Update in spondylarthropathy

Ankylosing spondylitis (AS) is the most common type of spondylarthropathy seen in clinical practice. Several reports have appeared in the past 2 years that have significantly advanced our understanding of the pathogenesis and treatment of this disease. Two recent studies finally provide direct experimental data in support of the hypothesis that a key trigger for disease is the intracellular proteolysis of bacterial and self-proteins producing peptides that exhibit molecular mimicry, thereby leading to activation of cross-reactive T cells once these peptides are presented on the cell surface. Two additional studies have shown that once disease is established, tumor necrosis factor alpha (TNFα) is a pivotal cytokine from the perspective of therapeutic intervention. One study provides evidence in support of the use of bisphosphonates for the treatment of AS. The association between HLA–B27 and ankylosing spondylitis is among the strongest of any HLA antigen to human disease, although the pathogenetic mechanism(s) remains unknown. The importance of bacteria has also been highlighted by the presence of intestinal mucosal inflammation in up to 60% of patients, and the observation that intestinal bacteria are essential to the development of arthritis in B27 transgenic rats (1, 2). One currently favored hypothesis proposes that disease follows induction of cytotoxic T cell autoreactivity in the course of T cell-mediated defense to certain bacteria following presentation by B27 of an arthritogenic self-peptide(s), perhaps derived from joint/entheseal cartilage, in a process that could involve cross-reactivity with bacterial-derived peptides (3). However, until recently, evidence in support of this hypothesis has been circumstantial and largely based on epidemiologic evidence of differing B27 subtype associations with disease. A major advance in our understanding of the pathogenesis of AS has been the recognition that inflammation within subchondral bone marrow is an early and consistent feature of the pathology of AS at both axial and peripheral sites (4). In addition, osteoporosis is an important feature of AS with an increased cumulative risk for vertebral fracture that is related to disease activity (5, 6). This may be evident within the first few years of disease. For the past several decades, the management of this disorder has been largely restricted to the use of nonsteroidal antiinflammatory drugs (NSAIDs) and physical modalities, with little attention being paid to the increased bone resorption evident from the outset of disease. Limited controlled data also supported the use of sulfasalazine in AS patients with peripheral synovitis (7, 8). Nevertheless, clinical improvement with these agents is modest. Bisphosphonates have enjoyed widespread use in the treatment of a variety of disorders of bone metabolism, but it is less widely known that they may also suppress macrophage function, particularly at the high concentrations achieved at sites of high bone turnover typically associated with subchondral marrow inflammation as may be seen in AS (9). TNFα has proven to be a pivotal cytokine from the perspective of therapeutic intervention in rheumatoid arthritis and the finding of TNFα in the serum, synovium, and sacroiliac joints of AS patients (10) raised hopes that targeting TNFα might also be beneficial in the treatment of AS. In this study, the remarkable sensitivity of MALDI-TOF mass spectroscopy and nanoelectrospray MSMS, which allows the detection and sequencing of peptides in the femto molar (10-15M) range, has been used to analyze the spectrum of peptides bound to HLA–B27 on the surface of cultured lymphoblastoid cells. One such peptide was shown to be a 12-amino acid peptide derived from B27 itself corresponding to amino-acid residues 309-320. This particular peptide was shown to bind to disease-associated B27 subtypes, B*2702 and B*2704, but not at all (or only weakly) to non-associated subtypes, B*2706 and B*2709, expressed on the surface of cultured lymphoblastoid cells. It was further shown that this peptide demonstrated homology to proteins from several arthritogenic bacteria. A peptide synthesized from one such protein, the DNA primase from Chlamydia trachomatis, was shown to bind disease-associated subtypes B*2705 and B*2702, but not the non-associated B*2706 subtype in vitro. Furthermore, this chlamydial peptide was generated following proteolysis by the 20S proteasome, the major proteinase generating antigenic peptides in cells. This work, therefore, reveals that a self-peptide derived from B27 itself demonstrates molecular mimicry with a peptide synthesized from an arthritogenic organism and is a natural ligand of disease-associated B27 subtypes. It therefore provides direct evidence in support of the arthritogenic peptide hypothesis although there are several missing pieces. It is not known if this chlamydial peptide is actually generated in vivo following antigenic processing within chlamydia-infected cells, if this leads to activation of cross-reactive T cells, and whether this is germane to the induction of autoimmunity specifically directed towards joint/entheseal molecular targets. The 309-320 amino acid sequence of B27 is conserved among HLA-B molecules and although it differs from HLA-A and C molecules this work does not explain why only a small proportion of HLA–B27-positive individuals (2–5%) develop disease. In previous work, the authors demonstrated that the induction of T cell cross-reactivity between a peptide derived from B27 and a peptide from retinal-S antigen resulted in uveitis in Lewis rats. This could be prevented by oral tolerance induction with the B27-derived peptide, and led to the first successful clinical trial of this oral B27-derived peptide in patients with autoimmune uveitis. In this article, the authors describe a model of arthritis and uveitis in Lewis rats following immunization with a B27-derived peptide (aminoacid residues 60-72 that distinguish B27 from other B molecules), which demonstrates homology with a peptide derived from cytokeratin (aminoacid residues 333-353). Antigenic mimicry was demonstrated by cross-reactivity of an αβ T cell line derived from rats immunized with these peptides, although the specific involvement of CD8 versus CD4 T cells was not cited. Furthermore, oral tolerance induction with this B27-derived peptide significantly reduced the incidence and severity of arthritis and uveitis induced by immunization with the cytokeratin-derived peptide. Cytokeratins are found in the skin, eye, synovial vascular endothelial cells, and gut epithelium, which could explain the localization of disease to these sites. This work may therefore be complementary to the earlier study by Ramos et al (11) in providing a mechanism whereby the induction of autoimmunity to peptides derived from B27 might lead to autoimmunity specifically directed towards joint/entheseal molecular targets. In other words, disease is a consequence of a chain of cross-reactive events starting from bacterial proteins through B27 to anatomically restricted self-proteins. This work also raises the intriguing possibility that future therapeutic strategies for spondylarthritis might employ vaccination with peptide(s) derived from B27 demonstrating molecular mimicry with arthritogenic bacteria and other self-proteins. The primary limitation of this study is that there is no evidence that these peptides are actually generated following antigenic processing in vivo. This study examined the efficacy and safety of a monthly regimen of pamidronate given intravenously to 84 AS patients refractory to NSAIDs. Patients were randomized to groups receiving either 60 mg or 10 mg doses of pamidronate. The 10-mg dose of pamidronate was chosen as the comparator group because arthralgia and myalgia occur commonly as a reaction to the first infusion of pamidronate, creating difficulties in ensuring effective patient blinding. The primary outcome was the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score and the primary end point was at 6 months. Treatment was well tolerated with few adverse events beyond the typical arthralgias and myalgias experienced following the first intravenous infusion of pamidronate, and only 6 patients withdrew for adverse events. A delayed response was evident with significant efficacy being observed at 6 but not at 3 months. Approximately 60% of patients in the 60-mg group experienced a ≥25% reduction in the BASDAI as compared with 30% of those who received 10 mg. Forty percent of patients in the 60-mg group experienced a ≥50% reduction in the BASDAI versus only 9% of those in the 10 mg group. Significant improvement was also noted in function, as measured by the Bath Ankylosing Spondylitis Functional Index (BASFI), and patient global index in the 60-mg group. Assessment of individual parameters comprising the BASDAI indicated that there was a significant reduction in axial pain and degree/duration of morning stiffness although no significant reduction in peripheral pain. There was no effect on swollen joint count although numbers of patients with peripheral synovitis were small. Finally, there was no significant change in acute phase reactants. These results are encouraging and warrant further investigation with larger numbers of patients and multi-center design. The apparent absence of an effect on peripheral synovitis is not too surprising as the half life of intravenously administered drug is only 1 hour and peak dose attained with conventional dosing is only 10-5M. Significant antiinflammatory effects are more likely to be observed within bone marrow since selective localization of drug at the sites of high bone turnover typically associated with subchondral inflammation in AS will lead to sufficiently high concentrations capable of inducing alterations in macrophage function and/or cytotoxicity (10-3M). For routine practice, the use of oral bisphosphonates would be preferable and these studies should be encouraged. This study was a multicenter, placebo-controlled trial in which 70 AS patients refractory to NSAIDs were randomized to receive either 3 infusions of infliximab (5 mg/kg at 0, 2, and 6 weeks), or placebo. Dosing was based on the regimen used for Crohn's disease and followed on the heels of 2 earlier open label trials in spondylarthropathy using this regimen that showed early and dramatic clinical and biochemical responses to therapy (15, 16). The primary outcome was the Bath AS Disease Activity 50% improvement criterion. This was attained at 12 weeks by 53% of patients receiving infliximab versus 9% of those on placebo. Clinical response was rapid, with almost 80% of patients achieving a 20% or greater reduction in the BASDAI by 2 weeks. This was accompanied by significant improvement in function, metrology, and levels of acute phase reactants that were all significant by 2 weeks. A ≥50% reduction in NSAID intake was possible in 56% of infliximab versus 19% of placebo treated patients, and NSAIDs were completely stopped in 41% of the infliximab treated patients versus 13% of placebo treated patients. A significant improvement was also noted in the physical function subscale of the Short Form-36 (SF-36) quality of life instrument. Infliximab was effective independent of the grade of spinal destruction and ankylosis evident on plain radiographs. In subgroup analyses, elevated C-reactive protein levels at baseline were predictive of response to treatment. Treatment was generally well tolerated although one patient developed tuberculosis in a lymph node and spleen, and another patient developed bronchocentric allergic granulomatosis of the lung. A third patient developed transient leukopenia. A report of the extension of this trial to one year, where patients initially treated with placebo received the same induction regimen of infliximab starting at week 12 while those patients on infliximab continued to week 52, has since been published (17). Similarly beneficial responses were noted in placebo patients and these were maintained throughout the duration of the study. Several conclusions directly impacting rheumatology practice can be drawn from this as well as several open label studies that have now been reported evaluating infliximab. There is little doubt that this treatment is highly efficacious for both axial and peripheral manifestations of disease, even in patients with longstanding disease and substantial ankylosis evident radiologically. Optimal disease control requires continuous administration every 6 to 8 weeks. Concomitant administration of methotrexate to reduce the risk of antibody development, as proposed in rheumatoid arthritis (RA), is not necessary as the administration of infliximab alone in patients with spondylarthropathy has not been associated with an increased incidence of infusion reactions. The development of antinuclear antibodies in the SpA trials has varied from 0% to 57%, although no clinical features indicative of lupus have been documented. There are 3 case reports of atypical tuberculosis developing in patients with spondylarthropathy treated with infliximab, reinforcing the need for pretreatment screening for exposure to tuberculosis, as recommended for RA. Perhaps the key challenge that needs to be addressed in view of the substantial costs associated with long-term therapy is the identification of prognostic factors related to structural damage so that patients can be more appropriately selected for this therapy. Prognostic data in AS is largely retrospective in origin. The optimal schedule of administration for long-term use also requires further study. A dose of 5 mg/kg every 6 weeks may be excessive and maintenance using the rheumatoid arthritis dosing regimen, i.e., 3 mg/kg at 0, 2, 6, and every 8 weeks, may be sufficient based on the results of one observational study and more cost-beneficial (18). Finally, there is as yet no data that infliximab prevents structural progression of disease. This was a small study that evaluated the efficacy of etanercept, a dimeric fusion protein of the human 75 kd (p75) tumor necrosis factor alpha receptor linked to the Fc portion of human IgG1, in 40 AS patients recruited from rheumatology practices in northern California. Patients were allowed to continue taking stable second line therapy (gold, methotrexate, sulfasalazine), corticosteroids, and NSAIDs. The primary outcome was a composite treatment response defined as 20% or greater improvement in at least 3 of duration of morning stiffness, degree of nocturnal spinal pain, the BASFI, the Patient's Global Assessment of Disease Activity, and the score for joint swelling with no worsening in any of the measures. The primary endpoint was at 4 months following which there was a 6-month open label extension in which all patients received etanercept. At baseline, patients in the etanercept group had a lower mean SF-36 score for physical functioning and were more likely to be receiving corticosteroids than those in the placebo group. Despite this, 80% of patients receiving etanercept achieved a response as compared with 30% in the placebo group. This was accompanied by significant improvement in chest expansion, enthesitis score, physical functioning scale of the SF-36 and acute phase reactants. In the open label phase, a similar response to etanercept was noted in those patients previously treated with placebo and the therapeutic response in etanercept-treated patients was maintained throughout the entire 10-month study period. Dosage reduction or discontinuation of medication was possible in 66% of patients receiving corticosteroids, 63% of those receiving methotrexate, 63% of those receiving sulfasalazine, and 73% of those receiving NSAIDs. Differences between treatment groups were already statistically significant by month 1. Treatment was well tolerated with only minor adverse events being noted and no significant treatment group differences. Although these results will require confirmation in a larger study, the data are consistent with observations in infliximab-treated patients identifying TNFα as a key molecular target for therapeutic intervention in AS. Choice of specific agent currently largely rests with patient preference as regards mode of administration. Recent advances in the management of SpA reported over the past 2–3 years now offer patients significant new therapeutic options with the potential for major clinical benefit. In addition, advances in diagnostic imaging now allow us to examine concepts that now constitute standard clinical practice in the management of rheumatoid arthritis. In particular, we need to explore how these therapies can best be used in early disease to prevent the development of disability, structural damage and to improve the quality of life.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.