D.M. Landewe, P. Ory, D. Vosse 1 , L. Zhou, W. Tsuji, J.C. Davis
The efficacy of etanercept in reducing the clinical signs and symptoms of ankylosing spondylitis (AS) has been demonstrated. However, it has not been previously shown whether etanercept is effective in modifying structural damage in AS.
Methods: Cases were subjects enrolled in a 24 week randomized, double-blind study of etanercept vs. placebo in patients with active AS despite therapy with NSAIDs and without complete spinal fusion. After the double-blind portion of the study concluded, all study subjects were invited to continue in an open-label extension phase. Plain radiographs of the lateral cervical and lumbar spine were obtained at baseline and at 2 years.
Controls derived from a natural history cohort of AS patients (OASIS) who had not received treatment with TNF inhibitors. From this cohort, baseline and 2 year radiographs of the lateral cervical and lumbar spine were also assessed. All radiographs were interpreted by 2 independent assessors without knowledge of treatment allocation or sequence using the modified Stoke AS Spinal Score (mSASSS).
Results: 257 cases who received at least one dose of etanercept were compared to 175 control subjects from the OASIS cohort. Subjects were primarily male (69-76%) with a mean age of approximately 42 years. In general, disease characteristics, as measured by the Patient Global Assessment of Disease Activity, Bath Ankylosing Spondylitis Functional Index (BASFI), and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) were higher in the etanercept treated subjects compared to those from the OASIS cohort. NSAID use was comparable between the two groups.
There was no statistically significant difference in the mean change in mSASSS over two years in the etanercept treated subjects compared to those in the OASIS cohort (0.91 + 2.45 vs. 0.95 + 3.18, respectively; p = 0.996). When only OASIS subjects with comparable baseline disease characteristics to those enrolled in the etanercept trial were included (n=76), a numeric but not statistically significant difference in mSASSS score was noted in the etanercept treated subjects compared to those in the OASIS cohort (0.91 + 2.45 vs. 1.27 + 3.64, respectively; p = 0.561). These results were similar when probability plots of radiographic progression in the two groups were compared.
Conclusions: Despite clinical efficacy, radiographic progression of AS at two years was not reduced in etanercept treated patients compared to those not receiving TNF inhibitor treatment.
Editorial Comment: These are compelling results that raise as many questions as are answered. Importantly, some design and analysis issues may serve to qualify the assurance of the authors regarding the lack of radiographic damage benefit with etanercept. Most notably, the results are post-hoc comparisons of two disparate cohorts, not intended a priori for comparison in this way. Though some attempt to equalize the characteristics of the two cohorts was made in a sub-analysis, this is not as methodologically sound as randomization. In addition, the decreased statistical power may contribute to the lack of significance in this sub-analysis. Other confounding issues are that many of the subjects in the etanercept treated group were not treated with a TNF inhibitor for the entire 2 year period, the almost ubiquitous presence of NSAIDs in both groups (known to have inhibitory effects on radiographic progression in this disease), and the possibility that either plain radiography or the mSASSS scoring method is not sensitive enough to detect treatment differences over two years (a relatively short time interval in this slowly progressive disorder).
OP0038 Adalimumab Reduces Spinal and Sacroiliac Joint Inflammation in patients with Ankylosing Spondylitis (AS) 52 Week Magnetic Resonance Imaging (MRI) Results from the Canadian AS Study
Maksymowych, Lambert, Salonen, et al
In this Phase III study, Maksymowych et al evaluated the efficacy of adalimumab (40 mg every other week) vs placebo in patients with ankylosing spondylitis (AS) who had failed at least one DMARD.
Methods: The main endpoints were spinal and sacroiliac (SI) joint inflammation as measured by the Spondyloarthritis Research Consortium (SPARCC) MRI score. MRIs were performed on a 11.5 Tesla and read by two blinded readers at baseline and at weeks 12 and 52. Patients in the placebo group were crossed over to adalimumab at week 24 so all patients at the week 52 time point were on adalimumab.
Results: Baseline mean spinal SPARCC scores for the placebo group (n=44) and the adalimumab group (n=38) were 19.9 and 16, respectively; SI scores were 7.5 and 5.7, respectively. At week 12, spinal and SI scores in the adalimumab group are reduced by a mean 54% (mean of 6.7) and 53% (mean of 2.1) respectively, p=<0.0001, compared to a mean reduction of 9% (mean of 18.6) and 13% (mean of 6.4) respectively in the placebo group. Improvements in CRP levels was significantly associated with changes in spinal SPARCC scores in the adalimumab group when compared to the placebo group (p=0.018). By week 52, the placebo crossover group had reductions in scores similar to the adalimumab group.
Conclusion: Adalimumab is efficacious in AS patients at significantly reducing spinal and SI joint inflammation as measured by SPARCC scoring before and after 12 weeks of therapy. Improvement was associated with CRP reduction and was sustained through 52 weeks.
Editorial Comment: Although there are now a number of clinical trials in patients with AS showing that TNF inhibitors improve symptoms and reduce CRP levels, there is a paucity of spinal imaging information. This study, though small in number, is a well designed randomized placebo controlled trial that clearly shows improvement in spinal and SI inflammation in response to TNF inhibition, but not with placebo treatment. Presumably, reduction in inflammation in these areas will reduce or prevent spinal fusion in the long-term. However, this will be much harder to prove because the ossification process occurs slowly over many years, and it is unlikely that many of these patients will remain on the same drug for 10 years (or that the investigators will continue to follow them for 10 years !!).
((top of page)) (next page)