Abstract
Biological therapies and new imaging techniques have changed the therapeutic and diagnostic approach to spondyloarthritis. In patients with axial spondyloarthritis, tumor necrosis factor α (TNFα) inhibitor treatment is currently the only effective therapy in patients for whom conventional therapy with nonsteroidal anti-inflammatory drugs (NSAIDs) has failed. TNFα inhibitor treatment is more effective in preventing articular damage in peripheral joints than in axial ones. It is important to treat patients at an early stage of disease to reduce disease progression; moreover it is necessary to identify causes of therapy inefficacy in preventing joint damage in the axial subset.
Introduction
Spondyloarthritis (SpA), a group of different diseases, consists of psoriatic arthritis (PsA), reactive arthritis, arthritis related to inflammatory bowel disease (IBD), a subgroup of juvenile idiopathic arthritis, and ankylosing spondylitis (AS), according to Moll and colleagues’ criteria of 1974 [Moll et al. 1974]. Amor and colleagues [Amor et al. 1990], the European Spondyloarthropathy Study Group [Dougados et al. 1991] and the Assessment of SpondyloArthritis International Society (ASAS) [Rudwaleit et al. 2009b, 2011] established their classification criteria over time, reaching the present classification that identifies two large SpA subtypes, axial and peripheral SpA.
Biological therapies and new imaging techniques have changed both the therapeutic and the diagnostic approach. Magnetic resonance imaging (MRI) in SpA reveals a new subset of patients with early and nonradiographic phases and verifies the efficacy and structural damage of therapies [Soscia et al. 2009; Pedersen et al. 2012; Peluso et al. 2012].
In this review, our purpose is to evaluate and update the opportunity of biological therapies in SpA.
Tumor necrosis factor α inhibitors
Five different tumor necrosis factor α (TNFα) inhibitors, etanercept, infliximab, adalimumab, golimumab and certolizumab pegol, differ in their molecular structure and administration route. Only infliximab is administered intravenously and has an induction regimen at 0, 2 and 6 weeks followed by a maintenance regimen for every 6–8 weeks. The others TNFα inhibitors are administered subcutaneously. A particular case is golimumab, also experimentally available in an intravenous formulation.
TNFα inhibitors have the same efficacy in PsA [Ritchlin et al. 2009; Atteno et al. 2010] and axial SpA [Baraliakos et al. 2012; Migliore et al. 2012] and are usually administrated after the failure of traditional disease-modifying antirheumatic drugs (DMARDs).
It is not yet known if methotrexate (MTX) gives an additive or synergistic benefit when used in combination with TNFα inhibitors in PsA; nevertheless a recent study [Fagerli et al. 2014] revealed that MTX has a role in retention rate of biological therapy but not in response rate and it is useful only in association with infliximab for its potential value to suppress antibody formation against biological therapies [Mease, 2013; Soriano, 2012; De Vries et al. 2007].
Certolizumab is the latest biological drug to add another choice for patients with SpA. In the RAPID-ax-SpA study [Landewé et al. 2014], certoluzimab was demonstrated to be effective in reducing signs and symptoms of axial involvement in patients with SpA, AS and nonradiographic axial SpA (nr-axSpA), with similar efficacy when given either as 200 mg every 2 weeks or 400 mg every 4 weeks. Although only adalimumab has been approved in Europe for nr-ax-SpA [Haibel et al. 2008; Sieper et al. 2013], etanercept [Song et al. 2013], certolizumab pegol [Landewè et al. 2014; Song and Rudwaleit, 2013] and infliximab (in a subgroup of the INFAST study) [Sieper et al. 2014a, 2014b] show efficacy and safety in patients with this SpA subtype compared with patients with AS. In addition, certolizumab pegol, in the RAPID–PsA study, showed inhibition of radiographic progression in patients with PsA compared with placebo after 24 weeks of therapy [van der Heijde et al. 2014].
In arthritis related to IBD, infliximab, adalimumab and golimumab are used to treat bowel disease and joint involvement [Denmark and Mayer, 2013]. Etanercept is effective for spinal symptoms in patients with SpA but not for IBD, and new manifestations of IBD might occur even during etanercept treatment [Sandborn et al. 2001; Song et al. 2008].
Extra-articular manifestations and TNFα inhibitors
Extra-articular manifestations are a common clinical feature of SpA. They vary widely in terms of both frequency and severity. The most common extra-articular manifestations are uveitis, bowel disease, and lung, heart, skin and kidney involvement.
Recently some authors have reported the efficacy of therapy with TNFα inhibitors in the treatment of uveitis in AS; in particular, it seems that TNFα inhibitors (infliximab and adalimumab) decrease the rate of recurrences [Levy-Clarke and Nussenblatt, 2006; van der Horst-Bruinsma and Nurmohamed, 2012]. A meta-analysis of studies about the use of TNFα inhibitors (etanercept and infliximab) revealed that they significantly reduce the incidence of uveitis flares compared with placebo in patients with AS [Braun et al. 2005]. A retrospective study of patients with SpA further confirms the efficacy of TNFα inhibitors in reducing acute uveitis flares [Guignard et al. 2006]. However, this analysis demonstrated a clear difference between etanercept and the anti-TNF antibodies (infliximab and adalimumab); the incidence of uveitis remained unchanged with etanercept treatment, whereas it was dramatically reduced after anti-TNF antibody treatment. Reports on the efficacy of adalimumab on uveitis are mainly based on retrospective analysis of placebo-controlled trials which show satisfying results [Rudwaleit et al. 2009a]. In a prospective study a significant decrease (73%) in recurrence rate of uveitis during adalimumab treatment was observed [van der Horst-Bruinsma et al. 2010]. There has been a recent observational report of the utility of golimumab in SpA-associated refractory uveitis [Miserocchi et al. 2013].
TNFα inhibitors (infliximab, etanercept, adalimumab and golimumab) are effective in the treatment of skin and nail lesions of psoriasis [Hoy and Scott, 2007; Barra et al. 2009; Brodszky et al. 2008; Gladman, 2008]. However, paradoxically between 1.5% and 5% of patients may present an onset or exacerbation of psoriasis during treatment with TNFα inhibitors [Kary et al. 2006; Wendling et al. 2008].
In contrast to the excellent efficacy of TNFα inhibitors on articular involvement in SpA, infliximab, adalimumab, golimumab and etanercept exert quite different effects on the gut. In fact, infliximab and adalimumab are effective in SpA and IBD, and golimumab is not yet registered for IBD treatment [Hanauer et al. 2002; Rutgeerts et al. 2005; Braun et al. 2007; Sandborn et al. 2007]. It has not been proven that etanercept is effective in IBD; while etanercept is effective in treating the axial involvement and peripheral arthritis associated with AS, case reports suggest that associated Crohn’s disease remains persistent or flares during etanercept therapy.
A meta-analysis of trials on the use of TNFα inhibitors in patients with AS investigated the incidence of flares or new onset of IBD; in this study there was a significant difference in favor of infliximab over etanercept and adalimumab [Braun et al. 2007]. Furthermore, in patients with a history of IBD flares, flares were more likely to occur in patients with AS receiving etanercept or adalimumab than in those treated with infliximab. This difference in clinical efficacy profiles has attracted great interest, but its scientific basis remains uncertain [Tracey et al. 2008].
TNFα inhibitors might decrease the cardiovascular manifestations and atherosclerotic cardiovascular risk in SpA [Di Minno et al. 2011, 2012; Costa et al. 2012].
The effect of TNFα inhibitors on lung involvement in SpA has never been evaluated. However, many recent case reports have described contrasting effects on pulmonary fibrosis in patients with rheumatoid arthritis (RA) [Ostor et al. 2004; Nouijai et al. 2009].
Some case reports suggest the potential role of TNFα inhibitors in improving Amyloid A (AA) amyloidosis [Kobak et al. 2007].
Initiating anti-TNFα therapy
According to the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis guidelines (GRAPPA) [Ritchlin et al. 2009], in patients with peripheral PsA, administration of at least one DMARD (e.g. sulfasalzine, leflunomide, MTX or cyclosporine) for more than 3 months is useful; more than 2 months with a standard target dose. Even if there is no evidence for using a combined therapy, a combination of two or more agents could be used in patients whose condition fails to respond to a single agent or who present joint damage progression in spite of treatment.
In patients with high disease activity persistence or joint damage progression despite therapy, TNFα inhibitors are recommended. Although no head-to-head trials have been published to directly compare efficacy and safety, all TNFα inhibitors have shown efficacy in phase III randomized controlled trials [Ritchlin et al. 2009].
Furthermore, in axial SpA, as well as in axial PsA, no DMARD therapies are recommended as first-line treatment, only nonsteroidal anti-inflammatory drugs (NSAIDs). TNFα inhibitors are recommended only after NSAID failure or in case of severe disease [Baraliakos et al. 2012] (Table 1).
When to start anti-TNF therapy.
BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C-reactive protein; DMARD, disease-modifying antirheumatic drug; ESR, erythrocyte sedimentation rate; MRI, magnetic resonance imaging; NSAID, nonsteroidal anti-inflammatory drug; SpA, spondyloarthritis; TNF, tumor necrosis factor.
Obviously, before starting TNFα inhibitor therapy, screening each patient for latent tuberculosis [Sanduzzi et al. 2012] and hepatic viral infections is suggested to promptly identify people at high risk of reactivating the disease. In this subgroup of patients, when suggested, it is useful to start adequate prophylaxis.
The highest response rate during TNFα inhibitor therapy in patients with PsA was achieved in those with elevated C-reactive protein (CRP) level at baseline [Glintborg et al. 2011], young age and low Bath Ankylosing Spondylitis Functional Index (BASFI) [Iervolino et al. 2012], while male sex was predictive of longer treatment continuation; in addition, a European League Against Rheumatism good response [Song et al. 2008], such as a high Health Assessment Questionnaire and visual analogue scale patient score was associated with a higher discontinuation risk [Gratacos et al. 2007]. Also in AS, younger age and male sex, and higher levels of inflammatory markers such as CRP were identified as independent baseline predictors of response or continuation of TNFα inhibitors. In contrast, higher baseline Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score was independently associated with treatment discontinuation [Arends et al. 2011; Mirjam et al. 2009]. Furthermore, in a recent study, TNFα inhibitor therapy seemed to reduce radiographic progression in patients with AS, especially in the case of early treatment and longer duration of follow up [Haroon et al. 2013]. In this study, the authors supported the hypothesis that the timing of treatment initiation is important in the ultimate effect on the rate of damage to the spine of patients with AS, as revealed by Maksymowych and colleagues in an MRI study [Maksymowych et al. 2013]. This study confirms that new bone formation is more likely in advanced inflammatory lesions, supporting a window of opportunity concept for disease modification.
Long-term data
Long-term results for biological treatment of SpA are available only for infliximab, etanercept and adalimumab in AS. Overall, biological therapy in AS shows only partial remission or low disease activity, based on ASAS criteria [Anderson et al. 2001].
The first TNFα inhibitor available for SpA was infliximab. Recent data show a drug survival rate of 48% after 8 years, with 88% having partial remission of disease or low disease activity [Baraliakos et al. 2011] and a low rate of radiographic progression despite therapy [Baraliakos et al. 2013b].
Etanercept shows a drug survival rate of 62% after 7 years, with 31% having partial remission and 44% complete remission of disease [Baraliakos et al. 2013a]. Five years of treatment with adalimumab in patients with AS resulted in 65% retention rate, and 51% and 61% of Ankylosing Spondylitis Disease Activity Score (ASDAS) partial remission and ASDAS remission. In this open-label extension phase of the ATLAS study, the strongest predictor of remission at 1 and 5 years and of sustained remission was achieving remission at 12 weeks of treatment [Poddubnyy and Rudwaleit, 2013].
Other mechanisms of action
Abatacept
Open-label studies have shown that abatacept has no major influence on clinical features of AS in the short term. Song and colleagues carried out an open-label study of 30 patients with AS (15 patients who were TNFα inhibitor naive and 15 whose disease failed to respond to TNFα inhibitor) who were treated with abatacept 10 mg/kg on days 1, 15, 29 and then every 4 weeks up to 6 months [Song et al. 2011]. At 24 weeks, 13% of the TNFα inhibitor naive patients while none of those whose disease failed to respond to TNFα inhibitor were classified as ASAS-40 responders. A total of 27% of the TNFα inhibitor-naive patients (
Furthermore, only in a case report was a patient with active SpA successfully treated with abatacept. This patient had disease refractory to TNFα inhibitors and all outcome measures including MRI, BASDAI, BASFI and inflammatory markers showed an improvement [Olivieri et al. 2009].
Tocilizumab and sarilumab (interleukin-6 receptor inhibitors)
One open-label study failed to demonstrate any clinical improvement in patients with axial SpA treated with tocilizumab [Lekpa et al. 2012a]. Dudler and Aubry-Rozier studied 18 patients with axial SpA refractory to anti-TNFα therapy and treated with tocilizumab 8 mg/kg/month [Dudler and Aubry-Rozier, 2011]. Of these 18 patients, 9 had classical AS, 6 had undifferentiated SpA and 3 were affected by PsA. Five out of the eight patients who fulfilled the modified New York criteria discontinued treatment due to the lack of improvement. Neutropenia occurred in four patients (22.4%), while a similar proportion developed abnormal liver function tests; 27.7% and 16.6% had high levels of low-density lipoprotein and triglycerides, respectively. Infections occurred in three patients. After the third infusion, there was an improvement in both erythrocyte sedimentation rate (ESR) and CRP. Recently, Sieper and colleagues demonstrated that there was no difference between patients with AS treated with tocilizumab or placebo at 12 weeks, although a reduction of CRP levels was observed in the tocilizumab arm [Sieper et al. 2014c]. However, the fall in inflammatory markers was not associated with any clinical improvement.
Sarilumab, a human immunoglobulin G1 that targets the interleukin-6 (IL6) receptor is currently in two phase III studies as a treatment for RA (RA-MOBILITY study). A phase III study was started in June 2010 to evaluate the long-term safety and efficacy of sarilumab in patients with AS; it was stopped due to lack of improvement in efficacy [Sieper et al. 2012].
Rituximab
The efficacy of rituximab needs to be examined in a larger randomized placebo-controlled trial with longer duration of follow up. The biggest experience is by Song and colleagues who performed an open-label study in which 20 patients with active AS were treated with two pulses of rituximab separated by 2 weeks [Song et al. 2010]. Of these 20 patients, 10 were TNFα inhibitor naive. At 6 months, 40% achieved ASAS-20 and BASDAI-20, while 25% were ASAS-40 and BASDAI-50 responders. Patients who were TNFα inhibitor naive were found to be better responders, with 50% and 60% achieving ASAS-20 and BASDAI-20 respectively. There were no new safety signals during the study. The positive response observed in the patients who were TNFα inhibitor naive was similar to that reported in patients with AS treated with infliximab [van der Heijde et al. 2005], etanercept [Davis et al. 2003] and adalimumab [van der Heijde et al. 2006]. However, a French study of eight patients with SpA receiving rituximab showed less favorable outcomes [Nocturne et al. 2010].
Another observational study reported moderate efficacy in SpA that was more marked in patients who were TNFα inhibitor naive [Wendling et al. 2012].
Secukimumab
Baeten and colleagues conducted a study on 30 patients with AS who were assigned randomly to receive secukinumab (2 × 10 mg/kg) or placebo infusion given 3 weeks apart [Baeten et al. 2013]. At 6 weeks, 61% of patients receiving secukinumab achieved ASAS-20 compared with 17% of patients treated with placebo. Thirty-five percent of patients treated with secukinumab achieved ASAS-40 and ASAS-5/6 responses respectively. This randomized controlled trial with short duration of follow up showed that secukinumab was clinically effective. However, the efficacy needs to be confirmed in a larger randomized placebo-controlled trial with longer duration of follow up.
Ustekinumab
Ustekinumab is a fully human immunoglobulin monoclonal antibody against a common subunit of IL12 and IL23, approved for patients with moderate to severe plaque psoriasis. Six large clinical trials have shown ustekinumab to be an excellent drug in the management of psoriasis [Gottlieb and Narang, 2013], reporting only few moderate adverse events (i.e. upper respiratory infection, liver enzyme increase or neutropenia). In addition, ustekinumab significantly improved active PsA compared with placebo, and might offer an alternative therapeutic mechanism of action to approved biological treatments [McInnes et al. 2013]. Recently, in an open-label clinical trial, ustekinumab was associated with a reduction in signs and symptoms in active AS. In this study, ASAS-20, ASAS-5/6 and ASAS-40 were reached in 75%, 50% and 65% of the patients respectively. Moreover, a significant improvement in other patient-reported outcome parameters and active inflammation as detected by MRI, as well as a significant reduction of NSAID intake, occurred during ustekinumab treatment. Furthermore, ustekinumab was well tolerated [Poddubnyy et al. 2014].
Apremilast
Apremilast is an orally available small molecule that specifically targets phosphodiesterase 4. Its effect on lipopolysaccharide-stimulated human peripheral blood mononuclear cells demonstrated a reduction in the number of cytokines, including interferon γ, TNFα, IL12 and IL23 [Palfreeman et al. 2013]. Apremilast has been tested in a number of psoriasis, PsA pilot study and phase II trials in order to evaluate its efficacy and safety. More recently, larger double-blind and randomized multicenter studies demonstrated that apremilast is effective in the treatment of psoriasis, PsA and AS, with significantly higher numbers of patients treated with apremilast achieving endpoints compared with baseline [Schett et al. 2012; Pathan et al. 2013].
Anakinra
Anakinra is an IL1 receptor antagonist currently approved for patients with active RA and it is useful in many rheumatic inflammatory diseases. There are no large trials evaluating its role in SpA but, among available data, some reports showed a clinical benefit only in a small number of treated patients, without MRI improvement. In these studies, carried out in patients with PsA [Jung et al. 2010] and AS [Haibel et al. 2005; Bennett et al. 2008], adverse events were mainly mild and the most frequent were injection site reactions.
Conclusion
In axial SpA, TNFα inhibitor therapy is currently the only effective therapy for patients with disease that has failed to respond to conventional therapy [Sieper, 2012], although these drugs show different side effects [Peluso et al. 2013].
TNFα inhibitor treatment is more effective at preventing articular damage in peripheral joints than in axial ones. It is important to treat patients at an early stage of disease to reduce disease progression; moreover it is necessary to identify causes of therapy inefficacy in preventing joint damage in axial subsets [Barr and Keat, 2010].
Furthermore, although encouraging results from the study of new molecules are in many subtypes of SpA, there is a need for larger and longer studies with imaging results to confirm their efficacy, safety and utility in SpA.
Footnotes
Conflict of interest statement
All the authors revised and approved the present version of this manuscript. No funding and economic support have been received to perform this study. Professor Raffaele Scarpa did not receive specific funding for the preparation of this manuscript. During the last 5 years he served on advisory boards and received honoraria and grants for research unrelated to this study. The other authors have nothing to declare.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
