Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

MEDICAL OPTIONS in Ankylosing Spondylitis

October 2008

Differentiating Inflammatory from Mechanical Back Pain: A Probability Analysis for Earlier Intervention in AS

Dr. Martin Rudwaleit, Charité University Hospital, Berlin, Germany

Assessing Disease Activity and Alternatives for Guiding Therapy

Dr. Xenofon Baraliakos and Dr. Juergen Braun, Ruhr-University Bochum, Herne, Germany

Beneficial Long-term Effects of TNF-aInhibition on Disease Activity

Dr. Nick Barkham, University of Leeds, Leeds, UK

Secondary Effects of TNF-aInhibition: Exploring Changes in Histopathological and Molecular Markers

Dr. Dominique L. Baeten, University of Amsterdam, Amsterdam, The Netherlands

Strategies to Modify Disease Progression and Long-term Complications

Dr. Filip Van den Bosch, Ghent University Hospital, Ghent, Belgium

DIFFERENTIATING INFLAMMATORY FROM MECHANICAL BACK PAIN

Martin Rudwaleit, MD

Consultant in Rheumatology, Charité University Hospital, Campus Benjamin Franklin, Berlin, Germany

Of the etiologies of lower back pain, ankylosing spondylitis (AS) is among the most likely to lead to significant disability. There are several potential advantages for early diagnosis, but no characteristic or combination of characteristics so far reliably differentiates AS from other conditions before the disease advances to radiological signs. In a series of publications over the last four years, a system of probabilities has been described that provides a high confidence index for the presence of AS. Since its introduction, the system has proven useful in accelerating the time to diagnosis and access to therapy. Although rapidly identifying a treatable cause of lower back pain has obvious clinical advantages, this methodology may have even greater clinical utility if tumour necrosis factor alpha (TNF-a) inhibitors and other highly targeted therapies fulfill their promise to modify disease progression.

Rationale for Earlier Diagnosis

Most individuals experience one or more episodes of prolonged lower back pain during their lifetime. In a health survey conducted among adults in the US, the 12-month prevalence of lower back pain lasting at least one month was 17.8%.¹ The low yield from radiographic studies means that the majority of these episodes remain idiopathic.² Spondyloarthritides have a prevalence variably reported as ranging from 1% to 2%, making them at least as common as rheumatoid arthritis.³ Although AS is the underlying etiology in a relatively small proportion of patients with lower back pain, it is a treatable disorder and one of the most important causes of back pain leading to permanent disability. The average delay to diagnosis of AS exceeds eight years.⁴

There is general agreement about the presence of AS in symptomatic patients who demonstrate sacroiliitis on radiographic examination,⁵ but strategies to identify probable disease much earlier in its course are needed. There are several potential advantages of early diagnosis, including more rapid introduction of therapies to modify symptoms. Patients with a probable diagnosis can be reassured that their back pain is not psychological in origin, and they can be protected from unnecessary diagnostic and therapeutic procedures, saving costs and the potential for harm. More recently, assertions that TNF-a inhibitors may be diseasemodifying, delaying or preventing irreversible joint damage,⁶ provide another potential motivation for early diagnosis.

Previous attempts to develop clinical criteria for diagnosis of AS in advance of radiographic signs of disease have been problematic due to lack of accuracy. For example, although the ESSG (European Spondyloarthropathy Study Group) criteria, published in 1991,⁷ have been widely cited, one study found that only 46% of patients identified as having AS with these criteria had confirmed AS after five years of follow-up.⁸ Other diagnostic methodologies using clinical criteria for early diagnosis of AS have also been criticized for lack of reliability.⁹

Diagnosis by Association of Clinical Signs: The Likelihood Ratios

While no clinical feature of AS is fully unique to this disease, the accumulation of associated clinical signs would reasonably be expected to increase the probability of having AS. Based on this concept, a methodology for early diagnosis of AS was developed that employed likelihood ratios (LRs) to permit clinicians to better evaluate the probability of the presence of AS. The original description of the methodology was published in 2004.¹⁰ The methodology was developed for the early diagnosis of axial spondyloarthritis, which includes both AS and undifferentiated spondyloarthritis. The methodology employs clinical findings, laboratory tests, and skeletal imaging but does not require positive findings of all three for a diagnosis.

Common clinical features of AS include inflammatory back pain that may include alternating buttock pain, enthesitis, arthritis, dactylitis, acute anterior uveitis, a positive family history of spondyloarthropathies (Spa), or uveitis, psoriasis, Crohn’s disease, and symptom relief with NSAIDs. Common laboratory findings include elevated levels of acute-phase reactants and positive HLA-B27. Although radiological findings in early disease are insufficiently specific to identify spondyloarthropathy or AS, the pattern of abnormalities can provide sufficient support within the context of other findings to make a diagnosis.

By evaluating these features in a large sample of AS patients and patients with other musculoskeletal disorders, probability scores were derived from the clinical, laboratory, and radiological signs. When entered into a decision tree designed to separate low- and high-probability patients, this methodology was effective for stratification. For the purposes of an early diagnosis, a probability of =90% was considered the threshold for a diagnosis. For example, while one to two clinical features of AS provided a probability of only 35% to 70%, two clinical features were sufficient to reach the =90% threshold when HLA-B27 positivity was also confirmed.

The LRs are not achieved simply by tallying the number of positive features. Rather, individual features are given different weight based on their predictive value in the initial analysis. The LRs combine the sensitivity and specificity of any given diagnostic characteristic into a single figure. Even though the values are approximate, the relative weight is important for demonstrating that one feature differs from others. For example, HLAB27 has a higher LR than any specific clinical feature, including arthritis or uveitis, and therefore, has greater weight for arriving at a probable diagnosis.

The algorithm and probability calculations were created to identify early AS in patients without definite radiographic sacroiliitis as defined in the modified New York criteria. One of the findings from this probability analysis is that no single symptom or finding is sufficient for making a diagnosis of AS because only 5% of chronic back pain sufferers have AS or undifferentiated axial SpA. In general, at least three parameters are needed to be confident in the diagnosis.

Proposed Criteria for Inflammatory Back Pain (IBP) in young to Middle-aged Adults with Chronic Back Pain (age <50 years old)

Individual parameters of IBP criteria:

• Morning stiffness lasting more than 30 minutes

• Improvement in back pain with exercise, but not with rest

• Awakening because of back pain during second half of the night

• Alternating buttock pain

Application as a classification criteria:

If at least two of the above four parameters are present then IBP is present with a sensitivity of 70.3% and a specificity of 81.2% (best tradeoff between sensitivity and specificity)

Adapted from Rudwaleit et al. Arthritis Rheum 2006;54:678-81.

In a subsequent modification of the algorithm, negative LRs were added to the methodology.¹¹ By adding the negative LRs, the proportion of patients with a probable diagnosis decreased because the proposed diagnostic threshold of 90% was more difficult to reach. On a clinical basis in which the goal is to select patients with a high probability of having AS, the methodology as described originally and in a subsequent publication is preferred.¹²

From Best Estimates to Clinical Application

The probability figures provide a best estimate of the presence of AS and also narrow the number of candidates for more expensive imaging studies. In patients suspected of AS, some form of imaging should be obtained at baseline. Radiographic images are the most widely used, but computed tomography (CT) may also be helpful, particularly for detecting sacroiliitis in patients with equivocal standard radiographs. However, the radiation exposure associated with CT is a concern, particularly in young women. The most useful method for evaluating the sacroiliac joints is magnetic resonance imaging (MRI), but it may not be an option at some centres due to availability, cost or both. As for alternative forms of imaging, scintigraphy has limited sensitivity and specificity, while positron emission tomography (PET) is not a well established imaging tool in SpA and both are therefore inferior to MRI. In addition to clinical signs, a trial of NSAIDs at full doses for two to three days should also be considered in patients to provide reassurance of the diagnosis when there is sufficient probability.

In a patient with HLA-B27 positivity, a good response to NSAIDs and an episode of enthesitis, it is reasonable to consider a diagnosis of AS regardless of the results of radiographs which may be normal during the first years of disease. However, it is important to emphasize the reliance on probabilities. In addition to alternative diagnoses, clinicians should consider co-existing diagnoses. For example, lumbar disc herniation may be the cause of back pain even in patients with AS. In addition, while a 90% probability was the threshold used in an analysis of this approach, a lower figure may be reasonably interpreted as strong evidence of AS when other variables are supportive. For example, an 80% LR for AS may be diagnostic in a young man with a three- to five-year history of inflammatory back pain and HLA-B27 positivity. Conversely, expensive imaging tests are not warranted in a patient over the age of 65 with a 20-year history of chronic back pain without radiographic sacroiliitis or syndesmophytes. The likelihood of AS in this patient is very low (<0.5%).

A definitive early diagnosis of AS facilitates therapy. While treatment with NSAIDs can be initiated without a definitive diagnosis, there is increasing resistance to this approach because of complications that include about a fourfold increase in the risk of upper gastrointestinal bleeding.¹³ TNF-a inhibitors are highly effective in AS patients who do not respond sufficiently to NSAIDs, but again, an acceptable benefit-to-risk ratio is dependent on first establishing the diagnosis of AS with a high probability. An important issue not yet resolved is whether TNF-a inhibitors can alter the course of AS by preventing progressive disease. Protection from progression may alter current treatment algorithms, particularly aggr
early in its course.

<img2209|center>

Summary

A high probability of AS can be derived from a combination of characteristic symptoms, laboratory studies, and imaging. By establishing a high probability of AS in patients with lower back pain, additional diagnostic tests can be avoided and therapy can be initiated. The most important goal of early initiation of appropriate therapy is to reduce symptoms to improve quality of life. If established, a relationship between early disease control and a delay in the complications of AS will provide an additional rationale for early therapy. The long average delay between the onset of symptoms and a definitive diagnosis emphasizes the need for probability analysis to guide early intervention for one of the most serious diseases that produce lower back pain.

References

1. Dillon C, Paulose-Ram R, Hirsch R, Gu Q. Skeletal muscle relaxant use in the United States: data from the Third National Health and Nutrition Examination Survey (NHANES III). Spine 2004;29:892-6.

2. Van den Bosch MA, Hollingworth W, Kinmonth AL, Dixon AK. Evidence against the use of lumbar spine radiography for low back pain. Clin Radiol 2004;59:69-76.

3. Braun et al. Prevalence of spondylarthropathies in HLA-B27-positive and -negative blood donors. Arthritis Rheum 1998;41:58-67.

4. Feldtkeller et al. Age of disease onset and diagnosis delay in HLA-B27 negative vs. positive patients with ankylosing spondylitis. Rheumatol Int 2003;23:61-6.

5. Van der Linden SM, Valkenburg HA, Cats A. Evaluation of the diagnostic criteria for ankylosing spondylitis. A proposal for the modification of the New York criteria. Arthritis Rheum 1984;27:361-8.

6. Baraliakos et al. Radiographic progression in patients with ankylosing spondylitis after four years of treatment with the anti-tumor necrosis factor alpha antibody infliximab. Rheumatology 2007;46:1450-3.

7. Dougados et al. The European Spondyloarthropathy Group preliminary criteria for the classification of spondyloarthropathy. Arthritis Rheum 1991; 34:1218-27.

8. Amor B. Usefulness of criteria for spondyloarthropathies. Joint Bone Spine 2000; 67(6):502-3.

9. Amor et al. Are classification criteria for spondyloarthropathy useful as diagnostic criteria? Rev Rhum Engl Ed 1995;62:10-15.

10. Rudwaleit et al. How to diagnose axial spondyloarthritis early. Ann Rheum Dis 2004; 64:535-43.

11. Rudwaleit M, Feldtkeller E, Sieper J. Easy assessment of axial spondyloarthritis (early ankylosing spondylitis) at the bedside. Ann Rheum Dis 2006;65:1251-2.

12. Rudwaleit M, Khan MA, Sieper J. The challenge of diagnosis and classification in early ankylosing spondylitis: do we need new criteria? Arthritis Rheum 2005;52:1000-8.

13. Langman et al. Risks of bleeding peptic ulcer associated with individual non-steroidal antiinflammatory drugs. Lancet 1994;343:1075-8.

ASSESSING DISEASE ACTIVITY AND ALTERNATIVES FOR GUIDING THERAPY

Editorial Overview:

Xenofon Baraliakos, MD, and Juergen Braun, MD

Rheumazentrum Ruhrgebiet, Ruhr-University Bochum, Herne, Germany

Ankylosing spondylitis (AS) is the prototype of the spondyloarthritides (SpA). Radiographic evidence of structural damage in the sacroiliac (SI) joints is the current gold standard for making the diagnosis of AS on the basis of the modified New York criteria.¹ However, the disease process usually starts long before radiographs show clear signs of structural damage. Since current guidelines usually require radiographic evidence of structural changes in the SI joints for a definitive diagnosis of AS, problems arise when patients in earlier stages of SpA are in need for potent antiinflammatory therapy. Sacroiliitis, a hallmark of AS, can be visualized by magnetic resonance imaging (MRI) mucht earlier in the course of the disease.^2,3 In patients with an established diagnosis, disease activity should be followed closely, by assessing clinical symptoms (inflammatory back pain, morning stiffness and others); laboratory parameters for disease activity,
ge.php?id=2210" />values; and imaging procedures for disease activity, such as MRI.

<img2210|center>

<img2211|center>

Patients showing definite signs of AS earlier before structural stages have occurred are usually diagnosed as undifferentiated SpA (uSpA) or axial uSpA. Recently, MRI has also been found valuable for identifying patients at high risk of developing progressive structural changes, indicating for the first time a possible relationship between spinal inflammation at baseline and future radiographic progression in AS.⁴

Since NSAIDs work usually very well for the initial control of symptoms of inflammatory lower back pain in those patients, physicians may not proceed with further diagnostic procedures. However, a definitive diagnosis is always necessary to allow for more potent therapies according to current treatment guidelines.⁵

Importance of Detecting Active Inflammation Early and the Impact of Imaging

The average delay from the first disease-related symptoms to a diagnosis of AS is currently between five and seven years, which is much longer than in other inflammatory rheumatic diseases.⁶ This may lead to persistent pain and functional impairments already presenting in young patients. Therefore, objective tools to evaluate disease activity in those patients are very helpful in early and latter stages of disease.⁷ Imaging of disease activity by MRI in early and established disease stages has a sensitivity for detecting active inflammation in the sacroiliac joints and the spine of approximately 80%.⁸ Other imaging modalities potentially considered for detecting disease activity, such as scintigraphy, are much less sensitive, with values between 50% and 70%.

From Symptom Control to the Importance of Disease Activity

Biologic therapies have contributed substantially to the treatment of AS patients.^9-11 Besides significant short-term reduction in disease activity, treatment with anti-TNF agents has shown sustained clinical efficacy in long-term studies with good tolerability and no major safety issues.^12-15

As stated in current guidelines, such as those issued from a collaboration of the Assessment in AS (ASAS) Working Group and the European League Against Rheumatism (EULAR),⁵ the treatment of AS is guided by clinical symptoms. While symptoms are an appropriate immediate measure of the impact of the disease on quality of life, they may not be representative of pathologic progression at the level of the involved structures in patients with AS. This can be also shown by analyses of MRI data from patients treated with biologics: long-term treatment leads to significant decrease of inflammatory activity, parallel to significant decrease of clinical assessments of disease activity. However, even after two years, there are still some inflammatory lesions in the spine of those patients (Figure 1).^16-18

Thus, persistent inflammation may permit disease to progress in patients who have reasonably well controlled symptoms. There is increasing evidence that the rate of progression of AS differs markedly between patients.¹⁹ Although a mean rate of radiographic progression can be calculated by using historical cohorts,¹⁹ there are still many patients showing markedly higher progression than the mean, while some show no progression at all. Recent studies have shown that this is related to the individual propensity of each patient to develop new syndesmophytes over the years. According to recent knowledge, the rate of future radiographic progression can only be predicted according to the number of syndesmophytes already present at the time of diagnosis.²⁰

On the other hand, although there is no definite evidence that biologic agents stop structural damage in patients with advanced AS,^21-23 these agents have been associated with substantial reductions in spinal disease activity, as assessed by MRI.^16-18 More importantly, they improve function and spinal mobility over time even if minor structural damage has occurred within two years.²⁴

Summary

Early detection of AS and SpA is critically important for the management of these patients, including an early initiation of effective therapy. The control of persistent spinal inflammation is a major aim of therapeutic strategies in SpA. Although radiographic evidence of AS is still the gold standard for making a definitive diagnosis of AS, MRI reveals inflammation and persistent disease
than radiographs. Clinical symptoms and objective indicators of inflammation, such as CRP and MRI, are important guides in the management of AS patients.

<img2212|center>

References

1. van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361-8.

2. Baraliakos et al. Assessment of acute spinal inflammation in patients with ankylosing spondylitis by magnetic resonance imaging (MRI): a comparison between contrast enhanced T1 and short-tau inversion recovery (STIR) sequences. Ann Rheum Dis 2005;64(8):1141-4.

3. Baraliakos et al. Inflammation in ankylosing spondylitis: a systematic description of the extent and frequency of acute spinal changes using magnetic resonance imaging. Ann Rheum Dis 2005;64(5):730-4.

4. Baraliakos et al. The relationship between inflammation and new bone formation in patients with ankylosing spondylitis. Arthritis Res Ther 2008;10(5):R104.

5. Zochling et al. ASAS/EULAR recommendations for the management of ankylosing spondylitis. Ann Rheum Dis 2006;65(4):442-52.

6. Feldtkeller E, Bruckel J, Khan MA. Scientific contributions of ankylosing spondylitis patient advocacy groups. Curr Opin Rheumatol 2000;12(4):239-47.

7. Rudwaleit et al. How to diagnose axial spondyloarthritis early. Ann Rheum Dis 2004;63(5):535-43.

8. Battafarano et al. Comparison of bone scan, computed tomography, and magnetic resonance imaging in the diagnosis of active sacroiliitis. Semin Arthritis Rheum 1993;23(3):161-76. 9. Braun et al. Treatment of active ankylosing spondylitis with infliximab: a randomised controlled multicentre trial. Lancet 2002;359(9313):1187-93.

10. Brandt et al. Six-month results of a double-blind, placebo-controlled trial of etanercept treatment in patients with active ankylosing spondylitis. Arthritis Rheum 2003;48(6):1667-75.

11. van der Heijde et al. Efficacy and safety of adalimumab in patients with ankylosing spondylitis: Results of a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2006;54(7):2136-46.

12. Braun et al. Persistent clinical efficacy and safety of anti-tumour necrosis factor alpha therapy with infliximab in patients with ankylosing spondylitis over 5 years: evidence for different types of response. Ann Rheum Dis 2008;67(3)340-5.

13. Baraliakos et al. Outcome of patients with active ankylosing spondylitis after two years of therapy with etanercept: Clinical and magnetic resonance imaging data. Arthritis Rheum 2005;53(6)856-63.

14. Baraliakos et al. Persistent clinical efficacy and safety of infliximab in patients with ankylosing spondylitis over 7 years - evidence for different response to anti-TNFa therapy. Ann Rheum Dis 2008;67(suppl 2):375.

15. Baraliakos et al. Persistent clinical response in patients with ankylosing spondylitis after 5 years of treatment with etanercept. Ann Rheum Dis 2008; 67(suppl 2):376.

16. Baraliakos et al. Magnetic resonance imaging examinations of the spine in patients with ankylosing spondylitis before and after therapy with the tumor necrosis factor alpha receptor fusion protein etanercept. Arthritis Rheum 2005; 52(4):1216-23.

17. Braun et al. Major reduction in spinal inflammation in patients with ankylosing spondylitis after treatment with infliximab: results of a multicenter, randomized, double-blind, placebo-controlled magnetic resonance imaging study. Arthritis Rheum 2006;54(5)1646-52.

19. Baraliakos et al. Radiographic progression in ankylosing spondylitis (AS) – the natural course. A retrospective cohort study. American College of Rheumatology Scientific Meeting. Oral presentation, November 2007.

20. Baraliakos et al. Progression of radiographic damage in patients with ankylosing spondylitis: defining the central role of syndesmophytes. Ann Rheum Dis 2007;66(7):910-5.

21. Baraliakos et al. Radiographic progression in patients with ankylosing spondylitis after 4 yrs of treatment with the anti-TNF-alpha antibody infliximab. Rheumatology (Oxford) 2007;46(9)1450-3.

18. Sieper et al. Persistent reduction of spinal inflammation as assessed by magnetic resonance imaging in patients with ankylosing spondylitis after 2 yrs of treatment with the anti-tumour necrosis factor agent infliximab. Rheumatology (Oxford) 2005;44(12):1525-30.

22. Baraliakos et al. Less radiographic progression in patients with active ankylosing spondylitis after 2 years of Anti-TNF therapy. American College of Rheumatology Scientific Meeting. Abstract, November 2005.

23. van der Heijde et al. Radiographic progression of ankylosing spondylitis after up to two years of treatment with etanercept. Arthritis Rheum 2008;58(5):1324-31.

24. Braun et al. Infliximab improves spinal mobility in patients with ankylosing spondylitis: two-year findings from the ASSERT clinical trial. American College of Rheumatology Scientific Meeting. Abstract, November 2007.

BENEFICIAL LONG-TERM EFFECTS OF TNF-a INHIBITION ON DISEASE ACTIVITY

Editorial Overview:

Nick Barkham, MB, ChB, MRCP(UK)

Academic Section of Musculoskeletal Diseases, University of Leeds, Leeds, UK

Results of the ASSERT (Ankylosing Spondylitis Study for the Evaluation of Recombinant Infliximab Therapy) trial associated early use of a biologic therapy in ankylosing spondylitis (AS) with a reduction in disease progression. The diseasemodifying effect of the biologic, a tissue necrosis factor alpha (TNF-a), has not previously been observed with NSAIDs or other AS therapies. The ability of biologics to alter the pathophysiologic process of inflammatory diseases has implications for modifying the risk of the most serious complications, including irreversible disability. There are issues yet to be resolved about the cost efficiency and benefit-to-risk ratio of early use of TNF-a inhibitors in AS, but results of ASSERT suggest the need for a reorientation that includes earlier diagnosis and closer monitoring of inflammation, shifting the focus of therapy from symptom to disease control.

ASSERT Control of Disease Activity

Inhibitors of TNF-a, a cytokine that is an important trigger of the inflammatory cascade, have demonstrated efficacy against a broad range of inflammatory processes, including rheumatoid arthritis,¹ inflammatory bowel diseases (IBD)2 and psoriasis.³ In AS, clinical trials demonstrating large reductions in symptoms have been conducted with the TNF-a inhibitors infliximab,⁴ etanercept⁵ and adalimumab.⁶ The clinical responses in these studies have often been dramatic even among patients previously refractory to first-line therapies. The initial studies primarily measured response in regard to symptom relief, the most important immediate outcome in a slowly progressive disease, but a series of studies has begun to evaluate the effect of these agents on disease activity. The promise of tight control of the inflammatory cascade is to provide protection against progression of AS, shifting treatment goals from an improvement in quality of life to an improvement in long-term outcome.

Several studies now support the premise that TNF-a inhibitors alter the course of AS. Of these, ASSERT, a placebo-controlled, doubleblind randomized study that employed magnetic resonance imaging (MRI) to evaluate inflammatory activity in involved joints, is the largest. In the initial publication, the randomized groups were compared at the end of 24 weeks.⁷ Follow-up now extends to seven years.⁸ At the earlier timepoint, the TNF-a inhibitor was associated with a reduction in inflammation relative to placebo as assessed with MRI. In subsequent follow-up, the primary end point has been sustained clinical improvement in spinal mobility and other outcomes captured with Assessments in AS (ASAS) response criteria.

In ASSERT, 276 patients were randomly assigned in an 8:3 ratio to the TNF-a inhibitor infliximab (5 mg/kg) or placebo in a scheduled administration at weeks 0, 2 and 6, and then every six weeks thereafter. About 80% of the patients had at least one active spinal lesion at baseline. When the MRI activity was evaluated at 24 weeks by two readers blinded to treatment assignment, there was a highly significant advantage for the 194 patients randomized to infliximab when compared to the 72 patients assigned to placebo (P<0.001). Indeed, while inflammatory activity was persistent in the placebo group over this study period, there was almost complete resolution of spinal inflammation in the group randomized to infliximab regardless of the baseline level.

At the completion of the 24-week evaluation, placebo patients were permitted to cross over to infliximab, which was again administered in a dose of 5 mg/kg at weeks 0, 2 and 6 and then every six weeks thereafter. Those who were in the infliximab group were permitted to remain on this therapy, although patients who had not achieved a Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score of =3 in two consecutive evaluations were permitted an increase in the every-six-week dose to 7.5 mg/kg. Of those initially randomized to infliximab and who remained in the study, 106 patients (53%) received a dose increase. There were a total of 161 patients who were assessed at the end of 102 weeks with several clinical scoring methods.⁹

MRI at the end of 102 weeks demonstrated a persistent reduction but not complete eradication of spinal inflammation in all patients. Although disease activity parameters and MRI were not perfectly correlated, they were consistent. The authors concluded that both types of information might be useful for the definition of response to anti-TNF therapy. As for clinical measures, the improvements measured according to BASDAI and ASAS were similar at the end of two years among those maintained on infliximab to those observed at 24 weeks. Moreover, spinal mobility remained improved when compared to baseline scores. In the placebo patients crossed over to infliximab, the relative improvements in clinic
from baseline were similar to that observed in those who initiated therapy on infliximab. The clinical results were consistent with persistent disease control provided by suppression of inflammation.

<img2213|center>

At the seven-year follow-up, a BASDAI value <4 was seen in 78% of patients. ASAS 20 response was observed in 82% of patients and ASAS 40 response was achieved in 62%. In the 18% classified as non-responders at seven years, all but one patient achieved an ASAS 20 response at least once within the previous three years and might be better characterized as partial responders. The time from onset of symptoms to initiation of infliximab was important for likelihood of partial response. Only 23% of those treated late vs. 56% of those treated early achieved this level of clinical response. Based on this outcome, the likelihood that early treatment is reflected in protection from long-term structural damage deserves to be further explored.

Reduction of Inflammatory Lesions: An MRI Confirmation

A smaller MRI-based randomized study of infliximab in AS patients generated similar results.¹⁰ In that study, 40 previously untreated HLA B27-positive patients with inflammatory back pain consistent with AS underwent MRI scanning of the spine and sacroiliac joints. Patients were randomized equally to infusions of placebo or infliximab 5 mg/kg at 0, 2, 6 and 12 weeks. Patients in both groups were permitted to take concomitant NSAIDS but adjustments in the doses were not permitted. The mean age of the participants was 28.8 years and the mean duration of symptoms was approximately 15 months. The primary end point was change in MRI scans of the spine and sacroiliac joints taken at 16 weeks when compared to baseline. Paired scoring was conducted for lesions identified at baseline and reassessed at 16 weeks. All MRI scans were assessed by two readers blind to the treatment assignment.

The average total MRI score for sacroiliac joints was 5.1 at baseline. At 16 weeks, there was no change in patients randomized to placebo but an average score reduction of 2.0 (range 0 to 6.5) in those randomized to infliximab (P=0.033). Of sacroiliac lesions identified on MRI at baseline, 62.7% had resolved at 16 weeks in the group receiving infliximab vs. 29.4% in the placebo group (P=0.001). New lesions were observed at 16 weeks in only one patient receiving infliximab vs. 11 in the placebo group (P=0.004). Nine patients had spinal lesions at baseline. Of the five in the infliximab group, three had resolved by 16 weeks vs. one of the four in the placebo group. At week 16, 10 out of 18 (55.6%) infliximab patients vs. two of 16 (12.5%) placebo patients (P=0.009) achieved a partial remission as defined by ASAS criteria.

BASDAI and ASAS Supporting Data

The relative differences in the control of inflammation were reflected in a variety of clinical scoring methods employed to gauge clinical effects over the same time period. For example, BASDAI scores were reduced 3.41 in the infliximab group vs. 0.75 in the placebo group (P=0.002). The proportion of patients who achieved ASAS 20 was 77.8% in the infliximab group vs. 31.3% (P=0.006) in the placebo group. The proportions reaching ASAS 70 were 55.6% and 12.5% (P=0.009), respectively.

The data from this study, as well as from ASSERT, demonstrated that infliximab relieves back pain by providing a rapid reduction of inflammatory disease activity as measured with MRI. In a substudy of ASSERT, an association was made between the reduction in inflammation and suppression of inflammatory cytokines, including C-reactive protein (CRP) and interleukin-6 (IL-6) as well as in vascular endothelial growth factor (VEGF).¹¹ No previous therapy has demo
of the inflammatory lesions on MRI in very early AS. While the current data suggest that infliximab inhibits the underlying pathophysiologic process of AS, further studies are needed to determine whether early treatment will also halt ankylosis.

<img2214|center>

Summary

Therapies targeted at TNF-a have the potential to change the natural history of AS and other progressive inflammatory diseases. The previous orientation in AS has been on controlling symptoms to improve quality of life. The biologics introduce the potential to control the pathophysiologic process to alter outcome. For example, new data from the ASSERT trial associated early biologic therapy with increased likelihood of an ASAS 70 response at seven years. Although the dynamics of the benefit-to-risk ratio of biologics and their cost suggest that these treatments may not be warranted in all AS patients, those destined for severe joint involvement may benefit from treatment before the disease reaches advanced stages of progression. While these issues are being resolved, the data from the current studies suggest that new strategies are needed to accelerate diagnosis of AS and to monitor disease activity. While acute symptom control is essential for achieving an acceptable quality of life, treating the disease process has the potential to reduce the risk of long-term disability.

References

1. Lipsky et al. Infliximab and methotrexate in the treatment of rheumatoid arthritis. Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. N Engl J Med 2000;343:1640–1.

2. D’haens et al. Endoscopic and histological healing with infliximab anti-tumor necrosis factor antibodies in Crohn’s disease: A European multicenter trial. Gastroenterology 1999; 116:1029–34.

3. Mease et al. Etanercept in the treatment of psoriatic arthritis and psoriasis: a randomised trial. Lancet 2000;356:385–90.

4. Brandt et al. Successful treatment of active ankylosing spondylitis with the anti-tumor necrosis factor monoclonal antibody infliximab. Arthritis Rheum 2000;43:1346–52.

5. Marzo-Ortega H, McGonagle D, O’Connor P, Emery P. Efficacy of etanercept in the treatment of the entheseal pathology in resistant spondylarthropathy. Arthritis Rheum 2001; 44:2112–7.

6. Van der Heijde et al. Adalimumab effectively reduces the signs and symptoms of active ankylosing spondylitis in patients with total spinal ankylosis. Ann Rheum Dis 2007;Epub ahead of print.

7. Braun et al. Treatment of active ankylosing spondylitis with infliximab: a randomised controlled multicentre trial. Lancet 2002; 359:1187–93.

8. Baraliakos et al. Persistent clinical efficacy and safety of infliximab in patients with ankylosing spondylitis over 7 years – evidence for different response to anti-TNF-a therapy. 2008 Annual Meeting of the European League Against Rheumatism (EULAR). Abstract FRI0290.

9. Sieper et al. Persistent reduction of spinal inflammation as assessed by magnetic resonance imaging in patients with ankylosing spondylitis after two years of treatment with the anti-tumor necrosis factor infliximab. Rheumatology 2005; 44:1525-30.

10. Barkham et al. A Randomized Controlled Trial of Infliximab shows clinical and MRI efficacy in patients with HLA B27 positive very early ankylosing spondylitis. 2008 Annual Meeting of the American College of Rheumatology, Abstract FRI0310.

11. Visvanathan et al. Inflammatory biomarkers, disease activity, and spinal disease measures in patients with ankylosing spondylitis after treatment with infliximab. Ann Rheum Dis 2008;67:511-7.

SECONDARY EFFECTS OF TNF-a INHIBITION: EXPLORING CHANGES IN HISTOPATHOLOGICAL AND MOLECULAR MARKERS

Editorial Overview:

Dominique L. Baeten, MD, PhD

Clinical Immunology and Rheumatology, Academic Medical Centre/University of Amsterdam, Amsterdam, The Netherlands

Tumour necrosis factor alpha (TNF-a) has proven to be an important therapeutic target across a broad range of inflammatory diseases. Although the primary benefit of TNF-a inhibitors appears to be suppression of a key mediator of the inflammatory cascade, there are a variety of downstream and ancillary effects that may vary in importance across disease states. In ankylosing spondylitis (AS) and other spondyloarthropathies, a series of studies with infliximab and other TNF-ainhibitors have demonstrated disease-modifying effects at the level of the synovium where improvements in vascularity and favourable structural changes appear to be the product of secondary effects of TNF-ainhibition. In AS, these actions may be more important than T-cell inhibition. Differences, if any, between infliximab, alternative TNF-ainhibitors, or inhibitors of other pro-inflammatory cytokines may provide additional insights into the mechanisms of disease responsive to antiinflammatory biologicals.

Involvement of Other Inflammatory Mediators

The efficacy of biologics targeted at TNF-a has confirmed that substantial clinical benefits can be derived from therapies directed at specific mediators of the inflammatory cascade. However, the relative mechanisms of benefit may not be the same across different inflammatory states for several reasons. One is that the relationship of inflammatory mediators and disease progression is unlikely to be identical in different tissues, such as the small bowel and the large joints. While leukocyte activation is common to the diseases in which TNF-a inhibitors show activity, involvement of other inflammatory mediators differ, and the pathological changes in affected tissues are not necessarily analogous. Although interruption of the inflammatory cascade is central to the activity of anti-TNF-a therapies, the isolation of additional mechanisms has the potential to yield insight into the molecular steps that promote and sustain the disease state.

The relative activities of infliximab in inflammatory bowel disease (IBD) and AS provide a basis for exploring the relative mechanism of benefit in diseases characterized by the activated inflammatory state. Although the effects of TNFa inhibition at the cellular level in the two disease states are likely to be similar, the relative importance of these effects for disease control may vary substantially. For example, T-lymphocytes have a clear pathogenic role in Crohn’s disease but not in AS. Therefore, the direct effects of infliximab on T-cells, which includes inhibition of T-cellmediated production of granulocytemacrophage colony-stimulating factor (GM-CSF) and potentially T-cell apoptosis¹ may be more relevant for the former, while the diminution of inflammatory activity and tissue remodelling, whether or not directly mediated by T-cells may be more important for the latter.

Downregulation of Endothelial Activation and Hypervascularity

In AS, substantial clinical improvements associated with infliximab, a monoclonal inhibitor of TNF-a,² and etanercept, a TNF-a receptor antagonist,³ verify a reduction in inflammatory activity, but subsequent studies have now demonstrated other activities that may be important to disease control. In a small study of the effect of infliximab on serial synovial biopsies from eight patients with a spondyloarthropathy, striking reductions in the thickness of the synovial lining were accompanied by downregulation of endothelial activation and hypervascularity.⁴

These changes, which correlated with a reduction in the inflammatory cell infiltrate, prompted a second and larger analysis, which included a small placebo control group.5 In this analysis, two populations were included. One was a cohort of 10 patients treated with infliximab 5 mg/kg at weeks 0, 2 and 6 and three patients treated with placebo. The other was a pooled cohort of 20 patients also treated with 5 mg/kg at weeks 0, 2 and 6. In the first group, infliximab significantly reduced synovial lining thickness (P=0.015), endothelial activation (P=0.034), and inflammatory cell infiltration with neutrophils (P=0.041), macrophages (P=0.034) and T-cells (P=0.026) relative to baseline and the placebo-treated patients. In the second group, the same reduction in inflammation was seen, but there was also a normalization of lining thickness and a significant reduction in blood vessels (P=0.039). These changes paralleled the clinical improvements in symptoms and were independent of spondyloarthropathy subtype.

Relative Activity Along Different Pathways

A trend toward normalization of histology after 12 weeks of infliximab in this study was supported by favourable changes in a variety of molecular markers, such as downregulation of VCAM-1 and CD146 expression. Trials of longer duration are needed to determine whether this normalization progresses. Such studies are particularly attractive because of speculation that remodelling of synovium results in structural changes in bone and cartilage. Relative effects of other TNFa inhibitors are also of interest. Although etanercept has also demonstrated favourable changes in remodelling at the synovial level,⁶ other biologics such as adalimumab have not yet been evaluated. When infliximab and etanercept have been compared, a variety of histologic differences, such as macrophage infiltration,⁷ have been observed, suggesting that TNF-a blockers do differ in relative activity along different pathways.

Surrogate markers for treatment benefit in spondyloarthropathies, such as sequential synovial tissue analysis employed in the pilot studies described above, have substantial potential utility in understanding the relative activity of targeted therapies. Serial synovial biopsies have already been validated as a surrogate marker
ge.php?id=2215" />s.8 In AS, the serial sampling of synovium in the pilot studies suggested comparable value as a surrogate marker and to yield new information about histopathological and molecular changes that define this disease.9 Additional studies to validate this or other surrogate markers are needed.

<img2215|center>

<img2216|center>

If the mechanism of benefit from TNF-a inhibition varies across disease states, the relative effects of different agents with activity in spondyloarthropathy may be important not only for understanding this condition but other inflammatory diseases, such as uveitis, that may occur independently or concomitantly. Association studies have already established that IBD and uveitis are more closely associated with peripheral spondyloarthritis than with pure axial AS, which may be meaningful for identifying shared mechanisms of clinical benefit from TNF-a inhibitors. Similarly, psoriasis is less commonly associated with gastrointestinal inflammatory disorders but more commonly associated with joint and axial disorders. The utility of TNF-a inhibitors such as infliximab across multiple inflammatory diseases suggests common pathophysiologic features, but differences in histopathology and molecular signalling may be more useful for unravelling details of how tissue-specific abnormalities in immunologic function can be reversed.

In progressive inflammatory disorders such as AS, one of the most pressing questions concerning the use of targeted biologicals is whether the natural history of the disorder can be altered. Symptomatic relief is an important goal, but the pilot histopathology studies suggesting normalization of the affected synovium raises the possibility that it may not only be possible to halt disease progression but also to reverse the structural changes induced by the disease. Currently, targeted therapies are generally withheld for late stages of inflammatory disease processes, including AS, but earlier treatment holds the potential for preventing advanced disease if it can be confirmed that structural changes can be avoided.

Summary

The development of TNF-a inhibitors has been instrumental in the effort to isolate the molecular characteristics of inflammatory diseases, including AS. The activity of these agents in a broad range of diseases at diverse tissue sites validates the instrumental role of TNF-a in the inflammatory response. However, signalling systems interrelating with TNF-a to drive specific inflammatory diseases are likely to differ. In AS, recent studies have suggested that the clinical benefits from infliximab appear to accrue from downregulation of vascularity and improvement in tissue remodelling. Distinct from IBDs, where T-cell apoptosis may play a far more important role in inducing disease quiescence, the specific histological and molecular effects of TNF-a blockade in AS promise a more detailed understanding of the inflammatory cascade in different tissues where TNF-a inhibitors have exhibited activity.

References

1. Angholt et al. Increased production of granulocytemacrophage colony stimulating factor (GM-CSF) in Crohn’s disease, a possible target for infliximab treatment. Eur J Gastroenterol Hepatol 2004; 16:649-55.

2. Van den Bosch et al. Randomized double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (infliximab) versus placebo in active spondyloarthropathy. Arthritis Rheum 2002;46:755-65.

3. Gorman JD, Sack KE, Davis JC. Treatment of ankylosing spondylitis by inhibition of tumour necrosis factor alpha. N Engl J Med 2002; 346:1349-56.

4. Baeten et al. Immunomodulatory effects of antitumour necrosis factor alpha therapy on synovium in spondyloarthropathy: histological findings in eight patients from an open-label pilot study. Arthritis Rheum 2001;44:186-95.

5. Kruithof et al. Histological evidence that infliximab treatment leads to downregulation of inflammation and tissue remodelling of the synovial membrane in spondyloarthropathy. Ann Rheum Dis 2005; 64:529-36.

6. Kruithof et al. Immunomodulatory effects of etanercept on peripheral joint synovitis in the spondyloarthropathies. Arthritis Rheum 2005; 52:3898-909.

7. De Rycke et al. Differential expression and response to anti-TNFalpha treatment of infiltrating versus resident tissue macrophage subsets in autoimmune arthritis. J Pathol 2005;206:17-27.

8. Taylor et al. Reduction of chemokine levels and leucocyte traffic to joints by tumour necrosis factor alpha blockade in patients with rheumatoid arthritis. Arthritis Rheum 2000;43:38-47.

9. Kruithof et al. Identification of synovial biomarkers of response to experimental treatment in earlyphase clinical trials in spondylarthritis. Arthritis Rheum 2006;54:1795-804.

STRATEGIES TO MODIFY DISEASE PROGRESSION AND LONG-TERM COMPLICATIONS

Editorial Overview:

Filip Van den Bosch, MD

Department of Rheumatology, Ghent University Hospital, Ghent, Belgium

The hypothesis that tumour necrosis factor alpha (TNF-a) inhibitors may modify the course of ankylosing spondylitis (AS) is based partly on their remarkable clinical efficacy and partly on favourable changes in cellular and molecular processes that are considered to be surrogates of disease progression. Increasing lengths of followup in AS patients maintained on a TNF-a inhibitor will provide increasing opportunities to evaluate whether these agents are truly disease-modifying as defined by radiography and other objective imaging modalities. While the acute clinical benefits of TNF-ainhibitors in AS have been well demonstrated in controlled trials, it may be necessary to reconsider treatment algorithms if these agents demonstrate protection against disease progression in long-term studies. Currently, most guidelines recommend TNF-a inhibitors in AS patients unresponsive to NSAIDs. If the long-term risk of disability can be attenuated, the focus of treatment will expand from adequate symptom control to preventing long-term structural damage and associated functional impairments.

Sustaining Disease-modifying Activity

TNF-a inhibitors are associated with high rates of efficacy in the treatment of AS. In placebocontrolled studies first conducted with infliximab, clinical improvements were accompanied by highly significant reductions in laboratory measures of disease activity.^1,2 Subsequent studies with the TNF-a inhibitors etanercept and adalimumab have demonstrated similar degrees of activity against AS.^3,4 The efficacy of these agents in AS is attributed to the importance of TNF-a as a mediator of the inflammatory response. The hypothesis that TNF-a inhibitors may halt or even reverse the disease process is supported by a series of clinical studies in AS as well as in diseases that appear to share pathological mechanisms, such as rheumatoid arthritis (RA). In both RA and AS, speculation that infliximab may prevent disease progression is suggested by favourable changes in serial synovial biopsies, including normalization of synovial lining, reduction in inflammatory cell infiltration and downregulation of vascularization.^5,6

These short-term changes in surrogate markers are encouraging, but AS is a chronic, lifelong condition characterized by progressive structural damage, rendering protection from clinical progression over long-term follow-up as the definitive test of a disease-modifying therapy. From the patient perspective, this protection will be best demonstrated through stabilization or improvement in reproducible clinical measures such as the Bath Ankylosing Spondylitis Functional Index (BASFI). From the view of pathology, favourable changes in structural appearance documented with imaging and histology will provide the most compelling evidence of disease-modifying benefits.

On a clinical basis, improvements in BASFI have been documented in every study yet conducted with TNF-a inhibitors, including several studies that have followed patients for up to five years. In a published trial with a fiveyear follow-up, 38 patients in an initial cohort of 69 who participated in a double-blind randomized trial have continued to receive infliximab at a dose of 5 mg/kg every six weeks. Of these, 78% had a Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) of <4, and 84% had an Assessment in Ankylosing Spondylitis (ASAS) 20 response at five years (64% with an ASAS 40 response).⁷ In abstract form, similarly favourable longterm control has been reported by other groups.

However, continuous therapy with a TNF-a inhibitor appears to be important for sustained benefit. In a three-year analysis of benefit from the same group that published their five-year experience, treatment discontinuation after three years led to relapse in 41 of 42 patients in a mean interval of 17.5 weeks.⁸ Although responses were comparable to those achieved after initial therapy when therapy was restarted, the relatively rapid return of disease activity
nance courses at regular intervals are required to sustain disease control.

Radiographic studies have provided evidence of disease modification with TNFa inhibitors in certain diseases of the spondyloarthritis concept. In psoriatic arthritis, inhibition of radiographic progression has been observed within six months of starting infliximab (Figure 1).
^{;img2217|center>}

In AS, preliminary studies showed less structural damage at two years in patients taking infliximab than a second group not taking infliximab over the same period of time.¹⁰ In a second AS study with follow-up out to four years, radiographic progression on infliximab was observed, but it was less than that anticipated from historical controls (Figure 2).¹¹

<img2218|center>

However, neither of the phase III studies with infliximab or etanercept associated TNF-a inhibitor therapy with a significant reduction in progression in structural damage as assessed with the modified Stoke Ankylosing Spondylitis Spine Score (mSASSS).^12,13 While more data are needed, placebo-controlled studies may be impractical over a period of time sufficient to demonstrate protection from radiographic progression. The best opportunity to confirm disease modification may therefore be accumulation of a substantial pool of evidence that correlates improved outcome with an attenuation of structural damage relative to historical controls. Additional studies, such as serial MRI, may be a useful adjunct to radiographic studies to confirm disease-modifying activity.

Histological studies in AS are problematic because of the difficulty of obtaining tissue samples from the axial skeleton, sacroiliac joints, and enthesis.¹⁴ However, serial synovial biopsy studies associating infliximab with a reduction in hypervascularity and synovial lining hyperplasia have provided some initial support for a protective effect in the involved tissue.⁵ Infliximab has also been associated with downregulation of synovial matrix metalloproteinases, which are implicated in neovascularisation, matrix degradation, and cartilage and bone destruction.¹⁵ As an isolated end point, histological evidence of disease control may be a less compelling demonstration of protection against disease progression than radiographic control, but it does provide reassurance of favourable activity at the cellular and molecular level.

From Symptom Control to Altering Long-term Complications: A Critical Consideration

Growing clinical experience has generated confidence that TNF-a inhibitors can be administered effectively and safely in longterm maintenance regimens, but it is important to recognize that the equation for calculating the safety, efficacy, and cost of these agents relative to alternatives is dependent on treatment goals. Strategies with the potential to modify disease, altering the risk of longterm complications, such as disability, generate different benefit-to-risk and cost-to-benefit ratios than strategies targeted at symptom control. For the same reason, re
inst extra-articular disease is also relevant to the calculation of clinical utility. In ASPECT (Ankylosing Spondylitis Patients Epidemiological Cross-sectional Trial), 42% of 847 patients had at least one extraarticular manifestation (Figure 3).¹⁶ This included anterior uveitis in 27%, psoriasis in 11% and inflammatory bowel disease (IBD) in 10%. The risk of extra-articular manifestations appears to increase with disease duration.

Figure 3.

<img2219|center>

Prevention of new-onset extra-articular complications through early initiation of a disease-modifying agent may be a more attractive management strategy than intensifying treatment after these develop.

The evidence to date that biologicals may reduce disease progression has shifted an orientation that was once narrowly focused on the acute control of symptoms. AS, like many other of the inflammatory conditions responsive to TNF-a inhibitors, is a chronic condition for which the effect of therapeutic strategies should be considered over the long term. Evidence that one strategy provides protection against progressive disease relative to alternatives will be a critical consideration for optimal treatment. As five-year outcome data accumulate for infliximab, etanercept, adalimumab and other TNF-a inhibitors, there will be an opportunity to assess diseasemodifying effects and their ability to alter long-term outcome.

Summary

There are clinical, radiographic and histological data supportive of the hypothesis that TNF-a inhibitors can modify the course of AS. Definitive data are anticipated from the growing pool of patients who have been treated for five or more years, a length of time that will provide more convincing evidence of a change in the course of a disease that has relatively slow progression in objective terms. Although TNF-a inhibitors have been largely reserved for AS patients refractory to alternative therapies, evidence that these agents can prevent the irreversible consequences of progression may lead to broader use or earlier introduction. The ability to modify disease progression will have particular importance if diagnostic tools can be developed for early identification of patients with rapidly progressing disease or who otherwise have a poor prognosis. A change in the natural history of AS with TNF-a inhibitors will be an important validation of the value of targeting a key molecular pathway of the inflammatory cascade.

References

1. Van den Bosch et al. Randomized double-blind comparison of chimeric monoclonal antibody to tumor necrosis alpha (infliximab) versus placebo in active spondyloarthropathy. Arthritis Rheum 2002;46:755-65.

2. Braun et al. Treatment of active ankylosing spondylitis with infliximab: a randomized, controlled multicenter study. Lancet 2002;359:1187-93. 3. Gorman JD, Sack KE, Davis JC. Treatment of ankylosing spondylitis by inhibition of tumour necrosis factor alpha. N Engl J Med 2002; 346:1349-56.

4. Van der Heijde et al. Adalimumab effectively reduces the signs and symptoms of active ankylosing spondylitis in patients with total spinal ankylosis. Ann Rheum Dis 2008;67:1218-21.

5. Ulfgren et al. Systematic anti-tumor necrosis factor alpha therapy in rheumatoid arthritis downregulates synovial tumor necrosis factor alpha synthesis. Arthritis Rheum 2000;43:2391-6.

6. Kruithof et al. Histological evidence that infliximab treatment leads to downregulation of inflammation and tissue remodelling of the synovial membrane in spondyloarthropathy. Ann Rheum Dis 2005;64:529-36.

7. Braun et al. Persistent clinical efficacy and safety of anti-tumour necrosis factor therapy with infliximab in patients with ankylosing spondylitis over 5 years: evidence for different types of response. Ann Rheum Dis 2008;67:340-5.

8. Baraliakos et al. Clinical response to discontinuation of anti-TNF therapy in patients with ankylosing spondylitis after 3 years of continuous treatment with infliximab. Arthritis Res Ther 2005;7:R439-R444.

9. Van der Heijde et al. Infliximab inhibits progression of radiographic damage in patients with active psoriatic arthritis through one year of treatment: results from the induction and maintenance psoriatic arthritis trial 2. Arthritis Rheum 2007;56:2698-707.

10. Baraliakos et al. Radiographic progression in patients with ankylosing spondylitis after 2 years of treatment with tumor necrosis factor alpha antibody infliximab. Ann Rheum Dis 2005;64:1462-6.

11. Baraliakos et al. Radiographic progression in patients with ankylosing spondylitis after four years of treatment with the anti-tumor necrosis factor alpha antibody infliximab. Rheumatology 2007;46:1450-3.

12. Braun et al. Treatment of active ankylosing spondylitis with infliximab; a randomized controlled multicentre study. Lancet 2002;359:1187-93.

13. van der Heijde et al. Two-year etanercept therapy does not inhibit radiographic progression in patients with ankylosing spondylitis. Ann Rheum Dis 2006;65(suppl. II):81.

14. Sieper J, Appel H, Braun J, Rudwaleit M. Critical appraisal of assessment of structural damage in ankylosing spondylitis: implications for treatment outcomes. Arthritis Rheum 2008;58:649-56.

15. Vandooren et al. Involvement of matrix metalloproteinases and their inhibitors in peripheral synovitis and down regulation by tumor necrosis factor-a blockade in spondyloarthropathy. Arthritis Rheum 2004;50:2942-53.

16. Vander Cruyssen et al. The epidemiology of ankylosing spondylitis and the commencement of anti-TNF therapy in daily rheumatology practice. Ann Rheum Dis 2007;66:1072-7.